NorFor – The Nordic feed evaluation system
EAAP – European Federation of Animal Science
NorFor Nordic feed evaluatio...

Author:
Harald Volden

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

NorFor – The Nordic feed evaluation system

EAAP – European Federation of Animal Science

NorFor Nordic feed evaluation system

The European Association for Animal Production wishes to express its appreciation to the Ministero per le Politiche Agricole e Forestali and the Associazione Italiana Allevatori for their valuable support of its activities

NorFor – The Nordic feed evaluation system

EAAP publication No. 130

edited by: Harald Volden

Wageningen Academic P u b l i s h e r s

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned. Nothing from this publication may be translated, reproduced, stored in a computerised system or published in any form or in any manner, including electronic, mechanical, reprographic or photographic, without prior written permission from the publisher: Wageningen Academic Publishers P.O. Box 220 6700 AE Wageningen The Netherlands www.WageningenAcademic.com [email protected]

ISBN: 978-90-8686-162-0 e-ISBN: 978-90-8686-718-9 DOI: 10.3920/978-90-8686-718-9 ISSN 0071-2477

Photographer: Jens Tønnesen from Dansk Landbrugs Medier

First published, 2011

©Wageningen Academic Publishers The Netherlands, 2011

The individual contributions in this publication and any liabilities arising from them remain the responsibility of the authors. The designations employed and the presentation of material in this publication do not imply the expression of any opinion whatsoever on the part of the European Association for Animal Production concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The publisher is not responsible for possible damages, which could be a result of content derived from this publication.

Table of contents Preface Anders H. Gustafsson

9

Acknowledgements

11

Glossary

13

1. Introduction H. Volden and A.H. Gustafsson

21

2. Overall model description H. Volden

23

3. Animal input characteristics. M. Åkerlind, N.I. Nielsen and H. Volden

27

4. Feed fraction characteristics H. Volden

33

5. Feed analyses and digestion methods M. Åkerlind, M. Weisbjerg, T. Eriksson, R. Tøgersen, P. Udén, B.L. Ólafsson, O.M. Harstad and H. Volden

41

6. Feed calculations in NorFor H. Volden

55

7. Digestion and metabolism in the gastrointestinal tract H. Volden and M. Larsen

59

8. Energy and metabolizable protein supply H. Volden and N.I. Nielsen

81

9. Animal requirements and recommendations N.I. Nielsen and H. Volden

85

10. Prediction of voluntary feed intake H. Volden, N.I Nielsen, M. Åkerlind, M. Larsen, Ø. Havrevoll and A.J. Rygh

113

11. Chewing index system for predicting physical structure of the diet P. Nørgaard, E. Nadeau Å. Randby and H. Volden

127

12. Prediction of milk yield, weight gain and utilisation of N, P and K M. Åkerlind and H. Volden

133

13. Standard feed value M. Åkerlind and H. Volden

137

NorFor – The Nordic feed evaluation system

7

14. System evaluation H. Volden, N.I. Nielsen, M. Åkerlind and A.J. Rygh

141

15. The NorFor IT system, ration formulation and optimisation H. Volden, H.D. Rokkjær, A. Göran and M. Åkerlind

167

Keyword index

179

8

NorFor – The Nordic feed evaluation system

Preface The present volume is the first comprehensive, published description of the Nordic Feed Evaluation System, NorFor. It includes the results of extensive work initiated as a project in 2001 by the farmers’ dairy cooperatives in Denmark, Norway, Iceland and Sweden. The overall aim was to create an identical, common feed evaluation system for all four countries to facilitate communication between farmers, consultants and feed industry representatives. Additional goals were to make dissemination of relevant research findings faster and more efficient, both within the four countries and internationally. Towards that end, a group of experts in ruminant nutrition was commissioned to develop such a system. Since we regarded this as a request to overhaul the current national systems comprehensively, we concluded that it provided an opportunity to develop a completely new feed evaluation system based robustly on current, published knowledge. NorFor was a development project that ran from 2002 to 2006 and included comparisons of various feed evaluation systems applied in western countries. The development also included validation of the models based on available research data from the Nordic countries. Significant parts of the development work involved close cooperation with scientists at the agricultural universities and research institutes in the Nordic countries. In addition, we have shared information and have had constructive discussions with experts in the feed industry and feed laboratories. We greatly appreciate the past and ongoing cooperation. From 2007 onwards, NorFor has been jointly funded and managed by the four organizations mentioned above through a cooperative organization (NorFor F.M.B.A.). The present description of NorFor includes all parts of the model. References to the most important sources of information are also given. We thereby provide a description of the biological basis for NorFor. The NorFor project group hopes this volume will be useful for dairy farmers, advisors, scientists and all other actors in the dairy industry in the four NorFor countries. We also recommend the reader to use the IT tools, since many of the nutritional interactions in the model are easier to use and understand if they are applied than if the texts, tables and graphs are read. NorFor Anders H. Gustafsson Leader of NorFor 2002-2009

NorFor – The Nordic feed evaluation system

9

Acknowledgements This publication summarizes the entire work of NorFor, including development and documentation. Numerous people have contributed to the compilation of this written document. NorFor has been developed over several years, thus the nature of their contributions has differed greatly. However, despite these differences, every contribution was equally appreciated. Thus, NorFor F.M.B.A. and the authors would like to sincerely thank everyone who has contributed to the project, and to this first version of the NorFor scientific publication. The main author of the present publication was Harald Volden, of TINE and the Norwegian University of Life Sciences, who has been the leader of the NorFor scientific development group. In addition, various people have played important roles in preparing the publication, including: Nicolaj I. Nielsen of AgroTech and the Danish Cattle Association; Maria Åkerlind of the Swedish Dairy Association; and Peder Nørgaard of the University of Copenhagen. Other persons, listed in the relevant chapters in this publication, have also contributed as authors. In addition, the following persons (in alphabetical order of family names) have contributed in different working groups during the development and/or documentation of NorFor: • Feed evaluation and feeding standards: Ole Aaes, Jan Bertilsson, Anders H. Gustafsson, Øystein Havrevoll, Mårten Hetta, Mogens Larsen, Maria Mehlqvist, Elisabeth Nadeau, Nicolaj I. Nielsen, Peder Nørgaard, Åshild Randby, Arnt Johan Rygh, Ingunn Schei, Harald Volden and Maria Åkerlind • Feed analysis and tables: Lars Bævre, Julia Beck, Margareta Emanuelson, Torsten Eriksson, Odd Magne Harstad, Erica Lindberg, Maria Mehlqvist, Jens Møller, Bragi Lindal Olafsson, Rudolf Thøgersen, Peter Udén, Harald Volden, Martin Riis Weisbjerg and Maria Åkerlind • IT-system: Johannes Frandsen, Henrik D. Rokkjær, Anders Göran (Tomlab), Niels Jafner, Ågot Ligaarden, Aud Elin Rivedal and Harald Volden • Communication: Marie Liljeholm, Maria Mehlqvist and Arnt Johan Rygh • National advisory tools: Åse Marit Flittie Anderssen, Henrik Martinussen, Patrik Nordgren, Berglind Ósk Óðinsdóttir and Maria Åkerlind. We would also like to thank all scientists who have contributed to fruitful discussions and made valuable comments on the model and documents. The present documentation was reviewed by Torben Hvelplund, University of Aarhus, Denmark, and Peter Udén, Swedish University of Agricultural Sciences, Sweden and they are greatly acknowledged for their work. Responsibility for the final content of the model and this present EAAP publication of the NorFor systems rests entirely with the authoring group and NorFor F.M.B.A.

NorFor – The Nordic feed evaluation system

11

Glossary Abbreviation AA AA-N AAj AAj_AATN AATN AATN_bal AATN_dep AATN_Eff AATN_gain AATN_gest AATN_maint AATN_milk AATN_mob AATN_NEG AATN_NEG_Min AATN_NEL AATN20 AATN8 ACF ADG ADG_calc Age Age_end Age_start ALF APL Avail_AAT b_car BCS BCS_change BCS_kg BUF BW BW_birth BW_calc BW_end BW_mat BW_start C12:0 C14:0 C16:0 C18:0 C18:1 C18:2 C18:3 C20:5

Unit g N/100 g N g/d % g/d % g/d g/d g/d g/d g/d g/MJ g/MJ g/MJ g/kg DM g/kg DM g/kg DM g/d g/d days days days g/kg DM g/d mg/kg DM BCS/d kg/BCS g/kg DM kg kg kg kg kg kg g/100 g FA g/100 g FA g/100 g FA g/100 g FA g/100 g FA g/100 g FA g/100 g FA g/100 g FA

Parameter name Amino acids Amino acid nitrogen in feedstuff Individual amino acid Individual amino acid absorbed in small intestine Amino acids absorbed in the small intestine AAT balance AAT for deposition in cows AAT efficiency of utilisation AAT requirement for weight gain in primiparous cows and growing cattle AAT requirement for gestation in cows and heifers AAT requirement for maintenance AAT requirement for milk production AAT for mobilisation in cows Available AAT to energy ratio in growing cattle AAT/NEG minimum recommendation Available AAT to energy ratio in cows Standard feed value for AAT at 20 kg DMI Standard feed value for AAT at 8 kg DMI Acetic acid in feedstuff Average daily body weight gain Estimated daily weight gain Current age Age at calving or sale Age at the start of a feeding period Alcohol in feedstuff Animal production level Available AAT for milk production Beta-carotene in feedstuff Body condition score, from 1 to 5 Daily change in body condition score Weight per unit body condition score Butyric acid in feedstuff Current body weight Body weight at birth Calculated body weight at a given age Body weight at calving or sale Body weight for a mature animal Body weight at the start of a feeding period Lauric acid in feedstuff Myristic acid in feedstuff Palmitic acid in feedstuff Stearic acid in feedstuff Oleic acid in feedstuff Linoleic acid in feedstuff Linolenic acid in feedstuff EPA, Eicosapentaenoic acid in feedstuff

NorFor – The Nordic feed evaluation system

13

Abbreviation

Unit

Parameter name

C22:6 Ca Ca_gain

g/100 g FA g/kg DM g/d

Ca_gest Ca_intake_Min Ca_main Ca_milk CAD CFat CFatD CHO CHOD CI Cl Cl_gain

g/d g/d g/d g/d mEq/kg DM g/kg DM %

Cl_gest Cl_intake_Min Cl_main Cl_milk Co conc_share corrNDF_fac CP CP_intake CPcorr CPD CTo Cu DIM DH DM DM1 DM2 DMcorr DMuncorr DMI DMIc DMIstd

g/d g/d g/d g/d mg/kg DM % of DM

EB EBW EBWG ECM ECM_response ECMherd e_Comp eCP ED

% kg g/d kg/d kg/d kg/d g/100g g/d

DHA, Docosahexaenoic acid in feedstuff Calcium in feedstuff Calcium requirement for weight gain in growing cattle and primiparous cows Calcium requirement for gestation Calcium minimum recommendation Calcium requirement for maintenance Calcium requirement for milk production Cation anion difference in feedstuff Crude fat in feedstuff Apparent total digestibility of crude fat Carbohydrates Apparent total digestibility of carbohydrates Chewing index of feedstuff Chloride in feedstuff Chloride requirement for weight gain in growing cattle and primiparous cows Chloride requirement for gestation Chloride minimum recommendation Chloride requirement for maintenance Chloride requirement for milk production Cobalt in feedstuff Concentrate share in ration Correction factor for kdNDF Crude protein in feedstuff Daily intake of crude protein Crude protein corrected in feedstuff Apparent total digestibility of crude protein Observed chewing time Copper in feedstuff Days in milk Danish Holstein Dry matter in feedstuff Dry matter, first step Dry matter, second step Dry matter, corrected for volatile losses Dry matter, not corrected for volatile losses Dry matter intake Dry matter intake of concentrate Dry matter intake when calculating standard feed values Energy balance Empty body weight Empty body weight gain Energy corrected milk Predicted ECM response Average daily ECM yield per cow in the herd Amino acid composition in endogenous amino acids Endogenous crude protein Efficient degradability

14

% min/kg DM g/kg DM g/d

g/kg DM g/d g/kg DM % min/kg NDF mg/kg DM days g/kg g/kg g/kg DM1 g/kg g/kg kg DM/d kg DM/d 8 or 20 kg DMI

NorFor – The Nordic feed evaluation system

Abbreviation

Unit

Parameter name

EFOS

% of OM

EI Ep erd erd_CP erd_NDF erd_RestCHO erd_ST ETo f_milk FA FA6mm from pool 1 to 2 Rumen passage rate of NDF particles>6mm from pool 2 Rumen passage rate of protein and starch particles6mm Microbial synthesis of amino acids in the rumen Microbial crude fat synthesis in the rumen Microbial protein synthesis in the rumen Microbial synthesis of organic matter in the rumen Microbial starch synthesis in the rumen Passage of organic matter to small intestine Red Danish Crude fat degraded in the rumen Crude protein degraded in the rumen Corrected crude protein degraded in the rumen Fermentation products from feed degraded in the rumen NDF degraded in the rumen NH3-crude protein degraded in the rumen Rumen degraded organic matter

18

g/kg rd OM

g/d g/d g/d g/d

NorFor – The Nordic feed evaluation system

Abbreviation

Unit

Parameter name

rd_pdCP

g/d

rd_pdNDFc

g/d

rd_pdNDFr

g/d

rd_pdST rd_RestCHO rd_sCP rd_sST rd_ST RestCHO RestCHOcorr RFA RI RLI

g/d g/d g/d g/d g/d g/kg DM g/kg DM g/kg CFat min/kg DM g/g NDF

Rough_share roughage_appetite RTo RUP S s+pdCP sCP Se si_outOM sid_AA

% of DM % min/kg NDF

Potentially degradable crude protein degraded in the rumen Potentially degradable NDF in concentrate degraded in the rumen Potentially degradable NDF in roughage degraded in the rumen Potentially degradable starch degraded in the rumen Rest CHO fraction degraded in the rumen Soluble crude protein degraded in the rumen Soluble starch degraded in the rumen Total starch degraded in the rumen Rest fraction in feedstuff Rest fraction corrected in feedstuff Residual fatty acids Rumination index Rumen load index, rumen degraded starch and rest fraction per unit of NDF Roughage percentage in the diet Roughage appetite proportion Observed ruminating time Rumen undegradable protein Sulphur in feedstuff Sum of sCP and pdCP in feedstuff Soluble crude protein in feedstuff Selenium in feedstuff Passage of organic matter to the large intestine Amino acids from feed digested in the small intestine Individual amino acids from feed digested in the small intestine Crude fat from feed digested in the small intestine Crude protein from feed digested in the small intestine Endogenous amino acids digested in the small intestine Microbial amino acids digested in the small intestine Microbial crude protein digested in the small intestine Microbial fatty acids digested in the small intestine Microbial starch digested in the small intestine Starch from feed digested in the small intestine Swedish Holstein Swedish Red Soluble starch in feedstuff Starch in feedstuff Content of starch and sugar in the diet Total intake of starch and sugar Apparent total digestibility of starch Sugar in feedstuff

g/kg DM g/kg CP g/kg CP mg/kg DM g/d g/d

sid_AAj sid_CFat sid_CP

g/d g/d

sid_eAA

g/d

sid_mAA sid_mCP

g/d g/d

sid_mFA sid_mST sid_ST SH SR sST ST ST_SU_DM ST_SU_intake STD SU

g/d g/d g/d g/kg ST g/kg DM g/kg DM g/d % g/kg DM

NorFor – The Nordic feed evaluation system

19

Abbreviation

Unit

Parameter name

SubR TAF TCL td_CFat td_CHO td_CP td_CPcorr td_OM TMR uNDF uNDS uOM Urea N VFA VitA VitD VitE VOS

none g/kg DM mm g/d g/d g/d g/d g/d g/kg DM g/kg DM g/kg DM g/g

YHerd Zn WPC

kg ECM/year mg/kg DM weeks

Substitution rate Total acids in feedstuff Theoretical cutting length Total apparently digested crude fat Total apparently digested carbohydrates Total apparently digested crude protein Total apparently digested crude protein corrected Total apparently digested organic matter Total mixed ration Undegraded NDF Undegraded neutral detergent solids Undegraded organic matter Urea nitrogen Volatile fatty acids Vitamin A in feedstuff Vitamin D in feedstuff Vitamin E in feedstuff Organic matter digestibility of feedstuff in vitro (VOS method) Yield level in the herd Zinc in feedstuff Weeks post calving

20

1000 IU/kg DM 1000 IU/kg DM IU/kg DM % of OM

NorFor – The Nordic feed evaluation system

1. Introduction H. Volden and A.H. Gustafsson Feed is one of the major expenses in modern cattle production. In addition to feed prices, its overall costs are affected by the efficiency of feed utilization and the output of animal products to be marketed. Hence, there is a clear need to evaluate feed quality in order to maximise profitability. This requires information on both animal requirements and nutrient supply, since formulation of an appropriate ration involves balancing available feeds in proportions that match the amounts of nutrients supplied to the animals’ nutrient requirements as closely as possible. There are two principal methods used to describe animal nutrition: those based on mechanistic approaches, which describe responses to nutrients from chemical and physiological processes in the gastrointestinal tract and intermediary, and empirical approaches describing simple relationships between nutrient intakes and production responses. The challenge in the development of new feed evaluation systems is to accurately predict responses to nutrients so that any difference in product income and feed costs can be maximized, while improving overall feed efficiency. Feed efficiency is also of great importance due to its impact on enteric methane emission to the atmosphere and nitrogen and phosphorus passing into the environment. Feed evaluation for cattle has long traditions. Important milestones are introductions of the Weende analysis by Henneberg and Stohmann (1864), and the starch equivalent system by Kellner (1912). In the Nordic countries the introduction of the milk production value by Hansson (1913) and the Scandinavian feed unit by Møllgaard (1929) were of great magnitude for modern cattle production. Extensive research in the 1950’s and 60’s further improved our knowledge in feed evaluation, and in the period from 1970 to 1990 most countries introduced new systems for energy based either on metabolizable energy (ME) or net energy for lactation (NEL) or growth (NEG) (Van der Honing and Alderman, 1988). Also the knowledge of protein evaluation for cattle have increased considerably during the last 40 years, from use of total or digestible crude protein to the protein evaluation systems that predict the host animal amino acid (AA) supply from dietary protein escaped ruminal degradation and from microbial protein synthesized in the rumen (Madsen, 1985; NRC, 1985; Vérité et al., 1987; AFRC, 1992; Tamminga et al., 1994). Traditional feed evaluation systems are additive and generally do not take into account interactions in digestion and nutrient metabolism. When interactions and non-linear relationships are considered, in attempts to describe feed metabolism, individual feeds will no longer have fixed values. Hence, the value of a given feedstuff will depend on how the feed is used. This means that the feed value cannot be determined until we know which feed ration will be used and the production situation in which it will be applied. Therefore, the development of a new feed evaluation system must consist of a ration evaluation system which can be used to optimize nutrient supply and production responses, rather than a system that can predict individual feed values. Development of new feed evaluation systems (Russell et al., 1992; Sniffen et al., 1992; Fox et al., 1992; NRC, 2001) and mechanistic models (Baldwin, 1995; Danfær et al., 2006), which describe nutrient supply and the nutritional requirements of cattle, are important for understanding ruminant nutrition and constitute an important step towards implementing more sophisticated nutritional strategies to optimize production responses in cattle. The use of non-linear semi-mechanistic feed evaluation systems for ration optimization requires powerful computing tools. Recent improvements in non-linear optimization tools and algorithms enable the development of more complex feed evaluation systems that can also be used in practice. The aim of this book is to provide a detailed description of the semi-mechanistic feed evaluation system NorFor, which has already been implemented by advisory services in four Nordic countries, i.e. Denmark, Iceland, Norway and Sweden.

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_1, © Wageningen Academic Publishers 2011

21

References AFRC (Agricultural and Food Research Council), 1992. Nutritive requirements of ruminant animals: protein. AFRC Technical Committee on Response to Nutrients. Report No. 9. Nutrition Abstracts and Reviews: Series B 62: 787-835. Baldwin, R.L., 1995. Modelling Ruminant Digestion and Metabolism. Chapman and Hall, London, UK. 592 pp. Danfær, A., P. Huhtanen, P. Udén, J. Sveinbjörnsson and H. Volden, 2006. The Nordic dairy cow model, Karoline – Description. In: Kebreab, E., J. Dikstra, A. Bannink, J. Gerrits and J. France (eds.). Nutrient Digestion and Utilisation in Farm Animals: Modelling Approaches. CAB International, Wallingford, UK, pp. 383-406. Fox, D.G., C.J. Sniffen, J.D. O’Connor, J.B. Russell and P.J. Van Soest, 1992. A net carbohydrate and protein system for evaluating cattle diets. III. Cattle requirements and diet adequacy. Journal of Animal Science 70: 3578-3596. Hansson, N., 1913. En ny metod för beräkning av fodermedlens produktionsvärde vid utfodring af mjölkkor. Meddelande Nr. 85 från centralanstalten för försöksväsendet på jordbruksområdet. Ivar Hæggströms Boktryckeri Aktiebolag, Stockholm, Sweden. (In Swedish). Henneberg, W. and F. Stohmann, 1864. Begründung einer rationellen fütterung der wiederkauer. Schwetske and Söhne, Braunschweig, Germany. (In German). Kellner, O., 1912. Die Ernährung der Landwirtschaftlichen nutztiere. Raul Parey, Berlin, Germany. (In German). Madsen, J., 1985. The basis for the proposed Nordic protein evaluation system for ruminants. The AAT-PBV system. Acta Agriculturae Scandinavica Supplement 25: 9-20. Møllgaard, H., 1929. Fütterungslehre des milchviehs. Verlag von M. and H. Schalper, Hannover, Germany. (In German). NRC (National Research Council), 1985. Ruminal nitrogen usage. National Academy Press, Washington, DC, USA. 138 pp. NRC, 2001. Nutrient requirements of dairy cattle. Seventh revised edition, National Academy Press, Washington, DC, USA. 408 pp. Russell, J.B., J.D. O’Connor, D.G. Fox, P.J. Van Soest and C.J. Sniffen, 1992. A net carbohydrate and protein system for evaluating cattle diets. I. Ruminal fermentation. Journal of Animal Science 70: 3551-3561. Sniffen, C.J., J.D. O’Connor, P.J. Van Soest, D.G. Fox and J.B. Russell, 1992. A net carbohydrate and protein system for evaluating cattle diets. II. Carbohydrate and protein availability. Journal of Animal Science 70: 3562-3577. Tamminga, S., W.M. Van Straalen, A.P.J. Subnel, R.G.M. Meijer, A. Steg, C.J.G. Wever and M.C. Block, 1994. The Dutch protein evaluation system: the DVE/OEB-system. Livestock Production Science 40: 139-155. Van der Honing, Y. and G. Alderman, 1988. Feed evaluation and nutritional requirements. III. Ruminants. Livestock Production Science 19: 217-278. Vérité, R., P. Chapoutot, B. Michalet-Doreau, J.L. Peyraud and C. Poncet, 1987. Révision du système des protéines digestibles dans l’intestin (PDI). Bulletin Technique C.R.Z.V. Thei1, INRA 70: 19-34.

22

NorFor – The Nordic feed evaluation system

2. Overall model description H. Volden The NorFor system is a semi-mechanistic, static and science-based model, which predicts nutrient supply and requirements for maintenance, milk production, growth and pregnancy in cattle. The model can be divided into five parts: (1) an input section describing characteristics of the animal and feeds available; (2) a module simulating processes in the digestive tract and the intermediary metabolism, termed the feed ration calculator (FRC); (3) a module predicting feed intake; (4) a module predicting the physical structure of the diet; and (5) an output section describing nutrient supply, nutrient balances and production responses (Figure 2.1). The input variables for the model are animal and feed characteristics. For dairy cows, the main input variables are body weight (BW), stage of lactation, pregnancy day and planned or potential daily milk production. For growing animals (bulls, steers and heifers) input variables are BW and average daily weight gain (ADG). The feed dry matter (DM) is separated into ash, crude protein (CP), crude fat (CFat), neutral detergent fibre (NDF), starch (ST), sugar (SU), fermentations products (FPF) such as organic acids and alcohols, and a residual fraction (RestCHO). The CP is divided into soluble (sCP), potentially degradable (pdCP), indigestible (iCP) and ammonia (NH3N). The NDF is divided into a total indigestible (iNDF) and a potentially degradable (pdNDF) fraction. The ST is divided into soluble (sST), potentially degradable (pdST) and indigestible (iST) fractions. The FPF are separated into lactic acid (LAF), volatile fatty acids (VFA) and alcohols. Fractional degradation rates (kd) of the soluble and potentially degradable feed fractions are also required for the model. The FRC consists of four sections: (1) the rumen, (2) the small intestine, (3) the large intestine and (4) metabolism (Figure 2.2). Feed organic matter (OM) entering the rumen is either fermented and used for microbial production, or it escapes from the rumen for further digestion in the lower digestive tract. Ruminal degradation of CP, ST and NDF in concentrate feeds are assumed to follow first-order single-compartment kinetics, while degradation of NDF in roughage is modelled as a Input Feed characteristics (chemical composition and particle length) Animal characteristics (body weight, breed, stage of lactation)

Physical structure (chewing time)

Gastro-intestinal tract and intermediary metabolism Nutrient calculator Nutrient digestion and metabolism

Feed intake

Output Nutrient supply Ration energy and protein value. Predicted milk yield, protein production, Nutrient balances in the rumen, etc.

Figure 2.1. Overview of the NorFor model. H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_2, © Wageningen Academic Publishers 2011

23

FEED

24

RUMEN

SMALL INTESTINE

HIND GUT

iNDF

pdNDF

pdNDF

pdNDF

F ib r e

iNDF

iST

iST

iST

pdST

pdST

pdST

St a r ch

sST

sST

Rest CHO

Rest CHO

R est FPF

FA

FA

iCP

Glycerol , galactose

CFat

Lipids iCP

sCP

sCP

mCFat

mCP

Microbial growth

Energy and nitrogen for microbial growth

pdCP

pdCP

P r ot ein

Energy and nitrogen for microbial growth

mST

mST

mCFat

mCP

mCP

Microbial growth

NH 3

eCP

eCP

PBV N

Net energy

Energy Energy

Energy

of CP_intake

UREA 4.6%

Milk or gain

AA in milk or gain

AAT N to milk or gain Utilization

AAT N

dep. / mob.

AAT N for hair and skin

Endogenous N

Endogenous urinary-N

INTERMEDIARY METABOLISM

Figure 2.2. Flow diagram of the gastrointestinal tract and intermediary compartments of the NorFor feed evaluation system. For all abbreviations check the abbreviation list.

FAECES

NorFor – The Nordic feed evaluation system

two-compartment system, with a non-escapable and an escapable pool. The nutrients available for microbial growth come from ruminally degraded NDF, ST, RestCHO, glycerol, CP and LAF. The efficiency of microbial synthesis depends on the level of feed intake and diet composition. The input to the small intestine consists of OM from microbes, unfermented feed fractions escaping from the rumen and endogenous secretions. These components are partly digested in the small intestine and are either metabolised or enter the large intestine. The OM passing into the large intestine is subjected to microbial fermentation, and the digested OM not used for microbial synthesis is absorbed and metabolised. Faecal excretion consists of OM from microbes synthesised in the large intestine, feed that has escaped previous digestion and undigested rumen microbial material. The intermediary metabolism section yields ME calculated from total tract digestible OM. Net energies for maintenance, lactation, growth and pregnancy are predicted from the ME. Different coefficients are used to calculate NE for maintenance, lactation, and growth. Net energy for lactation is used for dairy cows, while NEG is used as the energy measurement for growing cattle. The nitrogen (N) fractions entering the intermediary metabolism consist of NH3N absorbed from the rumen, dietary, microbial and endogenous amino acids (AA) absorbed from the small intestine, and NH3N from the large intestine. The absorbed AAs are utilized for maintenance, growth, pregnancy and milk production. The efficiency of AA utilization is specific for each production/process. The metabolizable protein available for animal production is assigned as amino acids absorbed from the small intestine (AATN). The N which is not used for maintenance or production is excreted in the urine. Predicting nutritive value is only one part of ration formulation as formulation involves both the selection of feed ingredients and the prediction of feed intake. Therefore, the NorFor system contains a module to predict the intake of feeds. For prediction of feed intake dietary fill values (FV) and animal intake capacity (IC) are applied. In roughages, FV is calculated from OM digestibility (OMD) and NDF content, and in ensiled forages the basic FV is also corrected for content of VFA, LAF and NH3N. Animal IC is dependent on BW, milk yield, stage of lactation, lactation number, ADG and physical activity. The model uses a combination of dietary physical effects and metabolic factors to impact feed intake, and the effect of easily fermentable carbohydrates on roughage intake is accounted for by using a substitution rate factor (SubR). A minimum amount of large particles is essential for optimal rumen function. Hence, a module to evaluate the physical structure of the diet is included in NorFor. The dietary physical effect is described by a total chewing index (CI), which is calculated as the sum of an eating (EI) and ruminating (RI) index for each individual feed. The EI value reflects the associated chewing activity as feed is consumed and is calculated from the particle length and NDF content of the feed. The RI value is calculated from particle length, NDF content and a hardness factor, which is dependent on the iNDF content of the feed. The hardness factor reflects the lignification of the structural fibre of the feed and the associated physical force required for the comminution of large particles. The output from a model calculation in NorFor describes the intake of the individual feeds in the total ration. It consists of variables describing efficiencies of digestion and nutrient utilization, production (milk, ADG), N excretion, nutrient balances, energy (NEL or NEG) and protein (AATN) values of the ration.

NorFor – The Nordic feed evaluation system

25

3. Animal input characteristics. M. Åkerlind, N.I. Nielsen and H. Volden Animal characteristics are needed for calculation of nutrient requirements and IC for both cows and growing cattle. Information on animal parameters, such as BW, is also needed when calculating ruminal variables, such as passage rates of feed fractions out of the rumen and efficiency of microbial protein synthesis. Moreover, the physical activity of the animal (simply classified according to whether it is loose or tied up) will affect the IC and energy requirements.

3.1 Input data for cows All required input data for cows are listed in Table 3.1 and default values for different breeds are compiled in Table 3.2. The information needed for cows in the system is breed, lactation number, BW, stage of lactation (days in milk) and milk production. Pregnancy and whether the cow is dry or lactating are also factors required for determining IC and requirements for gestation. When calculating energy and protein requirements for gestation, information on mature body weight (BW_mat) is needed since this variable is breed-specific (see Table 3.2 and Sections 9.1.3 and 9.2.5). BW_mat is also needed to determine protein requirements for growth in primiparous cows (Section 9.2.3). Animal weight changes can be estimated as changes in body condition score (BCS), where the BW per condition scores depends on the breed. The BCS in NorFor uses a scale from 1 to 5, where 1 is thin and 5 is fat (Gillund et al., 1999). One unit of BCS corresponds to 60 kg BW except for Jersey which is set to 45 kg BW; these values are based on data from several previous studies (e.g. Enevoldsen and Kristensen,1997; Gillund et al., 1999; Nielsen et al., 2003; Bossen, 2008). Table 3.1 Input data for cows in NorFor. Input data

Unit

Dry cow Lactation number Days in milk Daily milk yield Protein content in milk Fat content in milk Lactose content in milk Yield level of the herd Body weight, current Body weight, mature2 Weight gain in primiparous cows Days of gestation Daily change in body condition score Weight per unit body condition score2 Activity Breed3

none1 No. days kg/day g/kg g/kg g/kg kg ECM kg kg kg/day days BCS/day kg/BCS none1 none1

1

No unit. Default values can be taken from Table 3.2. 3 Abbreviations for breeds are explained in Table 3.2. 2

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_3, © Wageningen Academic Publishers 2011

27

429 428 430 429 431 430 432 431 433 432 434 433 435 434 435 436 436 437

compiled in Table 3.2. The information needed for cows in default the system is breed, lactation number, All required input data for cows are listed in Table 3.1 and values for different breeds are BW, stageinofTable lactation in milk) andneeded milk production. whether the cow is dry compiled 3.2. (days The information for cows inPregnancy the systemand is breed, lactation number, or lactating arelactation also factors IC andPregnancy requirements gestation. When BW, stage of (daysrequired in milk)for anddetermining milk production. andfor whether the cow is dry calculating protein requirements for gestation, information on mature body weight or lactatingenergy are alsoand factors required for determining IC and requirements for gestation. When (BW_mat) needed since this requirements variable is breed-specific Table 3.2on andmature Sections 9.1.3 and calculating isenergy and protein for gestation,(see information body weight 9.2.5). BW_mat is also needed to determine protein requirements for3.2 growth in primiparous cows (BW_mat) is needed since this variable is breed-specific (see Table and Sections 9.1.3 and Table 3.2. Default values for mature body weight (BW_mat) and the weights corresponding to a (Section 9.2.3). is also needed to determine protein requirements for growth in primiparous cows 9.2.5). BW_mat body condition score unit (BCS_kg) for dairy cows of different breeds. (Section 9.2.3). Table Abbr.3.1 3.1 Table 3.2

Dairy breed

BW_mat kg

BCS_kg kg/BCS

437 438 439 438 439 440 441 440 442 441 443 442 444 443 445 444 445 446

DH 3.2 Danish Holstein 640 60 Table IB Icelandic breed 470 60 where the Animal weight changes can be estimated as changes in body condition score (BCS), JER Jersey 440 45 BW per weight condition scorescan depends on the breed. The BCS in NorFor usesscore a scale fromwhere 1 to 5,the Animal changes be estimated as changes in body condition (BCS), NR Norwegian Red 600 where 1 is thin and 5 is fat (Gillund et al., 1999). One unit of BCS corresponds to 60 kg BW BW per condition scores depends on the breed. The BCS in NorFor uses a scale60 from 1 to 5, RD Danish Red 660 60 except for Jersey which is set to 45 kg BW; these values are based on data from several previous where 1 is thin and 5 is fat (Gillund et al., 1999). One unit of BCS corresponds to 60 kg BW studies (e.g.Jersey Enevoldsen and Kristensen,1997; Gillund et are al., based 1999; on Nielsen et al., 2003; previous Bossen, SH for Swedish 640 60 except which is setHolstein to 45 kg BW; these values data from several 2008). SR Swedish Red 620 60 studies (e.g. Enevoldsen and Kristensen,1997; Gillund et al., 1999; Nielsen et al., 2003; Bossen, 2008).

446 447 448 447 449 448 449 450

The energy requirement for milk production is based on the production of energy corrected milk The energy requirement for milk production is equations based on (Sjaunja the production1990) of energy corrected (ECM), which can be calculated either of depending on milk The energy requirement for milkfrom production is two based on the productionetofal., energy corrected milk (ECM), which can be calculated from either of two equations (Sjaunja et al., 1990) depending on whether information milk lactose is available: (ECM), which can beon calculated fromcontent either of two equations (Sjaunja et al., 1990) depending on whether information on milk lactose content is available: whether information on fmilk lactose content _ milk p _ milkis available: l _ milk ⎞ ⎛ [Eq. 3.1]

450 451 451 452 453 452 454 453 454

ECM = MY ⋅ ⎜ 0.01 + 0 .122 ⋅ + 0 .077 ⋅ + 0.053 ⋅ ⎟ ⎝⎛ f _10 milk p _10 milk l _10 milk ⎠⎞ ECM = MY ⋅ ⎜ 0.01 + 0 .122 ⋅ + 0 .077 ⋅ + 0.053 ⋅ ⎟ 10 10 10 ⎠ ⎝

f _ milk p _ milk ⎞ ⎛ ECM = MY ⋅ ⎜ 0 .25 + 0 .122 ⋅ + 0 .077 ⋅ ⎟ ⎝⎛ f _10 milk p _10 milk ⎠⎞ ECM = MY ⋅ ⎜ 0 .25 + 0 .122 ⋅ + 0 .077 ⋅ ⎟ 10 ⎠ where ECM⎝ is the energy corrected milk,10 kg/day;

[Eq. 3.1]

3.1

[Eq. 3.2] [Eq. 3.2]

3.2

MY is the daily milk yield, kg/day; and f_milk,

where the energy corrected kg/day; MYand is the daily milk yield, in kg/day; f_milk, g/kg. p_milkECM and isl_milk are the contentsmilk, of fat, protein anhydrous lactose milk, and respectively, p_milk and l_milk are the corrected contents ofmilk, fat, protein in milk, respectively, where ECM is the energy kg/day; and MYanhydrous is the dailylactose milk yield, kg/day; and f_milk, g/kg. p_milk and l_milk are therations contents fat, protein and the anhydrous lactose in milk, respectively, When formulating feed forofgroups of cows, daily ECM yield (ECMherd) can be estimated g/kg. from breed-specific lactation curves: 23 ECMherd = a + b ∙ YHerd – c ∙ DIM + ln(DIM) ∙ d

23

3.3

where ECMherd is the daily estimated ECM yield, kg/d; YHerd is the herd’s average ECM yield per cow, kg/305 d; DIM is days in milk; a, b, c and d are regression coefficients presented in Table 3.3 for primiparous and multiparous cows of different breeds. Standard lactation curves are available for different dairy breeds and lactation numbers. The ECMHerd value refers to the herd’s 305-d lactation yield, which is based on national herd recording data. Danish cow recording data (Danish Cattle Association) have been used to parameterize the standard lactation curves for the dairy breeds Jersey, Danish Holstein and Danish Red. Icelandic, Norwegian and Swedish national herd recording data are the basis for the standard lactation curves for the Icelandic breed, Norwegian Red, Swedish Holstein and Swedish Red cattle, respectively (Farmers Association of Iceland; Tine Dairies in Norway; Swedish Dairy Association). Daily milk protein yield is calculated from milk yield and milk protein content according to Equation 3.4. Protein production is essential for calculating the AA requirement for milk production: MPY = MY ∙ p_milk

3.4

where MPY is the milk protein yield, g/day; MY is the milk yield, kg/day; and p_milk is the milk protein content, g/kg.

28

NorFor – The Nordic feed evaluation system

Table 3.3. The multiple regression coefficients a, b, c and d is used to predict daily ECM yield from standardised lactations curve in Equation 3.3. Breed1

Lactation

a

b

c

d

DH DH IB IB JER JER NR NR RD RD SH SH SR SR

primiparous multiparous primiparous multiparous primiparous multiparous primiparous multiparous primiparous multiparous primiparous multiparous primiparous multiparous

-3.10 3.17 0.88 6.34 -8.15 -2.98 -7.09 -0.59 -6.20 1.45 -4.05 2.93 -4.46 -0.21

0.00325 0.00338 0.00288 0.00269 0.00324 0.00335 0.00314 0.00310 0.00335 0.00330 0.00299 0.00299 0.00304 0.00317

0.01685 0.06040 0.04009 0.06234 0.02810 0.05630 0.06440 0.09100 0.02138 0.06973 0.03560 0.06100 0.03970 0.06920

1.140 1.025 1.627 1.569 2.654 2.305 3.741 3.378 1.750 1.844 2.591 1.997 2.677 2.520

1 The

abbreviations of the different breeds are shown in Table 3.2.

3.2 Input data for growing cattle When calculating nutrient requirements and the IC for growing cattle, information on BW, ADG, sex, breed and activity are needed, and for heifers the days of gestation are also required (Table 3.4). Table 3.4. Input data for growing cattle. Parameter

Unit

Sex Breed2 Activity Body weight3 Average daily gain3 Days of gestation (heifers) Body weight, birth4 Body weight, start4 Body weight, end4 Body weight, mature4 Age, start4 Age, current4 Age, end4

none1 none1 none1 kg g/day days kg kg kg kg days days days

1

No unit. classifications of different breeds are shown in Table 3.5. 3 Body weight and daily gain are required parameters and can be estimated from the parameters that are marked by 4. 4 These parameters are required if current body weight and average daily gain data are not available. 2 The

NorFor – The Nordic feed evaluation system

29

485 486

sex, breed and activity are needed, and for heifers the days of gestation are also required (Table 3.4).

487 488 489

Table 3.4

490

3.2.1 Estimation of body weight and daily weight gain from a growth function

491 492 493 494 497 495 498 496 499 497 500 498 501 499 502 500 503 501 504 502 505 503 506 504 507 505 508 506 509 510 507 508 509 510 511

Whenplanning planningfeed feedrations rations animals of different ages, of BW_mat, ageBW andatBW When forfor animals of different ages, datadata of BW_mat, age and the at the start of the period andand planned ageage and BW at atthetheend start ofrearing the rearing period planned and BW endofofthe therearing rearingperiod period are are needed. needed. Expected Expected BW and can be estimated from the following logistic growth equation based BW and ADG canADG be estimated from the following logistic growth equation based on: on:

512 511 512 513 514 513 515 516 514 517 515 518 516 519 517 520 518 521 519 522 520 523 521 524 522 525 523 526 524 527 525 528 526 529 527 530 528 531 529 532 530 533 531 534 532 533 535 534 535

3.2.1 Estimation of body weight and daily weight gain from a growth function

(Age−for )))) age, kg; BW_start is the BW[Eq. ( A⋅(1−e( −B⋅BW Agethe _ start where BW_calc the⋅ eestimated current at the start, kg; 3.5] BW _ calc = BW _is start 3.5 A is described in Equation 3.6; B is described in Equation 3.8; Age is the current age, days; Age_start is the age at start, days. IfBW Age_start iscurrent 0 the BW_start the same is asthe BW_birth. whereBW_calc BW_calc theestimated estimated kg; isBW_start the kg; start, kg; A where isisthe BW forfor thethe current age,age, kg; BW_start is the BW atBW the at start,

is is described 3.7; in1)))) Equation3.8; 3.8;Age Age the current age, days; Age_start A describedininEquation Equation 3.6; B B is described described in Equation isisthe current age, days; ⋅(1−e(−B⋅(Age−Age _ start+24 A⋅(1−e(−B⋅(Age−Age _ start)))) ⋅1000 ADG calcis =atthe BW _ start e (AIf − BW _ is start ⋅ e (is Age_start age at ⋅start, days. If Age_start 0 the BW_start the same as BW_birth. is the_ age start, days. Age_start is 0 theisBW_start the same as BW_birth. [Eq. 3.6]

) ( ADG _ calc = (BW _ start ⋅ e (A⋅(1−e(−B⋅(Age−Age _ start+1)))) − BW _ start ⋅ e (A⋅(1−e(−B⋅(Age−Age _ start)))) )⋅1000 where ADG_calc is the estimated average daily gain for the current age, g/day; BW_start is the

3.6

[Eq. 3.6] BW at the start, kg; A is described in equation 3.7; B is described in equation 3.8; Age is the where ADG_calc is the estimated average daily gain for the current age, g/day; BW_start is the BW current age, days; Age_start is the age at start, days. If Age_start is 0 the BW_start is the same as where ADG_calc is is thedescribed estimatedin average daily3.7; gainBfor the current in age, g/day; BW_start at the start, kg; A Equation is described Equation 3.8; Ageisisthe the current BW_birth. BW at the start, kg; A is described in equation 3.7; B is described in equation 3.8; Age is the

age, days; Age_start is the age at start, days. If Age_start is 0 the BW_start is the same as BW_birth.

current age, days; Age_start is the age at start, days. If Age_start is 0 the BW_start is the same as Factor A and B in Equations 3.5 and 3.6 are calculated as: BW_birth. Factor A and B in Equations 3.5 and 3.6 are calculated as: _ mat .1 ⎞ Factor⎛ABW and B in⋅ 1Equations 3.5 and 3.6 are calculated as:

A = ln⎜⎜ [Eq. 3.7] 3.7 ⎟⎟ ⎝ BW _ start ⎠ ⎛ BW _ mat ⋅ 1.1 ⎞ ⎟ A = ln⎜⎜ [Eq. 3.7] BW__start mat ⋅ 1⎟⎠.1 / BW _ start ) ⎞ ⎛⎝ ln(BW ⎟⎟ ln ⎜⎜ ln( BW _ mat ⋅ 1.1 / BW _ end ) ⎠ [Eq. 3.8] 3.8 B= ⎝ _ start ⎛ ln(Age BW __ end mat −⋅ 1Age .1 / BW _ start ) ⎞ ⎟⎟ ln ⎜⎜ ln(and BW _Bmat .1 / BW _used end ) ⎠in Equations 3.5 and 3.6; BW_mat is the mature body weight, kg; where A are⋅ 1factors [Eq. 3.8] B = ⎝A where and B are −factors used in Equations 3.5 and 3.6; BW_mat is the mature body weight, kg; end Ageweight _ start BW_startAge is _the body at start or at birth, kg; BW_end is the body weight at the end of the

BW_start is the body weight at start or at birth, kg; BW_end is the body weight at the end of the feedingperiod; period;Age_start Age_start at start, days; Age_end theatage theofend feeding is is thethe ageage at start, days; and and Age_end is theisage theat end the of the feeding where and BIfare factors used Equations 3.5 and 3.6; BW_mat the mature body weight, kg; period,Aperiod, days. Age_start is 0in days, the BW_start is the same asis BW_birth. feeding days. BW_start is the body weight at start or at birth, kg; BW_end is the body weight at the end of the

feeding period; Age_start is the age at start, days; and Age_end is the age at the end of the Anexample exampleofofthe thelogistic logistic growth function of ADG during the rearing is shown in An growth function of ADG and and BW BW during the rearing periodperiod is shown feeding period, days. in Figure3.1. 3.1. Figure An example of the logistic growth function of ADG and BW during the rearing period is shown Figure 3.1of the feeding period depends on whether the animal will be sold, slaughtered or used for The end in Figure 3.1.

replacement. Default values for birth weights (BW_birth) and BW_mat that can be used for estimating

The of the feeding period depends on whether the animal will be sold, slaughtered or used for BWend and Figure 3.1ADG for different breeds and gender are shown in Table 3.5. BW_birth values for beef replacement. Default values for birth weights (BW_birth) and BW_mat that can be used for breeds were collected from Danish, Norwegian and Swedish national recording data compiled by estimating BW and ADG for different breeds and gender are shown in table 3.5. BW_birth values The end of the feeding period depends on whether animal will Research be sold, slaughtered used for thebeef Danish Cattle Norwegian Meattheand Poultry andordata Cattle statistics for breeds wereAssociation, collected from Danish, Norwegian and Swedish nationalCentre recording replacement. Default values for birth weights (BW_birth) andDanish, BW_mat that can beand used for (2007), respectively. Values for BW_mat were taken from Norwegian Swedish slaughter compiled by the Danish Cattle Association, Norwegian Meat and Poultry Research Centre and estimating BW and ADG for different breeds and gender are supplied shown inby table 3.5. BW_birth data for animals over 4respectively. years of ageValues over the 10 years the Danish Cattlevalues Association; Cattle statistics (2007), forlast BW_mat were taken from Danish, Norwegian for beef breeds were collected from Danish, Norwegian and Swedish national recording data Norwegian and data Poultry Research Centre and Dairy respectively. and Swedish Meat slaughter for animals over 4 years of Swedish age over the lastAssociation, 10 years supplied by the compiled by the Danish Cattle Association, Norwegian Meat and Poultry Research Centre and Danish Cattle Association; Norwegian Meat and Poultry Research Centre and Swedish Dairy Cattle statistics (2007), respectively. Values for BW_mat were taken from Danish, Norwegian Association, respectively. and Swedish slaughter data for animals over 4 years of age over the last 10 years supplied by the Danish Cattle Association; Norwegian Meat and Poultry Research Centre and Swedish Dairy Association, Table 3.5 respectively. Table 3.5 25 25

30

NorFor – The Nordic feed evaluation system

ADG_calc, g/day BW_calc, kg

1000 900 800 700 600 500 400 300 200 100 0

ADG, 24mo ADG, 27mo BW, 24mo BW, 27mo

0

5

10 15 Age, months

20

25

30

Figure 3.1. A logistic growth function predicting BW (BW_calc) and estimated weight gain (ADG_ calc) during the rearing of a heifer which is scheduled to calve at a live weight of 560 kg at either 24 or 27 months of age. To achieve a final weight of 560 kg, the weight gain is faster for a heifer finished at 24 months of age (ADG, 24mo; BW, 24mo) than for a heifer finished at 27 months (ADG, 27mo; BW, 27mo). Table 3.5. Default values for birth weight (BW_birth) and mature body weight (BW_mat) for heifers, bulls and steers of different breeds. Breeds Early maturing dairy breeds Danish Holstein Danish Red Icelandic breed Jersey Norwegian Red Swedish Holstein Swedish Red Early maturing beef breeds Aberdeen Angus Dexter Galloway Hereford Highland cattle Tiroler Grauvieh Late maturing beef breeds Belgian Blue Blonde d’Aquitaine Charolais Chianina Limosin Piemontese Saler Simmenthal Early·late maturing breed Crossbred1 1

BW_birth BW_birth BW_mat BW_mat BW_mat for heifer kg for bulls kg for heifer kg for bulls kg for steers kg 40 40 33 28 39 39 39

41 41 33 30 41 41 41

640 660 470 440 600 640 620

950 950 800 650 950 950 950

750 750 700 550 750 750 750

36 21 34 40 29 39

38 24 35 42 30 42

700 340 550 700 500 700

950 450 850 950 700 950

750 400 750 750 600 750

44 44 46 50 41 41 39 44

47 47 49 55 43 43 41 46

850 850 750 850 750 750 750 750

1,200 1,200 1,200 1,200 1,200 1,200 1,200 1,200

1,050 1,050 1,050 1,050 1,050 1,050 1,050 1,050

42

44

750

1,050

950

Crossbreeds of early and a late maturing breeds are assigned specific factors in some of the equations in Chapter 9.

NorFor – The Nordic feed evaluation system

31

References Bossen, D., 2008. Feeding strategies for dairy cows. Individual feed allocation using automatic live weight registrations as management parameter. PhD thesis. University of Copenhagen, Faculty of Life Sciences, Denmark. 159 pp. Enevoldsen, C. and T. Kristensen, 1997. Estimation of body weight from body size measurements and body condition scores in dairy cows. Journal of Dairy Science 80: 1988-1995. Gillund, P., O. Reksen, K. Karlberg, Å.T. Randby, I. Engeland and B. Lutnaes, 1999. Testing of a body condition score method in Norwegian cattle. Norsk Veterinærtidsskrift 111: 623-632. (In Norwegian with English summary). Nielsen, H.M., N.C. Friggens, P. Løvendahl, J. Jensen and K.L. Ingvartsen, 2003. Influence of breed, parity, and stage of lactation on lactational performance and relationship between body fatness and live weight. Livestock Production Science 79: 119-133. Sjaunja, L.-O., L. Bævre, I. Junkkarainen, J. Pedersen and J. Setälä, 1990. A Nordic proposition for an energy corrected milk (ECM) formula. 26th session of the International Committee for recording the productivity of Milk Animals (ICRPMA).

32

NorFor – The Nordic feed evaluation system

4. Feed fraction characteristics H. Volden Absorbed nutrients are provided from fermentation and digestion in the gastrointestinal tract, and the predictions are sensitive to the nutrient profiles of the feed. NorFor has an extensive feed table (www.norfor.info) that lists chemical composition and digestion characteristics of typical Nordic feedstuffs. The feed table values are continuously updated as new information becomes available.

4.1 Definition of roughage and concentrates Roughage and concentrates are feedstuff classes that are generally used and several criteria characterize both groups of feedstuffs, e.g. fibre content, energy density, moisture content and particle length. In NorFor it is necessary to identify a feedstuff either as ‘roughage’ or ‘concentrate’. This is decided from information on particle length; feedstuffs with particle lengths greater or less than 6 mm are characterized as roughages and concentrates, respectively. The identification of feedstuff class is necessary when describing the feed FV and in the calculation of the CI. This information is also necessary to predict the fractional passage rate of a feed fraction out of the rumen. Feedstuffs are further classified as liquid, fine, coarse, rolled, chopped, and unchopped, since this is used for identification of most frequent particle length (PL) which affects the calculated CI values.

4.2 Division of organic matter The DM is separated into OM and ash. The OM is further divided into the following components: CP, NDF, ST, CFat, LAF, VFA, alcohols and a calculated RestCHO fraction including SU. The main chemical components are further divided into sub-fractions (Figure 4.1) that have uniform rates of kd in the rumen.

Soluble Potentially degradable Total indigestible

Protein NDF

‘Soluble’ Potentially degradable Total indigestible

Amino acids Potentially degradable Total indigestible

Starch

Fermentation pr. Sugar Residual CHO Crude fat

Fatty acids

Ash

Figure 4.1. The feed fractionation scheme in NorFor.

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_4, © Wageningen Academic Publishers 2011

33

4.3 Crude protein fractions and amino acids The composition and properties of dietary proteins strongly influence protein metabolism in the gastrointestinal tract of ruminants, hence information on these variables is essential for robust feed evaluation. Feed CP (N·6.25) consists primarily of AA and smaller amounts of non-amino nitrogen, which is composed of NH3N, urea and nucleic acids. In NorFor CP is partitioned into three fractions: (1) soluble (sCP), (2) potentially degradable in the rumen (pdCP) and (3) total tract-indigestible (iCP). As example, Table 4.1 presents average CP fractions for a limited number of common feedstuffs. The sCP fraction contains soluble proteins, peptides, free AA, and non-amino nitrogen. When calculating ruminal degradation of the sCP fraction instantaneous degradation of NH3N and urea is assumed and a fractional degradation rate of 150%/h is used for the soluble AA fractions (Russel et al., 1992; Volden et al., 1998, 2002). The pdCP fraction is assumed to be insoluble, but degradable by microbes in the rumen. Individual rates of degradation (kdCP) for the pdCP fraction are estimated from ruminal degradation profiles using the in sacco technique (see Section 5.2.2 for a description of the method), assuming that the degradation profile follows first-order kinetics. Intestinal digestibility of the undegraded rumen feed protein (RUP) may vary considerably between feeds (Hvelplund, 1985; Van Straalen and Tamminga, 1990; Volden and Harstad, 1995). The in sacco mobile bag technique (MBT) provides an easy and fast method to determine intestinal digestibility of protein (Hvelplund et al., 1992, 1994) (see Section 5.2.4 for a description of the method). In NorFor, the MBT technique is used to determine feed iCP. The iCP represents a protein fraction that is completely resistant to both microbial degradation in the rumen and digestion in the small intestine. However, ruminal pre-incubation has been shown to have important effects on the intestinal digestion of protein in several feeds (Volden and Harstad, 1995;Vanhatalo et al., 1996). Consequently, iCP is determined in feed residues after 16 and 24 h of ruminal incubation for concentrates and roughages, respectively. Volden and Harstad, 1995 observed that pre-incubation in the rumen increased intestinal digestion and thus reduced the iCP. This implies that the feed protein contains a fraction that is not degradable Table 4.1. Crude protein (CP) fractions in selected feeds and the corresponding fractional degradation rate of the potential degradable protein fraction. Feedstuff

Barley Wheat Maize Peas Rape seed Rape seed meal Soybean meal Maize silage, high digestibility Maize silage, low digestibility Grass silage, high digestibility Grass silage, low digestibility

NorFor feed code

CP g/kg DM

Protein fractions, g/kg CP sCP1

pdCP2

iCP3

001-0016 001-0020 001-0014 003-0006 002-0007 002-0042 002-0053 006-0307 006-0309 006-0461 006-0463

113 131 96 239 218 388 516 78 75 173 144

290 220 114 711 280 216 160 464 437 668 628

670 750 886 289 660 734 840 432 459 267 228

36 29 51 28 85 58 11 140 140 38 49

kdCP4 %/h 11.3 14.3 2.5 9.1 9.5 9.5 7.9 4.6 4.6 12.1 8.0

1

sCP=soluble crude protein fraction. pdCP=potentially degradable crude protein. 3 iCP=total indigestible crude protein. 4 kdCP=fractional degradation rate of pdCP. 2

34

NorFor – The Nordic feed evaluation system

in the rumen, but is digestible in the small intestine, explaining why the sum of sCP + pdCP + iCP in most feedstuffs is not 1000 g/kg CP. The remaining CP fraction is added to the protein fraction digested in the small intestine. The amount of RUP is one of the primary factors determining the amount of dietary AAs entering the small intestine. NorFor predicts individual AAs absorbed from the small intestine. Thus, information regarding the AA profile of the RUP is required. A variable fraction of the feed protein is degraded in the rumen and the AA profile of the RUP may differ from the AA profile of the intact feed protein. Several authors have compared the AA profiles of intact feed protein and in sacco residues after ruminal incubation (Skórko-Sajko et al., 1994; Skiba et al., 1996; Weisbjerg et al., 1996; Zebrowska et al., 1997; Prestløkken, 1999; Harstad and Prestløkken, 2000; Harstad and Prestløkken, 2001). Although a limited number of feeds were evaluated, the acquired data suggest that there are only minor differences between the AA profiles of intact feed protein and RUP. Therefore, the AA profile and the sum of AA nitrogen (AA-N) in RUP are assumed to be the same as those of the original feedstuff. The NorFor feedstuff table (www.norfor.info) has information on the AA profiles of individual feeds.

4.4 Carbohydrate fractions Carbohydrates (CHO) are heterogeneous feed constituents that collectively comprise the largest component of cattle diets, and make the major contribution to supporting ruminal microbial growth. CHO digestion in the gastrointestinal tract varies considerably and the end-products from ruminal fermentation and intestinal digestion are the major nutrient supply for animal production. Moreover, the rate of CHO degradation in the rumen is a major factor affecting ruminal pH and thus the rumen environment and feed utilization. A good balance between the different CHO fractions is essential to maintain normal rumen function and optimize rumen fermentation. Carbohydrates can be divided into two main fractions, structural and non-structural The structural CHO originates from plant cell wall material (consisting of cellulose, hemicellulose and lignin), which in NorFor is defined as NDF (Van Soest, 1994). The NDF fraction is further divided into a potentially degradable fraction (pdNDF) and a total indigestible fraction (iNDF). The iNDF is determined from ruminal in sacco incubation for 288 h (see Section 5.2.3 for a description of the method), and the pdNDF fraction is estimated as total NDF minus iNDF. In concentrates, the rate of pdNDF degradation (kdNDF) is estimated by in sacco degradation, assuming first-order degradation kinetics (see Section 5.2.2 for a description of the method). In roughage, kdNDF is predicted from the combination of in vivo OMD and iNDF and using the Lucas principle applied to the excretion of non-fibre matter (Weisbjerg et al., 2004a, b). A complete description of the method and calculations is presented in Section 6.1. Examples of NDF content and degradation characteristics in commonly used feedstuffs are presented in Table 4.2. The non-structural carbohydrates consist of carbohydrate-based cell contents, including ST, SU, β-glucans and some of the pectins. In NorFor ST and SU are determined analytically, while β-glucans and pectins are part of the RestCHO fraction. Starch is an important feed constituent for high-yielding cattle due to its high digestibility and importance as a source of substrates for propionic acid fermentation in the rumen. Inadequate ST intake may depress microbial activity and thus reduce microbial protein synthesis. However, excessive levels of ST may depress fibre digestibility and roughage utilization, as well as causing ruminal acidosis and abnormalities in the rumen tissue (Owens et al., 1998). Starch digestibility in the rumen depends on grain type, grain processing and particle length (Mills et al., 1999; Offner and Sauvant, 2004; Larsen et al., 2009). In NorFor, ST is partitioned into three fractions (Table 4.3): ‘soluble’ (sST), potentially degradable (pdST) and indigestible (iST). The ST fractions are calculated from in sacco degradation profiles (see Section 5.2.2). The sST fraction represents small starch particles NorFor – The Nordic feed evaluation system

35

Table 4.2. NDF fractions in selected feeds and the corresponding fractional degradation rate of the potential degradable NDF fraction. Feedstuff

Barley Wheat Maize Peas Rape seed Rape seed meal Soybean meal Maize silage, high digestibility Maize silage, low digestibility Grass silage, high digestibility Grass silage, low digestibility

NorFor feed code

NDF g/kg DM

NDF fractions, g/kg NDF pdNDF1

iNDF2

001-0016 001-0020 001-0014 003-0006 002-0007 002-0042 002-0053 006-0307 006-0309 006-0461 006-0463

198 127 111 126 188 290 133 338 397 501 597

836 831 913 984 686 500 939 828 803 888 793

164 169 87 16 314 500 61 172 197 112 207

kdNDF3 %/h 3.2 3.5 3.0 10.4 13.3 6.7 5.0 3.4 3.1 5.0 3.7

1

pdNDF=potentially degradable NDF. iNDF=total indigestible NDF. 3 kdNDF=fractional degradation rate of pdNDF. 2

Table 4.3. Starch fractions in selected feeds and the corresponding fractional degradation rate of the potential degradable starch fraction. Feedstuff

Barley Wheat Maize Peas Maize silage, high digestibility Maize silage, low digestibility

NorFor feed code

Starch g/kg DM

Starch fractions, g/kg ST sST1

pdST2

iST3

001-0016 001-0020 001-0014 003-0006 006-0307 006-0309

615 667 712 511 347 299

350 399 230 230 500 500

650 601 770 770 500 500

3 5 30 30 10 10

kdST4 %/h 39.2 59.5 9.0 9.0 40.0 40.0

1

sST=small starch granules (‘water soluble starch fraction’). pdST=potentially degradable starch. 3 iST=total indigestible starch. 4 kdST=fractional degradation rate of pdST. 2

that are lost from nylon bags washed in cold water. It is assumed that the sST fraction follows the ruminal liquid pool and the fractional degradation rate is set to 150%/h, irrespective of feed source. The pdST fraction is calculated as the difference between the asymptotic value obtained from in sacco incubation and the sST value. The iST fraction is assumed to be completely resistant to digestion in the small intestine and is determined by the MBT (Norberg et al., 2007). The fractional degradation rates (kdST) of pdST are feed-specific and highly variable, ranging from 9%/h (maize) to 79%/h (oats) (Table 4.3). The kdST is estimated from in sacco degradation profiles.

36

NorFor – The Nordic feed evaluation system

704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 720 720 719 720 721 721 721 722 722 722 723 723 723 724 724 724 725 725 725 726 726 726 727 727 727 728 728 728 729 729 729 730 730 730 731 731 731 732 732 732 733 733 733 734 734 734 735 735 735 736 736 736 737 737 737 738 738 738 739 739 739 740 740 740 741 741 741 742 742 742 743 743 743 744 744 744 745 745 745 746 746 746 747 747 747 748 748 748 749 749 749 750 750 750 751 751 751 752 752 752 753 753 753 754 754 754 755 755 755 756 756 756 757 757 757 758 758 758 759 759 759

MBT (Norberg et al., 2007). The fractional degradation rates (kdST) of pdST are feed-specific and highly variable, ranging from 9%/h (maize) to 79%/h (oats) (Table 4.3). The kdST is estimated from in sacco degradation profiles. Table 4.3

The RestCHO fraction includes β-glucans, pectic substances and SU. Although the SU fraction

The RestCHO fraction includes β-glucans, pectic substances and SU. Although the SU fraction maybe beanalytically analyticallydetermined, determined, it categorized is categorized a part of RestCHO the RestCHO fraction. This is because may it is as aaspart of the fraction. This is SU is not analyzed in all in samples and tabulated values areare normally used. The because SUroutinely is not routinely analyzed all samples and tabulated values normally used. TheRestCHO is estimated by: RestCHO is estimated by: Re stCHO = 1000 − Ash − CP − CFat − NDF − ST − FPF

[Eq. 4.1]

4.1

where residual carbohydrate fraction in the g/kg DM; the ash content whereRestCHO RestCHOisisthethe residual carbohydrate fraction infeed, the feed, g/kg Ash DM;isAsh is the ash content in the feed, g/kg DM; CPCP is the crude protein content in theinfeed, g/kg DM; CFat is CFat the crude fatcrude fat in the feed, g/kg DM; is the crude protein content the feed, g/kg DM; is the content in the feed, g/kg DM; and is sum of products in content in in the the feed, feed, g/kg g/kg DM; DM; NDF and FPF FPF is the the sum of the theinfermentation fermentation products in isthe thethefeed, feed, content is the NDF content the feed, g/kg DM; ST starch content inthe the feed, g/kg DM; NDF is the NDF content in the feed, g/kg DM; ST is the starch content content Equation Equationin4.7. 4.7. feed, g/kg DM; and FPF is the sum of the fermentation products in the feed, in the feed, the fermentation products in the feed, Equation 4.7. Equation 4.7.g/kg DM; and FPF is the sum of 34 If N, the the RestCHO RestCHO If the the feedstuff feedstuff contains contains low low molecular molecular weight weight N N fractions, fractions, such such as as urea urea and and NH NH33N, the3RestCHO If contains low molecular weight Nformulas: fractions, such such as urea If the thefeedstuff feedstuff contains low molecular weight N fractions, asand ureaNH and N, the RestCHO is corrected (RestCHOcorr) using the 3N,NH is corrected (RestCHOcorr) using the following following formulas: is using thethe following formulas: is corrected corrected(RestCHOcorr) (RestCHOcorr) using following formulas: For For urea urea correction: correction: For For urea ureacorrection: correction: UreaN UreaN CFat − NDF − ST − FPF Re Re stCHOcorr stCHOcorr = = 1000 1000 − − Ash Ash − − CPcorr CPcorr − − CP CP ⋅⋅ UreaN − − CFat − NDF − ST − FPF 2915 − CFat − NDF − ST − FPF Re stCHOcorr = 1000 − Ash − CPcorr − CP ⋅ 2915 2915

[Eq. [Eq. 4.2] 4.2] [Eq. 4.2]

CP NH 3 N CFat − NDF − ST − FPF Re Re stCHOcorr stCHOcorr = = 1000 1000 − − Ash Ash − − CPcorr CPcorr − − CP ⋅⋅ NH 33 N − − CFat − NDF − ST − FPF 6 1000 − CFat − NDF − ST − FPF Re stCHOcorr = 1000 − Ash − CPcorr − .25 ⋅ 1000 6.25 1000

[Eq. [Eq. 4.3] 4.3] [Eq. 4.3]

⎛⎛ NH NH3N N ⎞⎞ CPcorr CPcorr = CP = ⎜⎛⎜11 − − NH33N ⎟⎟⎞ ⋅⋅ CP 1000 ⎠⎟⎠ ⋅ CP CPcorr = ⎝⎜⎝1 − 1000 1000 ⎠ ⎝

[Eq. [Eq. 4.4] 4.4] [Eq. 4.4]

4.2

For ammonia correction: For For ammonia ammoniacorrection: correction: For ammonia correction:

4.3

whereRestCHOcorr RestCHOcorr is the residual carbohydrate fraction corrected for low molecular where is residual carbohydrate fraction corrected for nitrogen where RestCHOcorr is the the residual carbohydrate fraction corrected for low low molecular molecular nitrogen nitrogen fractions in the feed, g/kg DM; Ash is the ash content in the feed, g/kg DM; CP is the crude protein where RestCHOcorr is the residual carbohydrate fraction corrected for low molecular nitrogen fractions in the feed, g/kg DM; Ash is the ash content in the feed, g/kg DM; CP is the crude fractions in the feed, g/kg DM; Ash is the ash content in the feed, g/kg DM; CP is the crude fractions the feed, feed, the content infat thecontent feed, g/kg DM; CPg/kg is the crude contentcontent ininthe DM; CFat theash fat content in the g/kg DM; NDF is the protein in feed, g/kg DM; CFat is the crude in the feed, DM; NDF is protein content in the theg/kg feed,DM; g/kgAsh DM;isis CFat iscrude the crude fat content in feed, the feed, g/kg DM; NDF is neutral protein content incontent thefibre feed, g/kg DM; CFat isDM; the crude incontent the content feed, DM;feed, NDF isDM; FPF the neutral detergent in feed, g/kg DM; ST is the in the g/kg detergent fibre incontent the feed, g/kg ST isfat the starch ing/kg the g/kg the neutral detergent fibre content in the the feed, g/kg DM; STcontent is the starch starch content in feed, the feed, g/kg the neutral fibre content in products theinfeed, g/kg ST is the starch content inCPcorr the feed, g/kg protein DM; FPF the sum fermentation the feed, Equation 4.7, DM; is the sumis of products the in feed, Equation 4.7, g/kgg/kg DM; CPcorr isis DM; FPF isdetergent thefermentation sum of of fermentation products in theDM; feed, Equation 4.7, g/kg DM; CPcorr iscrude DM; FPF the sum of products themolecular feed, Equation 4.7, g/kg CPcorr crude protein content in the corrected for low N g/kg DM; NH the Nisis isammonia the crude protein content infermentation the feed feedfor corrected for in low molecular N fractions, fractions, g/kgDM; DM; NH3N content inisthe feed corrected low molecular N fractions, g/kg DM; NH N 3N is 3the crude protein content in the feed corrected for low molecular N fractions, g/kg DM; NH ammonia N content in feed, g/kg N; UreaN is the urea N content in feed, g/kg N; 2915 3N is the ammonia N content in feed, g/kg N; UreaN is the urea N content in feed, g/kg N; 2915 content in feed, g/kg N; UreaN is the urea N content in feed, g/kg N; 2915 corresponds to the CP ammonia N content in content feed, g/kg N; UreaN is the urea N content in feed, g/kg N; 2915 corresponds to of kg corresponds to the the CP content of one one kg of of urea. urea. content of one kgCP of urea. corresponds to the CP content of one kg of urea. The CPcorr is calculated as: The calculated The CPcorr CPcorris calculatedas: The CPcorr isiscalculated as:as: 4.4

where CPcorr is the crude protein content in the feed corrected for low molecular weight N fractions,

where the crude N where CPcorr CPcorr is crude protein protein content content in in the the feed feed corrected corrected for for low low molecular molecular weight N g/kg DM; CPisisthe theCP crude protein content in the g/kg DM; and NH N isweight the or urea where CPcorr the crude protein content the feedfeed, molecular fractions, g/kg is crude protein in the g/kg DM; and N is ammonia theN 3 fractions, g/kgisDM; DM; CP is the the crude proteinincontent content incorrected the feed, feed, for g/kglow DM; and3NH NHweight 3N is the N content in the feed, g/kg N. fractions, g/kg DM; CP is the crude protein content in the feed, g/kg DM; and NH N is the ammonia or urea N content in the feed, g/kg N. 3 ammonia or urea N content in the feed, g/kg N. ammonia or urea N content in the feed, g/kg N. The RestCHO RestCHOfraction fractionrepresents representsthe therapidly rapidlyfermented fermentedwater-soluble water-soluble CHO fractions; The CHO fractions; aa a heterogeneous The RestCHO fraction represents the rapidly fermented water-soluble CHO fractions; The RestCHO fraction rapidly CHO fractions; arate (Weisbjerg et heterogeneous group carbohydrates, ranging from SU, has ruminal degradation group of carbohydrates, rangingthe from SU, fermented which awhich high ruminal degradation heterogeneous group of ofrepresents carbohydrates, ranging from has SU,water-soluble which has aa high high ruminal degradation heterogeneous group of carbohydrates, ranging from SU, which has a high ruminal degradation rate (Weisbjerg et al., 1998) to soluble fibre fractions, which have moderate degradation al., 1998) to soluble fibre fractions, which have moderate degradation rates (Sniffen et al., 1992; rate (Weisbjerg et al., 1998) to soluble fibre fractions, which have moderate degradation rates rates rate (Weisbjerg et al.,Lanzas 1998) to fibre which have rates (Sniffen al., 1992; et al., In NorFor the fractional degradation of (Sniffen etal., al.,2007). 1992; Lanzas et soluble al., 2007). In fractions, NorFor the fractional degradation rate of RestCHO RestCHO Lanzas etet In NorFor the2007). fractional degradation rate ofmoderate RestCHOdegradation inrate roughage (kdRestCHO) (Sniffen et al., 1992; Lanzas al., 2007). in (kdRestCHO) is calculated as: in roughage (kdRestCHO) is et calculated as:In NorFor the fractional degradation rate of RestCHO is roughage calculated as: in roughage (kdRestCHO) is calculated as: SU ⎛⎛ SU ⎞⎞⎟ ⋅ 150 SU ⎞⎞⎟ ⋅ 10 + ⎛⎛⎜ SU kd = ⎜⎜⎛11 − − kd Re Re stCHO stCHO = SU ⎟⎠⎞ ⋅ 150 SU ⎟⎠⎞ ⋅ 10 + ⎜⎝⎛ Re stCHO Re stCHO ⎝ Re stCHO Re stCHO ⎠⎟ ⋅ 10 + ⎝⎜ ⎠⎟ ⋅ 150 kd Re stCHO = ⎝⎜1 − ⎝ Re stCHO ⎠ ⎝ Re stCHO ⎠

[Eq. [Eq. 4.5] 4.5] [Eq. 4.5]

4.5

where kdRestCHO is the degradation rate of the rest carbohydrate fraction, %/h; SU is the sugar

where kdRestCHO is the rate of fraction, %/h; SU is where kdRestCHO is the degradation degradation ratecarbohydrate of the the rest rest carbohydrate carbohydrate fraction,4.1; %/h; is the the sugar sugar content, g/kg DM;RestCHO RestCHO is therest rest fraction, Equation theSU fractional degradation where kdRestCHO is the degradation of the rest carbohydrate fraction, SU is the sugar content, g/kg is carbohydrate fraction, 4.1; the content, g/kg DM; DM; RestCHO is the the restrate carbohydrate fraction, Equation Equation 4.1;%/h; the fractional fractional rate of SU is 150%/h; and the fractional degradation rate of the non-sugar part of the RestCHO content, g/kg DM; RestCHO is the rest carbohydrate fraction, Equation 4.1; the fractional

fraction is 10%/h. In concentrate feeds the kdRestCHO is set to 150%/h. 35 35 35

NorFor – The Nordic feed evaluation system

37

760 761 762 763

degradation ratefat of SU is 150 %/h;acids and the fractional degradation rate of the non-sugar part of the 4.5 Crude and fatty RestCHO fraction is 10 %/h.

764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781

richCrude than CPfatorand CHO, although 4.5 fatty acids the energy value is dependent on the proportion of fatty acids (FA)

782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808

Fatconcentrate normally feeds constitutes a low proportion of cattle diets. However, fat is often used as an energy In the kdRestCHO is set to 150%/h. source for high productivity cattle. On a weight basis, fat is approximately 2.2 times more energy-

in the CFat. constitutes However, arumen microorganisms cannot toleratefat high levels of as fatan(Jenkins, Fat normally low proportion of cattle diets. However, is often used energy 1993). Therefore, bothproductivity the level and type FA are basis, important to consider when optimizing cattle source for high cattle. Onofa weight fat is factors approximately 2.2 times more diets. The NorFor (www.norfor.info) includes information on FA profiles of individual energy-rich than CPfeedstuff or CHO,table although the energy value is dependent on the proportion of fatty feedstuffs. FA metabolism in the cannot gastro tolerate intestinal tract is not and is, acids (FA) inHowever, the CFat.individual However, rumen microorganisms high levels of expressed fat (Jenkins, Therefore, both and typemodule. of FA areConsequently, important factors consider when therefore,1993). not implemented in the the level metabolism onlytoinformation on dietary includes onlevel FA of the optimizing cattle diets. NorFor feedstuff (www.norfor.info) FA composition can beThe evaluated, which istable important when focusing on theinformation optimal FA profiles individual feedstuffs. However, individual FA metabolism intestinal tract feeds. diet, on of milk fat quality and milk off-flavours. The proportion of FAininthe thegastro CFat varies between is notmain expressed andlipids is, therefore, implemented in are the metabolism module. Consequently, The storage in plantnot seeds and grains triacylglycerols, which consist of three fatty only information on dietary FA composition can be evaluated, which is important acids attached to glycerol. In these feeds the proportion of FA in CFat is 80 towhen 85%focusing (Danfær et al., on the optimal FA level of the diet, on milk fat quality and milk off-flavours. The proportion of 2006). In contrast, forage lipids are mainly in the form of galactolipids, which contain galactose in FA in the CFat varies between feeds. The main storage lipids in plant seeds and grains are addition to glycerol and unsaturated Thus, their In FAthese proportions triacylglycerols, which consist of threefatty fatty acid acidsmoieties. attached to glycerol. feeds theare substantially lower: approximately 45% (Danfær et al., 2006). When the rumen OM fermentation is predicted in proportion of FA in CFat is 80 to 85% (Danfær et al., 2006). In contrast, forage lipids are mainly NorFor, theofglycerol and galactose are added to the feed fractions, and serve as energy in the form galactolipids, which contain galactose in fermentable addition to glycerol and unsaturated fatty sources for microbial growth. acid moieties. Thus, their FA proportions are substantially lower: approximately 45% (Danfær et al., 2006). When the rumen OM fermentation is predicted in NorFor, the glycerol and galactose are to the fermentable feed fractions, and serve as energy sources for microbial growth. 4.6added Fermentation products

4.6 Fermentation products Ensiled forages dominate as winter roughages in the Nordic countries. The feedstuffs have highly

Ensiled dominate as winter in the Nordic The feedstuffs highly acids) variableforages FPF (i.e. VFA, LAF and roughages alcohols) contents. Thecountries. VFA (acetic, propionic have and butyric variable FPF (i.e. account VFA, LAF alcohols) contents. Themajor VFA (acetic, can collectively forand up to 70 g, and LAF (the organicpropionic acid) for and 40 tobutyric 150 g per kg DM acids) collectively account to 70 g, and lactic acid (the major organic for 40 tosources for of the can silage. The VFA, whichfor areupend-products of ruminal fermentation, areacid) not energy 150 g per kg DM of the silage. The VFA, which are end-products rumen microorganisms. Lactic acid represents a minor sourceofofruminal energy fermentation, for microbialare growth, and not sources forlactic rumenacid microorganisms. a minor energy the energy ATP yield from metabolism isLactic only acid 25%represents of the yield fromsource CHOof fermentation (Van for microbial growth, and the ATP yield from lactic acid metabolism is only 25% of the yield Soest, 1994). A high content of fermentation products in silages has a negative effect on forage intake from CHO fermentation (Van Soest, 1994). A high content of fermentation products in silages (Huhtanen et effect al., 2002, 2007). Therefore, NorFor relative NorFor silage index (Huhtanen has a negative on forage intake (Huhtanen et al.,incorporates 2002; 2007).the Therefore, et al., 2002)the when calculating feed(Huhtanen intake (seeet Section 6.2). The silage index information incorporates relative silage index al., 2002) when calculating feed requires intake (see on fermentation acids (TAF), calculated as: section 6.2). The silage index requires information on fermentation acids (TAF), calculated as: TAF = LAF + ACF + PRF + BUF

[Eq. 4.6]

4.6

where of of total fermentation acidsacids in theinfeed, g/kg DM; the content whereTAF TAFisisthe thecontent content total fermentation the feed, g/kgLAF DM;isLAF is the of content of lactic g/kg DM; ACF is the acid content in the in feed, PRF is the lacticacid acidininthe thefeed, feed, g/kg DM; ACF is acetic the acetic acid content theg/kg feed,DM; g/kg DM; PRF is the propionic feed, g/kg DM; BUF is the acid acid content in thein feed, propionicacid acidcontent contentininthethe feed, g/kg DM; BUF is butyric the butyric content the g/kg feed, g/kg DM. DM

Ensiled forage also contains variable amounts of alcohol (Randby et al., 1999). This is taken into

Ensiled forage also contains variable amounts of alcohol (Randby et al., 1999). This is taken into accountwhen whenFPF FPFisiscalculated: calculated: account FPF = TAF + FOF + ALF

[Eq. 4.7]

4.7

36 where content of fermentation products in thein feed, TAF is the content where FPF FPFisisthethe content of fermentation products theg/kg feed,DM; g/kg DM; TAF is theofcontent of fermentation acids in the feed, Equation 4.6, g/kg DM; FOF is the content of formic acid in the fermentation acids in the feed, Equation 4.6, g/kg DM; FOF is the content of formic acid in the feed, feed, g/kg DM; the content of alcohol feed, g/kgDM. DM. g/kg DM; ALFALF is theis content of alcohol in in thethe feed, g/kg

809 810 811 812 813 814 815 816

4.7 Minerals

817

4.8 Vitamins

4.7 Minerals

Mineral elements may adversely affect production, animal health and reproduction (NRC, 2001). Several elements are also important for maintaining normal rumen function. NorFor predicts Mineral elements may adversely affect production, animal health and reproduction (NRC, 2001). mineral supply and incorporates requirements for the following elements: Calcium (Ca), Several elements are also important for maintaining normal rumen function. NorFor(S), predicts Phosphorus (P), Magnesium (Mg), Potassium (K), Sodium (Na), Chlorine (Cl), Sulfur Iron mineral (Fe), Manganese (Mn), Zinc (Zn), Copper (Cu), Cobalt (Co), Selenium (Se), Molybdenum (Mo) and Iodine (I). Animal mineral requirements are described in Chapter 10. The NorFor feedstuff 38 (www.norfor.info) includes information on the mineral NorForcontent – The of Nordic feedfeedstuffs. evaluation system individual table

supply and incorporates requirements for the following elements: Calcium (Ca), Phosphorus (P), Magnesium (Mg), Potassium (K), Sodium (Na), Chlorine (Cl), Sulfur (S), Iron (Fe), Manganese (Mn), Zinc (Zn), Copper (Cu), Cobalt (Co), Selenium (Se), Molybdenum (Mo) and Iodine (I). Animal mineral requirements are described in Chapter 9. The NorFor feedstuff table (www.norfor. info) includes information on the mineral content of individual feedstuffs.

4.8 Vitamins Rumen microbes synthesize adequate amounts of vitamin K to meet the requirements of most cattle, except in young calves. However, cattle require a dietary source of vitamins A and E. Vitamin D is also normally included in the diet. NorFor incorporates requirements for vitamins A, E and D (see Chapter 9) and the NorFor feedstuff table (www.norfor.info) provides information on the vitamin A, E and D contents of individual feedstuffs.

References Danfær, A., P. Huhtanen, P. Udén, J. Sveinbjörnsson and H. Volden, 2006. The Nordic dairy cow model, Karoline – Description. In: Kebreab, E., J. Dijkstra, A. Bannink, J. Gerrits and J. France (eds.). Nutrient digestion and utilisation in farm animals: Modelling approaches. CAB International, Wallingford, UK, pp. 383-406. Harstad, O.M. and E. Prestløkken, 2000. Effective degradability and intestinal digestibility of individual amino acids in solvent-extracted soybean meal (SBM) and xylose-treated SBM (SoyPass®) determined in sacco. Animal Feed Science and Technology 83: 31-47. Harstad, O.M. and E. Prestløkken, 2001. Rumen degradability and intestinal digestibility of individual amino acids in corn gluten meal, canola meal and fish meal determined in sacco. Animal Feed Science and Technology 94: 127-135. Huhtanen, P., H. Khalili, J.I. Nousiainen, M. Rinne, S. Jaakkola, T. Heikkilä and J. Nousiainen, 2002. Prediction of the relative intake potential of grass silage by dairy cows. Livestock Production Science 73: 111-130. Huhtanen, P., M. Rinne and J. Nousiainen, 2007. Evaluation of the factors affecting silage intake of dairy cows: a revision of the relative silage dry-matter index. Animal 1: 758-770. Hvelplund, T., 1985. Digestibility of rumen microbial protein and undegraded dietary protein estimated in the small intestine of sheep and by the in sacco procedure. Acta Agriculturae Scandinavica Supplement 25: 132-144. Hvelplund, T., M.R. Weisbjerg and L.S. Andersen, 1992. Estimation of the true digestibility of rumen undegraded protein in the small intestine of ruminants by the mobile bag technique. Acta Agriculturae Scandinavica, section A: Animal Science 42: 34-39. Hvelplund, T., F.D. DeB. Hovell, E.R. Ørskov and D.J. Kyle, 1994. True intestinal digestibility of protein estimated with sheep on intragastric infusion and with the mobile nylon bag technique. Proceedings of the Society of Nutrition Physiology, 3. pp. 64. Jenkins, T.C., 1993. Lipid metabolism in the rumen. Journal of Dairy Science 76: 3851-3863. Lanzas, C., C.J. Sniffen, S. Seo, L.O. Tedeschi and D.G. Fox, 2007. A revised CNCPS feed carbohydrate fraction scheme for formulating rations for ruminants. Animal Feed Science and Technology 136: 167-190. Larsen, M., P. Lund, M.R. Weisbjerg and T. Hvelplund, 2009. Digestion site of starch from cereals and legumes in lactating dairy cows. Animal Feed Science and Technology 153: 236-248. Mills, J.A.N., J. France and J. Dijkstra, 1999. A review of starch digestion in the lactating dairy cow and proposals for a mechanistic model: 1. Dietary starch characterization and ruminal starch digestion. Journal of Animal and Feed Sciences 8: 291-340. Norberg, E., O.M. Harstad and H. Volden, 2007. Assessment of recovery site of mobile nylon bags for measuring ileal digestibility in dairy cows. Journal of Dairy Science 90: 418-421. NRC (National Research Council), 2001. Nutrient requirements of dairy cattle. Seventh revised edition, National Academy press, Washington, DC, USA. Offner, A. and D. Sauvant, 2004. Prediction of in vivo starch digestion in cattle from in sacco data. Animal Feed Science and Technology 111: 41-56. Owens, F.N., D.S. Secrist, W.J. Hill and D.R. Gill, 1998. Acidosis in cattle – a review. Journal of Animal Science 76: 1275-1286.

NorFor – The Nordic feed evaluation system

39

Prestløkken, E., 1999. Protein value of expander-treated barley and oats for ruminants. Ph.D. Dissertation No. 5. Department of Animal Science, Agricultural University of Norway, Aas, Norway. Randby, Å.T., I. Selmer-Olsen and L. Bævre, 1999. Effect of ethanol in feed on milk flavour and chemical composition. Journal of Dairy Science 82: 420-428. Russell, J.B., J.D. O’Connor, D.G. Fox, P.J. Van Soest and C.J. Sniffen, 1992. A net carbohydrate and protein system for evaluating cattle diets. I. Ruminal fermentation. Journal of Animal Science 70: 3551-3561. Skórko-Sajko, H., T. Hvelplund and M.R. Weisbjerg, 1994. Rumen degradation and intestinal digestibility of amino acids in different roughages estimated by nylon bag techniques. Journal of Animal and Feed Sciences 3: 1-10. Skiba, B., M.R. Weisbjerg and T. Hvelplund, 1996. Effective degradability in the rumen and total tract digestibility of amino acids in different roughages determined in sacco. Journal of Animal and Feed Sciences 5: 347-363. Sniffen, C.J., J.D. O’Connor, P.J. Van Soest, D.G. Fox and J.B. Russell, 1992. A net carbohydrate and protein system for evaluating cattle diets. II. Carbohydrate and protein availability. Journal of Animal Science 70: 3562-3577. Vanhatalo, A., P. Dakowski and P. Huhtanen, 1996. Effects of stage of growth and duration of rumen incubation time on intestinal digestibility of rumen-undegradable nitrogen of grass by mobile-bag method in cows. Acta Agriculturae Scandinavica, Section A - Animal Science 46: 1-10. Van Soest, P.J, 1994. Nutritional Ecology of the Ruminant. (2nd ed.) Cornell University press, Ithaca, NY, USA. 476 pp. Van Straalen, W.M. and S. Tamminga, 1990. Protein degradation of ruminant diets. In: Wiseman, J. and D.J.A. Cole (eds.). Feedstuff Evaluation. Butterworths, London, pp. 55-72. Volden, H. and O.M. Harstad, 1995. Effect of rumen incubation on the true indigestibility of feed protein in the digestive tract determined by the nylon bag techniques. Acta Agriculturae Scandinavica, Section A - Animal Science 45: 106-115. Volden, H., W. Velle, O.M. Harstad, A. Aulie and Ø. Sjaastad, 1998. Apparent ruminal degradation and rumen escape of lysine, methionine and threonine - administered intraruminally in mixtures to high yielding cows. Journal of Animal Science 76: 1232-1240. Volden, H.L., T. Mydland, V. Olaisen, 2002. Apparent ruminal degradation and rumen escape of soluble nitrogen fractions in grass, and grass silage administered intraruminally to lactating dairy cows. Journal of Animal Science 80: 2704-2716. Weisbjerg, M.R., T. Hvelplund, S. Hellberg, S. Olsson and S. Sanne, 1996. Effective rumen degradability and intestinal digestibility of individual amino acids in different concentrates determined in sacco. Animal Feed Science and Technology 62: 179-188. Weisbjerg, M.R., T. Hvelplund and B.M. Bibby, 1998. Hydrolysis and fermentation rate of glucose, sucrose and lactose in the rumen. Acta Agriculturae Scandinavica, Section A - Animal Science 48: 12-18. Weisbjerg, M.R., T. Hvelplund and K. Søegaard, 2004a. Prediction of digestibility of neutral detergent solubles using the Lucas principle. Journal of Animal and Feed Sciences 13, Supplement 1: 239-242. Weisbjerg, M.R., T. Hvelplund and K. Søegaard, 2004b. Prediction of NDF digestibility based on assumptions about true digestibility and endogenous loss of NDS. Journal of Animal and Feed Sciences, 13, Supplement 1: 243-246. Zebrowska, T., Z. Dlugolecka, J.J. Pajak and W. Korczynski, 1997. Rumen degradability of concentrate protein, amino acids and starch, and their digestibility in the small intestine of cows. Journal of Animal and Feed Sciences 6: 451-470.

40

NorFor – The Nordic feed evaluation system

5. Feed analyses and digestion methods M. Åkerlind, M. Weisbjerg, T. Eriksson, R. Tøgersen, P. Udén, B.L. Ólafsson, O.M. Harstad and H. Volden Feed characteristics are determined via chemical analyses and digestion methods. Specific NorFor methods for determining parameters such as DM, sCP, iNDF and the in sacco methods are fully described in this chapter. Tables 5.1 and 5.2 present an overview of the feed analysis and digestion methods, respectively. Table 5.1. Recommended NorFor feed analysis methods. Parameter

Abbrev.

Unit

Reference method

Dry matter in concentrate Dry matter in roughage Ash Crude protein

DM

g/kg

EC No. 152/2009

DM

g/kg

Ash CP

g/kg DM g/kg DM

Soluble CP

sCP

g/kg CP

Ammonia nitrogen

NH3N

g N/kg N

Individual AA1 Crude fat Individual FA2

AAj CFat FAj

Neutral detergent fibre Starch

NDF

Free choice of MgO-method or Autoanalyzer (Broderick and Kang, 1980) g/100g CP EC No. 152/2009 g/kg DM EC No. 152/2009 g/100 g FA CEN ISO/TS 17764-1:2007 CEN ISO/TS 17764-2:2007 g/kg DM ISO 16472:2006 IDT

ST

g/kg DM

Lactic, propionic, butyric, formic acids and alcohol (ethanol) Sugar Calcium Phosphorus Magnesium Potassium Sodium Chloride Sulphur

LAF, PRF, g/kg DM BUF, FOF, ALF SU Ca P Mg K Na Cl S

g/kg DM g/kg DM g/kg DM g/kg DM g/kg DM g/kg DM g/kg DM g/kg DM

EC No. 152/2009 EC No. 152/2009 or Dumas

NorFor method

See Section 5.1.1 See Section 5.1.2 See Section 5.1.3

Spectrophotometric method or the plate count method described by Bach Knudsen (1997) and Bach Knudsen et al. (1987) HPLC or GC

EC No. 152/2009 ICP or free choice of method EC No. 152/2009 or ICP ICP or free choice of method ICP or free choice of method ICP or free choice of method EC No. 152/2009 or ICP ICP or free choice of method

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_5, © Wageningen Academic Publishers 2011

41

Table 5.1. Continued. Parameter

Abbrev.

Unit

Reference method

Iron, Copper, Manganese and Zinc Other micro minerals Vitamin A β-carotene Vitamin D Vitamin E

Fe, Cu, Mn, Zn

g/kg DM

EC No. 152/2009 or ICP

g/kg DM IU/kg DM IU/kg DM IU/kg DM IU/kg DM

Free choice of method EC No. 152/2009 or Jensen et al. (1998) EC No. 152/2009 or Jensen et al. (1998) Any appropriate method is acceptable EC No. 152/2009 or Jensen et al. (1999)

VitA b-car VitD VitE

NorFor method

1 The amino acids that can be reported in the NorFor feed tables are: alanine, arginine, asparagine, cysteine, glutamine,

glycine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine and valine. 2 The fatty acids that can be reported in the NorFor feed tables are FA50 MJ NEG/d) and vice versa for small animals. However, it has not been possible to improve this relationship. Table 9.2. NE requirement (MJ/d) for growth in primiparous cows related to BW and daily gain.

Figure 9.2

BW, kg

Daily weight gain, g/d

The utilisation coefficient of ME to NE for growth, corrected for the proportion of energy that originates from protein retention, is calculated according to the INRA (1989) equation:

250

500

750

k g _ corr = 0.35 + 0.25 ⋅ (1 − Ep )2

[Eq. 9.6] 400 4.4 7.5 10.6 500 4.5 7.6 10.7 for growth, corrected for 10.9 the proportion of where 600 kg_corr is the utilisation 4.7coefficient of ME to NE7.8 energy that originates from protein retention; and 700 in the total amount of 4.8energy retained per day 7.9 11.1 Ep is the daily NE requirement for protein retention relative to the total daily NE retention, Equation 9.7.

NorFor – The Nordic feed evaluation system 5.48 ⋅ gain _ prot

Ep =

5.48 ⋅ gain _ prot + 9.39 ⋅ gain _ fat

[Eq. 9.7]

87

⎛ 80_ prot + 9.39 ⋅ gain _ fat ) ⋅ 4.184 ⋅ k mg NEG _ gain = ⎜ (5.48 ⋅ gain ⎜⎛ 41000 .184 k gk_mg corr NEG _ gain = ⎝⎜ (5.48 ⋅ gain 70_ prot + 9.39 ⋅ gain _ fat ) ⋅ 1000 ⋅ k ⎜ g _ corr ⎝

NEG requirement (MJ/d)

2834 2834 2835 2835 2836 2837 2836 2838 2837 2839 2838 2840 2839 2841 2840 2842 2841 2843 2842 2844 2843 2845 2844 2846 2845 2847 2846 2848 2847 2849 2848 2850 2849 2851 2850 2852 2851 2853 2852 2854 2853 2855 2854 2856 2855 2857 2856 2858 2857 2859 2858 2860 2859 2861 2860 2862 2861 2863 2862 2863 2864 2864 2865 2866 2865 2867 2866 2868 2867 2869 2868 2870 2869 2870 2871 2871

⎞ ⎟ ⋅ 1.10 ⎟⎞ ⎠⎟ ⋅ 1.10 ⎟ ⎠

[Eq. 9.5] [Eq. 9.5]

where NEG_gain is the60 daily NE requirement for gain in growing cattle, MJ NEG/d; 5.48 and 9.39 the energyiscontent inNE protein and fat, respectively, kcal/g;cattle, gain_prot is the daily whereare NEG_gain the50 daily requirement for gain in growing MJ NEG/d; 5.48protein and retention, Equation 9.8; gain_fat is theand daily retention , Equation 9.9; the factor 9.39 are the energy content in protein fat,fatrespectively, kcal/g; gain_prot is the 4.184/1000 daily protein converts kcal to MJ;9.8; kmg is the combined utilisation coefficient of ME to the NE factor for maintenance retention, Equation is the daily fat retention , Equation 9.9; 4.184/1000and 40gain_fat growth, Equation 8.5kand is the utilisation coefficient of ME to NE for for growth, corrected g_corr converts kcal to MJ; combined utilisation coefficient of ME to NE maintenance and mg iskthe 30 k thatisoriginates for the proportion of energy from coefficient protein retention daily energy the utilisation of MEintothe NEtotal for growth, corrected growth, Equation 8.5 and g_corr retention, Equationof9.6. for the proportion energy 20 that originates from protein retention in the total daily energy retention, Equation 9.6. 10 kg_corr takes into account the dependence of the utilization of ME for The utilization coefficient growth on the proportion of energy retained as protein. Thus, the ratio between kmg and kg_corr of ME foris The utilization coefficient 0 kg_corr takes into account the dependence of the utilization used to on adjust NE requirement for gain toasaccount for variations in the proportion of kenergy growth the the proportion of energy retained protein. Thus, the ratio between k and is mg g_corr 0 10 20 30 40 50 60 70 80 that from retention theto total daily for energy retention (INRA, 1989).ofThe 10% usedoriginates to adjust the NEprotein requirement forin gain account variations in the proportion energy adjustment (factor in Equation 9.5) fordaily growth was added because a dataset NEG supply (MJ/d) that originates from1.1 protein retention inin theNE total energy retention (INRA, 1989).compiled The 10% from 11 Norwegian andinsix Swedish9.5) experiments bulls involving 120 treatments adjustment (factor 1.1 Equation in NE for with growth was added because a dataset(Homb, compiled 1974; Hole, 1985; Olsson Lindberg, 1985; 1987; Johansen, 1992; Skaara, 1994; from 11 Norwegian and sixand Swedish experiments with bulls 120 treatments (Homb, Figure 9.2. The relationship between energyOlsson, supply andinvolving energy requirement after a modification of Randby, 2000a; Randby, 2000b; Berg aand Volden, 2004; Berg, 2005; Berg, 1974; Hole, 1985; Olsson and Lindberg, 1985; Olsson, 1987; 1992; Skaara, 1994; in order to the original French system where 10% increase inVolden NEJohansen, forand growth was implemented unpublished; Havrevoll, unpublished; unpublished) showed underprediction of line shows Y=X Randby, Randby, 2000b; Berg Olsson, and Volden, 2004;and Volden andan Berg, 2005; Berg, obtain a2000a; closer relationship between energy supply energy requirements. Solid RMSE = 4.4; see Figure 9.2). Although, this modification was made, requirements Havrevoll, (r2 = 0.90 and unpublished; unpublished; Olsson, unpublished) showed an underprediction of 2=0.90 and and dotted line shows relationship between supply and requirement (r RMSE=4.4). The 2 that there is also a slope effect indicating that the estimated energy supply Figure 9.2 shows is too = 0.90 and RMSE = 4.4; see Figure 9.2). Although, this modification was made, requirements (r data are from the trials described in Table 9.11. high compared to that the energy larger animals that (>50the MJestimated NEG/d) and vicesupply versa for Figure 9.2 shows there isrequirement also a slope in effect indicating energy is too smallcompared animals. However, it has not been possible improve(>50 this MJ relationship. high to the energy requirement in largertoanimals NEG/d) and vice versa for small animals. However, it has not been possible to improve this relationship. requirement Figure 9.2 in larger animals (>50 MJ NEG/d) and vice versa for small animals. However, we were not able Figure 9.2to improve this relationship when using the described approach. The utilisation coefficient of ME to NE for growth, corrected for the proportion of energy that originates fromcoefficient protein retention, is NE calculated according tocorrected theforINRA (1989) Theutilisation utilisation coefficient of ME to for NEgrowth, for growth, for the equation: proportion The of ME to corrected the proportion of energy of thatenergy that originates retention, is calculated according to thetoINRA (1989) equation: originatesfrom fromprotein protein retention, is calculated according the INRA (1989) equation: [Eq. 9.6] k g _ corr = 0.35 + 0.25 ⋅ (1 − Ep )2 [Eq. 9.6] k g _ corr = 0.35 + 0.25 ⋅ (1 − Ep )2 9.6 where kg_corr is the utilisation coefficient of ME to NE for growth, corrected for the proportion of energy theistotal amount of coefficient energy retained perto day originates fromcorrected protein andof wherekin kg_corr is utilisation coefficient of ME NEgrowth, for growth, for the proportion of where thethe utilisation of ME NEtothat for corrected for theretention; proportion g_corr Ep is the NE amount requirement for protein retention relative to the total daily NE retention, energy in the retained perper day that originates from protein retention; and and Ep is energy indaily thetotal total amountofofenergy energy retained day that originates from protein retention; Equation 9.7. Ep the daily NE requirement for protein retention relative thetotal totaldaily dailyNE NEretention, retention, Equation 9.7. theisdaily NE requirement for protein retention relative totothe Equation 9.7. 5.48 ⋅ gain _ prot [Eq. 9.7] 9.7 Ep = + 9_.39 ⋅ gain _ fat 5.48 ⋅ gain5._48prot ⋅ gain prot [Eq. 9.7] Ep = 5.48 ⋅ gain _ prot + 9.39 ⋅ gain _ fat

where Ep is the daily NE requirement for protein retention relative to the total daily NE retention;

2872 2873 2872 2874 2873 2875 2874 2876 2875 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886

where Ep is the daily NE requirement for protein retention relative to the total daily NE retention; 5.48 and9.39 9.39are arethethe energy content in protein andrespectively, fat, respectively, kcal/g; gain_prot is the daily 5.48 content infor protein andretention fat, gain_prot the daily whereand Ep is the daily energy NE requirement protein relative tokcal/g; the total daily NEis retention; protein retention, Equation 9.8; and gain_fat is the daily fat retention, Equation 9.9. protein and gain_fat is and the daily fat retention,kcal/g; Equation 9.9. is the daily 5.48 andretention, 9.39 are Equation the energy9.8; content in protein fat, respectively, gain_prot protein retention, Equation 9.8; and gain_fat is the daily fat retention, Equation 9.9. The following followingequations equations used to calculate the daily retention of protein and fat (Robelin and The areare used to calculate the daily retention of protein and fat (Robelin & Daenicke,1980) 1980)and and examples of and fat and protein retention in growing cattle Daenicke, examples of fat protein retention in growing cattle are givenare in given Tablesin Tables 110 9.3 9.3 and and9.4. 9.4. 110

2887 2888 2889 2890

where to factor_5 are dependent on the type of system 88 gain_fat is the daily fat retention, g/d; factor_3 NorFor – The Nordic feed evaluation growing cattle (Table 9.5); EBW is the empty body weight, Equation 9.12; Fat_mass is the fat content in the EBW, Equation 9.13 and EBWG is the empty body weight gain, Equation 9.10.

gain _ prot = factor _ 1 ⋅ 1.06 ⋅ (EBWG − gain _ fat) ⋅ FFM0.06

[Eq. 9.8]

9.8

where daily protein retention, kg/d;kg/d; factor_1 is dependent on the on typethe of type growing wheregain_prot gain_protisisthethe daily protein retention, factor_1 is dependent of growing cattle is the empty bodybody weight gain, gain, Equation 9.10; gain_fat is the daily fat daily fat cattle(Table (Table9.5); 9.5);EBWG EBWG is the empty weight Equation 9.10; gain_fat is the retention, FFM is fat freefree mass, Equation 9.11.9.11. retention,Equation Equation9.9; 9.9;and and FFM is fat mass, Equation ⎛ ⎛ ⎛ 1000 3 + 2 ⋅_factor _ 4 ⋅ ln( EBW _) )5⋅ factor _ 5 ⎞⎞ ⋅ Fat _ mass⎛ ⎞ ⎛⎜ (factor ⎞⎞ _ 3 + 2_⋅ factor 4 ⋅ ln( EBW ) ) ⋅ factor [Eq. ⎟ ⎟ ⋅ ( EBWG⎟)⎟1,⋅78 gain __fat fat==⎜ ⎛⎜⎜1000 ( EBWG )1.78 ⋅ ⎜ ⋅ Fat _ mass ⎞⎟ ⋅ ⎜ (⎟factor gain ⎟⎟ ⎜ . 1 78 ⎜ ⎝⎜ ⎝ ⎟ ⎟ EBW ⎠ ⎜ ⎠ 1,78 _ 5 EBW Factor Factor _ 5 ⎠⎠ ⎝ ⎠⎠ ⎝ ⎝⎝

9.9]

9.9

Table 9.3. Estimated fat and protein retention in heifers of dairy breeds related to BW and ADG. BW, kg

200 300 400 500

800 g ADG

600 g ADG

Fat, g/d

Protein, g/d

Fat, g/d

Protein, g/d

142 207 270 333

134 125 114 102

85 124 162 199

105 101 95 88

Table 9.4. Estimated fat and protein retention in bulls of dairy breeds related to BW and ADG. BW, kg

ADG, g/d

Fat, g/d

Protein, g/d

100 200 300 400 500 100 200 300 400 500

1,200 1,500 1,400 1,200 1,000 1,000 1,300 1,200 1000 800

86 200 297 309 290 62 175 226 223 195

200 238 211 172 139 168 211 187 151 120

2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886

The following equations are used to calculate the daily retention of protein and fat (Robelin & Table 9.5.1980) Coefficients used for fatretention and protein retention in growing Daenicke, and examples of calculating fat and protein in growing cattle are given cattle. in Tables 9.3 and 9.4.

2887 2888 2889 2890

where onon thethe type ofof growing where gain_fat gain_fatisisthe thedaily dailyfat fatretention, retention,g/d; g/d;factor_3 factor_3totofactor_5 factor_5are aredependent dependent type growing cattle9.5); (Table 9.5);isEBW is the empty body weight, Equation Fat_mass is fat the content fat cattle (Table EBW the empty body weight, Equation 9.12;9.12; Fat_mass is the in the content in the EBW, 9.13 and EBWG is body the empty body weight gain, Equation EBW, Equation 9.13Equation and EBWG is the empty weight gain, Equation 9.10. 9.10.

2891

EBWG =

2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902

Animal

Factor_1

Factor_2

Factor_3

gain _ prot = factor _ 1 ⋅ 1.06 ⋅ (EBWG − gain _ fat) ⋅ FFM0.06

Heifer or steer 0.1616

-6.311

1.811

Factor_4 0

Factor_5 0.8

Factor_6 [Eq. 9.8]

1.046

Factor_7 -0.3939

1 Bull EM -1.68retention, 0.0189 0.1609 1.0 on the 1.023 -0.2855 where gain_prot is0.1541 the daily protein kg/d; factor_1 is dependent type of growing 1 9.5); 0.1541 cattle (Table EBWG is the-5.433 empty body1.5352 weight gain, is the daily fat Bull LM 0 Equation 9.10; 1.2 gain_fat1.024 -0.2704 retention, 9.9; and FFM is fat free1.3708 mass, Equation 9.11. 1.0 Bull CB1Equation0.1541 -5.7541 0.0442 1.023 0.27795 ⎛ ⎛ 1000 ⋅ Fat _ mass ⎞ ⎛ (factor _ 3 + 2 ⋅ factor _ 4 ⋅ ln( EBW ) ) ⋅ factor _ 5 ⎞ ⎞ 1 EM: early maturing; CB: cross breed. ⎜ ⎟ ⎟ ⋅ ( EBWG )1,78 _ fat = ⎜ ⎜ maturing; LM:⎟ ⋅late gain ⎜⎝ ⎟⎟ EBW ⎠ ⎜⎝ Factor _ 51,78 ⎠⎠ ⎝

EBW ⋅ factor _ 6 ⋅ ADG BW

[Eq. 9.9]

[Eq. 9.10]

9.10

whereEBWG EBWGisisthe thedaily daily empty body weight gain, the empty body weight, Equation where empty body weight gain, g/d; g/d; EBWEBW is theisempty body weight, 9.12; BW is the weight of the animal, kg. However, BW for primiparous cows is corrected with the Equation 9.12; BW is the weight of the animal, kg. However, BW for primiparous cows is followingwith termthe600/BW_mat; is dependent onisthe type of on growing (Table 9.5); and corrected following termfactor_6 ·600/BW_mat; factor_6 dependent the typecattle of growing ADG(Table is the average live weight, g/d.of live weight, g/d. cattle 9.5); anddaily ADGgain is theofaverage daily gain FFM = EBW − Fat _ mass

[Eq. 9.11]

where FFM is the fat free mass of the EBW, kg; EBW is the empty body weight, Equation 9.12; Fat_mass the fat content in the EBW,system Equation 9.13. and NorFor – TheisNordic feed evaluation EBW = e (factor _ 7 + factor _ 6⋅ln(BW ))

[Eq. 9.12]

9.11 89

2894 2894 2895 2895 2896 2896 2897 2897 2898 2898 2899 2899 2900 2900 2901 2901 2902 2902 2903 2903 2904 2904 2905 2905 2906 2906 2907 2907 2908 2908 2909 2909 2910 2910 2911 2911 2915 2915 2912 2912 2916 2916 2913 2915 2917 2913 2917 2916 2914 2918 2918 2914 2917 2919 2919 2918 2920 2920 2919 2921 2921 2920 2922 2922 2921 2922 2923 2923 2924 2924 2923 2925 2925 2924 2926 2926 2925 2927 2927 2926 2928 2928 2927 2929 2929 2928 2930 2930 2929 2931 2931 2930 2932 2932 2931 2933 2933 2932 2934 2934 2933 2935 2935 2934 2936 2936 2935 2937 2937 2936 2938 2938 2937 2939 2939 2938 2940 2940 2939 2941 2941 2940 2942 2942 2941 2943 2943 2942 2944 2944 2943 2945 2945 2944 2946 2946 2945 2947 2947 2946 2948 2948 2947 2949 2949 2948 2950 2950 2949 2951 2951 2950 2952 2952 2951 2953 2953 2952 2954 2954 2953 2955 2955 2954 2956 2956 2955 2957 2957 2956 2958 2958 2957 2959 2959 2958 2959

corrected corrected with with the the following following term term ·600/BW_mat; ·600/BW_mat; factor_6 factor_6 is is dependent dependent on on the the type type of of growing growing cattle cattle (Table (Table 9.5); 9.5); and and ADG ADG is is the the average average daily daily gain gain of of live live weight, weight, g/d. g/d. [Eq. [Eq. 9.11] 9.11]

FFM = EBW − Fat _ mass FFM = EBW − Fat _ mass

9.12; where kg; EBW is empty body weight, Equation whereFFM FFMis thefat fatfree freemass massof theEBW, EBW, EBW is the empty body weight, Equation where FFM isisthe the fat free mass ofofthe the EBW, kg;kg; EBW is the the empty body weight, Equation 9.12;9.12; and and Fat_mass is content the EBW, Equation 9.13 Fat_mass is the fatfat content inin the Fat_mass is the the fat content in theEBW, EBW,Equation Equation9.13. 9.13.. and _ 7 + factor _ 6⋅ ln(BW )) EBW = e ((factor EBW = e factor _ 7 + factor _ 6⋅ln(BW ))

[Eq. [Eq. 9.12] 9.12]

9.12

where body weight, kg; and are on of whereEBW EBWis theempty empty body weight, kg; factor_6 7 are dependent on the type of growing where EBW isisthe the empty body weight, kg; factor_6 factor_6 and 77and are dependent dependent on the the type type of growing growing cattle (Table 9.5); and BW is is thethe weight of the the animal, kg. However, However, BW for for primiparous cows is iscows is cattle is the weight of animal, kg. BW primiparous cows cattle(Table (Table9.5); 9.5);and andBW BW weight of the animal, kg. However, BW for primiparous corrected term ·600/BW_mat. BW_mat is BW 3.5). corrected with the following term ·600/BW_mat. BW_mat is the the maturity BW (Table (Table 3.5). 3.5). correctedwith withthe thefollowing following term 600/BW_mat. BW_mat is maturity the maturity BW (Table 2 + factor_ 3⋅ ln( EBW ) + factor _ 4 ⋅ (ln( EBW ) 22 )) Fat _ mass = ee (factor_ Fat _ mass = (factor_ 2 + factor_ 3⋅ ln( EBW ) + factor _ 4 ⋅ (ln( EBW ) )) 2

[Eq. [Eq. 9.13] 9.13]

9.13

where content in EBW, kg; EBW is body weight, Equation whereFat_mass Fat_massis thefat content in the EBW, EBW is empty the empty weight, Equation 9.12; where Fat_mass isisthe the fatfat content in the the EBW, kg; kg; EBW is the the empty bodybody weight, Equation 9.12; factor_2 to factor_4 are dependent on the type of growing cattle (see Table 9.5). 9.12; factor_2 to factor_4 are dependent on the type of growing cattle (see Table 9.5). factor_2 to factor_4 are dependent on the type of growing cattle (see Table 9.5). where Protein_mass is the protein content in the EBW, kg; factor_1 is dependent on the type where Protein_mass is the protein content in the EBW, kg; factor_1 is dependent on the type of of 9.11.. growing cattle cattle (see (see Table Table 9.5); 9.5);1.and and FFM is the fat free mass of the EBW, Equation 9.11 growing 06 FFM is the fat free mass of the EBW, Equation 1 . 06 [Eq. 9.14] Pr otein _ mass = factor _ 1 ⋅ FFM where protein content on the type of in the EBW, kg; factor_1 is dependent [Eq. 9.14] 9.14 Pr oteinProtein_mass _ mass = factoris_the 1 ⋅ FFM growing Table 9.3 9.3cattle (see Table 9.5); and FFM is the fat free mass of the EBW, Equation 9.11. Table

where Protein_mass is the protein content in the EBW, kg; factor_1 is dependent on the type of

Table 9.3 9.4 cattle (see Table 9.5); and FFM is the fat free mass of the EBW, Equation 9.11. Table 9.4 growing Table 9.5 Table 9.4 9.5

111 111

9.1.5 Mobilization and deposition Table 9.5 9.1.5 Mobilization Mobilization and deposition deposition 9.1.5 and NorFor considers mobilization deposition BCSsection3.1) (see Section 3.1)primifor both NorFor considers considers mobilization and and deposition usingusing BCS (see (see for both both and primi- and NorFor mobilization and deposition using BCS section3.1) for primi- and 9.1.5 Mobilization and deposition multiparous cows.The The amount of NE deposited when aiscow is fed its above its requirement energy requirement multiparous cows. amount of NE deposited when a cow fed above energy multiparous cows. The amount of NE deposited when a cow is fed above its energy requirement andhas hasconsiders positive energy balance (EB>100%) is BCS calculated as: NorFor mobilization and(EB>100%) deposition using (see for both primi- and and has positive energy balance (EB>100%) is calculated calculated as:section3.1) and aaapositive energy balance is as: multiparous cows. The amount of NE deposited when a cow is fed above its energy requirement and a positive balance (EB>100%) NELhas _ dep dep = NEL NEL − −energy NE __ ma ma int − − NEL NEL gain − − NE NEis__calculated gest − − NEL NELas: _ milk milk [Eq. NEL _ = NE int __ gain gest _ [Eq. 9.15] 9.15]

9.15

NELamount _ dep = of NELNE − NE _ ma intwhen − NEL gain − NE gestEB90%) values. The ratio of 500 ratio of were omitted from the dataset due to unrealistically high AATN_Eff (>90%) values. The500 AATN/NEG for growth was calculated from the amount of AATN available for growth and the AATN/NEG for growth was calculated from the amount of AATN available for growth and the NEG for growth. The relationships AAT efficiency and AAT _NEG, BW 500 requirement 181 between NEG requirement for106 growth. The relationships between AATNN efficiency and AATNN_NEG, BW and ADG are shown in Figures 9.7, 9.8 and 9.9. and Figures 9.7, 9.8 and 9.9. 600ADG are shown in94 170 113 197

700 9.11 Table Table 9.11 Figure 9.7 Figure 9.7

98

Figure 9.8 Figure 9.8

86

158

101

183

NorFor – The Nordic feed evaluation system

3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233

ADG ⎞ ⎛ ⎜1.88 − 0.03821⋅ AATN _ NEG − 0.00176 ⋅ BW − 0.2283 ⋅ ⎟ 1000 ⎟ ⋅100 AATN _ Eff = ⎜ [Eq. 9.34] 1 Table 9.11. Descriptive statistics of +the dataset ⋅used for developing the equation for predicting ⎜ + 0.0000019014 ⎟ 1.83 5 ⋅ BW 0 . 000000016 AAT _ NEG N ⎝ ⎠

the AATN efficiency for growth in growing cattle. The dataset included trials with Norwegian and Swedish bulls involving 120 treatments.

where AATN_Eff is the efficiency with which AATN is used for growth, %; BW is the weight of the animal, kg; ADG is the averageAverage daily live weightMinimum gain, g/d and AAT N_NEG is the ratio Maximum SD between AATN and energy available for growth, Equation 9.35.

DMI (kg/d) 6.4 2.8 9.9 1.8 BW (kg) 322 113 568 120 ADG (g/d) AAT − AAT _ ma int 1,144 1,567 196 − AATN _ gest 548 N N = AAT [Eq. 9.35] N _ NEG Protein retention (g/d) NEG_ gain173 114 226 26 PBVN (g/kg DM) 22 -29 77 22 AAT /NEG (g/MJ) 16.9 9.7 29.8 is the whereNAATN_NEG is the ratio between AATN and energy available for growth, g/MJ; AATN4.3 AATNNsupplied efficiency (%) 50.7 8.11; AAT 1.4 99.4 23.0 AAT from the feed ration, Equation N_maint is the AATN required for The ratio between MP and energy available for growth is calculated as:

maintenance, Equation 9.22; AATN_gest is the AATN required for gestation in heifers, Equation 1 (Homb, 1974; Hole, 1985; Olsson and Lindberg, 1985; Olsson, 1987; Johansen, 1992; Skaara, 1994; Randby, 2000a; 9.42; and NEG_gain is the amount of energy used for growth, Equation 9.5.

Randby, 2000b; Berg and Volden, 2004; Volden and Berg, 2005; Berg, unpublished data; Havrevoll, unpublished

The of AATN data). utilization for growth was calculated by dividing the amount of protein data;efficiency Olsson, unpublished retained by the amount of AATN available for growth. The amount of protein retained was estimated using Equation 9.8 (see section 9.1.4). The AATN available for growth was calculated as the totalAAT AATN_NEG supply is minus the AAT N required for maintenance and gestation. Some observations where the ratio between AAT and energy available for growth, g/MJ; AATN is were omittedNfrom the dataset due to unrealisticallyNhigh AATN_Eff (>90%) values. The ratio of the AATN supplied from the feed ration, Equation 8.11; AATN_maint is the AATN required for AATN/NEG for growth was calculated from the amount of AATN available for growth and the maintenance, Equation 9.22; AAT the AATAAT for gestation in heifers, Equation N_gest is between N required NEG requirement for growth. The relationships N efficiency and AATN_NEG, BW 9.42; andare NEG_gain the amount energy and ADG shown in is Figures 9.7, 9.8of and 9.9. used for growth, Equation 9.5.

The efficiency of AATN utilization for growth was calculated by dividing the amount of protein Table 9.11

retained by the amount of AATN available for growth. The amount of protein retained was estimated

Figure 9.7 using Equation 9.8 (see Section 9.1.4). The AATN available for growth was calculated as the total

AATN supply minus the AATN required for maintenance and gestation. Some observations were

Figure omitted9.8 from the dataset due to unrealistically high AATN_Eff (>90%) values. The ratio of AATN/NEG

for growth was calculated from the amount of AATN available for growth and the NEG requirement for growth. The relationships between AATN efficiency and AATN_NEG, BW and ADG are shown in Figures 9.7, 9.8 and 9.9. Based on the AAT supply from the feed ration and the AAT _Eff, the amount of AAT retained Figure 9.9

N

N

for growth can be calculated as:

N

Based on the AATN supply from the feed ration and the AATN_Eff, the amount of AATN retained for growth can be calculated as: ADG ⎞ ⎛ ⎜1.88 − 0.03821 ⋅ AATN _ NEG − 0.00176 ⋅ BW − 0.2283 ⋅ ⎟ 1000 ⎟ AATNretained = AATN ⋅ ⎜ ⎜ + 0.0000019014 ⋅ BW1.83 + 0.000000016 ⋅ AAT _ NEG 5 ⎟ N ⎝ ⎠

9.36

[Eq. 9.36] where AATNretained is the AATN retained for growth, g/d; AATN is AATN supplied from the feed ration, Equation 8.11; AATN_NEG is the ratio between AATN and energy available for growth; Equation 9.35; BW is the weight of the animal, kg; and ADG is the average daily live weight gain, g/d. 119

The maximum retention of AATN in growth can be calculated by putting the derivative of this equation equal to 0, which results in: AAT N _ NEG _ Min =

(1.88 − 0.00176 ⋅ BW − 0.2283 ⋅ ADG / 1000 + 0.0000019014 ⋅ BW 1.83 ) 0 .03821 ⋅ 2 − 0 .000000016 ⋅ 5 ⋅ AAT N _ NEG 4

9.37

However, the above equation is not suited for calculating the AATN requirement for growing cattle because it is not possible to calculate the minimum AATN requirement (AATN_NEG _Min) using AATN_NEG as an input, since AATN_NEG is a ration variable. Therefore, AATN_NEG was replaced NorFor – The Nordic feed evaluation system

99

AATN efficiency (%)

120 100 80 60 40 20 0

8

10

12

14

16

18

20

22

24

26

28

30

AATN (g/MJ NEG)

Figure 9.7. AATN efficiency (%) in growing cattle in relation to AATN supply. AATN supply is defined as the AATN available for growth per MJ NEG available for growth. A multiple regression with AATN supply, BW (see Figure 9.8) and ADG (see Figure 9.9) as significant explanatory variables explained the variation in AATN efficiency with high accuracy (r2=0.97 and RMSE=2.9). The data are from the trials described in Table 9.11.

AATN efficiency (%)

120 100 80

3234 3235 3236 3237 3238

60the AATN retained for growth, g/d; AATN is AATN supplied from the feed where AATNretained is ration, Equation 8.11; AATN_NEG is the ratio between AATN and energy available for growth; Equation 9.35; BW40 is the weight of the animal, kg; and ADG is the average daily live weight gain, g/d. 20

3239 3240

The maximum retention 0 of AATN in growth can be calculated by putting the derivative of this equation equal to 0, which results in: 50 100 150 200 250 300 350 400 450 500 550 600

3241

AAT N _ NEG _ Min =

3242

(1.88 − 0.00176 ⋅ BW − 0.2283 ⋅ ADG 1000 + 0 .0000019014 ⋅ BW 1 .83 ) BW /(kg) 0 .03821 ⋅ 2 − 0 .000000016 * 5 * AAT N _ NEG 4

Figure 9.8. AATN efficiency (%) in growing cattle in relation to BW based on the trials described in [Eq. 9.37] Table 9.11. A multiple regression with BW, AATN supply (see Figure 9.7) and ADG (see Figure 9.9) However, the above equation is not suited for calculating the AAT requirement for growing cattle N AAT efficiency as significant explanatory variables explained the variation in with high accuracy N because it is not possible to calculate the minimum AAT N requirement (AATN_NEG _Min) using 2 (r =0.97 and RMSE=2.9).

3243 3244 3245 3246 3247 3248 3249 3250

AATN_NEG as an input, since AATN_NEG is a ration variable. Therefore, AATN_NEG was replaced with a value of 14, which corresponds to the requirement of an animal with a BW of 300 with value corresponds to theoptimum requirement ofthe an economic animal with a BWthe of 300 kg and kg anda an ADGofof14, 800which g/d. Since the biological is rarely optimum, an ADG of 800 g/d. Since the biological optimum is rarely the economic optimum, the biological biological optimum (=maximum retention of AATN) was lowered by 1 g AATN/NEG. Therefore, optimum (=maximum of AAT ) was lowered 1 g AAT /NEG. Therefore, the following the following equation isretention a modification of NEquation 9.37, inby which 1 is subtracted and 14 replaces N equation is a modification of Equation 9.37, in which 1 is subtracted and 14 replaces AATN/NEG: AAT N/NEG:

3251

AAT N _ NEG _ Min =

3252

[Eq.is9.38] where AATN_NEG_Min is the minimum AATN requirement (g/MJ NEG); BW the weight of the animal, kg; and ADG is the average daily live weight gain, g/d.

3253 3254 3255 3256 3257 3258

(1.88 − 0.00176 ⋅ BW − 0.2283 ⋅ ADG / 1000 + 0.0000019014 ⋅ BW 1.83 ) − 1 0 .03821 ⋅ 2 − 0 .000000016 ⋅ 5 ⋅ 14 4

9.38

where AATN_NEG_Min is the minimum AATN requirement (g/MJ NEG); BW is the weight of the animal, kg; and ADG is the average daily live weight gain, g/d.

100 NorFor – The Nordic feed evaluation system In contrast to the AATN balance for dairy cows, the AAT N balance for growing cattle is calculated more simply: AATN _ NEG

3234 3235 3236 3237 3238

where AATNretained is the AATN retained for growth, g/d; AATN is AATN supplied from the feed ration, Equation 8.11; AATN_NEG is the ratio between AATN and energy available for growth; Equation 9.35; BW is the weight of the animal, kg; and ADG is the average daily live weight gain, g/d.

120 The maximum retention of AATN in growth can be calculated by putting the derivative of this equation equal to 0,100 which results in:

3241

AAT N _ NEG _ Min =

3242 3243 3244 3245 3246 3247 3248 3249 3250

[Eq. 9.37] 60 However, the above equation is not suited for calculating the AATN requirement for growing cattle because it is not possible 40 to calculate the minimum AATN requirement (AATN_NEG _Min) using AATN_NEG as an input, since AATN_NEG is a ration variable. Therefore, AATN_NEG was replaced with a value20of 14, which corresponds to the requirement of an animal with a BW of 300 kg and an ADG of 800 g/d. Since the biological optimum is rarely the economic optimum, the biological optimum (=maximum retention of AATN) was lowered by 1 g AATN/NEG. Therefore, 0 the following equation is a modification of Equation 9.37, in which 1 is subtracted and 14 replaces 500 600 700 800 900 1000 1,100 1,200 1,300 1,400 1,500 1,600 1,700 1,800 AATN/NEG: ADG (g/d)

3251

AAT N _ NEG _ Min =

3252 3253 3254 3255 3256

AATN efficiency (%)

3239 3240

(1.88 − 0.00176 ⋅ BW − 0.2283 ⋅ ADG / 1000 + 0.0000019014 ⋅ BW 1.83 )

80

0 .03821 ⋅ 2 − 0 .000000016 * 5 * AAT N _ NEG 4

(1.88 − 0.00176 ⋅ BW − 0.2283 ⋅ ADG / 1000 + 0.0000019014 ⋅ BW 1.83 ) − 1 4

0 .03821cattle ⋅ 2 − 0 .in 000000016 5 ⋅ 14 Figure 9.9. AATN efficiency (%) in growing relation⋅to ADG based on the trials described in [Eq. 9.38] Table 9.11. A multiple regression with ADG, AATN supply (see Figure 9.7) and BW (see Figure 9.8) as significant explanatory variables explained the variation in AATN efficiency with high accuracy (r2=0.97 and RMSE=2.9). where AAT N_NEG_Min is the minimum AATN requirement (g/MJ NEG); BW is the weight of

the animal, kg; and ADG is the average daily live weight gain, g/d.

3257 3258

In contrast contrasttotothe theAAT AAT dairy cows, the AAT for growing is calculated N balance N balance for for dairy cows, the AAT for growing cattle iscattle calculated In N balance N balance more simply: more simply:

3259

AATN _ bal =

3260 3261 3262 3263

N N N N where AATfor N_bal is the AATN balance, %; AATN_NEG is the ratio between AATN and energy available growth, Equation 9.35; and AATN_NEG_Min is the minimum AATN requirement, available for growth, Equation 9.35; and AATN_NEG_Min is the minimum AATN requirement, Equation 9.38. Equation 9.38.

AATN _ NEG ⋅100 AATN _ NEG _ Min

[Eq. 9.39]

9.39

where AAT _bal is the AAT balance, %; AAT _NEG is the ratio between AAT and energy

9.2.4 Mobilization Mobilizationand anddeposition deposition 9.2.4

3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275

In dairy cows, mobilized or deposited body protein will affect the amount of AATN available for In dairy cows, mobilizedInorNorFor deposited bodyN protein will affect amount of AATN available for milk mobilization (AATthe milk protein production. the AAT N_mob) or AATN deposition protein production. In to NorFor the AAT (AAT AAT deposition (AATN_dep) (AAT is related the animal's energy balance. This means that or when a Ncow is in negative N_dep) N mobilization N_mob) is related to theprotein animal’s balance.It This meansthat thatbody when a cowreserves is in negative energy balance willenergy be mobilized. is assumed energy containenergy 135 g balance protein/kg (NRC, 1985) and,Itbased on values from INRA (1989), 24.8 MJ energy is supplied per (NRC, protein will be mobilized. is assumed that body energy reserves contain 135 g protein/kg kg mobilized body on mass, and the energy requirement for MJ deposition MJ/kg. The 1985) and, based values from INRA (1989), 24.8 energyisis31.0 supplied per kgutilization mobilized body i.e. The 80%utilization (Tammingaofetmobilized al., 1994),protein of mobilized milk production is assumedisto31.0 be high, mass, and theprotein energyforrequirement for deposition MJ/kg. whereas utilizationisof AATN for deposition in80% adult(Tamminga animals has et been shown towhereas be low the utilization for milkthe production assumed to be high, i.e. al., 1994), (INRA, NRC, 2001); it is assumed to be 120 50%. The amount of AATN originating from of AAT1989; N for deposition in adult animals has been shown to be low (INRA, 1989; NRC, 2001); it mobilized protein is calculated as: is assumed to be 50%. The amount of AAT originating from mobilized protein is calculated as:

3276

AATN _ mob =

3277 3278 3279 3280 3281 3282 3283

from body mass, g/d; NEL_mob is the where N mobilized whereAAT AATNN_mob _mobisisthe theamount amountofofAAT AAT N mobilized from body mass, g/d; NEL_mob is the amount amount of energy mobilized from body Equation 9.18; is the AATcontent perkg kgbody mass; N contentper of energy mobilized from body mass,mass, Equation 9.18; 135135 is the AAT N body mass; 24.8 is the energyper supplied per kg mobilized body MJ NEL/kg; and 0.8torefers 24.8 is the energy supplied kg mobilized body mass, MJmass, NEL/kg; and 0.8 refers the efficiency to the efficiency of mobilized AATN for milk production.

3284

AATN _ dep =

3285 3286 3287 3288

N

135 ⋅ 0.8 ⋅ NEL _ mob 24.8

[Eq. 9.40]

9.40

of mobilized AATN for milk production.

When cows are in positive energy balance the AATN requirement for protein deposition is When cows calculated as:are in positive energy balance the AATN requirement for protein deposition is calculated

as:

135 ⋅ 2 ⋅ NEL _ dep 31.0

[Eq. 9.41]

is the feed amount of AATN required where AAT N_dep NorFor – The Nordic evaluation systemfor protein deposition, g/d; NEL_dep is the amount of energy deposited, Equation 9.17; 135 is the AATN content per kg body mass; 31.0 is the energy content per kg body mass, MJ NEL/kg; and two refers to an efficiency of 50% for AATN deposition.

101

3280 3281 3282 3283

When cows are in positive energy balance the AATN requirement for protein deposition is calculated as:

3284

AATN _ dep =

3285 3286 3287 3288 3289 3290

where g/d; NEL_dep is the whereAAT AATNN_dep _depisisthe theamount amountofofAAT AATNNrequired requiredfor forprotein proteindeposition, deposition, g/d; NEL_dep is the amount amount of energy deposited, Equation the AAT per kg body mass; 31.0isisthe energy N content of energy deposited, Equation 9.17; 9.17; 135 is135 theisAAT content per kg body mass; 31.0 N refers to an efficiency of 50% for the energy content per kg body mass, MJ NEL/kg; and two content per kg body mass, MJ NEL/kg; and 2 refers to an efficiency of 50% for AATN deposition. AATN deposition.

3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310

to the efficiency of mobilized AATN for milk production.

135 ⋅ 2 ⋅ NEL _ dep 31.0

[Eq. 9.41]

9.41

The AATN requirements for deposition and mobilization are illustrated in Table 9.12.

The AATN requirements for deposition and mobilization are illustrated in Table 9.12.

9.2.5 Gestation

Table 9.12

The AATN requirement for gestation in heifers and cows depends on the stage of pregnancy and includes the AATN required for the development and maintenance of all conceptus tissues, including 9.2.5 Gestation the foetus. The AATN requirement equation is modified from on NRC uses mature The AATN requirement for gestation in heifers and cows depends the(1985) stage ofand pregnancy and BW as a scaling factor instead of birth weight. includes the AATN required for the development and maintenance of all conceptus tissues,

including the foetus. The AATN requirement equation is modified from NRC (1985) and uses -0.00262 ∙ gest_day 8.5357 – (13.1201 ) - 0.00262 ∙ gest_day mature BW asBW_mat a scaling factor instead birth weight.∙ e ∙ 34.375 ∙ eof 600 −0.00262⋅ gest _ day AATN_gest = BW _ mat 8 . 5357 − ( 13 . 1201 ⋅ e ) − 0 .00262⋅gest _ day ⋅ 34.375 ⋅ e 0.5 600 [Eq. 9.42] AATN _ gest =

9.42

5 gestation, g/d; BW_mat is the mature BW, kg (Table where AATN_gest is the daily AAT required0.for 3.2 and 3.5); and gest_day is the actual day of gestation; 0.5 is the assumed AATN utilization for where AATN_gest is the daily AAT required for gestation, g/d; BW_mat is the mature BW, kg gestation. (Table 3.2 and 3.5); and gest_day is the actual day of gestation; 0.5 is the assumed AAT N

utilization for gestation.

The AATN requirement for gestation is illustrated in Figure 9.10.

The AATN requirement for gestation is illustrated in Figure 9.10.

9.3 PBVN

Figure 9.10

The minimum PBVN recommendation for dairy cows is related to milk yield according to the following equation, which is illustrated in Figure 9.11.

9.3 PBVN

-0.14 ∙ ECM) for dairy cows is related to milk yield according to the The PBVN recommendation PBVminimum N_DM_Min = 10 ∙ (1 – e following equation, which is illustrated in Figure 9.11.

9.43

where PBVN_Min is the minimum recommended level of PBVN, g/kg DM; and ECM is the daily 121 energy-corrected milk yield, Equation 3.1 and 3.2. Table 9.12. Weight change and AATN requirement for deposition and mobilization in cows depending on breed and change in BCS. Daily change in BCS1

Weight change, kg/d

AATN requirement, g/d

Large breed

Jersey

Large breed

Jersey

-0.02 -0.01 0 0.01 0.02

-1.2 -0.6 0 0.6 1.2

-0.9 -0.45 0 0.45 0.9

-154 -77 0 192 384

-115 -58 0 144 288

1

One BCS=60 kg for large breeds and 45 kg for Jersey, i.e. 0.01 BCS corresponds to 600 g BW for large breeds and 450 g BW for Jersey.

102

NorFor – The Nordic feed evaluation system

AATN requirement (g/d)

350

DH Jersey

300 250 200 150 100 50 0

0

50

100

150

200

250

300

Days in gestation

Figure 9.10. Estimated AATN requirement (g/d) for pregnancy depending on days of gestation and mature BW. Jersey has a mature BW of 440 kg; Holstein has a mature BW of 640 kg. Figure 9.11 shows that for cows producing more than 20 kg ECM per day, the minimum PBVN recommendation is 10 g/kg DM, while for cows producing less than 20 kg the minimum PBVN is reduced to zero when the milk production is 0, i.e. for dry cows. If recirculation of urea to the rumen is taken into account, then the minimum PBVN should theoretically be 0, implying that the N available for microbial growth is similar to the N required for microbial growth. The NorFor PBVN recommendation for lactating dairy cows is based on a former Danish recommendation, which was set to 0 g/d (Strudsholm et al., 1999) on the basis of several production trials (Kristensen, 1997; Madsen et al., 2003), but raised to 10 g/kg, for two reasons. Firstly, unlike the former Danish system, NorFor takes into account the recirculation of urea to the rumen in the calculation of PBVN (see Equation 7.35), and secondly the feed passage rate is lower than in the former AAT/PBV system (Madsen et al., 1985), which will result in a higher ruminal degradation of protein. A maximum of 40 g PBVN per kg DM is recommended for both lactating and dry cows

PBVN (g/kg DM)

12 10 8 6 4 2 0

0

10

20

30 40 ECM (kg/d)

50

60

Figure 9.11. PBVN recommendation for dairy cows as a function of ECM yield. NorFor – The Nordic feed evaluation system

103

to avoid increased energy expenditure for the excretion of N via urine (Danfær, 1983; Reynolds, 2000) and decreased fertility (Rajala-Schultz, 2001). For growing cattle the minimum PBVN recommendation is 0 g/kg DM. This recommendation is not based on production trials but on a theoretical approach, i.e. that there is a balance between CP entering the rumen and CP leaving the rumen. The maximum PBVN recommendation is set to 55 g/kg DM based on Olsson (1987). Olsson (1987) conducted several trials with intensively reared bulls and concluded a maximum recommendation of 4 g PBV/MJ ME. Assuming a 60% ME to NEG conversion efficiency, and an average content of 7.5 MJ NEG/kg DM, the maximum recommendation is 50 g PBVN/kg DM (4/0.6=6.7 g PBVN/MJ NEG·7.5). Since the energy content of a ration may be higher than 7.5 MJ NEG/kg DM, it was decided that NorFor has a maximum recommendation of 55 g PBVN per kg DM.

9.4 Rumen load index The maximum recommended RLI (see Chapter 7 and Equation 7.15 for calculation) is set to 0.6, which gives a moderate reduction of NDF degradation rate. A RLI of 0.6 corresponds to a diet with 240-280 g starch per kg DM depending on starch source. The sugar and starch content of a diet will often be 290-320 g/kg DM. These maximum levels of starch and sugar are in accordance with previous Danish recommendations (Strudsholm et al., 1999). The NRC (2001) recommends diets with a non-fibre carbohydrates content ranging from 360-440 g/kg DM, depending on the NDF content of the ration; 360 g/kg DM for diets with a low NDF content (250 g/kg DM), and 440 g/kg DM for diets with a high NDF content (330 g/kg DM). No health problems were reported in Danish trials with dairy cows fed diets containing 330-340 g starch/kg DM using maize silage and barley as starch sources (Hymøller et al., 2005).

9.5 Chewing time Chewing time index (CI) is used for cows and growing cattle in NorFor to ensure good rumen function, see further descriptions in Chapter 11. The minimum recommendation for CI is 32 minutes per kg DM for large breeds and 30 minutes per kg DM for Jersey cows. These recommendations are based on assessments in relation to the former Danish chewing index (Nørgaard, 1986). However, in the future the recommendations will be validated against data where e.g. rumen pH and milk fat content has been measured.

9.6 Intake capacity Calculations of the ration FV and animal IC are described in Chapter 10. The total FV of the ration should not exceed the IC of the animal. When optimising the feed ration FV should be as close to IC as possible in order to ensure the animals satiety. However for practical reasons NorFor has different minimum recommendations for IC for lactating cows, dry cows and growing cattle.

9.7 Fatty acids The type of fat, i.e. chain length, degree of saturation and physical form, is of importance for the response in milk yield (Børsting et al., 2003). Too high fat content in dairy cattle diets will reduce feed intake and fibre degradation in the rumen (NRC, 2001). The NRC (2001) recommends that the CFat content in a ration should not exceed 6-7% of DM, and the Danish maximum recommendation for growing cattle is 50-60 g/kg DM (Strudsholm et al., 1999). A relationship was established between ECM yield and FA/kg DM, based on Danish trials with fat supplementation suggesting that the maximal ECM yield is obtained with 46 g FA/kg DM for fat with a degree of saturation of approximately 50 (Børsting et al., 2003). If the fat supplementation consists of FAs with a high 104

NorFor – The Nordic feed evaluation system

3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381

The type of fat, i.e. chain length, degree of saturation and physical form, is of importance for the response in milk yield (Børsting et al., 2003). Too high fat content in dairy cattle diets will reduce feed intake and fibre degradation in the rumen (NRC, 2001). The NRC (2001) recommends that the CFat content in a ration should not exceed 6-7% of DM, and the Danish maximum recommendation for growing cattle is 50-60 g/kg DM CFat of DMI (Strudsholm et al., degreeAofrelationship saturation was thenestablished the maximum ECM yield can beFA/kg achieved 75 gonFA/kg DM (Børsting 1999). between ECM yield and DM,atbased Danish trials with supplementation suggesting that the ECM yield is obtained g FA/kg et al.,fat2003). NorFor has implemented a maximal maximum recommendation ofwith 45 g46FA/kg DM for both DM fat growing with a degree of saturation of approximately 50 content (Børstingofet64 al.,g/kg 2003). If the fat cowsforand cattle. This corresponds to a CFat DM, assuming a 50/50 supplementation consists of FAs with degree ofinsaturation the and maximum ECM ratio of roughage to concentrate, anda high FA contents the CFat then of 650 750 g/kg in yield roughage and can be achievedrespectively. at 75 g FA/kg DM (Børsting et al., 2003). NorFor has implemented a maximum concentrates, recommendation of 45 g FA/kg DM for both cows and growing cattle. This corresponds to a CFat content of 64 g/kg DM, assuming a 50/50 ratio of roughage to concentrate, and FA contents 9.8 Minerals in the CFat of 650 and 750 g/kg in roughage and concentrates, respectively.

3382 3383 3384 3385 3386 3387 3388

Mineral requirements are generally based on a factorial approach in which the requirements for 9.8 Minerals

3389 3390 3391 3392 3394 3394 3394 3393 3394 3394 3394 3394 3395 3394 3395 3394 3395 3395 3395 3395 3395 3395 3396 3395 3396 3396 3396 3396 3396 3396 3396 3396 3397 3397 3397 3397 3397 3397 3397 3397 3397 3398 3398 3398 3398 3398 3398 3398 3399 3398 3399 3398 3399 3399 3399 3399 3399 3400 3399 3400 3399 3400 3400 3400 3400 3400 3401 3400 3401 3400 3401 3401 3401 3401 3401 3401 3401 3402 3402 3402 3402 3402 3402 3402 3402 3402 3403 3403 3403 3403 3403 3403 3403 3404 3403 3404 3403 3404 3404 3404 3404 3404 3405 3405 3404 3405 3404 3405 3405 3405 3405 3406 3406 3405 3406 3405 3406 3406 3406 3406 3407 3406 3407 3406 3407 3407 3407 3407 3407 3408 3408 3407 3408 3407 3408 3408 3408 3408 3409

9.8.1 Macro minerals Recommendations for calcium, phosphorus, magnesium, sodium, potassium and chlorine for dairy Recommendations forcattle calcium, sodium, potassium and chlorine for dairy cows and growing arephosphorus, calculatedmagnesium, using the following equations. Note that the absorption cows and growing calculated using the following equations. the absorption coefficients rangecattle fromare16% for magnesium to 90% for sodiumNote andthat potassium:

maintenance, milk production, growth and gestation are considered. The requirements also include a

Mineral requirements are generally based on a factorial approach in which the requirements for safety margin, although this margin notgestation quantified, based onThe therequirements lowest absorption maintenance, milk production, growthisand are but considered. also coefficients from the literature. Furthermore, these low absorption coefficients also apply for mineral supplements, include a safety margin, although this margin is not quantified, but based on the lowest although they have higher true absorption coefficients (NRC, 2001). absorption coefficients from the literature. Furthermore, these low absorption coefficients also apply for mineral supplements, although they have higher true absorption coefficients (NRC, 9.8.1 Macro minerals 2001).

coefficients range_from Ca_maint = factor 1 ⋅ BW16% 0.50for magnesium to 90% for sodium and potassium: [Eq. 9.45] Ca_maint = factor _1 ⋅ BW 00..50 [Eq. 9.45] Ca_maint = factor 1 [Eq. 9.45] Ca_intake_ _ ma0int 9.44] Ca_maint = Min factor=_ _Ca 1 ⋅⋅ BW BW .50 50+ Ca _ milk + Ca _ gain + Ca _ gest [Eq. 9.45] Ca_maint = factor _ 1 ⋅ BW 0.50 [Eq. 9.45] Ca_maint = factor _ 1 ⋅ BW 0 . 50 [Eq. Ca_maint = factor _ 1 ⋅ BW 0.50 [Eq. 9.45] 9.45] Ca _ milk= =factor 1 . 22_⋅1ECM 0 . 50 [Eq. 9.46] Ca_maint ⋅⋅ BW 9.45] Ca _ milk 1 .. 22 50 [Eq. 9.46] Ca_maint = =factor _⋅1ECM BW 000....50 50 9.45] Ca [Eq. 9.46] Ca _ _ milk milk = =1 1 . 22 22 ⋅⋅ ECM ECM 0 0 . 50 50 [Eq. 9.46] Ca _ milk = 1 . 22 ⋅ ECM 0 . 50 [Eq. 9.46] Ca .. 50 [Eq. 9.46] 9.46] Ca _ _ milk milk = =1 1 .. 22 22 ⋅⋅ ECM ECM 0 00.22 50 [Eq. − 0.22 ⋅ ADG 123 Ca = 1 ⋅⋅ ECM .. 50 [Eq. 9.46] Ca_gain ⋅ BW_mat 0.50 9.47] [Eq. − 0.22 Ca _ _ milk milk= =9.83 1 .. 22 22 ECM 0 00.22 50 ⋅⋅ BW ADG 9.46] 0.22 − 0.22 1000 ADG Ca_gain = 9.83 ⋅ BW_mat BW ⋅ 0.50 [Eq. 9.47] 0.22 ⋅⋅ BW − 0.22 ⋅⋅ ADG1000 0.50 Ca_gain = 9.83 ⋅⋅ BW_mat [Eq. 9.47] Ca_gain = 9.83 BW_mat BW 0.50 [Eq. 9.47] 0.22 − 0.22 1000 ADG Ca_gain = 9.83 ⋅ BW_mat ⋅ BW ⋅ 0.50 1000 [Eq. 9.47] 0.22 − 0.22 ADG 0.22 − 0.22 Ca_gain = = 9.83 9.83 ⋅⋅ BW_mat BW_mat BW ⋅ ADG1000 0.50 0.50 [Eq. 9.47] 9.47] Ca_gain ⋅⋅ 0BW [Eq. .00007 ⋅gest⋅⋅)ADG ⋅gest 1000 0.22 − 1000 0.50 ⎛ 9.83 ⎞ 0.22 − − 0.22 0.22 Ca_gain = ⋅⋅ BW_mat ⋅⋅ 00BW 0.02456 ⋅ e(((000...05581 [Eq. 9.47] 05581 − ..00007 ⋅⋅gest ))ADG ⋅⋅gest ⎛ ⎞ Ca_gain = 9.83 BW_mat BW ⋅ 0.50 05581 − 00007 gest gest [Eq. 9.47] ⎜ ⎟ 1000 ⎛ ⎞ 0.02456 e ⋅ Ca_gest = ⎜⎛⎜ 0.02456⋅ e(0.05581− 0.00007⋅gest)⋅gest 1000 .50 [Eq. 9.48] ⎟⎞⎟⎞⎟ 0 0.02456 e ⋅ ( 0 . 05581 − 0 . 00007 ⋅ gest ) ⋅ gest = Ca_gest 0 . 50 ( 0 . 05581 − 0 . 00007 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) ⎜ ⎛ [Eq. 9.48] Ca_gest 0 . 50 = [Eq. 9.48] 0.02456 e ⋅ ( 0 . 05581 − 0 . 00007 ⋅ gest ) ⋅ gest 0.02456 e − ⋅ ⎜ ⎟ ⎛ ⎞ Ca_gest − 0− .00007 ⋅gest )⋅gest 00..05581 00..00007 ⋅⋅((gest − 1)⋅(gest − 1) ⎠⎟⎞ 0..50 ⎝⎜⎛ −0.02456 [Eq. 9.48] ⋅⋅ ee⋅⋅ (ee0(((.05581 − − 0.02456 0.02456 Ca_gest = = [Eq. 0.05581 05581 −.00007 0.00007 00007 ⋅(gest gest −11))⋅⋅((gest gest − −11)) ⎟ 0 0.02456 − ⋅⋅gest ))⋅⋅gest Ca_gest = ⎝⎜⎜⎝⎜⎛⎝⎛ − 0..50 50 [Eq. 9.48] 9.48] 0.02456 0.05581 0.00007 ⋅(gest −1)⋅(gest −1) ⎠⎟⎠⎟⎞⎠⎟⎞ 0 Ca_gest = 50 05581 − 00− .00007 gest gest 0.02456 ⋅⋅ ee⋅⋅ ((ee00((..05581 [Eq. 9.48] ⎜⎜⎝ − 0.02456 − 0 . 05581 − 0 . 00007 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) 0.02456 ⎟ (0.05581− 0.00007⋅(gest −1)⋅(gest −1) ⎠⎟ 0.50 Ca_gest = ⋅⋅ eeint [Eq. 9.49] 9.48] 0.02456 −in0.02456 ⎠ 0.50 = ⎜⎝⎜⎝ − Ca_gest P_intake_M = P _ ma + P _ milk + P _ gain + P _ gest [Eq. 9.48] ⎟ ⎠ ( 0 . 05581 − 0 . 00007 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) ⎜⎝ −in (0.05581 0.00007 ⋅_ (gest −1+ )⋅(P gest −1) ⎟⎠ 0.02456 ⋅⋅ eeint P_intake_M = P _ ma + P __ −milk + P gain __ gest [Eq. 9.49] P_intake_M in = P _ ma int + P milk + P _ gain + P gest [Eq. 9.49] − 0.02456 P_intake_M in = P _ ma int + P _ milk + P _ gain + P _ gest ⎝ ⎠ [Eq. 9.49] P_intake_M in [Eq. P_intake_M in = =P P_ _ ma ma int int + +P P ___ milk milk + +P P ___ gain gain + +P P ___ gest gest [Eq. 9.49] 9.49] P_intake_M in = P _ ma int + P milk + P gain + P gest [Eq. 9.49] P_maint = 1.0in⋅ DMI 0ma .70int + P _ milk + P _ gain + P _ gest [Eq. P_intake_M = P _ [Eq. 9.50] 9.49] P_maint = 1 . 0 ⋅ DMI 0 . 70 P_intake_M in = P _ ma int + P _ milk + P _ gain + P _ gest [Eq. 9.50] 9.49] P_maint = = 11..00 ⋅⋅ DMI DMI 0 0..70 70 [Eq. 9.50] 9.50] P_maint [Eq. P_maint = 1 . 0 ⋅ DMI 0 . 70 [Eq. 9.50] 9.50] [Eq. P_maint = 1 . 0 ⋅ DMI 0 . 70 P_maint = 1.0 ⋅ DMI 0.70 [Eq. 9.50] P_milk==factor _ 2 ⋅ MY 0.70 [Eq. 9.50] 9.51] P_maint 1 . 0 ⋅ DMI 0 . 70 [Eq. P_milk = factor _ 2 ⋅ MY 0 . 70 P_maint = 1.0 ⋅ DMI 0.700.70 9.51] [Eq. 9.50] P_milk _ 2 ⋅ MY [Eq. P_milk= = factor factor MY 70 [Eq. 9.51] 9.51] P_milk = factor___ 222 ⋅⋅⋅ MY MY 000...70 70 [Eq. 9.51] P_milk = factor [Eq. P_milk= factor_ 2 ⋅ MY 0.70 0.22 − 0.22 ADG [Eq. 9.51] 9.51] P _ gain =factor 1.2 + ⋅ BW_mat ⋅ BW −− 0.22 ⋅ ADG1000 0.70 [Eq. 9.52] 0.22 P_milk = __ 4.635 22 ⋅⋅ MY 00..70 [Eq. 0.22 P _ gain 1.2 + 4.635 0.70 9.52] P_milk == factor MY⋅⋅ BW_mat 70 [Eq. 9.51] 9.51] 0.22 ⋅⋅⋅ BW − 0.22 0.22 ⋅⋅⋅ ADG P _ gain = 1.2 + 4.635 BW 0.70 9.52] ADG 1000 P _ gain = 1.2 + 4.635 ⋅⋅ BW_mat BW_mat BW [Eq. 9.52] 0.22 − 0.22 1000 0.70 ADG 0.22 − 0.22 P _ gain = 1.2 + 4.635 BW_mat ⋅ BW ⋅ 0.70 1000 [Eq. 9.52] ADG 0.22 − 0.22 [Eq. 9.52] ADG P _ gain = 1.2 + 4.635 ⋅ BW_mat ⋅ BW ⋅ 0.70 1000 0.70 P _ gain = 1.2 + 4.635(⋅0BW_mat ⋅ BW ⋅ [Eq. 9.52] 1000 0.22 − 0.22 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest 1000 ⎛ ⎞ ADG 0.22 − 0.22 P _ gain = 1.2 + 4.635 ⋅ BW_mat ⋅ BW ⋅ 0.70 0.02743 e ⋅ [Eq. 9.52] ADG ( 0 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest ⎛ ⎜ ⎟ ⎞ P _ gain==⎛ 1.2 + 4.635 ⋅(00BW_mat ⋅ BW ⋅ 0.70 [Eq. 9.52] ( . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest 1000 ⎞ ⋅ 0.02743 e P_gest 0 . 70 − 0.000075⋅gest)⋅gest ⋅⋅ ee (0..05527 1000⎟⎟⎞ ⎜⎜⎛⎜⎛ 0.02743 [Eq. 9.53] 0.02743 = 05527 − 0 . 000075 ⋅ gest ) ⋅ gest P_gest 0 . 70 ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) ⎟ ⎞ [Eq. 9.53] P_gest 0 . 70 = [Eq. 9.53] e ⋅ ( 0 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest ⎟ ⎛⎜⎛ −0.02743 ⎞ P_gest 0 . 70 = ⋅ 0.02743 e ( 0 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) ⎝ ⎠ [Eq. 9.53] ⎜ ⎟ ⎞ 0.70 0.02743 e ⋅ ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) − ⋅ 0.02743 e ⋅ P_gest = 0.02743 e ⎜ ⎟ [Eq. 9.53] ⎝⎝⎜⎜⎛ − ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) 0.02743 e ⋅ ⎠ ⎟ ( 0 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest P_gest 0 . 70 = ⎞ ⎠ ⎟ [Eq. 9.53] 0.02743 e ⋅ ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) P_gest = ⎝⎜⎛ − 0 . 70 ( 0 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest 0.02743 e ⋅ [Eq. 9.53] ⎞ ⎠ ⎟ 0.02743 e − ⋅ ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) 0.02743 e ⋅ ⎜ ⎟ ⎝ ⎠ ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) ⎜− ⎟ 0 P_gest ..70 = 0.02743 e ⋅ [Eq. 0.02743 e − ⋅ ⎝ ⎠ P_gest 0 70 = [Eq. 9.53] 9.53] ⎝⎜⎜ Min = Mg _ ma ⎠⎟⎟ Mg_intake_ int + Mg _ milk + Mg _ gain + Mg _ gest 9.54] ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) Mg_intake_ (0.int 05527 − 0_ ⋅Mg (gest −1)⋅(gest −1_ 0.02743 ⋅⋅__eema = Mg + Mg milk + __ gain + Mg [Eq. 9.54] Mg_intake_ Min = Mg int + Mg __.000075 milk + Mg + Mg __) gest gest ⎝⎝ − ⎠⎠ [Eq. 9.54] 0.02743 −Min Mg_intake_ Min = Mg __ ma ma int + Mg milk + Mg __ gain gain + Mg gest [Eq. 9.54] Mg_intake_ Min = Mg ma int + Mg _ milk + Mg gain + Mg _ gest [Eq. 9.54] Mg_intake_ Min = Mg _ ma int + Mg _ milk + Mg _ gain + Mg _ gest [Eq. Mg_intake_Min = Mg _ ma int+ Mg _ milk + Mg _ gain + Mg _ gest [Eq. 9.54] 9.54] [Eq. 9.55] Mg_maint = ⋅ 0 . 003 BW 0 . 16 Mg_intake_ Min = Mg ma 9.54] Mg_maint = 0 ..003 [Eq. 9.55] Mg_intake_ Min =⋅⋅ BW Mg 00__..16 ma int int+ + Mg Mg __ milk milk + + Mg Mg __ gain gain + + Mg Mg __ gest gest 9.54] [Eq. 9.55] Mg_maint = 0 003 BW 16 [Eq. 9.55] Mg_maint= 0.003⋅ BW 0.16 [Eq. 9.55] Mg_maint= 0.003⋅ BW 0.16 [Eq. 9.55] Mg_maint = 0 . 003 ⋅ BW 0 . 16 [Eq. 9.55] Mg_maint = 0..15 003 ⋅ BW 00..16 9.56] Mg_milk = ⋅⋅ ECM 16 9.56] Mg_milk ==0 00...15 0...16 16 [Eq. Mg_maint 003 ⋅⋅ BW 9.56] Mg_milk ⋅⋅ ECM ECM 16 [Eq. 9.55] 9.55] Mg_maint 003 BW 000 9.56] Mg_milk= ==0 00..15 15 ECM 0..16 16 [Eq. 9.56] Mg_milk= 0.15⋅ ECM 0.16 [Eq. 9.56] Mg_milk = 0 . 15 ⋅ ECM 0 . 16 9.56] Mg_milk 0.16 0.16 [Eq. 9.57] Mg_gain == 00.15 .45⋅ ECM ⋅ ADG ADG 1000 [Eq. 9.57] 9.56] Mg_milk = 0 . 15 ⋅ ECM 0 . 16 ADG Mg_gain = 0 . 45 ⋅ 0 . 16 [Eq. 9.57] Mg_gain = 0 . 45 ⋅ 0 . 16 9.56] Mg_milk 01000 .16 0.16 [Eq. 9.57] Mg_gain == 00.15 .45⋅ ECM ⋅ ADG 1000 [Eq. 9.57] 1000 0.16 Mg_gain = 0.45 ⋅ ADG ADG [Eq. 9.57] 1000 ADG Mg_gain = 0 . 45 ⋅ 0 . 16 [Eq. 9.58] 9.57] Mg_gain .450⋅ .16 1000 0.16 Mg_gest == 00.33 [Eq. [Eq. 9.58] 9.57] Mg_gain ..45 ⋅ .ADG 0.16 Mg_gest == 000..33 16 1000 [Eq. ADG 9.57] Mg_gest 0 [Eq. 9.58] Mg_gain 450 Mg_gest = == 0 0.33 33 0⋅ ..16 16 1000 0.16 [Eq. 9.58] Mg_gest = 0.33 0.16 1000 [Eq. 9.58] Mg_gest = 0 . 33 0 . 16 [Eq. 9.58] NorFor – The Nordic feed evaluation system Mg_gest = 0Min .33 = 0.16 [Eq. 9.58] Na_intake_ Na _ ma int+ Na _ milk + Na _ gain + Na _ gest 9.59] [Eq. Na_intake_ Min = Na _ ma int+ Na _ milk + Na _ gain + Na _ gest [Eq. 9.59] Mg_gest = ..33 0 Na_intake_ Na [Eq. 9.58] 9.59] Mg_gest =0 0Min 33 = 0..16 16 [Eq. 9.58] Na_intake_ Min = Na _ _ ma ma int int+ + Na Na _ _ milk milk + + Na Na _ _ gain gain + + Na Na _ _ gest gest 9.59] Na_intake_Min = Na _ ma int+ Na _ milk + Na _ gain + Na _ gest [Eq. Na_intake_Min Min = = Na Na _ _ ma ma int int+ + Na Na _ _ milk milk + + Na Na _ _ gain gain + + Na Na _ _ gest gest [Eq. 9.59] 9.59] Na_intake_ [Eq. 9.59] Na_maint= factor_ 3 ⋅ BW 0.90 [Eq. 9.60]

(( ((( ( (( ((( ( ( (((( (

(( ((( (( (( ( (( (( (( ( (( (( ( ((

(( ((( (

(( ((( (

)) (( ))) ((( )(

(( ((( ( ( (((( (

)) ))) )

)) (( ))) ((( )(

(( ((( ( ( (((( (

)) )) ) ))

)) ))) )

)) ))) ) )) )) ))

)) ))) )

)) ))) )

)) ))) )

)) ))) )

9.44 9.45 9.46 9.47 9.48 9.49 9.50 9.51 9.52 9.53 9.54 9.55 9.56 9.57

105

3405 3405 3405

Mg_milk =00...15 15 ECM ..16 Mg_milk ECM 16 Mg_milk Mg_milk== 15⋅⋅⋅⋅⋅ECM ECM0000 0...16 16 Mg_milk ==000..15 15 ECM 16

3406 3406 3406

ADG ADG Mg_gain === 000..45 .45⋅⋅⋅ADG 016 .16 Mg_gain ADG1000 Mg_gain 1000 0000...16 Mg_gain == 00..45 45⋅⋅ADG .16 Mg_gain 45 1000 1000 16 1000

[Eq. 9.57] [Eq. 9.57] [Eq. 9.57] [Eq. 9.57] [Eq. 9.57]

3407 3407 3407

Mg_gest Mg_gest===000..33 .33 33000..16 .16 16 Mg_gest Mg_gest == 00..33 33 00..16 16 Mg_gest

[Eq. 9.58] [Eq. 9.58] [Eq. 9.58] [Eq. 9.58] [Eq. 9.58]

9.58

3408 3408

Na_intake_ Min = Na Na _ ma ma int + Na Na _ milk milk + Na Na _ gain gain Na gest Na_intake_ Min Na ma int Na milk Na gain Na _ gest Na_intake_ ++++ ___gest Na_intake_Min Min=== = Na Na___ _ma maint int+++ +Na Na___ _milk milk+++ +Na Na___ _gain gain +Na Na _gest gest Na_intake_ Min int Na

[Eq. 9.59] [Eq. 9.59] [Eq. 9.59] [Eq. 9.59] [Eq. 9.59]

9.59

3409 3409 3409

Na_maint = factor factor _ 3 ⋅ BW 0.90 Na_maint factor BW ..90 90 Na_maint Na_maint BW Na_maint=== = factor factor____3333⋅⋅⋅⋅BW BW0000..90 90

[Eq. 9.60] [Eq. 9.60] [Eq. 9.60] [Eq. 9.60] [Eq. 9.60]

9.60

3410 3410 3410

Na_milk MY 90 Na_milk Na_milk== =000..63 .63 63⋅⋅⋅⋅MY MY0000...90 .90 Na_milk Na_milk== 00..63 63⋅MY MY 0.90 90

[Eq. 9.61] [Eq. 9.61] [Eq. 9.61] [Eq. [Eq.9.61] 9.61]

9.61

3411 3411 3412 3411 3412 3412 3412 3412 3412 3412 3412 3413 3413 3413 3413 3413 3413 3413 3414 3414 3414 3414 3414 3414 3414 3415 3415 3415 3415 3415 3415 3415 3416 3416 3416 3416 3416 3416 3416 3416 3417 3417 3417 3417 3417 3417 3417 3418 3418 3418 3418 3418 3418 3418 3419 3419 3419 3419 3419 3419 3419 3420 3420 3420 3420 3420 3420 3420 3421 3421 3421 3421 3421 3421 3422 3422 3422 3422 3422 3422 3422 3423 3423 3423 3423 3424 3423 3424 3423 3424 3424 3425 3423 3424 3425 3424 3425 3425 3426 3424 3425 3426 3425 3426 3426 3427 3425 3426 3427 3426 3427 3427 3428 3426 3427 3428 3427 3428 3428 3428 3429 3427 3428 3429 3428 3429 3429 3429 3430 3428 3429 3430 3429 3430 3430 3430 3431 3429 3430 3431 3431 3431 3432 3430 3431 3432 3432 3432 3433 3431 3432 3433 3433 3433 3434 3432 3433 3434 3434 3434 3435 3433 3434 3435 3435 3436 3434 3435 3436 3436 3437 3435 3436 3437 3437 3438 3436 3437 3438 3438 3439 3437 3438 3439 3439 3439 3440 3438 3440 3440 3441 3439 3440 3441 3441 3441 3442 3440 3442 3442 3442 3443 3441 3442 3443 3443 3443 3442 3443 3443

ADG Na_gain 90 ADG Na_gain 000..90 Na_gest == = 1111...4.444⋅⋅⋅0⋅ADG .90 1000 Na_gain .90 1000 Na_gain = Na_gain =111..44.400⋅ADG 90 1000 00..90 Na_gest = ..ADG 90 Na_gest = 90 1000 1000 Na_gest = 1 . 4 0 . 90 Na_gest ==11..44 00..90 90 Na_gest Na_gest = 1 . 4 0 . 90 124 124 Na_gest = 1 . 4 0 . 90 Na_gest = 1.4in0.=90K _ ma int + K _ milk + K _ gain 124 K_intake_M + K _ gest 124 K_intake_M = _ int + _ + _ + K K_intake_Min in == K K __ ma ma int ++ K K __milk milk ++K K __gain gain +124 K_ _ gest gest K_intake_M in K ma int K milk K gain gest K_intake_Min in == K K__ma maint int++KK__milk milk++KK__gain gain+++KKK___gest gest K_intake_M K_intake_M in = K _ ma int + K _ milk + K _ gain + K _ K_intake_M gest K_intake_Min in ==KK__ma maint int++KK__milk milk++KK__gain gain++KK_ _gest gest

(((

[Eq. 9.56] [Eq. 9.56] [Eq. 9.56] [Eq. 9.56] [Eq. 9.56]

))))

(((

))))

( (((

) )))) )

( ((( (

) )))) )

[Eq. 9.62] 9.62 [Eq. 9.62] [Eq. 9.63] 9.62] [Eq. [Eq.9.62] 9.62] [Eq. 9.63] [Eq. 9.63] [Eq. 9.63] [Eq.9.63] 9.63] [Eq. [Eq. 9.63] 9.63 [Eq. 9.63] [Eq. 9.63] 9.64] [Eq. 9.64] [Eq. 9.64] [Eq. 9.64] [Eq. 9.64] [Eq. 9.64] [Eq. 9.64] [Eq. 9.64] 9.64 [Eq. 9.64] K_maint= (0.038⋅ BW + factor_ 4 ⋅ DMI) 0.90 [Eq. 9.65] ( ) K_maint = 0 ..038 ⋅⋅ BW + factor _ 4 ⋅⋅ DMI 0 ..90 [Eq. 9.65] K_maint = 0 038 BW + factor _ 4 DMI 0 90 [Eq. 9.65] ( ) K_maint== (00..038 factor___444⋅⋅DMI ⋅DMI DMI 90 [Eq. 9.65] 038⋅⋅⋅BW BW+ factor ..90 [Eq. 9.65] ))))000..090 K_maint +++factor [Eq. 9.65] K_maint BW factor _ 4 ⋅ DMI = (0.038 9.65 [Eq. 9.65] K_maint ) )0.090 90 [Eq. 9.65] K_maint 038 factor__44⋅ DMI ⋅ DMI .90 [Eq. 9.65] K_milk===1(.(500.⋅.038 MY⋅⋅BW 0BW .90++factor [Eq. 9.66] K_milk = 1 . 5 ⋅ MY 0 . 90 [Eq. 9.66] K_milk= MY 90 [Eq. 9.66] K_milk 90 [Eq. 9.66] =11 1...55 5⋅⋅⋅MY MY000 0....90 90 [Eq. 9.66] K_milk = [Eq. 9.66] K_milk 90 = [Eq. 9.66] 9.66 K_milk ==111...555⋅⋅⋅MY MY 0 . 90 [Eq. 9.66] K_milk MY 0 . 90 [Eq. 9.66] ADG K_gain = 1.6 ⋅ ADG 0.90 [Eq. 9.67] 1000 ADG K_gain = 1 . 6 ⋅ 0 . 90 [Eq. 9.67] K_gain = 1.6 ⋅ADG 00..90 [Eq. 9.67] ADG1000 K_gain 90 [Eq. 9.67] = 11..66⋅⋅ADG 90 1000 [Eq. 9.67] K_gain = [Eq. 9.67] 9.67 K_gain = 90 [Eq. 9.67] 1000 0000....90 1000 ADG1000 K_gain == 111...666⋅⋅⋅ADG 90 [Eq. 9.67] 1000 K_gain 0 . 90 [Eq. K_gest = 1.0 0.90 1000 [Eq. 9.67] 9.68] 1000 K_gest [Eq. 9.68] K_gest = = 11..00 00..90 90 [Eq. 9.68] 9.68 K_gest [Eq. 9.68] =111...000 000...90 90 [Eq. 9.68] K_gest = [Eq. 9.68] K_gest = 90 [Eq. 9.68] K_gest ==11..00 00..90 [Eq. 9.68] K_gest 90 [Eq. 9.68] Cl_intake_ Min = Cl _ ma int + Cl _ milk + Cl _ gain + Cl _ gest 9.69] Cl_intake_ [Eq. 9.69] Cl_intake_Min Min = = Cl Cl _ _ ma ma int int + + Cl Cl _ _ milk milk + + Cl Cl _ _ gain gain + + Cl Cl _ _ gest gest [Eq. 9.69] Min = = Cl Cl___ma ma int ++Cl Cl milk ++Cl Cl gain ++Cl Cl gest Cl_intake_Min maint int+ Cl____milk milk+ Cl____gain gain+ Cl____gest gest 9.69 [Eq. [Eq. 9.69] 9.69] Cl_intake_ [Eq. 9.69] Cl_intake_ Min = Cl ma int + Cl milk + Cl gain + Cl gest [Eq. 9.69] Cl_intake_ Min = Cl _ ma int + Cl _ milk + Cl _ gain + Cl _ gest [Eq. 9.69] Cl_intake_Min = Cl _ ma int + Cl _ milk + Cl _ gain + Cl _ gest [Eq. 9.69] Cl_maint = 00..0225 0225⋅⋅ BW BW 00..90 90 [Eq.9.70] 9.70] Cl_maint= = [Eq. 9.70 Cl_maint 0225 BW [Eq. 9.70] = 0000....0225 0225⋅⋅⋅⋅BW BW00000....90 90 Cl_maint = [Eq. 9.70] Cl_maint .90 90 [Eq. 9.70] [Eq. 9.70] Cl_maint = 0225 BW 90 [Eq. 9.70] Cl_maint ==00..0225 ⋅⋅BW 00.90 [Eq. 9.70] Cl_maint 0225 BW . 90 [Eq. 9.70] Cl_milk = 11..15 15⋅⋅ MY MY 00..90 90 [Eq.9.71] 9.71] Cl_milk= = [Eq. 9.71 Cl_milk 15 MY 90 [Eq. 9.71] Cl_milk =1111....15 15⋅⋅⋅⋅MY MY0000....90 90 [Eq. 9.71] [Eq. 9.71] Cl_milk= 90 [Eq. 9.71] Cl_milk = 15 MY 90 [Eq. 9.71] Cl_milk= 1.15⋅ MY 0.90 [Eq. 9.71] Cl_milk= 1.15⋅ ADG MY 0.90 [Eq. 9.71] Cl gain = = 11..00 ⋅⋅ ADG 90 [Eq.9.72] 9.72] ADG1000 00..90 9.72 Cl ___ gain gain [Eq. Cl = 1 . 0 ⋅ 0 . 90 [Eq. 9.72] ADG ADG 1000 Cl __ gain 0000...90 [Eq. 9.72] gain = = 111...000⋅⋅⋅ADG 90 1000 [Eq. 9.72] . 90 [Eq. 9.72] Cl gain = 90 [Eq. 9.72] ADG 1000 1000 0.90 Cl _ gain = 1.0 ⋅ ADG1000 [Eq. 9.72] 1000 Cl _ gain==11.0.00⋅ .90 1000 0.90 [Eq. 9.73] 9.72] Cl_gest [Eq. 1000 9.73 Cl_gest 90 [Eq. 9.73] Cl_gest == =111...000 000...90 90 [Eq. 9.73] Cl_gest [Eq. 9.73] =11..00 00..90 90 [Eq. 9.73] Cl_gest = Cl_gest [Eq. [Eq. 9.73] 9.73] Cl_gest = 1.0 0.90 [Eq. 9.73] Cl_gestCa_intake_Min = 1.0 0.90 [Eq. 9.73] where is the the calcium recommendation fordairy dairycows, cows, g/d;Ca_maint, Ca_milk, where Ca_intake_Min is calcium the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, where Ca_intake_Min is recommendation for g/d;Ca_maint, Ca_milk, where Ca_intake_Min is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, where Ca_intake_Min is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, Ca_intake_Min is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, where Ca_intake_Min is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, Ca_gain, Ca_gest is a cows requirement of Ca for maintenance, milk production, gain and gestation, where is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, Ca_gain, Ca_gest is a cows requirement offor Camaintenance, forfor maintenance, milk production, gain and gestation, Ca_gain, Ca_gest is aa cows requirement of Ca milk production, gain and gestation, where Ca_intake_Min is the calcium recommendation dairy cows, g/d;Ca_maint, Ca_milk, Ca_gain, Ca_gest is cows requirement of Ca for maintenance, milk production, gain and gestation, Ca_gain, Ca_gest is requirement of milk production, gain and gestation, Ca_gain, Ca_gest is aa cows cows requirement ofCa Ca for maintenance, milk production, gain and gestation, Ca_gain, Ca_gest is requirement of for maintenance, milk production, gain and gestation, respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are where Ca_intake_Min is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, Ca_gain, Ca_gest cows requirement Cafor formaintenance, maintenance, milk production, gain and gestation, respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are Ca_gain, Ca_gest isisaaacows cows requirement ofofCa Ca for maintenance, milk production, gain and gestation, respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and Ca_gain, Ca_gest is a cows requirement of Ca for maintenance, milk production, gain and gestation, respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and recommendations of phosphorus, magnesium, sodium, potassium and for dairy cows and recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and growing cattle, g/d; BW is the the weight of the the animal, animal, kg; BW_mat ischlorine the mature weight ofthe the respectively; P_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min areanimal, recommendations ofBW phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and growing cattle, g/d; is weight of kg; BW_mat is the weight of recommendations of phosphorus, magnesium, sodium, potassium and chlorine dairy cows and growing cattle, g/d; BW is the weight of the animal, kg; BW_mat is the mature weight of the growingcattle, cattle, g/d; BW isMg_intake_Min, the weight ofanimal, the animal, kg; BW_mat ismature thefor mature weight of the growing g/d; BW is the weight of the kg; BW_mat is the mature weight of the growing cattle, g/d; BW is the weight of the animal, kg; BW_mat is the mature weight of the growing cattle, g/d; BW is the weight of the animal, kg; BW_mat is the mature weight of the animal, kg (Table 3.2 and Table 3.5); ECM is the energy-corrected milk yield, Equation 3.1 and 3.2; recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and3.2; growing cattle, g/d; BW is the weight of the animal, kg; BW_mat is the mature weight of the animal, kg (Table 3.2 and Table 3.5); ECM is the energy-corrected milk yield, Equation 3.1 and 3.2; growing cattle, g/d; BW is the weight of the animal, kg; BW_mat is the mature weight of the animal, kg (Table 3.2 and Table 3.5); ECM is the energy-corrected milk yield, Equation 3.1 and kg (Table 3.2 and Table 3.5); ECM is theisisweight energy-corrected milk yield, Equation 3.1and and 3.2; MY is animal, kg (Table 3.2 and 3.5); milk Equation 3.1 3.2; animal, kg (Table 3.2 and Table 3.5); ECM the energy-corrected yield, Equation 3.1 and 3.2; animal, kg (Table 3.2 and Table 3.5); ECM the energy-corrected milk yield, Equation 3.1 and 3.2; MY is the milk yield, kg/d; ADG is theECM daily gain ofgrowing growing cattle including primiparous growing cattle, g/d; BW is Table the weight of theis animal, kg;of BW_mat ismilk theyield, mature weight of3.1 the animal, kg (Table 3.2 and Table 3.5); ECM isthe theenergy-corrected energy-corrected milk yield, Equation and 3.2; MY is the milk yield, kg/d; ADG is the daily gain cattle including primiparous animal, kg (Table 3.2 and Table 3.5); ECM isweight the energy-corrected milk yield, Equation 3.1 and 3.2; MY is the the milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous MY is milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous the milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous the milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous MY is the milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous cows, g/d; gest is the actual day in gestation from day 190 of gestation until calving; The animal, kg (Table 3.2 and Table 3.5); ECM is the energy-corrected milk yield, Equation 3.1 and MY isg/d; the milk milk yield, kg/d;ADG ADG isgestation thedaily dailyfrom weight gain of growing cattle including primiparous 3.2; cows, cows, gest is the actual day in day 190 gestation until calving; The MY is the yield, kg/d; is the weight gain of growing cattle including primiparous cows, g/d; gest is the actual day in gestation from day 190 of gestation until calving; The cows, g/d; gest is the actual day in gestation from day 190 ofof gestation until calving; The g/d; isgest is the actual day in gestation from day190 190 of gestation until calving; Theand denominators g/d; gest is the actual day in gestation from day 190 gestation until calving; The cows, g/d; is the actual day in gestation from day of until calving; The denominators equations are absorption coefficients; factor_1 is0.031 0.031 for lactating cows and MY the gest milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous g/d; gest is theequations actual day in gestation from day 190 ofgestation gestation until calving; The in the are absorption coefficients; is for lactating cows cows, g/d; gest is the actual day in gestation from day 190 offactor_1 gestation until calving; The denominators in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows and denominators in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows and denominators in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows denominators in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows and in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows and 0.0154 for non-lactating cows, g Ca/kg BW; factor_2 is 0.98 for large breeds and 1.17 for Jersey, cows, g/d; gest is the actual day in gestation from day 190 of gestation until calving; The denominators inthe theequations equations are absorption coefficients; factor_1 0.031 for lactating cows cows,are gCa/kg Ca/kg BW;factor_2 factor_2 is0.98 0.98 forlarge large breeds and 1.17for for Jersey, gg0.0154 for denominators in absorption coefficients; factor_1 is is 0.031 for lactating cows andand 0.0154 for non-lactating cows, Ca/kg BW; factor_2 is 0.98 for large breeds and 1.17 for Jersey, g 0.0154 for non-lactating cows, ggg BW; is for breeds and 1.17 Jersey, g 0.0154 for non-lactating cows, Ca/kg BW; factor_2 is for large breeds and 1.17 for Jersey, g 0.0154 for non-lactating cows, gglactating Ca/kg BW; factor_2 isis0.98 0.98 for large breeds and 1.17 for Jersey, ggP/kg non-lactating cows, g Ca/kg BW; factor_2 is 0.98 for large breeds and 1.17 for Jersey, g P/kg milk; 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, g milk; denominators in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows and 0.0154 for non-lactating cows, Ca/kg BW; factor_2 0.98 for large breeds and 1.17 for Jersey, 0.0154 for non-lactating cows, g Ca/kg BW; factor_2 is 0.98 for large breeds and 1.17 for Jersey, g factor_3 is 0.038 for cows and 0.015 for non-lactating cows and growing cattle, g P/kg milk; milk; factor_3 is 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, g P/kg factor_3 is 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, g P/kg milk; factor_3 is 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, g milk; factor_3 is 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, ggg Na/kg factor_3 is 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, Na/kg BW; factor_4 is 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, g 0.0154 for factor_3 non-lactating cows, glactating Ca/kg cows BW; factor_2 is for 0.98 for large cows breeds and 1.17 forcattle, Jersey, milk; factor_3 isis0.038 0.038 forlactating cows and 0.015 fornon-lactating non-lactating cows and growing cattle, P/kg milk; is for lactating 0.015 cows and growing cattle, g 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, g Na/kg BW; factor_4 6.1 for and 2.6 for non-lactating and growing g Na/kg BW; factor_4 is 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, g Na/kg BW; factor_4 is 6.1 for lactating cows and for non-lactating cows growing cattle, g BW; factor_4 is 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, gnonK/kg is dry matter intake which is2.6 set to 3.3% and1.67% 1.67% ofand BW for lactating and P/kg factor_3 0.038 for lactating cows and 0.015 for and non-lactating cows and growing g DM; Na/kg BW;and factor_4 isfor 6.1lactating for lactating cows and 2.6 for non-lactating cows and growing cattle, ggnonNa/kg BW; factor_4 is 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, gcattle, BW;milk; factor_4 is 6.1is cows and 2.6 for non-lactating cows and growing cattle, K/kg DM; and DMI dry matter intake which is set to 3.3% of BW for lactating and DM; and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonK/kg DM; DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonK/kg DM; and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonDM; and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonlactating cows, respectively. For growing cattle, DMI is calculated according to Equations 9.74 and Na/kg BW; factor_4 is 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, g DM; and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonK/kg DM; and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonrespectively. For growing cattle, DMI is calculated according to Equations 9.74 and and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and non-lactating lactating cows, respectively. For growing cattle, DMI is calculated according to Equations 9.74 and lactating cows, respectively. For growing cattle, DMI isiscalculated calculated according totoEquations Equations 9.74 and lactating cows, For cattle, DMI is according 9.74 and lactating cows, respectively. For growing cattle, calculated according Equations 9.74 and 9.75. K/kg and respectively. DMI isFor dry growing matter intake which isDMI set toiscalculated 3.3% andaccording 1.67% of to BW for lactating and non-9.75. lactating cows, respectively. For growing cattle, DMI according to Equations 9.74 lactating cows, respectively. Forgrowing growing cattle, DMI calculated according Equations 9.74 and cows,DM; respectively. cattle, DMI is is calculated toto Equations 9.74 and 9.75. 9.75. 9.75. lactating cows, respectively. For growing cattle, DMI is calculated according to Equations 9.74 and

The recommendations formacro macrominerals mineralsare arefrom fromNRC NRC(2001) (2001)with with modifications for phosphorus 9.75.recommendations for macro minerals are from NRC (2001) with modifications for phosphorus recommendations for macro minerals are from NRC (2001) with modifications for phosphorus The for modifications for phosphorus Therecommendations recommendations for macro minerals areNRC from NRC (2001) with modifications for phosphorus The for macro minerals are from (2001) with modifications for phosphorus recommendations for macro minerals are from NRC (2001) with modifications for phosphorus The for minerals are (2001) with modifications for phosphorus sincerecommendations content inmilk milk isdifferentiated differentiated between large breeds and Jersey based on the phosphorus content in milk is differentiated between large breeds and Jersey based on recommendations formacro macro minerals arefrom fromNRC NRC (2001) with modifications for phosphorus the phosphorus content in milk is differentiated between large breeds and Jersey based on since the phosphorus content in is between large breeds and Jersey based on since the phosphorus content in milk is differentiated between large breeds and Jersey based on since the phosphorus content in milk is differentiated between large breeds and Jersey based on the phosphorus content in milk is differentiated between large breeds and Jersey based on since the phosphorus content ininmilk isisSwedish differentiated between large breeds and Jersey based on Danish Aaes, 2004) and Swedish data(Lindmark-Månsson (Lindmark-Månsson et al.,.,2003). .,2003). 2003). In NRC The recommendations for macro minerals are from NRC (2001) with modifications forbased phosphorus et al In NRC (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson the phosphorus content milk differentiated between large breeds and Jersey on et al 2003). In NRC Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson Danish (Sehested and Aaes, 2004) and data et al ., In NRC Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson et al ., 2003). In NRC Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson et al ., 2003). In NRC Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson et al., 2003). In NRC (2001) Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson et al ., 2003). In NRC the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but inin (2001)the forabsorption absorption ofSwedish calciumand andphosphorus phosphorus depend on the but since phosphorus in milk is differentiated between large breeds and Jersey based on Danish (Sehested andcontent Aaes, 2004) and data (Lindmark-Månsson et al .,feedstuff, 2003). In NRC the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but in (2001) the coefficients for of calcium depend on the feedstuff, but in (2001) the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but in the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but inin the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but in NorFor (2001) the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but in these are fixed at 50% and 70%, respectively, based on values from NRC (2001). DMI is NorFor 50% and 70%, respectively, based on values from NRC (2001). DMI is Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson et al ., 2003). In NRC (2001) the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but NorFor these are fixed at 50% and 70%, respectively, based on values from NRC (2001). DMI NorFor these are fixed at 50% and 70%, respectively, based on values from NRC (2001). DMI isis NorFor these are at 50% and 70%, respectively, based from NRC (2001). DMI is NorFor these are fixed fixed at 50% and 70%, respectively, basedon onvalues values from NRC (2001). DMI is is required these are fixed at 50% and 70%, respectively, on values from NRC (2001). DMI NorFor these are at 50% and based from NRC DMI isin (2001) the coefficients for absorption ofrespectively, calcium andbased phosphorus depend on the(2001). feedstuff, but NorFor these arefixed fixed at 50% and70%, 70%, respectively, basedon onvalues values from NRC (2001). DMI is 125 125 125 125 NorFor these are fixed at 50% and 70%, respectively, based on values from NRC (2001). DMI is 125 125 125 125 125 NorFor – The Nordic feed evaluation system 106

3444 3444 3445 3445 3446 3446 3447 3447 3448 3448 3449 3449 3450 3450 3451 3451 3452 3452 3453 3453 3454 3454 3455 3455 3456 3456 3457 3457 3458 3458 3459 3459 3460 3460 3461 3461 3462 3462 3463 3463 3464 3464 3465 3465 3466 3466 3467 3467 3468 3468 3469 3469 3470 3470 3471 3471 3472 3472 3473 3473 3474 3474 3475 3475 3476 3476 3477 3477 3478 3478 3479 3479 3480 3480 3481 3481 3482 3482 3483 3483 3484 3484

required to estimate the maintenance requirement of phosphorus and potassium and was set to 3.3% to BW estimate thecows maintenance requirement of NorFor phosphorus and potassium and wasset settofor to3.3% 3.3% of BW required todairy estimate theaccording maintenance requirement of phosphorus andfor potassium and was of for to NRC (2001). equations DMI were developed of for dairybased cows according toofNRC (2001). NorFor equations for DMI were developed for for growing forBW dairy cows according to NRC NorFor equations for DMIsystem. were developed growing cattle, on a subset the(2001). data used to develop the feed intake growing cattle,on based on a subset the data to develop feedintake intakesystem. system. cattle, based a subset of theofdata usedused to develop thethefeed DMIHeifer Steers = BW / 100 ⋅ (0.000004 ⋅ BW 22 − 0.0049 ⋅ BW + 3.1033) DMIHeifer Steers = BW / 100 ⋅ (0.000004 ⋅ BW − 0.0049 ⋅ BW + 3.1033)

[Eq. 9.74] [Eq. 9.74]

DMIBulls = BW/100⋅ (2.9 − 0.002⋅ BW) DMIBulls = BW/100⋅ (2.9 − 0.002⋅ BW)

[Eq. 9.75] [Eq. 9.75]

9.74 9.75

where DMIHeifersSteers and DMIBulls are estimated DMI for heifers, steers and bulls, kg DM/d; BW where DMIHeifersSteers and DMIBulls are estimated DMI for heifers, steers and bulls, kg DM/d; BW theweight weight theanimal, animal, where DMI and DMI HeifersSteers Bulls are estimated DMI for heifers, steers and bulls, kg DM/d; BW isisthe ofofthe kg.kg. is the weight of the animal, kg. Thecalcium calciumrecommendation recommendation growing cattle calculated from tabulated in Pehrson The forfor growing cattle was was calculated from tabulated values values in Pehrson The recommendation forusing growing cattle was calculated etal al. (1975)and andisiscalculated calculated following equation: from tabulated values in Pehrson et .calcium (1975) using thethe following equation: et al. (1975) and is calculated using the following equation: ⎛ 4.744 + 0.0361⋅ BW + 0.00000565⋅ BW2 ⎞ Ca _ int ake _ Min = ⎜⎛ 4.744 + 0.0361⋅ BW + 0.00000565⋅ BW2 ⎟⎞ Ca _ int ake _ Min = ⎜ + 0.0106 ⋅ ADG + 0.00000622⋅ ADG2 ⎟ ⎝⎜ 2 ⎠⎟ ⎠ ⎝ + 0.0106⋅ ADG + 0.00000622⋅ ADG

9.76

[Eq. 9.76] [Eq. 9.76]

where is the calcium recommendation for growing cattle, g/d; BW isg/d; the BW weight whereCa_intake_Min Ca_intake_Min is the calcium recommendation for growing cattle, is of the weight where Ca_intake_Min is the recommendation for growing cattle, g/d; BW the weight of the animal, kg; and is calcium theisaverage daily live weight gain of gain the animal, g/d.isHeifers late of the animal, kg; ADG and ADG the average daily live weight of the animal, g/d.inHeifers in late the animal, kg; and ADG is theCa average daily live weight gain of 9.48. the animal, g/d. Heifers in late gestation require anan additional supply for for gestation, Equation gestation require additional Ca supply gestation, Equation 9.48. gestation require an additional Ca supply for gestation, Equation 9.48. The recommendation for sulfur is set to 2 g/kg DM for dairy cows and growing cattle (NRC, 2001). The recommendation sulfur isto set2 to 2 g/kg DMdairy for cows dairyand cows and growing cattle2001). (NRC, 2001). The forfor sulfur is set g/kg DM cattleon(NRC, The recommendation macro mineral recommendations for dairy cowsfor and growing cattlegrowing that depend the BW and The macro mineral recommendations for dairy cows and growing cattle that depend on the BW and The macro mineral recommendations dairy ECM or ADG are illustrated in Tablesfor 9.13 andcows 9.14.and growing cattle that depend on the BW and ECMororADG ADGareare illustrated in Tables and 9.14. ECM illustrated in Tables 9.13 9.13 and 9.14. Table 9.13 Table 9.8.2 9.13 Micro minerals Table 9.14 Table Micro9.14 mineral recommendations for dairy cows and growing cattle are shown in Table 9.15. The

9.8.2 Micro copper (Cu)minerals requirement in the ration is dependent on the molybdenum and sulfur contents; the 9.8.2 Micro minerals

Micro mineral recommendations cows(Table and growing cattle areifshown in Table 9.15. The recommendation of 10 mg Cufor perdairy kg DM 9.15) is valid the ration contains less than 1 mg Micro mineral recommendations for2dairy cowsper and cattle1989). are shown Table 9.15. of The copper (Cu) requirement in the ration dependent on the(NRC, molybdenum andinsulfur contents; the molybdenum (Mo) and less than g is sulfur kggrowing DM Higher amounts molybdenum copper (Cu) requirement inCu theper ration is dependent on the molybdenum andcontains sulfur contents; recommendation of 10 mg kgcoefficient DM (Table 9.15) is valid ifbetween the ration less thanthe 1 be below and sulfur decrease the absorption of Cu. The ratio Cu and Mo should not recommendation of 10 mg Cu per kg2DM (Table 9.15) is valid if the ration contains less than 1 mg molybdenum (Mo) and less than g sulfur per kg DM (NRC, 1989). Higher amounts of 4:1molybdenum (NRC, 1989).(Mo) Synthesis ofthan the hormone the thyroid gland is inhibited by goitrogenic mg and less 2absorption g sulfurthyroxine per kg DMin(NRC, 1989). Higher amounts of molybdenum and sulfur decrease the coefficient of Cu. The ratio between Cu and Mo substances, which are common in rape seed products thatofcontain (Hermansen molybdenum sulfur the absorption Cu.thyroxine Theglucosinolates ratioinbetween Cu and Mois et al., should not be and below 4:1 decrease (NRC, 1989). Synthesiscoefficient of the hormone the thyroid gland 1995). Since rape seed products are commonly used in feed rations for dairy cows in Nordic should notbybegoitrogenic below 4:1substances, (NRC, 1989). Synthesis of theinhormone thecontain thyroid gland countries, is inhibited which are common rape seedthyroxine products in that the recommended iodine level was set to 1.0 mg/kg DM for dairy cows, although NRC (2001) only inhibited by goitrogenic substances, which are common in rape seed products that contain glucosinolates (Hermansen et al., 1995). Since rape seed products are commonly used in feed rations recommends 0.6Nordic mg/kgcountries, DM. the substantial amounts ofmg/kg goitrogenic substrates, the glucosinolates et al.,If1995). Sincecontains rape seed products commonly used inDM feedforrations for dairy cows (Hermansen in the ration recommended iodine level are was set to 1.0 dairy for dairy cows inNRC Nordic countries, theincreased recommended iodine level wasration set tocontains 1.0 mg/kg DM1999). for dairy iodine recommendation should be 2.0 mg/kg (Strudsholm etsubstantial al., cows, although (2001) only recommends 0.6tomg/kg DM. IfDM the cows, although NRC (2001) only recommends 0.6 mg/kg DM.should If the ration containstosubstantial amounts of goitrogenic substrates, the iodine recommendation be increased 2.0 mg/kg DM amounts of goitrogenic substrates, the iodine recommendation should be increased to 2.0 mg/kg DM (Strudsholm et al., 1999). al., 1999). (Strudsholm Table 9.13.etRecommendations (g/day) for macro minerals for growing cattle gaining 750 g/day1.

The recommendation includes requirements for maintenance and gain. 126

BW, kg

Calcium

Phosphorus126 Magnesium

Sodium

Potassium

200 kg 400 kg 600 kg

24 32 40

12 15 18

5 8 11

22 37 52

6 10 13

1 Phosphorus, magnesium, sodium and potassium recommendations are from NRC (2001). The calcium recommendation

is from Pehrson et al. (1975).

NorFor – The Nordic feed evaluation system

107

Table 9.14. Recommendations (g/day) for macro minerals for dairy cows depending on BW and milk yield (MY) or ECM1. BW, kg

MY or ECM, kg/d

Calcium

Phosphorus2 Magnesium Sodium

Potassium

430 430 430 600 600 600

17/22 24/31 31/40 25/25 35/35 45/45

80 102 124 109 136 163

49 60 72 63 77 91

151 166 181 201 218 235

29 37 46 35 44 53

34 40 46 43 50 57

1 The recommendations are from NRC (2001) with a slight modification for phosphorus, for which the milk content is

from Lindmark-Månsson et al. (2003) and Sehested and Aaes (2004). 2 The P recommendation is dependent on breed. Jersey cows (BW=430 kg) have a content of 1.17 g P/kg milk whereas 0.98 g P/kg milk is used for large breed cows (BW=600 kg). MY and ECM yield were assumed to be equal for large breeds but for Jersey cows MY was multiplied by 1.3 to get ECM yield.

Table 9.15. Recommendations for micro minerals (mg/kg DM) for dairy cows1 and growing cattle2.

Dairy cows Growing cattle

Cobalt

Copper3

Iodine4

Iron

Manganese Selenium5 Zinc

0.1 0.1

10 10

1.0 0.5

50 50

40 20

0.2 0.1

50 30

1 The

recommendations are based on NRC (2001) with the exception of selenium. recommendations are based on NRC (2000). 3 The recommendation is dependent on the content of molybdenum and sulfur in the ration. 4 The recommendation considers some content of goitrogenic substrates, which are sometimes present in rape seed products (Hermansen et al., 1995). 5 The recommendation is from former Swedish recommendations (Spörndly, 2003). 2 The

9.8.3 Cation anion difference In order to reduce the risk of milk fever, the CAD has been introduced as a nutritional variable in feeding, especially for dry cows. The NorFor recommendation for CAD is -150 to 0 mEq/kg DM in the dry period, or at least for the last 3 to 4 weeks of the dry period (Horst et al., 1997; Moore et al., 2000; Kristensen, 2005). For lactating cows the recommendation is 200 to 450 mEq/kg DM since Borucki Castro et al. (2004) found that increasing CAD from 140 to 450 mEq/kg DM had no effect on DMI or milk yield. The CAD should never exceed 450 mEq/kg DM and this is important when adding buffer to the ration (Kristensen, 2005). Research results in the literature are equivocal about whether a reduction in dietary K and a moderate CAD in the dry period are sufficient to avert milk fever (Overton and Waldron, 2004). Potassium is therefore included in the calculation of CAD (see Chapter 6) and NorFor does not provide any recommendations for K in relation to CAD.

9.9 Vitamins Cattle require vitamins A, D, E and K. However, the dietary requirements are only absolute for vitamins A and E, since vitamin K is synthesized by ruminal and intestinal bacteria, and vitamin D is synthesized by the action of ultraviolet radiation in the skin (NRC, 2001). NorFor has 108

NorFor – The Nordic feed evaluation system

recommendations for the supply of fat soluble vitamins (A, D and E) to dairy cows and growing cattle, which are from NRC (2001) and shown in Table 9.16. The recommendations from NRC (2001) are based on supplemental vitamins and, thus, do not take into account the natural content of vitamins in feedstuffs. NorFor includes the natural content of vitamins A, D and E when calculating the supply of these vitamins from a feed ration and recommendations are therefore a total recommendation. The beta-carotene in feedstuffs is converted to vitamin A such that 1 mg beta-carotene corresponds to 400 IU of vitamin A (Strudsholm et al., 1999). Table 9.16. Recommendations for vitamins A, D and E supplies (IU/kg BW) to dairy cows and growing cattle are based on NRC (2001). Requirements are expressed as total amounts, i.e. supplemental plus vitamins provided by feedstuffs.

Dairy cows Growing cattle

Vitamin A1, 2

Vitamin D3

Vitamin E4

110 80/110

30 10

0.8/1.6 0.3/1.6

1

Beta-carotene can be converted to vitamin A and the total requirement can be met from beta-carotene. 1 mg betacarotene corresponds to 400 IU vitamin A (Strudsholm et al., 1999). 2 The requirement of 80 IU/kg BW is for bulls, steers and non-pregnant heifers, while the requirement of 110 IU/kg BW is for pregnant heifers (NRC, 2001). 3 The requirement is from Strudsholm et al. (1999). 4 The requirement of 0.8 IU/kg BW is for lactating cows, while the requirement of 1.6 IU/kg BW is for dry cows and heifers in late gestation. The requirement for growing cattle is modified from NRC (2001), which suggests a range from 0.26 to 0.8 IU/kg BW, where the lowest value refers to grazing animals and the highest value to animals fed stored forages.

References Berg, J. and T. Matre, 2001. Produksion av storfekjøtt. Landbruktforlaget. Norway. (In Norgwegian). Berg, J. and H. Volden, 2004. Økologisk storfekjøtt. Buskap nr 4. 16-17. (In Norwegian). Borucki Castro, S.I., L.E. Phillip, V. Girard and A. Tremblay, 2004. Altering dietary cation-anion difference in lactating dairy cows to reduce phosphorus excretion to the environment. Journal of Dairy Science 87: 1751-1757. Børsting, C.F., J.E. Hermansen and M.R. Weisbjerg, 2003. Fedtforsyningens betydning for mælkeproduktionen. In: Strudsholm, F. and K. Sejrsen (eds.). Kvægets ernæring og fysiologi. DJF-rapport nr. 54, Danmarks Jordbrugsforskning, Denmark, pp. 133-152. (In Danish). Danfær, A., 1983. Næringsstoffernes intermediære omsætning. In: Østergaard, V. and A. Neimann-Sørensen (eds.). Optimale foderrationer til malkekoen. Foderværdi, foderoptagelse, omsætning og produktion. Beretning nr. 551 fra Statens Husdyrbrugsforsøg, Denmark 12.1-12.65. (In Danish). Garcia, F., J. Agabriel and D. Micol, 2007. Alimentation des bovins en croissance et à l’engrais. In: INRA (ed.). Alimentation des bovin, ovins et caprin. Besoin des animaux - Valeurs des aliments. Edition Quæ, Versaille, France, pp. 89-120. (In French). Hermansen, J., O. Aaes, S. Ostersen and M. Vestergaard, 1995. Rapsprodukter til malkekøer – mælkeydelse og mælkekvalitet. Forskningsrapport nr. 29 fra Statens Husdyrbrugsforsøg, Denmark. (In Danish). Hole, J.R, 1985. The nutritive value of silage made from Poa pratensis ssp. Alpigena and Phlenum pratense. III. Growing bulls fed silage made from the regrowth of Poa pratensis or Phlenum pratense. Report No. 232. Department of Animal Nutrition, Agricultural University of Norway, Aas, Norway. 17 pp. Homb, T., 1974. Grovfôr-kraftfôr-forholdet ved intensivt oppdrett av 14 måneder gamle slakteokser. Institutt for husdyrernæring og fôringslære, Norges landbrukshøgskole. Melding nr. 164, Norway. 18 pp. (In Norwegian). Horst, R.L., J.P. Goff, T.A. Reinhardt and D.R. Buxton, 1997. Strategies for preventing milk fever in dairy cattle. Journal of Dairy Science 80: 1269-1280.

NorFor – The Nordic feed evaluation system

109

Hvelplund, T. and J. Madsen, 1990. A study of the quantitative nitrogen metabolism in the gastro-intestinal tract, and the resultant new protein evaluation system for ruminants. The AAT-PBV system. The Royal Veterinarian and Agricultural University, Copenhagen, Denmark. Hymøller, L., M.R. Weisbjerg, P. Nørgaard, C.F. Børsting, and N.B. Kristensen, 2005. Majsensilage til malkekøer. DJF rapport nr. 65. Danmarks Jordbrugsforskning, Denmark. (In Danish). INRA (Institut National de la Recherche Agronomique), 1989. Ruminant nutrition: Recommended allowances and feed tables. John Libbey and Co Ltd, London, UK. 389 pp. Johansen, A., 1992. A comparison between meadow fescue and timothy silage. Doctor Scientarium Thesis. Agricultural University of Norway, Aas, Norway. 108 pp. Kristensen, N.B., 2005. Brug CAB til optimering af foderrationer til malkekøer. I: Fodringsdag - Temadag om aktuelle fodringsspørgsmål. 30 august. Dansk Kvæg, Denmark, pp. 23-42. (In Danish). Kristensen, V.F., 1997. Optimal proteinforsyning. In: Malkekøernes ernæring. Intern rapport nr. 88. Statens Husdyrbrugsforsøg, Denmark, pp. 46-55. (In Danish). Kristensen, V.F., 1995. Forudsigelse af foderoptagelsen hos malkekøer. Intern rapport nr. 61. Statens Husdyrbrugsforsøg, Denmark. 28 pp. (In Danish). Kristensen, V.F., 1983. Styring af foderoptagelsen ved hjælp af foderrationens sammensætning og valg af fodringsprincip. In: Østergaard, V. and A. Neimann-Sørensen (eds.). Optimale foderrationer til malkekoen. Foderværdi, foderoptagelse, omsætning og produktion. Beretning nr. 551 fra Statens Husdyrbrugsforsøg, Denmark. 7.1-7.35. (In Danish). Lindmark-Månsson, H., R. Fondén and H.E. Pettersson, 2003. Composition of Swedish dairy milk. International Dairy Journal 13: 409-425. MacRae, J.C., P.J. Buttery and D.E. Beever, 1988. Nutrient interactions in the dairy cow. In: P.C. Garnsworthy (ed.). Nutrition and lactation in the dairy cow. Butterworths, London, UK, pp. 55-75. Madsen, J., 1985. The basis for the proposed Nordic protein evaluation system for ruminants. The AAT-PBV system. Protein Evaluation for Ruminants. Acta Agriculturae Scandinavica Supplement 25: 13-19. Madsen, J., L. Misciatteli, V.F. Kristensen and T. Hvelplund, 2003. Proteinforsyning til malkekøer. In: Strudsholm, F. and K. Sejrsen (eds.). Kvægets ernæring og fysiologi. DJF-rapport nr. 54. Danmarks Jordbrugsforskning, Denmark, pp. 113-121. (In Danish). Minde, A. and A.J. Rygh, 1997. Metoder for å bestemme nedbrytingshastighet, passasjehastighet og vomfordøyelighet av NDF hos mjølkeku. Master thesis. Norges landbrukshøgskole, Norway. 118 pp. (In Norwegian). Moore, S.J., M.J. Van de Haar, B.K. Sharma, T.E. Pilbeam, D.K. Beede, H.F. Bucholtz, J.S. Liesman, R.L. Horst and J.P. Goff, 2000. Effects of altering dietary cation-anion difference on calcium and energy metabolism in peripartum cows. Journal of Dairy Science 83: 2095-2104. Mydland, L.T., 2005, Molecular and chemical characterization of the rumen microbiota and quantification of rumen microbial protein synthesis. Ph.D. thesis. Norwegian University of Life Sciences, Aas, Norway. 155 pp. Nielsen, H.M., N.C. Friggens, P. Løvendahl, J. Jensen and K.L. Ingvartsen, 2003. Influence of breed, parity, and stage of lactation on lactational performance and relationship between body fatness and live weight. Livestock Production Science 79: 119-133. NRC (National Research Council), 1985. Nutrient Requirement of Dairy Cattle. Fifth revised edition. National Academy Press, Washington, DC, USA. NRC, 1989. Nutrient Requirement of Dairy Cattle. Sixth revised edition. National Academy Press, Washington, DC, USA. NRC, 2000. Nutrient Requirement of Beef Cattle. Seventh revised edition. National Academy Press, Washington, DC, USA. NRC, 2001. Nutrient Requirement of Dairy Cattle. Seventh revised edition. National Academy Press, Washington, DC, USA. Nørgaard, P., 1986. Physical structure of feeds for dairy cows - A new system for evaluation of the physical structure in feedstuffs and rations for dairy cows. In: Neimann-Sørensen A. (ed.). New developments and future perspectives in research of rumen function. Brussels, Belgium, pp. 85-107. Olsson, I., 1987. Effect of protein supply on the performance of intensively reared bulls - evaluation of the DCP and the Nordic AAT-PBV protein evaluation system. Dissertation. Report 164. Swedish University of Agricultural Sciences, Uppsala, Sweden. Olsson, I. and J.E. Lindberg, 1985. Performance of bull calves fed rations with different calculated amino acid flow to the duodenum. Acta Agriculturae Scandinavica Supplement 25.

110

NorFor – The Nordic feed evaluation system

Overton, T.R. and M.R. Waldron, 2004. Nutritional management of transition dairy cows: Strategies to optimize metabolic health. Journal of Dairy Science 87: E105-E119. Pehrson, B., V. Kossila, J.T. Hansen and N.E. Hansen, 1975. Förslag til normer för makro- och mikromineraler till nötkreatur och svin. Foderjournalen, pp. 55-106. (In Swedish). Prestløkken, E., 1999. Protein value of expander-treated barley and oats for ruminants. Ph.D. thesis. Agricultural University of Norway, Aas, Norway. 137 pp. Rajala-Schultz, P.J., W.J.A. Saville, G.S. Frazer and T.E. Wittum, 2001. Association between milk urea nitrogen and fertility in Ohio dairy cows. Journal of Dairy Science 84: 482-489. Randby, Å.T., 2000a. Ensilert eller propionsyrekonservert bygg til mjølkekyr og kastrater. Husdyrforsøksmøtet 2000. Institutt for husdyrfag, NLH, Aas, Norway, pp. 409-412. (In: Norwegian). Randby, Å.T., 2000b. Effekten av økende dosering med et maursyreholdig ensileringsmiddel på næringsverdien av surfôret i produksjonsforsøk med kyr og kastrater. Hellerud rapport nr 25, Aas, Norway. 37 pp. (In Norwegian). Randby, Å.T., 2003. Høstetid og fôrkvalitet. Grønn kunnskap. 7: 27-43. (In Norwegian). Reynolds, C.K., 2000. Forage evaluation using measurements of energy metabolism. In: Givens, D.I., E. Owen, R.F.E. Axford and H.M. Omed (eds.). Forage evaluation in ruminant nutrition. CAB International, Wallingford, UK, pp. 95-111. Robelin, J. and R. Daenicke, 1980. Variations of net requirements for cattle growth with live weight, live weight gain, breed and sex. Annales de Zootechnie 29: 99-118. Sehested, J. and O. Aaes, 2004. Fosfor til kvæg. DJF-rapport nr. 60. Danmarks Jordbrugsforskning, Denmark. (In Danish). Schei, I., H. Volden and L. Bævre, 2005. Effects of energy balance and metabolizable protein level on tissue mobilization and milk performance of dairy cows in early lactation. Livestock Production Science 95: 35-47. Sjaunja, L.O., L. Bævre, L. Junkkarainen, J. Pedersen and J. Setälä, 1990. A Nordic proposition for an energy corrected milk (ECM) formula. 26th session of the International Committee for Recording the Productivity of Milk Animals (ICRPMA). Skaara, E., 1994. The effect of N-fertilization on the protein value of silage assist by the AAT/PBV system. Doctor Scientarium Thesis. Agricultural University of Norway, Aas, Norway. 95 pp. Spörndly, R., 2003. Fodermedelstabeller för idisslare. Rapport nr. 257. Department of Animal Nutrition and Management, Swedish University of Agricultural Sciences, Uppsala, Sweden. 96 pp. (In Swedish). Strudsholm, F., O. Aaes, J. Madsen, V.F. Kristensen, H.R. Andersen, T. Hvelplund and S. Østergaard, 1999. Danske fodernormer til kvæg. Rapport nr. 84. Landbrugets Rådgivningscenter, Århus, Denmark. (In Danish). Subnel, A.P.J., R.G.M. Meijer, W.M. Van Straalen and S. Tamminga, 1994. Efficiency of milk protein production in the DVE protein evaluation system. Livestock Production Science 40: 215-224. Tamminga, S., W.M. van Straalen, A.P.J. Subnel, R.G.M Meijer, C. Steg, J.G. Wever and M.C. Block, 1994. The Dutch protein evaluation system: the DVE/OEB-system. Livestock Production Science 40: 139-155. Thuen, E., 1989. Influence of source and level of dietary protein on milk yield, milk composition and some rumen and blood components in dairy cows. Doctor Scientarium Thesis. Agricultural University of Norway, Norway. 151 pp. Van Es, A.J.H., 1975. Feed evaluation for dairy cows. Livestock Production Science 2: 95-107. Van Es, A.J.H., 1978. Feed evaluation for ruminants. I. The system in use from May 1978 onwards in the Netherlands. Livestock Production Science 5: 331-345. Van Straalen, W.M., 1994. Modelling of nitrogen flow and excretion in dairy cows. Ph.D. thesis, Department of Animal Nutrition, Wageningen Agricultural University, Wageningen, the Netherlands. 205 pp. Volden, H., 1990. Response of dietary AAT density on milk yield - Review of production trials in Norway. Proceedings of the seminar on fat and protein feeding to the dairy cow. October 15-16, Eskilstuna, Sweden, pp. 81-89. Volden, H., 1999. Effects of level of feeding and ruminally undegraded protein on ruminal bacterial protein synthesis, escape of dietary protein, intestinal amino acid profile, and performance of dairy cows. Journal of Animal Science 77: 1905-1918. Volden, H., 2001. Utvikling av et mekanistisk system for vurdering av fôr til drøvtyggere, AAT-modellen. In: Fôropptak og fôrmiddelvurdering hos drøvtyggere. Fagseminar 18.-19. september 2001. Quality Hotell Halvorsbøle, Jevnaker, Norway. 30 pp. (In Norwegian). Volden, H. and O.M. Harstad, 2002. Effects of duodenal amino acid and starch infusion on milk production and nitrogen balance in dairy cows. Journal of Animal Science, Supplement 1 80: 321. Volden, H. and J. Berg, 2005. Nytt fôrvurderingssystem. Buskap nr 5: 12-14. (In Norwegian).

NorFor – The Nordic feed evaluation system

111

3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955

10. Prediction of voluntary feed intake 10.0 Prediction of voluntary feed intake

H. Volden, Volden,N. N.II Nielsen, Nielsen,M. M.Åkerlind, Åkerlind,M.M. Larsen, Havrevoll and A.J. Rygh H. Larsen, Ø. Ø. Havrevoll, A. J. Rygh Describingthe thenutrient nutrient variables in aand diettheir andinteractions their interactions is an important part of ration Describing variables in a diet is an important part of ration formulationasasproductivity productivity of the dairy is sensitive the profile of the nutrients formulation of the dairy cow cow is sensitive to the to profile of the nutrients absorbed.absorbed. However, ofof feed intake is probably the most important determinant of production. However,the theprediction prediction feed intake is probably the most important determinant of production. Feed by by animal andand feedfeed characteristics. The most Feedintake intakeisisprimarily primarilyinfluenced influenced animal characteristics. The important most important animal animal characteristics areand BWphysiological and physiological including stage lactation, milkproduction, stage characteristics are BW state,state, including stage of of lactation, milk production, stage of gestation, live weight gain and body condition score. Feed characteristics of gestation, live weight gain and body condition score. Feed characteristics such as digestibility such as digestibility and fiber content have both a strong influence on rumen fill and, hence, feed and fibre content have both a strong influence on rumen fill and, hence, feed intake (Kristensen, intake (Kristensen, 1983). However, several studies (Rinne et al., 2002; Garmo et al., 2007) have 1983). However, several studies (Rinne et al., 2002; Garmo et al., 2007) have shown that cows may shown that cows may stop eating before the fill capacity of the rumen is reached. This has been stop eating before theregulation fill capacity of which the rumen reached. This hastobeen attributed attributed to metabolic (MR), is alsoisan important factor consider when to metabolic regulationfeed (MR), which is alsointake an important factor to consider when predicting intake. predicting intake. Physical regulation is related to the ruminal NDF poolfeed (Bosch et Physical intake regulation ruminal pool (Bosch al., 1992; Rinne et of al.,the 2002) and is al ., 1992; Rinne et is al.,related 2002) to andthe is partly an NDF indirect effect of theetenergy concentration partly an indirect effect ofinthe energy concentration the diet because DMI declines in a curvilinear diet because DMI declines a curvilinear manner withofincreasing energy density of the diet manner with increasing energy density of the diet (Mertens, 1994). (Mertens, 1994). Several toto predict DMI forfor practical ration formulation. These Severalsystems systemshave havebeen beendeveloped developed predict DMI practical ration formulation. These systems systems vary both complexity in their complexity and of choice of variables included (animal, nutritional and vary both in their and choice variables included (animal, nutritional and environmental). environmental). In the Danish1983) (Kristensen, 1983)(Jarrige, and French (Jarrige, 1986) systems, intake from In the Danish (Kristensen, and French 1986) systems, feed intakefeed is predicted is predicted from dietary fill values and animal fill capacity. In the system developed by Mertens dietary fill values and animal fill capacity. In the system developed by Mertens (1987), intake is (1987), intake is regulated both physically by the diet (NDF) and metabolically by the energy regulated both physically by the diet (NDF) and metabolically by the energy demand of the animal. demand of the animal. A similar approach is also used to predict DMI in NorFor. Each individual A similar approach is also used(FV; to predict Each individual is capacity assigned a basic feed is assigned a basic fill value sectionDMI 6.2) in andNorFor. the animal is assigned an feed intake fill value (FV; Section 6.2) and the animal is assigned an intake capacity (IC) expressed in the same (IC) expressed in the same units as the feed. When predicting DMI, the following general units as the When predicting DMI, the following general equation must be fulfilled: equation mustfeed. be fulfilled: IC= FV_ intake

[Eq.10.1]

10.1

whereIC ICisisthe theanimal animalintake intake capacity FV_intake is total the total intake expressed where capacity andand FV_intake is the feedfeed intake expressed in fill in fill units, whichwhich is calculated as described in Section 6.2. units, is calculated as described in section 6.2. Despite thethe feed FVFV andand thusthus the FV_intake is notisconstant; it varies Despitethis thisgeneral generalrelationship, relationship, feed the FV_intake not constant; it varies with with the concentrate substitution rate(SubR), (SubR),which which affects affects roughage if the the concentrate substitution rate roughageintake. intake.Moreover, Moreover, if the energy energy concentration the ration is high, relative the animal’s requirements, above concentration of the of ration is high, relative to thetoanimal’s requirements, the the above general equation general equation will overestimate feed intake. An MR factor should therefore be included in the will overestimate feed intake. An MR factor should therefore be included in the equation when equation when formulating rations to meet specific animal production levels (e.g. weight gain or formulating rations to meet specific animal production levels (e.g. weight gain or milk yield). milk yield).

In NorFor, NorFor,the theICICofofeach each animal category estimated feeding experiments in which In animal category has has beenbeen estimated from from feeding experiments in which animals were fed total mixed rations ad libitum or fed only roughage ad libitum with fixed animals were fed total mixed rations ad libitum or fed only roughage ad libitum with fixed levelslevels of concentrate. tried to toevaluate evaluatehow howdifferent different types of roughage of concentrate.These Theseexperiments experiments have tried types of roughage are are utilized and also the interaction between roughage and differentcategories categories utilized and also the interaction between roughage andconcentrate concentrate by by different of of animals. From each experiment, the FVs of the different feeds were calculated assuming that the FV_intake represented the IC of the animals.

10.1. Dairy cows

149

A database of 183 dietary treatments was compiled from Norwegian, Danish, Swedish and Icelandic production experiments. (Table 10.1) in which grass or maize silages constituted the sole or dominant forage. The breeds represented in the database were the dominant Nordic dairy breeds, i.e. DH, SR, SH, NR and IB (Table 3.2). Animal and feed characteristics for the large dairy breeds, JER and IB are presented in Table 10.2, 10.3 and 10.4, respectively. The total DMI varied from 9.7 to 32.3 kg/d H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_10, © Wageningen Academic Publishers 2011

113

Table 10.1. Studies used to develop the feed intake module for dairy cows in NorFor. Bertilsson and Murphy, 2003 Bossen et al., 2010a Bossen et al., 2010b Hetta et al., 2007 Hymøller et al., 2005 Johansen, 1992 Kristensen, 1999 Kristensen et al., 2003 Kristensen et al., unpublished data Misciattelli et al., 2003 Mo and Randby, 1986 Mould, 1996 Nordang and Mould, 1994 Randby, 1992

Randby, 1997 Randby, 1999a Randby, 1999b Randby, 2000 Randby, 2002 Randby, 2007 Randby and Mo, 1985 Randby and Selmer-Olsen, 1997a Randby et al., 1999 Rikarðsson, 1997 Rikarðsson, 2002 Schei et al., 2005 Sveinbjörnsson and Harðarsson, 2008 Weisbjerg et al., unpublished data

Table 10.2. Descriptive statistics of data used to develop the equation for predicting intake capacity of dairy cows: large dairy breeds. Item

Average

SD1

Maximum

Minimum

Days in milk Body weight, kg ECM2, kg/d Dry matter intake, kg/d Concentrate proportion, kg/kg DM Roughage intake, kg DM/d Roughage basis fill value, FV/kg DM Starch + sugars, kg/d

123 577 29.9 19.7 0.42 11.2 0.51 2.7

65 59 5.8 3.4 0.101 2.3 0.031 0.7

328 775 48.9 32.3 0.84 19.4 0.76 5.0

7 437 15.1 9.7 0.14 3.7 0.42 1.5

1 SD=standard

2 ECM=energy

deviation. corrected milk.

Table 10.3. Descriptive statistics of data used to develop the equation for predicting intake capacity of dairy cows: Jersey cows. Item

Average

SD1

Maximum

Minimum

Days in milk Body weight, kg ECM2, kg/d Dry matter intake, kg/d Concentrate proportion, kg/kg DM Roughage intake, kg DM/d Roughage basis fill value, FV/kg DM Starch + sugars, kg/d

119 458 25.8 16.0 0.40 9.6 0.44 3.6

77 43 6.0 2.8 0.09 2.3 0.020 0.7

294 598 47.2 23.8 0.59 18.7 0.48 5.9

7 308 4.0 5.3 0.11 3.1 0.41 1.2

1SD=standard

2ECM=energy

114

deviation. corrected milk.

NorFor – The Nordic feed evaluation system

Table 10.4. Descriptive statistics of data used to develop the equation for predicting intake capacity of dairy cows: Icelandic cows. Data from Baldursdóttir (2010). 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000

1

Item From each experiment, the FVs of the different Averagefeeds were SD calculatedMaximum animals. assuming thatMinimum the FV_intake represented the IC of the animals.

Days in milk 58 34 147 7 Body weight, kg 439 47 619 300 10.1. ECM2Dairy , kg/d cows 20.4 4.7 41.0 7.6 Dry matterofintake, kg/d treatments was compiled15.1 2.1 Danish, Swedish 25.3 and 7.1 A database 183 dietary from Norwegian, Icelandic production experiments. (Table 10.1) in which maize silages constituted the 0.28 Concentrate proportion, kg/kg DM 0.41 grass or0.09 0.49 sole or dominant forage. The breeds represented in the Roughage intake, kg DM/d 9.0 database were 1.8 the dominant 15.6Nordic dairy4.0 breeds, i.e., basis DH, SR, NRFV/kg and IBDM (Table 3.2). Animal characteristics0.62 for the large 0.48 Roughage fill SH, value, 0.53 and feed 0.025 dairy breeds, JER kg/d and IB are presented in Table 10.2, The total DMI Starch + sugars, 2.510.3 and 10.4, 0.5 respectively.3.1 0.5

varied from 9.7 to 32.3 kg/d and the concentrate intake from 2.7 to 19.8 kg DM/d. The roughage FV ranged from 0.42 to 0.76 per kg DM. A multiple regression approach was used to derive the 1SD=standard deviation. equation for predicting animal IC. 2 ECM=energy corrected milk.

Table 10.1 Table 10.2 and the concentrate intake from 2.7 to 19.8 kg DM/d. The roughage FV ranged from 0.42 to 0.76 Table 10.3 per kg10.4 DM. A multiple regression approach was used to derive the equation for predicting animal IC. Table

Based 1995), thethe following modified multiple regression Basedon onthe thework workofofKristensen Kristensen(1983, (1983, 1995), following modified multiple regression equation equation used to describe IC: was usedwas to describe IC: IC _ cow = ( a ⋅ DIM b ⋅ e c ⋅ DIM − DIM − d + e ⋅ ECM + ( BW − f ) ⋅ g )

10.2

[Eq. 10.2]

where intake capacity of lactating dairy dairy cows; cows; DIM isDIM days is in milk; ECM is the whereIC_cow IC_cowisisthethe intake capacity of lactating days in milk; ECM is the energy is is thethe animal body weight, kg; and a, b, a, c, b, d, c, e, d, f, ge,are energycorrected correctedmilk, milk,kg/d; kg/d;BW BW animal body weight, kg; and f, g are regression regression coefficients Table 10.3). coefficients (see Table(see 10.5). ® The et et al.,al., 1998) in Microsoft Excel, whichwhich employs a generalized reduced reduced ® Excel, The Solver Solvertool tool(Fylstra (Fylstra 1998) in Microsoft employs a generalized gradient non-linear optimization code (Lasdon et al., 1978; Sveinbjörnsson et al., 2006), was gradient non-linear optimization code (Lasdon et al., 1978; Sveinbjörnsson et al., 2006), was used used to parameterize the equation and fit regression coefficients to the above model by to parameterize themean equation and fit regression coefficients the above model by minimizing the minimizing the root square prediction error (RMSPE) for to animal IC. The regression root mean square prediction error (RMSPE) for animal IC. The regression coefficients for predicting coefficients for predicting the IC of lactating dairy cows are presented in Table 10.5.

the IC of lactating dairy cows are presented in Table 10.5.

Table 10.5

Figure 10.1 illustrates how the IC coefficients of large dairyused breeds varies with lactation period, with a 305-d Table 10.5. Multiple regression to predict dairy cow intake capacity (IC). yield potential of between 6 000 and 10 000 kg ECM. For Icelandic cows the same approach as 1 in Table 10.5 is from described above was used to parameterize the IC,regression and the parameter values Cow category Multiple coefficients the work of Baldursdóttir (2010).

a

b

c

d

e

f

-0.0006 -0.0006 -0.0004 -0.0004 -0.0014 -0.0013

0.55 0.55 0.25 0.15 0.65 0.14

0.091 500 0.091 575 [Eq. 10.3] 0.110 360 0.110 405 0.015 400 0.035 523

The DMI is also affected by animal activity, and when cows are on pasture, or confined in a loose-house system, an IC exercise value is added (Kristensen, 1995):

Primiparous large dairy breeds Multiparous large breeds ) ⋅ factorR IC = (IC _ animal + ICdairy _ exercize Jersey primiparous cows Jersey multiparous cows Icelandic primiparous cows Icelandic multiparous cows

1

2.59 2.82 2.25 2.40 4.07 4.77

0.134 0.134 0.134 0.134 0.087 150 0.071

g 0.006 0.006 0.006 0.006 0.002 0.0013

IC=(a·DIMb·ec·DIM – DIM–d + e·ECM + (BW – f)·g).

NorFor – The Nordic feed evaluation system

115

4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043

used to parameterize the equation and fit regression coefficients to the above model by minimizing the root mean square prediction error (RMSPE) for animal IC. The regression coefficients for predicting the IC of lactating dairy cows are presented in Table 10.5. Table 10.5

Figure 10.1 illustrates how the IC of large dairy breeds varies with lactation period, with a 305-d

Figure 10.1 illustrates how the6,000 IC ofand large10,000 dairy breeds varies with lactationcows period, a 305-d yield potential of between kg ECM. For Icelandic thewith same approach as yield potential of between 6 000 and 10 000 kgthe ECM. For Icelandic cows the sameinapproach as is from described above was used to parameterize IC, and the parameter values Table 10.5 described above was used to parameterize the IC, and the parameter values in Table 10.5 is from the work of Baldursdóttir (2010). the work of Baldursdóttir (2010).

The DMI DMIisisalso alsoaffected affected by ⋅animal when arepasture, or confined (0.45activity, ) ⎞or confined − activity, 0.09 ⋅ pH +when ACF ⋅ (1.cows 5− 0.223 ⋅on pHpasture, ⎛ LAF The and)and cows are on in a in a loose⎟ ⎜ by animal house system, an ICanexercise value is added (Kristensen, 1995): loose-house system, IC exercise value is added (Kristensen, 1995): [Eq. 5.5] DM corr = DM uncorr + ⎜ + PRF ⋅ (1.4 − 0.182 ⋅ pH) + BUF ⋅ (1.9 − 0.272 ⋅ pH)⎟ ⎟ ⎜ ⎟ ⎜ ) ⋅ factorR IC = (IC _ animal + is ICeither _ exercize [Eq. 10.3] ⋅ 0.6in Equation 10.2 or IC_dry described 10.3 ALF + NH 3N where IC_animal described in Equation ⎠ ⎝ +IC_cow 10.5 or 10.6; IC_exercize is 0 if the cows are tied up and 0.15 if the cows are in a loose housing where IC_animal is either IC_cow in Equation 10.2 or IC_dryindescribed Equation 10.5 system or on pasture; factorR is an addescribed libitum correction factor as described Equationin 10.4. or 10.6; IC_exercize is 0 if the cows are tied150 up and 0.15 if the cows are in a loose housing system Figure 10.1 or on pasture; factorR is an ad libitum correction factor as described in Equation 10.4. ⎛ ACF ⋅ (1.5 − 0.223 ⋅ pH ) + PRF ⋅ (1.4 − 0.182 ⋅ pH )⎞ ⎟ 5.6] = DMfed +⎜ DM corrare If libitum, e.g.e.g. , in in situations where roughage intake is not optimal, due to due to[Eq. ⎜ + BUF If cows cows arenot notuncorr fedadad libitum, situations not optimal, feed )where ⋅ (1.4do − 0not ⋅ pHmaximal + ALFroughage +roughage .182 NH 3 N ⋅ intake 0intake, .6 ⎟⎠ is ⎝ feed shortage or feeding systems that allow an adjustment factor shortage or feeding systems that do not allow maximal roughage intake, an adjustment factor is is introduced: introduced: ⎛ roughage _ appetite ⎞ − 13 ⎟ + 0.8502 factorR = 0.0214 ⋅ ⎜ 5 ⎝ ⎠

[Eq.10.4]

sid _ mAA = r _ mAA ⋅ 0.85

10.4

[Eq. 7.43]

where factorR is an adjustment factor for the intake capacity; roughage_appetite is the appetite

where factorR is an adjustment factor for the intake capacity; roughage_appetite is the appetite proportion,%%ofofexpected expected libitum intake. adad libitum intake. proportion,

In non-lactating non-lactatingcows, cowsone an of equation on the work of based Kristensen used to predict IC: In the twobased alternative equations, on the (1995) work ofisKristensen (1995) are used to predict IC: IC _ dry = (5 + ((BW − f ) ⋅ 0.006)) ⋅ factor _ h IC _ dry = (5 + ((BW − 575) ⋅ 0.006)) ⋅ factor _ h

[Eq.10.5]

[Eq.10.5] 10.5

where IC_dry is the intake capacity of non-lactating cows, FV/d; BW is the body weight, kg; f is the [Eq.10.6] Jersey=0.83). IC _ dry = (factor _ i f+in (YHerd 6500)factor_h ⋅ 0.0003 + (is BW f ) ⋅ 0.006 ) ⋅ factor _ hfactor (large breeds=1; regression factor Table−10.5; the−breed correction DM _ int ake ⎛ ST _ SUof_ non-lactating ⎞ ⎛ ST _isSU ⎞ where IC_dry the+ intake factor_i the age and−gestation = 0is.97 − 0.2119 cows, ⋅⎜ FV _ SubR 0.562 ⋅ capacity 0.1932 5.122 ⎟ ⋅ 0.05 ⎜ ⎟ ⋅ 0.1 −FV/d; 1000 1000 ⎝ ⎠ ⎝ ⎠ stage factor (less than 260 days of gestation for primiparous cows = 5.3; multiparous cows = 5.8; 10.0 >260 days of gestation: primiparous cows = 4.8; multiparous cows = 5.3);IC BW is the body 10,000 kg ECM weight, kg; f 9.0 is the regression factor f in table 10.5; factor_h is the breed correction factor (large IC 6,000 kg ECM breeds = 1; Jersey = 0.83). YHerd is the yield level in the herd, 305-d average. [Eq.10.8] 8.0 When DMI is calculated, the IC is set equal to the FV_intake. Several factors affect the 7.0 FV_intake and it isPage generally accepted that increases in concentrate supplementation decrease the Equations on 85 in the proof: roughage intake 6.0 (Thomas, 1987). This is not only due to the filling effect of additional concentre but also by the accompanying reduction in ruminal NDF digestibility (Khalili and Huhtanen, 1991; Huhtanen 5.0 and Jaakkola, 1993; Stensig et al., 1998), which results in a higher roughage FV. 4.184FVk mg However, the concentrate substitution rate is not a constant (INRA, 1989); the true roughage ⋅ which for cows factor_1= 0.29256, for heifers and steers factor _ 1 = 90 ⋅ feed but 4.0 size varies with ration and composition. Moreover, MR is not related to each individual 1000 k m , for bulls instead to the entire ration. Bosch et al. (1992), Rinne et al. (2002) and Eriksen and Ness (2007) 3.0 k mgsize, indicating that the DMI is .184 pool have shown that cows do not eat to a constant ruminal4NDF factor _ 1concentration = 91⋅ ⋅ be used metabolically regulated and that dietary NDF can an indirect of of dairy breeds or crossbreed bulls ofmeasure beef breeds 2.0 1000 k m andasfor MR (Mertens, 1987; The same that was the development of animal 0 1994). 25 50 75 dataset 100 125 150 used 175in200 225 250 275 300 IC was k mg also used to establish a relationship between FV_intake and the effects of SubR and MR. When 4.184 Days factorinto _ 1 =account 100 ⋅ SubR⋅and MR, ; the 4.184/1000 converts to MJ;equation: the km is used to taking the factor FV_intake can in bemilk calculated fromkcal the general 1000 k m

Intake capacity (IC)

3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4001 4000 4002 4003 4004 4005 4006 4007 4008 4009 4010

Figure 10.1. throughout a 305-d lactation period at two production levels calculate FV_ intake = ∑Feed DMIi ⋅intake FVi + ∑capacity DMIj ⋅ FV(IC) [Eq.10.7] j ⋅FV_ SubR + FV_ MR (6,000 and 10,000 kg ECM). i j

4044 151

116

NorFor – The Nordic feed evaluation system

4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4045 4042 4046 4043 4047 4048 4044 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086

where IC_dry is the intake capacity of non-lactating cows, FV/d; factor_i is the age and gestation stage factor (less than 260 days of gestation for primiparous cows = 5.3; multiparous cows = 5.8; >260 days of gestation: primiparous cows = 4.8; multiparous cows = 5.3); BW is the body weight, kg; f is the regression factor f in table 10.5; factor_h is the breed correction factor (large breeds = 1; Jersey = 0.83). YHerd is the yield level in the herd, 305-d average.

When DMI is calculated, the IC is set equal to the FV_intake. Several factors affect the FV_intake

and it DMI is generally accepted increases supplementation decrease the roughage When is calculated, the ICthat is set equal to in theconcentrate FV_intake. Several factors affect the intake (Thomas, Thisaccepted is not only to the in filling effect ofsupplementation additional concentrate FV_intake and it is1987). generally thatdue increases concentrate decrease but the also by roughage intake (Thomas, 1987). is not onlydigestibility due to the filling effect of Huhtanen, additional concentre the accompanying reduction in This ruminal NDF (Khalili and 1991; Huhtanen but by the accompanying in ruminal NDF digestibility (Khalili and Huhtanen, andalso Jaakkola, 1993; Stensigreduction et al., 1998), which results in a higher roughage FV. However, the 1991; Huhtanen and Jaakkola, 1993; et al ., 1998), whichthe results a higherFV roughage concentrate substitution rate is not aStensig constant (INRA, 1989); true in roughage varies FV. with ration However, the concentrate substitution rate notrelated a constant (INRA, 1989); the FVthe entire size and composition. Moreover, MR is is not to each individual feedtrue butroughage instead to varies with ration size and composition. Moreover, MR is not related to each individual feed but ration. Bosch et al. (1992), Rinne et al. (2002) and Eriksen and Ness (2007) have shown that cows instead to the entire ration. Bosch et al. (1992), Rinne et al. (2002) and Eriksen and Ness (2007) do not eat to a constant ruminal NDF pool size, indicating that the DMI is metabolically regulated have shown that cows do not eat to a constant ruminal NDF pool size, indicating that the DMI is and that dietary NDF concentration be⋅ (used measure of MR (Mertens, 1987; ) + ACF ) ⎞ ⋅ (0.45 − 0.09 ⋅ pHcan 1.5 − 0.as 223an ⋅ pHindirect ⎛ LAF metabolically regulated ⎜ and that dietary NDF concentration can⎟be used as an indirect measure of 1994). The same dataset that was used in the development of animal IC wasofalso used establish a [Eq. 5.5] MR (Mertens, 1987; 1994). The same dataset that was used in the development animal ICtowas ⎟ ⎜ ( ) ( ) DM corr = DM uncorr + + PRF ⋅ 1.4 − 0.182 ⋅ pH + BUF ⋅ 1.9 − 0.272 ⋅ pH ⎟ ⎜ relationship between FV_intake and the effects of SubR and MR. When taking into account SubR also used to establish⎜ a relationship between FV_intake and the ⎟effects of SubR and MR. When + NH3 N ⋅ 0.6 ⎠equation: ⎝ + ALF and MR, FV_intake can be calculated fromcan thebe general taking intothe account SubR and MR, the FV_intake calculated from the general equation: where FV_intake is the feed intake expressed in fill units; FVi is the fill value of the i’th concentrate FV/kg DM; FVj is the basic fill value of the j’th roughage, Equation 6.10 and FV_ intake =feed, ∑ DMI [Eq.10.7] 10.6 i ⋅ FVi + ∑ DMIj ⋅ FVj ⋅FV_ SubR + FV_ MR 6.11; FV_SubR rate factor, 0 to1, Equation 10.8 and 10.15; FV_MR is the the i is the substitution j metabolic regulation ⎛factor, 10.9 and ⋅10.16. ) + PRF (1.4 − 0.182 ⋅ pH )⎞ is the fill value of the i’th concentrate ACF ⋅ (1Equations .5 − 0.223 ⋅ pH where ⎟i [Eq. 5.6] = DM uncorr is + ⎜the feed intake expressed in fill units; FV DM corrFV_intake ⎜ + BUF ⋅ (1.4 − 0.182 ⋅ pH ) + ALF + NH N ⋅ 0.6 ⎟

feed, FV/kg DM; the related basis fill value of the j’th Equation 6.10 and 6.11; FV_SubR 3 roughage, ⎝ j is ⎠ Traditionally, SubRFV has been to the effect 151of concentrate per se. However, it may be is the substitution rate factor, 0 to1, Equation and 10.14; FV_MR is the metabolic difficult to define a feedstuff as strictly ‘roughage’10.7 or ‘concentrate’. In NorFor, variations in theregulation factor, Equations 10.8 and 10.15. SubR are explained by changes in the NDF digestion in the rumen and the effect of rapidly degradable CHO’s on digestion in the rumen. This is similar to the effect used to explain the [Eq. sid _ mAA =ofr ruminal _ mAA .85 been correction NDF digestibility, RLI of (Equation 7.15)per in Chapter 7. Therefore, Traditionally, SubR⋅ 0has related toi.e. thethe effect concentrate se. However, it 7.43] may the be difficult suitability includingas dietary ST‘roughage’ and SU in the of theInsubstitution factor (FV_SubR) to define aoffeedstuff strictly or estimation ‘concentrate’. NorFor, variations in the SubR are was assessed, ExcelinSolver tool to develop a SubR function based on CHO’s explained by employing changes inthe theMicrosoft® NDF digestion the rumen and the effect of rapidly degradable the and SU supply and its effect on roughage FV:to explain the correction of ruminal NDF on dietary ruminalSTdigestion. This is similar to the effect used

digestibility, i.e. the RLI (Equation 7.15) in Chapter 7. Therefore, the suitability of including dietary _ SU _ DM_ h ⎞ ⎛ ST _ SU _ int ake ⎞ IC __dry + ((+in BW − f )⋅ ⎛⎜⋅estimation 0ST .006 )) ⋅ factor FV SR ==(SU 05.97 0.562 .2119 − 5.122 05 [Eq.10.5] ⎟ ⋅ 0.1 − 0.1932 ⋅factor ⎜ ⎟ ⋅ 0.assessed, ST and the of− 0the substitution (FV_SubR) was employing the 1000 1000 ⎝ ⎠ ⎝ ⎠ ® Microsoft Excel Solver tool to develop a SubR function based on the dietary ST and SU supply [Eq.10:8] and its effect on roughage FV:

where FV_SubR is the ⎛roughage substitution⎞ correction factor, 0 to 1; ST_SU_DM is the ST _ SU _ DM ⎛ ST _ SU _ int ake ⎞ 5.122 + 0.562 ⋅ ⎜ sugars in the−diet, ⋅⎜ −the FV _ SubR =of 0.97 0.2119 .1 − 0.1932 ⎟ ⋅ 0DM; sugar ⎟ ⋅ 0.05and proportion starch and g/g and ST_SU_intake is starch 1000 1000 ⎝ ⎠ ⎝ ⎠ intake, g/d.

10.7

where FV_SubR is the roughage substitution correction factor, 0 to 1; ST_SU_DM is the proportion

Figure 10.2 illustrates how FV_SubR is affected by the amount and starch proportion ST +intake, SU in g/d. of starch and sugars in thethe diet, g/g DM; and ST_SU_intake is the and of sugar [Eq.10.8] the ration; an increased proportion of SU + ST in the diet has a negative effect on the roughage FV and 10.2 this effect is alsohow related to the SU +isST intake.by Thethe latter effectand may explain why Figure illustrates theproof: FV_SubR affected amount proportion of ST + SU in the Equations on Pageconcentrate 85 in the replacing a starchy with a soluble fibre-based concentrate has been have a FV and ration; an increased proportion of SU + ST in the diet has a negative effect found on thetoroughage highly variable effects on forage intake (Huhtanen, 1993; Van Vuuren et al., 1993; Petit and this effect is also related to the SU + ST intake. The latter effect may explain why replacing a starchy Tremblay, 1995; Huhtanen et al., 1995; Hymøller et al., 2005). concentrate with a soluble fibre-based concentrate has been found to 4have variable effects .184 ak highly mg factorPetit _ 1 = and 90 ⋅ Tremblay, ⋅ which for cows factor_1= 0.29256, for heifers andetsteers on forage intake (Huhtanen, 1993; Van Vuuren al., 1993; 1995; Huhtanen et , for bulls 1000 k Figure 10.2 m

al., 1995; Hymøller et al., 2005).

4.184 k mg factor _ 1 = 91 ⋅ causing ⋅ In MR isordefined as a regulatory factor cows to stop eating before reaching their of NorFor, dairy breeds crossbreed 1000 k m and for bulls of beef breeds In NorFor, is defined as a regulatory causingrelated cows factor to stop eating before reaching their full ruminal MR FV capacity. Physiologically, thisfactor is an animal rather than a direct k therefore 4.capacity. 184 mg Physiologically, ruminal effect, and it is expected to influence animal IC. However, in NorFor it is full ruminal FV this is an animal related factor rather than a direct factor _ 1 = 100 ⋅ ⋅ ; the factor 4.184/1000 converts kcal to MJ; the km is used to addedruminal to the feed of the equation whentocalculating forIC. computational k m expected effect, andside it is 1000 therefore influence DMI animal However, inreasons. NorForThe it isFV_MR added toisthe feed

calculated side of theas: equation when calculating DMI for computational reasons. The FV_MR is calculated as: calculate ⎛ ⎜ 2.530 FV _ MR = ⎜1.453 − ⎜ ⎛ (0.466 − FV _ r ) ⎞ ⎜ ⎜ 0.065 ⎟⎠ 1 + e⎝ ⎝

⎞ ⎟ ⎟ ⋅ IC ⎟ 8 ⎟ ⎠

[Eq.10.9]

where FV_MR is the roughage metabolic correction factor; FV_r is the mean basis roughage fill value in the diet, FV/kg DM, Equation 6.10; IC is animal intake capacity, Equation 10.3; and the IC/8 ratio is the adjustment factor for animal IC level.

NorFor – The Nordic feed evaluation system 152

10.8

117

1.2

1.1

1 FV_SubRr

Starch + sugar (g/g DM)

0.05

0.40 0.35 0.30 0.25 0.20 0.15 0.10

0.9

1

2

0.8 8 7 5 6 4 3 Starch + sugar (kg/d)

Figure 10.2. A principle figure describing the effect of starch + sugar intake and their proportion in the diet on roughage substitution rate (FV_SubR) factor. where FV_MR is the roughage metabolic correction factor; FV_r is the mean basis roughage fill value in the diet, FV/kg DM, Equation 6.10; IC is animal intake capacity, Equation 10.3; and the IC/8 ratio is the adjustment factor for animal IC level. Dividing IC by 8 adjusts the standard IC so that FV_MR can be used across dairy breeds. Mertens (1994) demonstrated that the threshold between an energy regulated intake and a fill limited intake was dependent on the milk yield. Maximum fill and minimum energy concentration was achieved at a higher NDF diet concentration (440 g/kg DM versus 320 g NDF/kg DM) for low-yielding cows (20 kg ECM) than for high yielding cows (40 kg ECM). Figure 10.3 shows the relationship between FV_MR and FV_r at two levels of IC (6 and 8, which represent different milk yields). At an FV_r of 0.48 the FV_MR is zero, and with a higher and increasing FV_r a low yielding cow will reach its IC relative faster than a high yielding cow. However, at low FV_r values an opposite effect is achieved, which probably is explained by a higher diet energy density and thus a metabolic regulation of the intake (Mertens, 1994). Parameterization of Equation 10.6 showed that FV_MR makes a highly significant contribution to the prediction of roughage intake. The RMSPE was reduced by 19% when the FV_MR was introduced into the FV_intake equation. The need for this correction factor demonstrates that ruminal fill is not the only factor limiting roughage intake, as there are other important metabolic regulations (Forbes, 1995). Since FV_MR is related to roughage FV, and thus to the concentration of NDF in roughage, the rate of rumen particle size reduction may also partly explain this correction factor, since physical particle breakdown is not accounted for when calculating roughage FV. The introduction of FV_SubR and FV_MR in the prediction of DMI demonstrates that roughage FV is not a constant. Ration formulation is therefore performed by an iteration process in NorFor, using a computer program capable of solving non-linear optimization problems (see Chapter 15). The intake sub-model is further evaluated in Chapter 14.

118

NorFor – The Nordic feed evaluation system

Metabolic regulation factor (FV_MR)

1.5 1.3

IC=8 IC=6

1.0 0.8 0.5 0.3 0.0 0.25 0.30 -0.3

0.35

0.40

0.45

0.50

0.55

0.60

0.65

0.70

0.75

-0.5 -0.8 -1.0 -1.3

Roughage basis fill value (FV_r)

Figure 10.3. Relationship between roughage basis fill value (FV_r) and metabolic regulation factor (FV_MR) at two levels of intake capacity (IC=6 and 8).

10.2 Growing cattle The feeding strategies used in Nordic beef production vary widely among countries, from an intensively reared beef animals fed up to 90% concentrates to an extensive system where only small amounts (0 to10%) of concentrates are used. Hence, there are substantial variations in diet composition, which makes it challenging to develop a voluntary feed intake system that is relevant to all production systems. To develop equations for predicting the IC of bulls, steers and heifers, a database was compiled from 32 Danish, Swedish and Norwegian experiments (Table 10.6), in which cattle were fed a wide range of diets, including 145, 36 and 115 treatments for bulls, steers and heifers, respectively. Animal and feed characteristics for the bull, steer and heifer datasets are presented in Tables 10.7, 10.8 and 10.9, respectively. In the bull dataset, DMI varied from 4.5 to 11.7 kg/d (concentrate from 0 to 6.8 kg DM/d) and ADG ranged from 429 to 1,578 g, while in the heifers dataset, ADG varied from 378 to 1196 g/d and the DMI from 3.3 to 11.4 kg/d. 4131 4131 4132 4132 4133 4133 4134 4134 4135 4135 4136 4136 4137 4137 4138 4138 4139 4139 4140 4140 4141 4141 4142 4142 4143 4143 4144 4144 4145 4146 4147 4148 4149 4150 4151

For all all three three animal categories, categories,variations variationsininIC IC wereexplained explainedby bychanges changesininBW BW andADG. ADG.The The For ® changes inand Forequations all threeanimal animal categories, variations in were IC were explained by BW and ADG. The IC ® Excel using the same IC were parameterized using the Solver tool in Microsoft IC equationswere wereparameterized parameterized using the Solver tool in in Microsoft Excel usingusing the same ® Excel equations using the Solver tool Microsoft the same approach approach as for the dairy cows. In bulls the IC is calculated from: approach as for the dairy cows. In bulls the IC is calculated from:

as for the dairy cows. In bulls the IC is calculated from: − 20 IC__ bull bull == 00..006544 006544⋅⋅BW BW++00..0007337 0007337⋅⋅ADG ADG++ − 20 IC .9552⋅ ⋅BW BW 00.9552

[Eq.10.10] 10.10] [Eq.

10.9

whereIC_bull IC_bullisisisthe the feed intake capacity of bulls; BW the animal body weight, and ADG is where IC_bull the feed intake capacity ofbulls; bulls; BWis isthe theisanimal animal bodyweight, weight, kg;and andkg; ADG where feed intake capacity of BW body kg; ADG the average daily gain, g/d. is the average daily gain, g/d. is the average daily gain, g/d. The steer dataset included data obtained from only 3636 dietary treatments andand preliminary Thesteer steerdataset datasetincluded includeddata dataobtained obtained from only dietary treatments a preliminary evaluation The from only 36 dietary treatments and a apreliminary evaluation showed that the IC ICand ofsteers steers and heifers werecomparable. comparable. These twodatasets datasets were combined evaluation showed the of and heifers were These two were showed that the ICthat of steers heifers were comparable. These two datasets were therefore therefore combined in in the theof parameterization ofderived IC.The TheIC ICwas was derived from thesedata datafrom: from: therefore combined parameterization of IC. derived these in the parameterization IC. The IC was from these datafrom from: ⎛ −−33 ⎞ ⎞⎟ ⋅ IC _ gest IC __ heifer heifer == ⎛⎜⎜00..007236 007236⋅⋅BW BW++00..0005781 0005781⋅ ⋅ADG ADG++ IC ⋅ IC _ gest 0 . 9552 BW⎟⎠ ⎠ ⎝ 0 . 9552 ⋅ ⋅BW ⎝

[Eq.10.11] 10.11] [Eq.

where IC_heifer IC_heifer is is the the feed feed intake intakecapacity capacityof ofsteers steersand andheifers; heifers;BW BWisisthe theanimal animalbody bodyweight, weight, and ADG ADG is is the the average average daily dailygain, gain,g/d. g/d.IC_gest IC_gestisisthe theintake intakecapacity capacityadjustments adjustmentsfor forheifers heifers kg; and due to gestation gestation described described in in Equation Equation10.12. 10.12.

NorFor – The Nordic feed evaluation system

The gestation gestation correction correction factor factorisiscalculated calculatedas: as:

10.10

119

Table 10.6. Studies used to develop the feed intake module for growing cattle in NorFor.

4131 4131 4132 4132 4133 4133 4134 4134 4135 4135 4136 4136 4137 4137 4138 4138 4139 4139 4140 4140 4141 4141 4142 4142 4143 4143 4144 4144 4145 4145 4146 4146 4147 4147 4148 4148 4149 4149 4150 4150 4151 4151 4152 4152 4153 4153 4154 4154 4155 4155 4156 4156 4157 4157 4158 4158 4159 4159 4160 4160 4161 4161 4162 4162 4163 4163 4164

Bulls

Steers

Heifers

Andersen et al., 1993a Andersen et al., 1993b Berg, 2004 Berg and Volden, 2004 Damgaard and Hansen, 1992 Jørgensen et al., 2007 Kirkland et al., 2006 Matre, 1984 Nadeau et al., 2002 Olsson and Murphy, unpublished data Selmer-Olsen, 1994 Therkildsen et al., 1998 Vestergaard et al., 1993b Vestergaard et al., unpublished data Nadeau, unpublished data

Matre, 1984 Matre, 1987 Randby, 1999c Randby and Selmer-Olsen, 1997b Rustas et al., 2003 Selmer-Olsen, 1994

Andersen et al., 2001 Havrevoll, unpublished data Hessle et al., 2007 Ingvartsen et al., 1988 Mäntysaari, 1993 Rustas, 2009 Sejrsen and Larsen, 1978 Olsson, unpublished data Selmer-Olsen, 1994 Vestergaard et al., 1993a

For all three animal categories, variations in IC were explained by changes in BW and ADG. The For all10.7. three Descriptive animal categories, variations IC were explained changesfor in predicting BW and ADG. Thecapacity Table statistics of datainused to develop theby equation intake IC equations were parameterized using the Solver tool in Microsoft®®Excel using the same IC equations were parameterized using the Solver tool in Microsoft Excel using the same of bulls. approach as for the dairy cows. In bulls the IC is calculated from: approach as for the dairy cows. In bulls the IC is calculated from:

Item Average − 20 − 20 IC _ bull = 0.006544⋅ BW + 0.0007337⋅ ADG + IC _ bull = 0.006544⋅ BW + 0.0007337⋅ ADG +0.9552 ⋅ BW 0.9552 Age at start of experiment, d 224 ⋅ BW

SD1

Maximum

[Eq. 10.10] [Eq. 10.10]

Minimum

109 366 83 Age at end of experiment, d 392 127 676 188 where IC_bull is the feed intake capacity of bulls; BW is the animal body weight, kg; and ADG where IC_bull isweight, the feedkg intake capacity of bulls;362 BW is the animal body 130 body weight, 634kg; and ADG 178 isAverage the average daily gain, g/d. is the average daily gain, g/d. Average live weight gain, g/d 1,165 222 1,578 429 The dataset included obtained from only 367.3 dietary treatments a preliminary Drysteer matter intake, kg/d data 2.1 and 11.7 4.5 The steer dataset included data obtained from only 36 dietary treatments and a preliminary evaluation showed thatkg theDM/d IC of steers and heifers were comparable. These two datasets were 0 Concentrate intake, 3.0 1.9 6.8 evaluationcombined showed that theparameterization IC of steers andof heifers were Thesethese two datasets were therefore in the IC. The IC comparable. was derived from data from: Concentrate proportion, kg/kg DM 0.41 0.29 therefore combined in the parameterization of IC. The IC was derived from these0.95 data from: 0 Roughage intake, kg DM/d 4.3 2.1 9.5 0.2 −3 ⎞ ⎛ IC _ heifer = ⎜⎛basis 0.007236 0005781DM + 0.FV/kg ⋅ ADG + ⎟ ⎞⋅ IC _ gest 0.06 [Eq. 10.11] Roughage fill⋅ BW value, 0.68 0.41 − 3 0.54 [Eq. 10.11] 0.9552 ⋅ BW IC _ heifer = ⎝⎜ 0.007236 ⋅ BW + 0.0005781 ⋅ ADG + ⎠ ⎟ ⋅ IC _ gest 0.9552 ⋅ BW ⎠ ⎝

1

SD=standard deviation. where IC_heifer is the feed intake capacity of steers and heifers; BW is the animal body weight, where is the feed intake capacity of steersisand BW is the animal body kg; andIC_heifer ADG is the average daily gain, g/d. IC_gest theheifers; intake capacity adjustments for weight, heifers kg; and ADG is described the gain, g/d. IC_gest is the capacity heifers where IC_heifer is average the feed intake capacity of steers andintake heifers; BW isadjustments the animalfor body weight, kg; due to gestation in daily Equation 10.12. due gestation Equation 10.12. and to ADG is thedescribed average in daily gain, g/d. IC_gest is the intake capacity adjustments for heifers due The gestationdescribed correctionin factor is calculated to gestation Equation 10.11. as: The gestation correction factor is based on Kristensen and The gestation correction factor is calculated as: Ingvartsen (2003) and calculated as: 1 IC _ gest = [Eq. 10.12] 1 ⎛ gest _ day − 357 ⎞ IC _ gest = 10.11 [Eq. 10.12] ⎜⎛ gest _ day − 357⎟⎞ 36 ⎠⎟ 1 + e ⎝⎜ 36 ⎠ 1 + e⎝ where gestation correction on the capacity; gest_day is the day of gestation. whereIC_gest IC_gestisisthethe gestation correction on intake the intake capacity; gest_day is the day of gestation. where IC_gest is the gestation correction on the intake capacity; gest_day is the day of gestation. Evaluation 10.10 andand 10.11 showed that they for the Jersey breed. EvaluationofofEquations Equations 10.9 10.10 showed thatwere theyunsuitable were unsuitable for the Jersey breed. Evaluationthe of Equations 10.10 andwas 10.11 showed theyyoung were unsuitable for the Jersey Therefore, following equation modified forthat Jersey stock, irrespective of sex:breed. Therefore, the following equation was modified for Jersey young stock, irrespective of Therefore, the following equation was modified for Jersey young stock, irrespective of sex: sex: −2 IC _ jersey = 0.0067 ⋅ BW + 0.0005781⋅ ADG + 10.12 [Eq. 10.13] −2 IC _ jersey = 0.0067 ⋅ BW + 0.0005781⋅ ADG +0.9552 ⋅ BW [Eq. 10.13] 0.9552 ⋅ BW where IC_jersey is the feed intake capacity of Jersey young stock; BW is the animal body weight, 120and IC_jersey NorFor – The Nordic feed evaluation where is the feed daily intakegain, capacity stock; BW is the animal body weight,system kg; ADG is the average g/d. of Jersey young kg; and ADG is the average daily gain, g/d. The IC of growing cattle is also corrected for activity:

Table 10.8. Descriptive statistics of data used to develop the equation for predicting intake capacity of steers. 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170

1

Item Average SDchanges in BW Maximum For all three animal categories, variations in IC were explained by and ADG.Minimum The IC equations were parameterized using the Solver tool in Microsoft® Excel using the same approach as for dairy cows. from: Age at start of the experiment, d In bulls the IC is calculated 237 125 450 125 Age at end of experiment, d 349 127 555 208 − 20 Average body weight, kg 319 104 470 158 IC _ bull = 0.006544⋅ BW + 0.0007337⋅ ADG + [Eq. 10.10] Average live weight gain, g/d 0.9552 748 ⋅ BW 252 1,196 187 Dry matter intake, kg/d 6.3 1.8 9.5 4.1 where IC_bullintake, is the feed intake capacity of bulls; BW kg; and ADG0 Concentrate kg DM/d 1.0is the animal 0.9body weight,2.6 isConcentrate the average proportion, daily gain, g/d. kg/kg DM 0.21 0.18 0.54 0 Roughage intake, kg DM/d 5.3 2.5 9.5 2.2 The steer dataset included data obtained from only 36 dietary treatments and a preliminary Roughage basis fill value, FV/kg DM 0.58 0.07 0.67 0.48 evaluation showed that the IC of steers and heifers were comparable. These two datasets were therefore combined in the parameterization of IC. The IC was derived from these data from: 1 SD=standard deviation.

−3 ⎞ ⎛ IC _ heifer = ⎜ 0.007236 ⋅ BW + 0.0005781 ⋅ ADG + ⎟ ⋅ IC _ gest 0.9552 ⋅ BW ⎠ ⎝

[Eq. 10.11]

Table 10.9. Descriptive statistics of data used to develop the equation for predicting intake capacity of heifers. where IC_heifer is the feed intake capacity of steers and heifers; BW is the animal body weight, kg; and ADG is the average daily gain, g/d. IC_gest is the intake capacity adjustments for heifers due Itemto gestation described in Equation 10.12. Average SD1 Maximum Minimum The correction factor Agegestation at start of experiment, d is calculated as: 368

124 491 122 Age at end of experiment, d 468 103 634 310 1 Average 379 108 598[Eq. 10.12] 150 IC _ gest =body weight, kg day −g/d 357 ⎞ ⎛ gest _gain, Average live weight 798 205 1,196 378 ⎜ ⎟ 36 ⎠ Dry matter1intake, 7.3 1.6 11.4 3.3 + e ⎝ kg/d Concentrate intake, kg DM/d 1.3 1.3 5.1 0 where IC_gestproportion, is the gestation correction capacity; gest_day Concentrate kg/kg DM on the intake 0.19 0.20 is the day 0.75of gestation.0.0 Roughage intake, kg DM/d 6.0 2.1 11.4 1.2 Evaluation Equations 10.10 and 10.11 they were unsuitable for the0.67 Jersey breed.0.49 Roughageof basis fill value, FV/kg DM showed that 0.54 0.04 Therefore, the following equation was modified for Jersey young stock, irrespective of sex: 1

SD=standard deviation.

IC _ jersey = 0.0067 ⋅ BW + 0.0005781⋅ ADG +

−2 0.9552 ⋅ BW

[Eq. 10.13]

where IC_jersey is the feed intake capacity of Jersey young stock; BW is the animal body weight,

where IC_jersey is the feed intake capacity of Jersey young stock; BW is the animal body weight, kg; and andADG ADGisisthe theaverage average daily gain, kg; daily gain, g/d.g/d.

The IC ICofofgrowing growingcattle cattleis is also corrected activity: The also corrected for for activity: IC = IC _ animal ⋅ IC _ exercize

[Eq. 10.14]

10.13

where capacity; IC_animal is either IC_bull, IC_heifer or IC_jersey described in whereIC ICisisthe theintake intake capacity; IC_animal is either IC_bull, IC_heifer or IC_jersey described in Equation Equation10.10, 10.9, 10.11 10.10 and and10.13, 10.12,respectively. respectively.

Based on the range of animal characteristics in the datasets, it was decided that the IC equations are 154 with body weights of ≥80 kg, and for Jersey with valid for animals of large dairy and beef breeds body weights ≥60 kg. Figure 10.4 illustrates how the IC in growing bulls changes with BW and ADG over time up to the target slaughter weight of 300 kg at 16 months of age. The ADG changes in a curvilinear manner NorFor – The Nordic feed evaluation system

121

4191 4190 4192 4193 4194 4195 4191 4196 4192 4197 4193 4198 4194 4199 4195 4200 4196 4201 4197 4202 4198 4203 4199 4204 4200 4205 4201 4206 4202 4207 4203 4208 4204 4209 4205 4206 4210 4207 4208 4209 4210

1,400

5.0

Body weight ADG IC

4.5 4.0

1,200

3.5

1000

3.0 2.5

800

Intake capacity

Body weight (kg) or daily weight gain (g)

4171 4172 4173 4174 4175 4171 4176 4172 4177 4173 4178 4174 4179 4175 4180 4176 4181 4177 4182 4178 4183 4179 4184 4180 4185 4181 4186 4182 4187 4183 4188 4184 4189 4185 4186 4187 4190 4188 4189

1,600

Based on the range of animal characteristics in the datasets, it was decided that the IC 2.0equations 600 are valid for animals of large dairy and beef breeds with body weights of ≥ 80 kg, and for Jersey 1.5 with body weights ≥400 60 kg. 1.0 Figureon 10.4 how thecharacteristics IC in growinginbulls changes with and ADG over time up to 200 Based theillustrates range of animal the datasets, it wasBW decided that the IC equations 0.5 the valid targetfor slaughter ofdairy 300 kg at beef 16 months age.body Theweights ADG changes a curvilinear are animalsweight of large and breeds of with of ≥ 80inkg, and for Jersey 0 kg.highest ADG is achieved at an age of 330 days. Since the 0.0 manner with time and IC is with body weights ≥ 60the 91 and 122 ADG, 152 182 213 changes 243 274 over 304 time 334 365 426 456fashion. 486 dependent on both the BW it also in a 395 sigmoidal dayswith BW and ADG over time up to Figure 10.4 illustrates how the IC in growing bullsAge, changes Figure 10.4 the target slaughter weight of 300 kg at 16 months of age. The ADG changes in a curvilinear Figure 10.4. Changes inhighest intake ADG capacity (IC) over time bulldays. withSince a target slaughter weight of manner with time and the is achieved at an agefor ofa330 the IC is 300 kgdairy (587 kg body at 16 months of age.isover As for feed and intake in growing cattle dependent the SubR and MR effects. To dependent oncows, both the weight) BW ADG, it also changes time inon a sigmoidal fashion. parameterize these two variables for the growing cattle, the same approach was used as for lactating dairy However, in the cattle with time and cows. the highest ADGtheis concentrate achieved atcharacteristics an age of 330available days. Since thegrowing IC is dependent on both Figure 10.4 datasets incomplete, made it difficult develop a SubR equation based on starch and the BW were and ADG, it alsowhich changes over time intoa sigmoidal fashion. sugar Consequently, SubR equation developed from the information on the As forinformation. dairy cows, the feed intake the in growing cattle iswas dependent on the SubR and MR effects. To proportion of these in the diet. Thegrowing following equation is used for growing cattle, of all parameterize two for the cattle, theissame approach as and for As for dairy concentrate cows, thevariables feed intake in growing cattle dependent onwas theused SubR MR effects. types: dairy cows. However, the concentrate characteristics available in the growing cattle lactating To parameterize these two variables for the growing cattle, the same approach was used as for datasets were incomplete, which made it difficult to develop a SubR equation based on starch and lactating dairy cows. However, the concentrate characteristics available in the growing cattle 2 sugar information. Consequently, the SubR equation was developed from the information on the ⎛ ⎞ concit_ difficult share datasets were to develop a SubR equation based on starch and ⎜ incomplete, which made ⎟ 0 .9646 0 .706 The ⋅ proportion of⎜ concentrate in the− diet. following equation⎟ is used for growing cattle, of all 100 [Eq.information 10.15] sugar information. Consequently, the SubR equation was developed from the on the FV _ SubR = ⎜ types: 2 ⎟

⎜ concentrate concin_ the sharediet. The following ⎛ conc _ share ⎞ ⎟ proportion of equation is used for growing cattle, of all types: 1 − 0 .7512 ⋅ − 0 .2085 ⋅ ⎜ ⎟

⎜ ⎟ 100 100 ⎝ ⎠ ⎠2 ⎝ ⎛ ⎞ conc _ share ⎜ ⎟ 0 .9646 − 0 .706 ⋅ ⎜ ⎟ 100 correction factor, where FV_SubR is the roughage substitution FV _ SubR =⎜ 2 ⎟ conc _ share _ share proportion of⎜ 1concentrate in the diet on a DM basis, %. ⎛ conc ⎞ ⎟ − 0 .2085 ⋅ ⎜ ⎟ ⎟ ⎜ − 0 .7512 ⋅ 100 100 ⎝ ⎠ ⎠ ⎝

[Eq. 10.15] 0 to 1; and conc_share is the

10.14

Figure 10.5 shows how the FV_SubR factor changes with the proportion of concentrate in the where FV_SubR is the roughage substitution correction factor, 0 to 1; and conc_share is the proportion diet. The growingiscattle diets vary widely in the concentrate to 0roughage (0 to 90is%the where FV_SubR the diet roughage substitution correction factor, to 1; andratio conc_share of concentrate in the on a the DM basis, %. concentrate DM basis),inthus, correction proportion ofonconcentrate the diet FV_SubR on a DM basis, %. factor must be able to correct the roughage intake over a large range of concentrate intake values. The FV_SubR factor, and Figure10.5 10.5shows shows howthethe FV_SubR factor changes with theconcentrate proportion consequently, the roughage FV, increase considerably when inof theconcentrate diet exceeds Figure how FV_SubR factor changes with thethe proportion of concentrate in the in the diet. The The growing cattle diets vary widely in in thetheconcentrate totoroughage (0 to to 90% 60%. Thisgrowing indicates that the voluntary roughage intake is likely to be lowratio at high levels diet. cattle diets vary widely concentrate roughage ratio (0 90 %ofconcentrate on DM basis),on thus, FV_SubR correction mustfactor be able to correct roughage concentrate. concentrate DMthe basis), thus, the FV_SubRfactor correction must be able the to correct the intake over a large range of concentrate values. The FV_SubR factor, consequently, the roughage FV, roughage intake over a largeintake range of concentrate intake values. Theand FV_SubR factor, and Figure 10.5 increase considerably whenFV, theincrease concentrate in the diet exceeds 60%. Thisinindicates the voluntary consequently, the roughage considerably when the concentrate the diet that exceeds 60%. This intake indicates thetovoluntary intake likely to be low at high levels of roughage is that likely be low atroughage high levels of isconcentrate. The FV_MR significantly affected the prediction of DMI in growing cattle, and by including this concentrate. variable in the significantly prediction of FV_intake, MSPE wasofreduced 14% forcattle, the predicted The FV_MR affected thetheprediction DMI inbygrowing and by including this roughage intake. The FV_MR was parameterized independently of animal category and is Figure 10.5 variable in the prediction of FV_intake, the MSPE was reduced by 14% for the predicted roughage calculated as: intake. The FV_MR was parameterized independently of animal category and is calculated as: The FV_MR significantly affected the prediction of DMI in growing cattle, and by including this variable in the MSPE was reduced by 14% ⎛for ⎞ ⎛ prediction of FV_intake,1the .2158 IC the ⎞ predicted ⎟⎟ ⋅ 1 − e − 0.029474088⋅ BW ⋅ ⎜ ⎟ [Eq. 10.16] FV _ MR = ⎜⎜ − 3.8301 + 2.6608 ⋅ FV _ r + 10.15 roughage intake. The FV_MR was parameterized independently of animal category and is FV _ r 3 ⎝ ⎠ ⎠ ⎝ calculated as:

(

)

(

)

⎛ 1.2158 ⎞ 155 − 0.029474088⋅ BW ⎛ IC ⎞ ⎟ ⋅ 1− e FV _ MR = ⎜⎜ − 3.8301 + 2.6608 ⋅ FV _ r + ⋅⎜ ⎟ FV _ r ⎟⎠ ⎝ 3⎠ ⎝

122

[Eq. 10.16]

NorFor – The Nordic feed evaluation system 155

FV_SubR

6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0

0

10

20

30

40

50

60

70

80

90

100

Concentrate in diet, %

Figure 10.5. Effect of proportion of concentrate in the diet on roughage substitution rate in growing cattle. where FV_MR is the roughage metabolic correction factor; FV_r is the diet basis roughage fill value, Equation 6.10; IC is the animal intake capacity, Equation 10.13; BW is the animal body weight, kg; and the IC/3 ratio is an adjustment factor for animal IC level. The computer program TableCurve 2D from SYSTAT was used to resolve the equation profile and the FVL_intake equation was parameterized using the Solver tool in Microsoft® Excel, as described for the dairy cow model. Based on sheep data, Mertens (1994) demonstrated that the dietary NDF concentration at the intersect of metabolic and physical regulation is dependent on animal BW. This may explain why animal BW significantly affected the FV_MR in Equation 10.15. The IC/3 ratio in the equation reflects the different animal categories.

10.3 Implications A primary objective of this work has been to develop a system that can be used to formulate diets that maximize roughage intake while meeting animals’ energy requirements with a minimal dietary energy concentration. It is important to ensure that there is a high proportion of roughage in diets to maintain a favourable rumen environment. The feed intake sub-model in NorFor is complex and consists of several non-linear equations. This comprehensive approach requires DMI to be calculated iteratively by a computer program. Although the system is a simplification of numerous factors and animal-diet interactions affecting voluntary feed intake, it is applicable to diets with a wide range of roughage qualities and diverse feeding situations.

References Andersen, H.R., J. Foldager, P.S. Varnum and S. Klastrup, 1993a. Effects of urea and soybean meal at two levels in a whole crop barley silage ration on performance, carcass and meat quality in young bulls. Forskningsrapport nr. 4 fra Statens Husdyrbrugsforsøg, Denmark. (In Danish with English summary). Andersen, H.R., B.B. Andersen, P. Madsen, P.S. Varnum and L.R. Jensen, 1993b. Effects of maize silage plus urea or soybean versus concentrate on performance, carcass and meat quality of young bulls. Forskningsrapport nr. 5 fra Statens Husdyrbrugsforsøg, Denmark. (In Danish with English summary). Andersen, H.R., B.B. Andersen and H.G. Bang, 2001. Kødracekrydsninger: Kvier kontra ungtyre fodret med forskelligt grovfoder-kraftfoderforhold og slagtet ved forskellig vægt. DJF rapport Husdyrbrug nr. 28. 82 pp. (In Danish).

NorFor – The Nordic feed evaluation system

123

Baldursdóttir, R., 2010. Development of the feed intake capacity of Icelandic cows in the NorFor feed evaluation system. Master Thesis, Department of Animal and Aquacultural Sciences, Norwegian University of Life Sciences, Norway. 63 pp. Berg, J., 2004. Sluttrapport. Produksjonsegenskapene hos slakteokser av Charolais x NRF, Blond D’ Aquitaine x NRF og NRF ved to ulike fôrstyrkenivåer. Institutt for husdyr og akvakulturvitenskap, UMB. Intern rapport, pp. 11. (In Norwegian). Berg, J. and H. Volden, 2004. Økologisk storfekjøtt. Buskap nr 4. 16-17. (In Norwegian). Bertilsson J. and M. Murphy, 2003. Effects of feeding clover silages on feed intake, milk production and digestion in dairy cows. Grass and Forage Science 58: 309-322. Bosch, M.W., S.C.W. Lammers-Wienhoven, G.A. Bangma, H. Boer and P.W.M. Van Adrichem, 1992. Influence of stage of maturity of grass silages on digestion processes in dairy cows. 2. Rumen contents, passage rates, distribution of rumen and faecal particles and mastication activity. Livestock Production Science 32: 265-281. Bossen, D., M.R Weisbjerg, L. Munksgaard, and S. Højsgaard, 2010. Allocation of feed based on individual dairy cow live weight changes. I: Feed intake and live weight changes. Livestock Science 126: 252-272. Bossen, D. and M.R. Weisbjerg, 2010. Allocation of feed based on individual dairy cow live weight changes. II: Effect on milk production. Livestock Science 126: 273-285. Damgaard, P. and S. Hansen, 1992. Sammenligning af valset og pilleteret kraftfoder til ungtyre. LK-meddelelse nr. 189. 3 pp. (In Danish). Eriksen, J. and S. Ness, 2007. Effet av høstetid for surfôr og kraftfôrnivå på passasjekinetikk of NDF estimert med markørmetoden og vomtømmingsmetoden. Masteroppgave Universitetet for miljø- og biovitenskap. 87 pp. (In Norwegian). Forbes, J.M., 1995. Voluntary Food Intake and Diet Selection in Farm Animals. CAB International, Wallingford, UK. Fylstra, D., L. Lasdon, J. Watson and A. Waren, 1998. Design and use of the Microsoft Excel Solver. Interfaces 28: 29-55. Garmo, T.H., H. Volden, S. Krizsan and S.K. Nes, 2007. Effects of level of concentrate supplementation on nutrient digestion of lactating dairy cows grazing at two pasture allowances. Journal of Animal Science 85: 635, Supplement 1. Hessle, A., E. Nadeau and S. Johnsson, 2007. Beef heifer production as affected by indoor feed intensity and slaughter age when grazing semi-natural grasslands in summer. Livestock Science 111: 124-135. Hetta, M., J.W. Cone, G. Bernes, A.-M. Gustavsson and K. Martinsson, 2007. Voluntary intake of silages in dairy cows depending on chemical composition and in vitro gas production characteristics. Livestock Science 1006: 47-56. Huhtanen, P., 1993. The effects of concentrate energy source and protein content on milk production in cows given grass silage ad libitum. Grass and Forage Science 48: 347-355. Huhtanen, P. and S. Jaakkola, 1993. The effects of forage preservation method and proportion of concentrate on digestion of cell wall carbohydrates and rumen digesta pool size in cattle. Grass and Forage Science 48: 155-165. Huhtanen, P., S. Jaakkola and E. Saarisalo, 1995. The effects of concentrate energy source on the milk production of cows given a grass silage-based diet. Animal Science 60: 31-40. Hymøller, L., M.R. Weisbjerg, P. Nørgaard, C.F. Børsting and N.B. Kristensen, 2005. Majsensilage til malkekøer. DJF Rapport Husdyrbrug nr. 65. 71 pp. (In Danish). INRA (Institute National de la Recherche Agronomique), 1989. Feed intake: The fill unit system. In: Jarrige, R. (ed.). Ruminant Nutrition. Recommended allowances and feed tables. John Libbey and Co Ltd, London, UK, pp. 61-71. Ingvartsen, K.L., J. Foldager, J.B. Larsen and V. Østergaard, 1988. Growth and milk yield by Jersey reared at different planes of nutrition. Beretning nr. 645 fra Statens Husdyrbrugsforsøg. 72 pp. (In Danish with English summary and subtitles). Jarrige, R., 1986. Voluntary intake in cows and its prediction. International Dairy Federation Bulletin 196: 4-16. Johansen, A., 1992. A comparison between meadow fescue and timothy silage. Doctor Scientarium Thesis. Agricultural University of Norway, Aas, Norway. 108 pp. Jørgensen, K.F., J. Sehested and M. Vestergaard, 2007. Effect of starch level and straw intake on animal performance, rumen wall characteristics and liver abscesses in intensively fed Friesian bulls. Animal 1: 797-803. Kirkland R.M., T.W.J. Keady, D.C. Patterson, D.J. Kilpatrick and R.W.J. Steen, 2006. The effect of slaughter weight and sexual status on performance characteristics of male Holstein-Friesian cattle offered a cereal-based diet. Animal Science 82: 397-404. Kristensen, V.F., 1983. Styring av foderopptagelsen ved hjælp af foderrasjonens sammensætning og valg af fodringsprincip. In: Østergaard, V. and A. Neimann-Sørensen (eds.). Optimale foderrationer til malkekoen. Foderværdi, foderopptagelse, omsætning og produktion. Rapport nr. 551. Beretning fra Statens Husdyrbrugsforsøg. pp. 7.1-7.35. (In Danish).

124

NorFor – The Nordic feed evaluation system

Kristensen, V.F., 1995. Forudsigelsen af foderoptagelsen hos malkekøer. Intern rapport nr 61. Statens Husdyrbrugsforsøg. 28 pp. (In Danish). Kristensen, V.F., 1999. Grovfoderkildens betydning for malkekoens produktion og effektivitet. Temamøde vedr. Malkekøernes og kviernes fodring. Intern Rapport nr 118, Danmarks Jordbrugsforskning. pp. 18-33. (In Danish). Kristensen, V.F., M.R. Weisbjerg, C.F. Børsting, O. Aaes and P. Nørgaard, 2003. In Malkekoens energiforsyning og produktion. In: Strudsholm, F. and K. Sejrsen (eds.). Kvægets ernæring og fysiologi. Bind 2 - Fodring og produktion. DJF rapport nr. 54. pp. 73-112. (In Danish). Khalili, H. and P. Huhtanen, 1991. Sucrose supplements in cattle given grass silage-based diet. 2. Digestion of cell wall carbohydrates. Animal Feed Science and Technology 33: 263-273. Lasdon, J., A. Waren, A. Jain and M. Ratner, 1978. Design and testing of generalized reduced gradient code for nonlinear programming. ACM Transactions on Mathematical Software 4: 34-49. Mäntysaari, P., 1993. The effects of feeding level and protein source of the diet on growth and development at slaughter of pre-pubertal heifers. Acta Agriculturae Scandinavica. 43: 44-51. Matre, T., 1984. Kjøtproduksjon på oksar og kastratar. I. Forsøk med ulike proteinkjelder Institutt for husdyrernæring, Norges landbrukshøgskole. Melding nr 217. 19 pp. (In Norwegian). Matre, T., 1987. Kjøtproduksjon på oksar og kastratar. III. Ulik fôrstyrke og grovfôrkvalitet til kastratar. Institutt for husdyrernæring, Norges landbrukshøgskole. Melding nr 243. 29 pp. (In Norwegian). Mertens, D.R., 1987. Predicting intake and digestibility using mathematical models of ruminal function. Journal of Animal Science 64: 1548-1558. Mertens, D.R., 1994. Regulation of Forage Intake. In: Fahey Jr. G.C., M. Collins, D.R. Mertens and L.E. Moser (eds.). Forage quality, evaluation, and utilization. American Society of Agronomy, Inc., Crop Science Society of America, Inc. and Soil Science Society in America, Inc., Madison, WI, USA, pp. 450-493. Misciattelli, L., V.F. Kristensen, M. Vestergaard, M.R. Weisbjerg, K. Sejrsen and T. Hvelplund, 2003. Milk production, nutrient utilisation and endocrine responses to increased postruminal lysine and methionine supply in dairy cows. Journal of Dairy Science 86: 275-286. Mo, M. and Å. Randby, 1986. Wilted silage for lactating dairy cows. Proceedings 11th General Meeting of the EGF. Troia, Portugal 4-9. May. Mould, F., 1996. Fôropptak og produksjon hos melkekyr ved et lavt PBV I rasjonen. Husdyrforsølsmøtet. pp. 121127. (In Norwegian). Nadeau, E., A. Hessle, B.-O. Rustas and S. Johnsson, 2002. Prediction of silage intake by Charolais bulls. In: Durand, J-L., J-C Emile, C. Huyghie, and G. Lemaire (eds.). Multi-function grasslands quality forages, Animal products and landscapes. Grassland Science in Europe, Volume 7. Proceedings of the 19th General Meeting of the European Grassland Federation 27-30 May, La Rochelle, France, pp. 220-221. Nordang, L. and F. Mould., 1994. Surfôr med ulikt innhold av PBV til mjølkekyr. Husdyrforsøksmøtet 1994, pp. 99-103. (In Norwegian). Petit, H.V. and G.F. Tremblay, 1995. Ruminal fermentation and digestion in lactating cows fed grass silage with protein and energy supplements. Journal of Dairy Science 78: 342-352. Randby, Å. and M. Mo, 1985. Surfôr tilsatt foraform i sammenlikning med maursyresurfôr til melkekyr, Hellerud høsten 1984. M-141. Rapporter fra forsøksvirksomheten ved Institutt for husdyrernæring, s. 75-78. (In Norwegian). Randby, Å.T., 1992. Grass-clover silage for dairy cows. Proceedings of the 14th General Meeting, European Grassland Federation, Lahti, Finland, 8-11 June, pp. 272-275. Randby, Å.T., 1997. Feeding of silage effluent to dairy cows. Acta Agriculturae Scandinavica Section A. Animal Science 47: 20-30. Randby, Å.T. and I. Selmer-Olsen, 1997a. Formic acid treated or untreated round bale grass silage for dairy cows. 8th International Symposium Forage Conservation. Brno, Czech Republic, pp. 160-161. Randby, Å.T. and I. Selmer-Olsen, 1997b. Formic acid treated or untreated round bale grass silage for steers. Presented on Fifth Research Conference, BGS, University of Plymounth, Devon, UK, 8-10 Sept. 2 pp. Randby, Å.T., 1999a. Grass silage treated with two different acid-based additives for dairy cows. The Fifth International Symposium on the Nutrition of Herbivores. April 11-16, San Antonio, TX, USA. 8 pp. Randby, Å.T., 1999b. Effect of increasing levels of a formic acid based additive on ad libitum intake of grass silage and the production of meat and milk. Proceedings of the XII International Silage Conference, Uppsala, Sweden, pp. 205-206.

NorFor – The Nordic feed evaluation system

125

Randby, Å.T., 1999c. Virkning av økende dosering med GrasAAT på surfôropptak og produksjon av mjølk og kjøtt. Forsøk utført på Hellerud Forsøksgård i 1997/98. Institutt for husdyrfag, NLH. Rapport nr 24. Hellerud forsøks- og eliteavslgard. Det kgl. Selskap for Norges Vel. 24 pp. (In Norwegian). Randby, Å.T., 2000. Round bale grass silage treated with an additive containing both lactic acid bacteria and formate for dairy cows and steers. Sixth British Grassland Society Research Conference, SAC, Craibstone, Aberdeen, UK, pp. 121-122. Randby, Å.T., 2002. Silage fermentation by concentrate type interaction. Proceedings of the XIII International Silage Conference, Auchincruive, UK, pp. 312-313. Randby, Å.T., 2007. Effect of propanol and dimethylsulphide in grass silage on organoleptic milk quality. Journal of Animal and Feed Science 16, Suppl. 1, pp. 102-107. Randby, Å.T., I. Selmer-Olsen and L. Baevre, 1999. Effect of ethanol in feed on milk flavour and chemical composition. Journal of Dairy Science 82: 420-428. Rikarðsson, G., 1997. Mismunandi kjarnfóðurgjöf fyrir mjólkurkýr. Ráðunautafundur, pp. 242-254. (In Icelandic). Rikarðsson, G., 2002. Átgeta íslenskra mjólkurkúa. Ráðunautafundur, pp. 125-127. (In Icelandic). Rinne M., P. Huhtanen and S. Jaakkola, 2002. Digestive processes of dairy cows fed silages harvested at four stages of grass maturity. Journal of Animal Science 80: 1986-1998. Rustas, B.-O., 2009. Whole-crop cereals for growing cattle. Effects of maturity stage and chopping on intake and utilisation. Acta Universitatis Agriculturae Sueciae, Doctoral thesis no. 2009:74. Swedish University of Agricultural Sciences, Faculty of Veterinary Medicine and Animal Science, Uppsala, Sweden. Rustas, B.-O., E. Nadeau and S. Johnsson, 2003. Feed consumption and performance in young steers fed whole-crop barley silage at different levels of concentrate. In: Garmo, T.H. (ed.). Proceedings of the International symposium “Early harvested forage in milk and meat production”, 23-24 October 2003, Nannestad, Norway. Agricultural University of Norway, Department of Animal and Aquacultural Sciences, Aas, Norway. Schei, I., H. Volden and L. Bævre, 2005. Effects of energy balance and metabolizable protein level on tissue mobilization and milk performance of dairy cows in early lactation. Livestock Production Science 95: 35-47. Selmer-Olsen, I., 1994. Feed intake, milk yield and liveweight gain with silage from big bales or silos. Poster presented at the 45th Annual Meeting of EAAP, Edinburg, Scotland, 5-8 September. 8 pp. Sejrsen, K. and J.B. Larsen, 1978. Ensilage-kraftfoderforholdets indflydelse på kviers foderoptagelse og tilvækst samt mælkeydelse i første laktation. Beretning nr. 465 fra Statens Husdyrbrugsforsøg. 74 pp. (In Danish). Stensig, T., M.R. Weisbjerg and T. Hvelplund, 1998. Digestion and passage kinetics of fibre in dairy cows as affected by the proportion of wheat starch or sucrose in the diet. Acta Agriculturae Scandinavica Section A. Animal Science 48: 129-140. Sveinbjörnsson, J. and G.H. Harðarsson, 2008. Þungi og átgeta íslenskra mjólkurkúa. Fræðarþing landbúnaðarins 5: 336-344. (In Icelandic). Sveinbjörnsson, J., P. Huhtanen and P. Udén, 2006. The Nordic dairy cow model, Karoline – Development of volatile fatty acid sub-model. In: Kebreab E., J. Dikstra, A. Bannink, J. Gerrits and J. France (eds.). Nutrient Digestion and Utilisation in Farm Animals: Modelling Approaches. CAB International, Wallingford, UK, pp. 1-14. Therkildsen, M., M. Vestergaard, L.R. Jensen, H.R. Andersen and K. Sejrsen, 1998. Influence of feeding intensity, grazing and finishing on growth and carcass quality of young Friesian bulls. Acta Agriculturae Scandinavica 48: 193-201. Thomas, C., 1987. Factors affecting substitution rates in dairy cows on silage based rations. In: Haresign, W. and D.J.A. Cole (eds.) Recent Advances in Animal Nutrition. Butterworths, London, UK, pp. 205-216. Van Vuuren, A.M., C.J. Van Der Koelen and J. Vroons-De Bruin, 1993. Ryegrass versus corn starch or beet pulp fiber diet effects on digestion and intestinal amino acids in dairy cows. Journal of Dairy Science 76: 2692-2700. Vestergaard, M., K. Sejrsen, J. Foldager, S. Klastrup and D.E. Bauman, 1993a. The effect of bovine growth hormone on growth, carcass composition and meat quality of dairy heifers. Acta Agriculturae Scandinavica 43: 165-172. Vestergaard, M., M. Sommer, S. Klastrup and K. Sejrsen, 1993b. Effects of the beta-adrenergic agonist cimaterol on growth and carcass quality of monozygotic Friesian young bulls at three developmental stages. Acta Agriculturae Scandinavica 43: 236-244.

126

NorFor – The Nordic feed evaluation system

11. Chewing index system for predicting physical structure of the diet P. Nørgaard, E. Nadeau Å. Randby and H. Volden 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 4607 4608 4609 4610 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623 4624 4625 4626 4627 4628

An important goal in dairy cow management is, among other things, to develop a feeding strategy

thatChewing ensures goodindex rumen system function, which is vitally important for efficient milk production and healthy 11 for predicting physical structure of the diet

animals. A certain intake physically effective fibre isother essential fortostimulating rumination, salivation An important goal in dairyofcow management is, among things, develop a feeding strategy and rumen motility, and avoiding milk fat depression (Mertens, 1997; De Brabrander et al., that ensures good rumen function, which is vitally important for efficient milk production and 2002). A high intake of rapidly fermentable carbohydrates results production of a high level of volatile healthy animals. A certain intake of physically effective fibreinisthe essential for stimulating fatty acids salivation in the rumen and a motility, subsequent in pH, which increases the1997; risk of rumination, and rumen and reduction avoiding milk fat depression (Mertens, De subacute ruminal acidosis (Krause Infermentable addition to carbohydrates the structural results fibre content Brabrander et al., 2002). A and highOetzel, intake of2006). rapidly in the of a feed, the production of dietary a high level of volatile acids inand theaffects rumenthe andeating a subsequent reduction in pH,time (RT) length of the particles is alsofatty important time (ET), rumination which increases the riskand of subacute ruminal acidosis and Oetzel, 2006). In addition to and, thus, salivation thereby buffering of the (Krause rumen environment (Krause and Oetzel, 2006). the contentobserved of a feed,that the one length the dietary particles is also important and effective for De structural Boever etfibre al. (1993) kgofNDF from late cut grass silage was more affects the eating time (ET), (RT) thus, salivation buffering of stimulating rumination thanrumination NDF fromtime early cutand, grass silage (whichand hasthereby low levels of lignification), the rumen environment (Krause and Oetzel, 2006). De Boever et al. (1993) observed that one kg demonstrating that fibre type affects chewing activity. NDF from late cut grass silage was more effective for stimulating rumination than NDF from early cut grass silage (which has low levels of lignification), demonstrating that fibre type affects The formulation chewing activity. of diets for dairy cows with respect to the fibre content cannot be based solely on

a defined roughage to concentrate ratio or a minimum amount or proportion of roughage NDF in the ration (Mertens, 1997). Balch (1971) fibrousnesses different The formulation of diets for dairy cows withproposed respect toto therank fibrethe content cannot beof based solelyfeeds by a chewing index value that accounts for ETorplus RT. In this chapter, we present model for predicting on a defined roughage to concentrate ratio a minimum amount or proportion of aroughage NDF thethe diet eating index (EI), rumination index (RI) and chewing index (CI). These characteristics in ration (Mertens, 1997). Balch (1971) proposed to total rank the fibrousnesses of different feeds by chewing index value that accounts for ET plus RT. Inofthis we intention present a model areaused in NorFor to optimize the physical structure thechapter, diet. The of thisfor system is to predicting thean diet eating index rumination index and total chewing index (CI).toThese characterize additive CI for(EI), individual feeds and (RI) hence ranking feeds according their ability to characteristics are mastication used in NorFor to optimize physical structure of the diet. The intention of stimulate particle when cows arethe eating and during rumination. The time spent masticating this system is to characterize an additive CI for individual feeds and hence ranking feeds and the intensity of mastication are used to quantitatively rank the biological fibrousnesses of feeds, according to their ability to stimulate particle mastication when cows are eating and during which is closely related to the stimulation of salivation, the frequency of rumen motility and the rumination. The time spent masticating and the intensity of mastication are used to quantitatively formation of a stable flowing layer system in the reticulo-rumen. The NorFor chewing system has rank the biological fibrousnesses of feeds, which is closely related to the stimulation of been developed from the Danishmotility chewing system (Nørgaard, 1986) and usinginnew data salivation, the frequency of rumen andindex the formation of a stable flowing layerby system from Nørgaard et al. the reticulo-rumen. The(2010). NorFor chewing system has been developed from the Danish chewing index system (Nørgaard, 1986) and by using new data from Nørgaard et al. (2010).

11.1 Predicting the chewing index from eating and rumination

11.1 Predicting the chewing index from eating and rumination

Eachfeed feedisisgiven givena aCICIvalue value (min/kg DM), which is estimated the of sum and RI: Each (min/kg DM), which is estimated as theassum EI of andEIRI: CI = EI + RI

[Eq. 11.1]

11.1

where index for for a feedstuff, min/kg DM; DM; EI is the index asindex described in whereCI CIisisthe thechewing chewing index a feedstuff, min/kg EI eating is the eating as described in equation DM, asas described in in equation Equation11.2, 11.2,min/kg min/kgDM; DM;and andRIRIisisthe therumination ruminationindex, index,min/kg min/kg DM, described Equation 11.4. 11.4.

EI and RI describe the recorded ET and RT, respectively, and in NorFor the feed EI and RI values

EI and RI describe the recorded ET and RT, respectively, and in NorFor the feed EI and RI refer to those for a standard cow of a large multiparous dairy breed with a BW of 625 kg, a NDF values refer to those for a standard cow of a large multiparous dairy breed with a BW of 625 kg, a roughage (NDFr) intake of 0.7% of BW andand a daily rumination time of of 400 min. This corresponds to NDF roughage (NDFr) intake of 0.7% of BW a daily rumination time 400 min. This a daily DMI of 20 kg with a CI value of 30 min/kg DM, assuming that rumination time is equivalent corresponds to a daily DMI of 20 kg with a CI value of 30 min/kg DM, assuming that rumination to 2/3 the totaltochewing use time. of a standardized cow impliescow thatimplies the unit time is of equivalent 2/3 of thetime. total The chewing The use of a standardized that‘minutes/kg the DM’“minutes/kg reported inDM” the practical tool to atool standard and thus, unit reported in therefers practical refers tocow a standard cowdoes and not thus,necessarily does not reflects

the actual mean chewing time (CT) (min/kg DM) for that particular cow. The equations for EI and 172 RI were obtained from a meta-analysis of recorded ET and RT from cattle fed 85 different diets consisting of unchopped grass silage, grass hay or lucerne hay with or without supplementation of ground concentrates or rolled barley (Nørgaard et al., 2010). Based on the meta-analysis, the NorFor EI and RI values were standardized to reference values of 50 and 100 min/kg NDFr for unchopped roughage, respectively. In concentrates, the standard EI was set to 4 min/kg DM. As described in H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_11, © Wageningen Academic Publishers 2011

127

4630 4631 4632 4633 4634 4635 4636 4637 4638 4639 4640 4641 4642 4643 4644 4645 4646 4647 4648

equations for EI and RI were obtained from a meta-analysis of recorded ET and RT from cattle fed 85 different diets consisting of unchopped grass silage, grass hay or lucerne hay with or without supplementation of ground concentrates or rolled barley (Nørgaard et al., 2010). Based on the meta-analysis, the NorFor EI and RI values were standardized to reference values of 50 and 100 min/kg NDFr for unchopped roughage, respectively. In concentrates, the standard EI was set to 4 min/kg DM. As described in Chapter 4, concentrates and roughage are defined on the Chapter 4, concentrates and roughage are defined on the basis of the most frequently measured basis of the most frequently measured particle length (PL) that they contain, such that feedstuffs particle (PL) theyas contain, such that feedstuffs with a PL≤6 mmmm areare defined as as concentrates with a PLlength ≤6 mm arethat defined concentrates while feedstuffs with a PL >6 defined while feedstuffs with a PL>6 mm defined as roughage. maize silage are chopped at roughage. Grass and maize silage areare chopped at harvest and theGrass effect and of physical processing on harvest and the effect of physical processing on the EI and RI values is based on the theoretical the EI and RI values is based on the theoretical chopping length (TCL), which corresponds to the chopping length (TCL), which corresponds to theobserved PL value (O´Dogherty, et al. PL value (O´Dogherty, 1984). Nørgaard et al. (2010) that the mean ET 1984). and RT Nørgaard per kg (2010)decreased observedwith thatincreasing the meanBW. ET and RT kgofNDFr decreased withinto increasing BW. Thus, the NDFr Thus, theper size the cow is not taken account when e.g. no distinctions in this respect are drawn between of calculating the values EI orinto RI (account size of the cow is notof taken when calculating the values of EI or RI (e.g.cows no distinctions large breeds small between breeds or cows primiparous vs.dairy multiparous This meansorthat in thisdairy respect arevs.drawn of large breeds breeds). vs. small breeds primiparous vs. Jersey cows have a higher meanJersey ET and RT per kgaNDFr than large breeds same multiparous breeds). Thisrecorded means that cows have higher recorded meanfed ETthe and RT per kg diet. Feeding high moisture silage containing less than 40% dry matter has been shown to NDFr than large breeds fed the same diet. Feeding high moisture silage containing less than 40% increase EThas (Debeen Boever et al.,to1993). However, NorFor et this is 1993). not reflected in a higher EI this is not dry matter shown increase ET (DeinBoever al., However, in NorFor value because the DM content of a feedstuff is not taken into account when calculating the values reflected in a higher EI value because the DM content of a feedstuff is not taken into account when of EI or RI.

4649 4650 4651 4652 4653 4654 4655 4656 4657 4658 4659

11.2 Calculation of the eating and rumination indices

4660

EI = 50 ⋅

4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4672

calculating the values of EI or RI.

11.2 Calculation of the eating and rumination indices

Table 11.1 presents an overview of the equations used to calculate EI and RI depending on PL. The EI describes the eating time per kg DM of a given feedstuff for the standard cow (Nørgaard Table 11.1 presents overview of thefeeds equations used EI DM. and RI depending on PL. The et al., 2010). Ground,an rolled or pelleted are given anto EIcalculate of 4 min/kg The recorded ET EI describes eating time per(Mertens, kg DM of a given feedstuff for the standard cow (Nørgaard et al., decreases as a the result of chopping 1997).

2010). Ground, rolled or pelleted feeds are given an EI of 4 min/kg DM. The recorded ET decreases

Table as a result 11.1 of chopping (Mertens, 1997).

EI of of roughage roughage and to the NDF content and and a particle length factorfactor for EI andby-products by-productsare areproportional proportional to the NDF content a particle length for eating (Size_E) (Nørgaardetetal., al.,2010). 2010). EI EI is is calculated calculated using equation: eating (Size_E) (Nørgaard usingthe thefollowing following equation: NDF ⋅ Size _ E 1000

[Eq. 11.2]

11.2

whereEI EIisisthe theeating eatingindex, index,min/kg min/kgDM; DM;NDF NDF NDF content in the feedstuff, g/kg DM; Size_E where is is thethe NDF content in the feedstuff, g/kg DM; is a correction factor that depends on particle length, as described in Equation 11.3; and Size_E is a correction factor that depends on particle length, as described in equation 11.3; and 50 is the standardized eating time, NDF. NDF. 50 is the standardized eatingmin/kg time, min/kg Size_E is a correction factor for the feed particle length, which ranges from 0.67 in finely chopped feedstuffs to 1 of in the unchopped roughage Figure 11.1). index Size_Esystem. is parameterized from Table 11.1. Overview calculations used(see in the chewing recorded ET data in cattle fed chopped (10 or 20 mm TCL) or unchopped silage (Nørgaard, unpublished) using the following equation: Feed type and is calculated Concentrates Roughages ⋅ PL Size _ E = 1 − 0 .52 ⋅ e − 0 .078 Processing Ground

PL or TCL (mm)1 Size_E Size_R Hardness factor Eating index (EI, min/kg DM) Rumination index (RI, min/kg DM) Chewing index (CI, min/kg DM)

Rolled

Finely

Coarsely

≤2.0

2-4

Chopped

[Eq. 11.3] Long or slightly

chopped

173

4-6

4

6-40 >40 1–0.52·exp(-0.078·TCL) 1 1–exp (-0.173·(TCL/0.7-1)) 1 0.75 + iNDF/1000 50·NDF/1000·Size_E

0

100·NDF/1000·Size_R·Hardness factor

EI + RI

1 PL is the most frequent particle length in a feedstuff. TCL (theoretical chopping length) corresponds to PL in roughage

(O´Dogherty, 1984).

128

NorFor – The Nordic feed evaluation system

4689 4690 4691 4692 4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4714 4715

where EI is the eating index, min/kg DM; NDF is the NDF content in the feedstuff, g/kg DM; Size_E is a correction factor that depends on particle length, as described in equation 11.3; and 50 is the standardized eating time, min/kg NDF.

Size_Eisisaacorrection correctionfactor factor feed particle length, which ranges 0.67 in finely chopped Size_E forfor thethe feed particle length, which ranges from from 0.67 in finely feedstuffs to 1 in unchopped roughage (see Figure 11.1). 11.1). Size_E is parameterized fromfrom recorded ET chopped feedstuffs to 1 in unchopped roughage (see Figure Size_E is parameterized recorded ET data cattle fed(10 chopped 20 mm or unchopped silage (Nørgaard, data in cattle fedinchopped or 20 (10 mmorTCL) or TCL) unchopped silage (Nørgaard, unpublished data) unpublished) and isusing calculated using the equation: following equation: and is calculated the following Size _ E = 1 − 0 .52 ⋅ e − 0 .078 ⋅ PL [Eq. 11.3] 11.3 where Size_E is the correction factor for the eating index due to the chopping or processing of the feedstuff. The Size_E rangesfactor from 0.67 for fine chopped to 1chopping for unchopped where Size_E is the factor correction for the eating indexroughage due to the or processing of roughages; and PL is the most frequent particle length in the feedstuff, mm. PL corresponds 173 the feedstuff. The Size_E factor ranges from 0.67 for fine chopped roughage to 1 fortounchopped TCL for roughage (O´Dogherty, 1984) and is therefore used as the input for the equation.

roughages; and PL is the most frequent particle length in the feedstuff, mm. PL corresponds to TCL for roughage (O´Dogherty, 1984) therefore usedthe as value the input for thewhich equation. Figure 11.1 shows how the length ofand feedisparticles affects of Size_E, approaches 1 for PL values > 50 mm, showing that slight chopping of roughage with TCL values higher than Figure 11.1 not shows how length of feed particles affects the value of Size_E, which approaches 1 50 mm does affect thethe value of EI.

for PL values >50 mm, showing that slight chopping of roughage with TCL values higher than 50

mm does Figure 11.1not affect the value of EI.

RI activity during rumination perper kg DM of aof given feedstuff in in the RI describes describesthe theassociated associatedchewing chewing activity during rumination kg DM a given feedstuff the standard cow. determinedfrom fromthe the feed feed NDF NDF content, factor forfor rumination standard cow. RIRI is is determined content,aaparticle particlelength length factor rumination (Size_R) and a hardness factor,iswhich is to related to the content iNDF content the feedstuff (Size_R) and a hardness factor, which related the iNDF in theinfeedstuff (Nørgaard et ., 2010). RI is using calculated using the following (Nørgaard al., 2010).etRIalis calculated the following equation:equation: RI = 100 ⋅

NDF ⋅ Size _ R ⋅ Hardness _ factor 1000

[Eq. 11.4]

11.4

where RI is the rumination index, min/kg DM; NDF is the NDF content in the feedstuff, g/kg

where RI is the rumination index, min/kg DM; NDF is the NDF content in the feedstuff, g/kg DM;Size_R Size_Risisa correction a correction factor depends on particle as described in Equation 11.5; DM; factor thatthat depends on particle lengthlength as described in equation 11.5; the Hardness_factor Hardness_factoris is described in Equation andis 100 is the standardized rumination time, the described in equation 11.6; 11.6; and 100 the standardized rumination time, min/kgNDF. NDF. min/kg If lower than 2.02.0 mmmm (as in processed concentrates), then the RI the RI is If the thePL PLofofthe thefeedstuff feedstuffis is lower than (asfinely in finely processed concentrates), then is assumedtotobebezero zero(see (seeTable Table 11.1). 11.1). Nørgaard Nørgaard and shown that thethe most assumed andSehic Sehic(2003) (2003)have have shown that most frequent frequent particle length in the faeces of cattle fed solely roughage is 0.7 mm, consequently, feedstuffs with a PL>0.7 mm are considered to stimulate rumination. However, some concentrates have 1.0 a PL>0.7 mm but are not considered to stimulate rumination. Therefore, the borderline value for PL was set to 2.0 mm to ensure that all finely ground concentrates are given an RI of zero. Thus, Size_R is zero for feeds with a PL ≤2.0 mm and their CI is 4 min/kg DM corresponding 0.8 to eating index (Table 11.1). Coarsely processed concentrates with a PL between 2 Size_R to and 6 mm (e.g. rolled grains, coarsely ground grains, oil cakes and hulls) are considered Size_E stimulate rumination. Several studies have shown that chopping roughage at TCL values higher 0.6 not decrease the recorded RT (De Boever et al., 1993; Garmo et al., 2008). than 20 mm does Furthermore, Nørgaard and Sehic (2003) showed that the PL of particles in swallowed boli from cows eating grass silage chopped at 19 mm TCL was the same as the PL of boli particles from 0.4 cows eating unchopped grass silage. Therefore, the RT for roughage chopped at a TCL higher than 20 mm is considered to have the same RT as unchopped roughage, and consequently Size_R at 20 mm TCL is set to 0.99. Size_R decreases exponentially when TCL is below 20 mm, as 0.2 11.1. illustrated in Figure

Correction factor

4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4673 4672 4674 4675 4676 4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688

⎛ ⎛ PL ⎞ ⎞ ⎜⎜0.0 −0.173⋅⎜ −1⎟ ⎟ 0.7 ⎠ ⎟⎠ ⎝ Size _ R = 1 − e 0 ⎝

10

20

30

40

50 [Eq. 11.5] 60

PL (mm)

Figure 11.1. Effect of the most frequent particle length (PL) in a feedstuff on the correction factors 174 Size_E and Size_R. Size_E and Size_R are used to adjust the eating and rumination index due to chopping or processing of feedstuffs (see Table 11.1). NorFor – The Nordic feed evaluation system

129

4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4716 4717 4714 4718 4715 4719 4716 4720 4717 4721 4718 4722 4719 4723 4720 4724 4721 4725 4722 4726 4723 4727 4724 4728 4725 4729 4726 4730 4727 4731 4728 4732 4729 4733 4730 4734 4731 4735 4732 4733 4736 4734 4737 4735 4738 4739 4736 4740 4737 4741 4738 4742 4739 4743 4740 4744 4741 4745 4742 4746 4743 4747 4744 4748 4745 4749 4746 4750 4747 4748 4751 4749 4750 4752 4753 4751 4754 4755 4752 4756 4753 4754

the Hardness_factor is described in equation 11.6; and 100 is the standardized rumination time, min/kg NDF. If the PL of the feedstuff is lower than 2.0 mm (as in finely processed concentrates), then the RI is assumed to be zero (see Table 11.1). Nørgaard and Sehic (2003) have shown that the most particle length the faeces cattle solely roughage is 0.7ismm, consequently, feedstuffs with a frequent particleinlength in the of faeces offed cattle fed solely roughage 0.7 mm, consequently, feedstuffs with PL>0.7 mmto arestimulate considered to stimulate rumination. However, some have a PL>0.7 mm PL>0.7 mm area considered rumination. However, some concentrates concentrates have a PL>0.7 mm but are not considered to stimulate rumination. Therefore, the set to 2.0 but are not considered to stimulate rumination. Therefore, the borderline value for PL was borderline valuethat for all PL finely was setground to 2.0 mm to ensure that finely concentrates are given mm to ensure concentrates are all given an ground RI of zero. Thus, Size_R is zero for an RI of zero. Thus, Size_R is zero a PLDM ≤2.0corresponding mm and their CI 4 min/kg DM feeds with a PL≤2.0 mm and theirforCIfeeds is 4 with min/kg to is eating index (Table 11.1). corresponding to eating index (Table 11.1). Coarsely processed concentrates with a PL between 2 Coarsely processed concentrates with a PL between 2 and 6 mm (e.g. rolled grains, coarsely ground and 6 mm (e.g. rolled grains, coarsely ground grains, oil cakes and hulls) are considered to grains, oil cakes and hulls) are considered to stimulate rumination. Several studies have shown that stimulate rumination. Several studies have shown that chopping roughage at TCL values higher chopping roughage TCL values higher than 20 mm does the recorded RT (De Boever et not al., decrease 1993; Garmo et al., 2008). than 20 mm does not at decrease the recorded RT (De Boever et al., 1993; Garmo et al., 2008). Furthermore, Nørgaard and Sehic (2003) showed Furthermore, Nørgaard and Sehic (2003) showed that the PL of particles in swallowed bolithat fromthe PL of particles in grass swallowed boli fromatcows eating silage mm TCL was cows eating silage chopped 19 mm TCL grass was the samechopped as the PLatof19 boli particles fromthe same as the PL of boli particlesgrass fromsilage. cowsTherefore, eating unchopped silage. Therefore, the RT for roughage cows eating unchopped the RT forgrass roughage chopped at a TCL higher than 20 mm considered have20 themm same as unchopped roughage, andRT consequently Size_R chopped at is a TCL highertothan is RT considered to have the same as unchopped roughage, at 20consequently mm TCL is setSize_R to 0.99.atSize_R exponentially when TCL is below 20 mm, as when TCL and 20 mmdecreases TCL is set to 0.99. Size_R decreases exponentially illustrated in Figure is below 20 mm, as11.1. illustrated in Figure 11.1. ⎛ PL ⎞ ⎞ factor for the rumination index, which depends on chopping or where Size_R⎛⎜ is the correction ⎜ −0.173⋅⎜ 0.7 −1⎟ ⎟⎟ ⎝ ⎠ ⎠ Size_R ⎝ processing The factor ranges from 0 for fine processed feeds[Eq. to 111.5] for 11.5 Size _ R = 1 −the e feedstuff. unchopped roughage or roughage chopped with a PL higher than 20 mm. PL is the most frequent particle length (mm) in correction the feedstufffactor and for PL corresponds to TCL depends (theoretical whereSize_R Size_R forroughage the rumination index,depends which on or chopping or where is is thethe correction factor for the rumination index, which on chopping chopping length, mm; O'Dogherty, 1984) andranges can therefore used as the inputfeeds for the equation; processing thefeedstuff. feedstuff. TheSize_R Size_R factor from 0be for fine processed 1 for unchopped processing the The factor ranges from 0 for fine processed feeds to 1tofor 0.7 is the most frequent particle length in faeces174 from cattle, meaning that feedstuffs with a unchopped or chopped roughagewith chopped a PL higher than PL 20 mm. is the most frequent roughage orroughage roughage a PLwith higher than 20 mm. is thePL most frequent particle length PL>0.7 stimulate rumination (Nørgaard and Sehic, 2003). particle length (mm) in the feedstuff and for roughage PL corresponds to TCL (theoretical (mm) in the feedstuff and for roughage PL corresponds to TCL (theoretical chopping length, mm; chopping length, mm;and O'Dogherty, 1984)beand can therefore be used the input for the equation; O’Dogherty, can therefore used the input theas 0.7relatively is the most frequent The structural 1984) NDF in immature spring grass andas chopped beetfor pulp isequation; soft and has 0.7 is thelength most frequent particle length inmeaning faeces from cattle, meaning that feedstuffsstimulate with a rumination particle in faeces from cattle, that feedstuffs with a PL>0.7 little resistance to physical breakdown during chewing compared with mature grass, barley straw, PL>0.7 stimulateSehic, rumination and Sehic, 2003). (Nørgaard 2003).(Nørgaard lucerne hay and cut after blooming or whole cotton seeds. The content of iNDF in a feedstuff is closely related to the lignification of the NDF fibres (Weisbjerg and Soegaard, 2008) and The structural NDF in immature spring grass and chopped beet pulp is soft and has relatively lignification is known increase resistance to physical breakdown. to has physical The structural NDF intoimmature spring grass and chopped beet The pulpresistance is soft and relatively little little resistance to physical breakdown during chewing compared with mature grass, barley straw, breakdown is physical ranked bybreakdown a hardness during factor, which hascompared been givenwith a value of 1 grass, for typical grass resistance to chewing mature barley lucerne hay cut after blooming or whole cotton seeds. The content of iNDF in a feedstuff isstraw, lucerne silage with an blooming iNDF content of 250cotton g/kg NDF. The hardness increases from 0.8 for hay cut after or whole seeds. The contentfactor ofand iNDF in a feedstuff is closely related closely related to the lignification of the NDF fibres (Weisbjerg Soegaard, 2008) and immature spring grass with an iNDF content of approximately 50 g/kg NDF to 1.25 for late cut to the lignification NDF resistance fibres (Weisbjerg and Soegaard,The 2008) and lignification lignification is knownoftothe increase to physical breakdown. resistance to physical is known lucerne hay with an iNDF content of approximately 500 g/kg NDF. The hardness factor is to increaseisresistance physicalfactor, breakdown. Thebeen resistance physical is ranked by breakdown ranked by to a hardness which has given a to value of 1 forbreakdown typical grass calculated as: silage with an iNDFwhich content of been 250 g/kg NDF. The hardness factor increases fromwith 0.8 for a hardness factor, has given a value of 1 for typical grass silage an iNDF content immature spring grass an iNDF content of approximately 50 immature g/kg NDF to 1.25 for latewith cut an iNDF of 250 g/kg NDF. Thewith hardness factor increases from 0.8 for spring grass ⎛ iNDF ⎞ [Eq. 11.6] Hardnesshay _offactor 0.75 + ⎜ content lucerne with= an iNDF of approximately 500 late g/kgcut NDF. The hardness factor is ⎟ g/kg content approximately 50 NDF to 1.25 for lucerne hay with an iNDF content of ⎝ 1000 ⎠ calculated as: 500 g/kg approximately NDF. The hardness factor is calculated as:

where the Hardness_factor is used for calculating the rumination index and ranges from 0.8 to ⎛ iNDF ⎞ 11.6] iNDF content in the feedstuff, g/kg[Eq. Hardness _ factor = 0.75 + ⎜feedstuffs; ⎟ 1.25 for typical Nordic iNDF is the NDF. 1000 ⎝

⎠

11.6

Table shows examples ofisEI, RI for andcalculating CI values for concentrates ranging where11.2 the Hardness_factor used thesome rumination indexand androughages ranges from 0.8 to 1.25 where the Hardness_factor is used for calculating the rumination index and ranges from 0.8 to from 4 to 22Nordic min/kgfeedstuffs; DM. The main reason foriNDF the much higher CI of dried beet pulpNDF. compared to for typical iNDF is the content in the feedstuff, g/kg 1.25 for typical Nordic feedstuffs; iNDF is the iNDF content in the feedstuff, g/kg NDF. barley and rape seed cake is its higher PL, which causes a higher RI. The higher CI of grass silage compared to examples maize silage is due to its CI higher TCL and NDF content, which result in higher Table 11.2 11.2shows shows EI,RI RI and values some concentrates roughages. The main Table examples ofofEI, and CI values forfor some concentrates and and roughages ranging EI and RI (Table 11.2). reason for the much higher CI of dried beet pulp compared to barley and rape seed cake is its from 4 to 22 min/kg DM. The main reason for the much higher CI of dried beet pulp compared to higher PL, which causes higher RI.higher The higher CI of grassa silage to maize barley and rape seeda cake is its PL, which causes higher compared RI. The higher CI of silage grass is due to its Table 11.2 silage maize silage iswhich due toresult its higher TCL and NDF highercompared TCL andtoNDF content, in higher EI and RIcontent, (Table which 11.2). result in higher EI and RI (Table 11.2). The EI and RI are additive values and the CI of a total diet (min/kg DM) is considered to be the The of EIthe andindividual RI are additive sum feed CIvalues values:and the CI of a total diet (min/kg DM) is considered to be the sum Table of the11.2 individual feed CI values: ∑ DMI ⋅ CI

i i The EI andi RI are additive values and the CI of a total diet (min/kg DM) is considered to be the [Eq. 11.7] CI _ DM = sum of the individual ∑ DMI i feed CI values:

11.7

i

∑ DMI i ⋅ CI i

where CI_DM is the chewing index for the total diet, min/kg DM; DMIi is the dry[Eq. matter intake 11.7] CI _ DM = i ∑ DMI i kg DM/day; and CI is the chewing index for the i-th feedstuff, min/kg DM for the i-th feedstuff, i i as described in Equation 11.1. 130 NorFor – The Nordic feed evaluation system where CI_DM is the chewing index for the total diet, min/kg DM; DMIi is the dry matter intake for the i-th feedstuff, kg DM/day; and CIi is the chewing index for the i-th feedstuff, min/kg DM

Table 11.2 Estimated chewing index values for different feedstuffs in the NorFor feed stuff table.

Barley RSC DBP GS MS

Feed code

NDF g/kg DM

iNDF g/kg NDF

PL1 mm

001-0008 002-0044 004-0020 006-0241 006-0308

206 268 382 417 364

158 511 89 151 187

2.1 2.0 4.0 20 10

Size_E Size_R Hardness EI2 RI2 CI2 min/kg min/kg min/kg DM DM DM 0.29 0.89 0.76

0.56 0.99 0.90

0.91 1.26 0.84 0.90 0.94

4 4 4 19 14

5 0 18 37 31

9 4 22 56 45

1

PL is the most frequent particle length in concentrates and corresponds to TCL (theoretical chopping length) in roughage (O´Dogherty, 1984). 2 See Equations 11.1 to 11.6. RSC, rape seed cake; DBP, dried beet pulp; GS, grass silage; MS, maize silage.

where CI_DM is the chewing index for the total diet, min/kg DM; DMIi is the dry matter intake for the i-th feedstuff, kg DM/day; and CIi is the chewing index for the i-th feedstuff, min/kg DM as described in Equation 11.1.

11.3 Implications The objectives of the work described in this chapter were to develop a system for ranking the structural value of different feedstuffs by calculating an additive chewing index, which could be used as a tool to minimize the risk of digestive disorders and ensure proper rumen function of a diet fed to dairy cows. The minimum recommended chewing index (32 min/kg DM) is intended to ensure a high degree of fibre digestion, a high acetate to propionate ratio and an acceptable milk fat content even at a high DMI.

References Balch, C.C., 1971. Proposal to use time spent chewing as an index to which diets for ruminants possess the physical property of fibrousness characteristic of roughages. British Journal of Nutrition 26: 383-392. De Boever, J.L., A. De Smet, D.L. De Brabander and C.V. Boucqué, 1993. Evaluation of physical structure. 1. Grass silage. Journal of Dairy Science 76: 140-153. De Brabrander, D.L., J.L. De Boever, J.M. Vanacker, C.V. Boucqué and S.M. Botterman, 2002. Evaluation of physical structure in cattle nutrition. Recent developments in ruminant nutrition. Wiseman, J. and P.C. Garnsworthy (eds.). Nottingham University Press, Nottingham, UK, pp. 47-80. Garmo, T.H., A.T. Randby, M. Eknaes, E. Prestløkken and P. Nørgaard, 2008. Effect of grass silage chop length on chewing activity and digestibility. In: Hopkins, A., T. Gustafsson, J. Bertilson, G. Dalin, N. Nilsdotter-Linde and E. Spörndly (eds.). Grassland Science in Europe, volume 13, pp. 810-812. Krause, K.M. and G.R. Oetzel, 2006. Understanding and preventing subacute ruminal acidosis in dairy herds: a review. Special Issue: Feed and animal health. Animal Feed Science and Technology 126: 215-236. Mertens, D.R., 1997. Creating a system for meeting the fiber requirements of dairy cows. Journal of Dairy Science 80: 1463-1481. Nørgaard, P., 1986. Physical structure of feeds for dairy cows. A new system for evaluation of the physical structure in feedstuffs and rations for dairy cows. New developments and future perspectives in research of rumen function, pp. 85-107.

NorFor – The Nordic feed evaluation system

131

Nørgaard, P. and A. Sehic, 2003. Particle size distribution in silage, boluses, rumen content and faeces from cows fed grass silage with different theoretical chopping length. Proceedings from the 6th International Symposium on the Nutrition of Herbivores, pp. 457-460. Nørgaard, P., E. Nadeau and A. Randby, 2010. A new Nordic structure evaluation system for diets fed to dairy cows: a meta analysis. In: Sauvant, D., J. Van Milgen, P. Faverdin and N. Friggens (eds.). Modelling nutrient digestion and utilization in farm animals. Wageningen Academic Publishers, Wageningen, the Netherlands, pp. 112-121. O’Dogherty, M, 1984. The description of chop length distribution from forage harvesters and a geometric simulation model for distribution generation. In: Kennedy, P. (ed.). Canadian society of animal science, occasional publication, Edmonton, Alberta, Canada, pp. 62-75. Weisbjerg, M.R. and K. Soegaard, 2008. Feeding value of legumes and grasses at different harvest times. In: Hopkins, A., T. Gustafsson, J. Bertilsson, G. Dalin, N. Nilsdotter-Linde and E. Spörndly (eds.). Biodiversity and animal feed: Future challenges for grassland production. Grassland Science in Europe, volume 13, pp. 513-515.

132

NorFor – The Nordic feed evaluation system

4825 4825 4826 4825 4826 4827 4827 4826 4828 4828 4827 4829 4828 4830 4829 4831 4830 4829 4832 4831 4830 4832 4831 4833 4832 4833 4834 4835 4834 4833 4836 4835 4834 4837 4836 4835 4837 4838 4836 4837 4838 4839 4838 4840 4839 4841 4840 4839 4842 4841 4840 4843 4842 4841 4844 4843 4842 4845 4844 4843 4846 4845 4844 4847 4846 4845 4848 4847 4846 4849 4848 4847 4850 4849 4848 4850 4851 4849 4850 4851 4852 4851 4853 4852 4854 4853 4852 4855 4854 4853 4856 4855 4854 4856 4855 4857 4856 4858 4857 4859 4858 4857 4860 4859 4858 4861 4860 4859 4862 4861 4860 4863 4862 4861 4864 4863 4862 4865 4864 4863 4865 4866 4864 4865 4866 4867 4866 4868 4869 4870 4871 4872 4873 4874 4875

12.0 Prediction of milk yield, weight gain and utilisation of N, P 12.0 milk yield, weight gain and utilisation of N, P and KPredictionofofmilk 12. Prediction yield, weight gain and utilisation of N, P and K and KPrediction 12.0 of milk yield, weight gain and utilisation of N, P M. Åkerlind and H. Volden M. and K and M. Åkerlind Åkerlind andH. H.Volden Volden

NorFor predicts ECM and protein yield for dairy cows and weight gain for growing cattle. M. Åkerlind Volden Excretion of and N, PH. and K can also be predicted for a given diet and production level. Keep NorFor NorForpredicts predictsECM ECMand andprotein proteinyield yieldfor fordairy dairycows cowsand andweight weightgain gainfor forgrowing growingcattle. cattle. Excretion track on excreted N, PK and Kalso are of interest when estimating plant fertilising value and also Excretion of N, P and can be predicted for a given diet and production level. Keep of N, P and K can also be predicted for a given diet and production level. Keep track on excreted NorFor predicts ECM and protein yield for dairy cows and weight gain for growing cattle. evaluating the impact of excreted N interest and P onwhen eutrophication of the fertilising environment. track on excreted N, P and K are of estimating plant value and also Excretion ofare N, Pofand K canwhen also be predictedplant for afertilising given diet value and production level. Keep N, P and K interest estimating and also evaluating the impact of evaluating the impact excreted N interest and P onwhen eutrophication the fertilising environment. track on excreted N, Pof and K are of estimating of plant value and also

excreted N and P on eutrophication of the environment. 12.1 Prediction of milk yield evaluating the impact of excreted N and P on eutrophication of the environment. 12.1 Prediction Prediction of ECMof yield milk is estimated yield from the total supply of NEL minus basal energy 12.1 Prediction offorismilk yield requirements corrected any changes in BCS (seesupply chapterof9). Since 1 kgbasal ECMenergy contains Prediction of ECM yield estimated from the total NEL minus 12.1 Prediction of milk yield

3.14 MJ (Sjaunja et al., for 1990), the estimated ECM isSince calculated as: contains requirements corrected changes in BCS (seeproduction chapterof9). 1 kg ECM Prediction of yield isisany estimated from total supply minus Prediction ofECM ECM yield estimated fromthe the total supply ofNEL NEL minusbasal basalenergy energy requirements 3.14 MJ (Sjaunja et al., 1990), the estimated ECM production is calculated as: requirements corrected for any changes in BCS (see chapter 9). Since 1 kg ECM contains

corrected for anyNEL changes (see _Chapter Since 1 kg ECM contains − NE _in maBCS int − NEL gain − NE9). _ gest − NEL _ dep + NEL _ mob 3.14 MJ (Sjaunja et

ECMMJ _ response = et al., 1990), the estimated ECM production is calculated as: 3.14 (Sjaunja al., 1990), the estimated ECM production is calculated as: ECM _ response =

14 _ gest − NEL _ dep + NEL _ mob NEL − NE _ ma int − NEL _ gain −3.NE 3 . 14 _ gest − NEL _ dep + NEL _ mob[Eq. 12.1] NEL − NE _ ma int − NEL _ gain − NE

12.1] 12.1 ECM _ response = [Eq. 3.14 where ECM_response is the predicted energy corrected milk yield, kg/day; NEL is the net [Eq. 12.1] energy lactation obtained from the diet,energy Equation 8.3; NE_maint, NEL_gain, NE_gest, whereECM_response ECM_response thepredicted predicted energy corrected milk yield, kg/day; NEL the net energy where isisthe corrected milk yield, kg/day; NEL is theisnet NEL_dep and NEL_mob is energy required of 8.3; maintenance, gestation, growth of primiparous energy lactation lactation obtained obtained from from the diet, the diet, Equation Equation 8.3; NE_maint, NE_maint, NEL_gain, NEL_gain, NE_gest, NE_gest, NEL_dep and where ECM_response is thefrom predicted energy corrected milk yield, kg/day; NEL9.18 is the net cows, and supply of energy mobilisation, Equation 9.1, 9.4, 9.3, 9.17 and NEL_dep and NEL_mob is energy required of maintenance, gestation, growth of primiparous NEL_mob is energy required of maintenance, gestation, growth of primiparous cows, and supply of energy lactation obtained from the diet, Equation 8.3; NE_maint, NEL_gain, NE_gest, respectively; and 3.14 is the energy content of 19.3, kg ECM. cows, and supply of energy from mobilisation, Equation 9.1, 9.4, 9.3, 9.17 and 9.18 energy from mobilisation, Equation 9.1, 9.4, 9.17 and 9.18 respectively; and 3.14 is the energy NEL_dep and NEL_mob is energy required of maintenance, gestation, growth of primiparous respectively; andECM. 3.14 is the from energy content of 1Equation kg ECM.9.1, 9.4, 9.3, 9.17 and 9.18 content ofsupply 1 kg cows, and of energy mobilisation, Milk protein response is calculated by multiplying available AATN by the efficiency of AA respectively; and is thesynthesis. energy content of 1 kg AAT ECM. is the total AAT minus basal utilisation forresponse milk3.14 protein The available N N Milk protein is calculated by multiplying AAT Milk protein response is calculated by multiplyingavailable available AATNNby bythe theefficiency efficiencyof ofAA AA utilisation AAT Response in milkThe protein production isiscalculated as: N minus basal N requirement. utilisation for milk protein synthesis. available AAT the total AAT N for milk protein synthesis. The available AATN isavailable the totalAAT AATNNbyminus basal AAT requirement. Milk protein response is calculated by multiplying the efficiency of NAA AAT N requirement. Response in milk protein production is calculated as: utilisation for milk protein protein production synthesis. The available AAT Response in milk is calculated as: is the total AAT minus basal N N AAT _ Eff [Eq. 12.2] AAT100 N _ Eff Pr otein _ response = (Avail _ AATN ) ⋅ [Eq. 12.2] 12.2 AAT100 _ Eff N where Pr oteinProtein_response _ response = (Availis_ the AAT predicted milk protein production, g/day; Avail_AAT [Eq.N 12.2] is the N )⋅ 100 where Protein_response is the in predicted milk protein g/day;Equation Avail_AAT the available available amino acid absorbed the small intestine forproduction, milk production, 9.26 is and

N Pr otein _ response = (Avail _ AAT ) ⋅ protein AAT Response inNmilk production is calculated as: N requirement.

where Protein_response is the predicted milk protein production, g/day; Avail_AATNNis the amino acid in theofsmall intestine for milkfor production, Equation 9.26 and AAT AAT isabsorbed theacid efficiency amino acid utilization for milkproduction, protein production, Equation N_Eff available amino absorbed in the small intestine milk Equation 9.26 andN_Eff is the where Protein_response isutilization the predicted milk protein production, g/day; Avail_AAT N is the efficiency of amino acid for milk protein production, Equation 9.24. 9.24. AAT N_Eff is the efficiency of amino acid utilization for milk protein production, Equation available amino acid absorbed in the small intestine for milk production, Equation 9.26 and 9.24. AAT _Eff is the efficiency of amino acid utilization for milk protein production, Equation

12.2 of weight gain gain 12.2NPrediction Prediction of weight 9.24. For cattle,of ADG can be predicted from either the energy available for growth or the 12.2growing Prediction weight gain AAT forof growth. Since the energy utilization kg_corr (Equation 9.6) and kormgthe AATN For growing growing cattle, ADG can bepredicted predicted from either the energy available growth N available For cattle, ADG can be from either thefactors energy available forfor growth or the 12.2 Prediction weight gain (Equation 8.5),growth. both depend on and ADG, they cannot used to (Equation predict daily weight available for SinceSince theBW energy utilization factorsbe kg_corr (Equation 9.6) and k kmg(Equation AAT for ADG growth. the energy utilization kg_corr 9.6) and N available For growing cattle, can be the predicted from either thefactors energy available for growth or mg the gain for a given diet.on Therefore, predicted requirement for ADG was assumed toweight vary with 8.5), both depend BW and ADG, they cannot be used to predict daily weight gain for a given (Equation 8.5), both depend on BW and ADG, they cannot be used to predict daily AAT N available for growth. Since the energy utilization factors kg_corr (Equation 9.6) and kmg animal type and diet. BW. gain for a given Therefore, the predicted requirement for ADG was assumed to vary diet. Therefore, the predicted requirement for ADG was assumed to vary with animal type (Equation 8.5), both depend on BW and ADG, they cannot be used to predict daily weightwithand BW. animal and diet. BW. Therefore, the predicted requirement for ADG was assumed to vary with gain fortype a given For growing bulls ofofdairy gain is is predicted from available energy: For growing bulls dairybreed breedweight weight gain predicted from available energy: animal type and BW. For growing bulls of dairy breed weight gain is predicted from available energy: NEG - NE_maint

gaingrowing _ response 1000breed ⋅ For bulls of =dairy weight gain is predicted from available energy: [Eq. 12.3] NEG 0 .0243 + 7.7315 NEG ⋅-BW NE_maint gain _ response NEG = 1000 ⋅ [Eq. 12.3]

12.3

0 .0243 ⋅-BW +gain 7.7315 For growing bulls of beef breed NEG weight NE_maint is predicted from available energy:

gaingrowing _ response 1000breed ⋅ For bulls of =beef weight gain is182 predicted from available energy: [Eq. 12.3] NEG 0 .0243 ⋅ BW + 7.7315 NEG - NE_maint 182 gain _ response NEG = 1000 ⋅ [Eq. 12.4] 0 .0116 ⋅ BW + 9.7214 182

where gain_response

12.4

is the predicted weight gain based on the energy supply, g/d; NEG is the

NEG where gain_responseNEG is the predicted weight gain based on the energy supply, g/d; NEG is energy gain, Equation 8.8;8.8; NEG_maint is the energy requirement of of maintenance, Equation 9.1; BW the energy gain, Equation NEG_maint is the energy requirement maintenance, is current9.1; body weight, kg; body and the numerator calculates the energy content 1 kg content of weight gain. Equation BW is current weight, kg; and the numerator calculates the of energy of 1 kg of weight gain.

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_12, © Wageningen Academic Publishers 2011 AAT =

⋅

133

4870 4872 4871 4871 4873 4872 4872 4874 4873 4873 4875 4874 4874 4876 4875 4875 4877 4876 4876 4877 4877 4878 4878 4879 4878 4880 4879 4879 4881 4880 4880 4882 4881 4881 4883 4882 4882 4884 4883 4883 4885 4884 4884 4885 4885 4886 4886 4887 4886 4888 4887 4887 4889 4888 4888 4890 4889 4889 4890 4890 4891 4891 4892 4891 4893 4892 4892 4894 4893 4893 4895 4894 4894 4895 4896 4895 4896 4896 4897 4898 4897 4897 4899 4898 4898 4900 4899 4899 4901 4900 4900 4902 4901 4903 4901 4902 4903 4902 4904 4903 4905 4904 4904 4906 4905 4905 4907 4906 4906 4908 4907 4907 4908 4909 4908 4909 4910 4910 4909 4911 4911 4912 4912 4913 4913 4914 4914 4915 4915 4916 4916 4917 4917 4918 4918 4919 4919 4920 4920 4921 4921 4922 4922 4923 4923 4924 4924 4925 4925 4926 4926 4927 4927 4928 4928 4929 4929

gain, Equation is thegain energy requirement of maintenance, the NEG_maint predicted weight based on the energy supply, g/d; NEG is where gain_response the energy NEG is8.8; the predicted gain on thecalculates energy supply, g/d; NEG is where gain_response Equation 9.1; BW is current body weight,weight kg; and thebased numerator the energy content the energy gain, Equation NEG_maint is the energy requirement of maintenance, NEG is8.8; the gain, Equation 8.8; NEG_maint is the requirement of maintenance, of 1energy kg of9.1; weight gain. Equation BW is current body weight, kg; andenergy the numerator calculates the energy content Equation BWgain. is current body weight, kg; and the numerator calculates the energy content of 1 kg of9.1; weight of 1 kg of weight gain. for heifers and steers are calculated from available AATN: Weight gain prediction Weight gain gainprediction predictionfor forheifers heifersand andsteers steers calculated from available AAT Weight areare calculated from available AAT N: N: Weight gain prediction for heifers and steers are calculated from available AAT _Eff AATN: N (AATN - AATN _maint - AATN _gest) ⋅ gain _ responseAAT = 1000 ⋅ (AAT - AAT _maint - AAT _gest) ⋅ AATN _Eff 100 [Eq. 12.5] N N − 0.1262 ⋅ BWN+ 194.2 12.5 gain _ responseAAT = 1000 ⋅ (AATN - AATN _maint - AATN _gest) ⋅ AATN _Eff 100 [Eq. 12.5] 100 − 0.1262 ⋅ BW + 194.2 [Eq. 12.5] gain _ responseAAT = 1000 ⋅ 0.1262gain ⋅ BW based + 194.2on the amino acid supply, g/day; where gain_ responseAAT is the predicted −weight where gain_responseAAT is the predicted weight gain based on the amino acid supply, g/day; AATN is the response amino acid in the small intestine, Equation AAT the AATN gain_ is the predicted weight gain based on the8.11; amino acid supply,isg/day; where N_maint AATabsorbed is the amino acidrequirement absorbed infor the small Equation 8.11; _maint isprotein theisanimal’s protein NAAT where is the predicted weight gain based onAAT theAAT amino acid supply, g/day; animal’s maintenance, Equation 9.22; is Nthe AAT is protein the response amino acid in the intestine, small intestine, Equation 8.11; _maint the AATabsorbed N_gest N gain_ requirement for maintenance, Equation 9.22; AAT _gest is the protein requirement of gestation for AATN is protein the of amino acid absorbed in Equation the small 9.42; intestine, 8.11; AAT _maint N Equation is the efficiency of theis the requirement gestation for heifers, AAT is Nthe protein animal’s requirement for maintenance, Equation 9.22; N_EffAAT N_gest heifers, Equation 9.42; AAT _Eff is the efficiency of the utilisation of the available amino acids protein animal’s protein requirement for Equation 9.22; utilisation ofof thegestation availablefor amino acids for protein gain, 9.34; BWisisthe the current BW, requirement heifers, Equation 9.42; AATEquation is the efficiency of the N maintenance, N_gest N_EffAAT for and protein gain, Equation 9.34; BW isfor theprotein current BW, kg; numerator calculates the protein requirement of for heifers, Equation 9.42;gain, AAT _Eff is the the efficiency of the kg; theofnumerator calculates the protein content in 1Equation kg ofand weight gain. utilisation thegestation available amino acids 9.34; BW is the current BW, N content inof 1numerator kg weight gain. the utilisation the of available amino acids for protein gain, 9.34;gain. BW is the current BW, kg; and the calculates protein content in 1Equation kg of weight kg; 12.3and Prediction the numerator of N, calculates P and K theutilisation protein content andinexcretion 1 kg of weight gain. 12.3 Prediction N, and K utilisation and 12.3amount Prediction ofKPN, P and utilisation excretion The of N, Pofand excreted areK calculated as excretion the and difference between ingested amounts 12.3 Prediction N, and K in utilisation and excretion and incorporated in milk, used and The incorporation of N, amounts P and K The that amount of N, Pofand KPexcreted aregestation calculated asfor thebody. difference between ingested The amount of PPand KKexcreted calculated thebody. difference between The amount ofN, N,gestation and excreted are as the difference between ingested amounts and in milk, gain and corresponds tocalculated the requirements ofThe protein, P and ingested K described in K and that incorporated in milk, used inare gestation andasfor incorporation of N, amounts P and and that incorporated in milk, used gestation andforforbody. body.of The incorporation ofN, N,PPand and K in milk, chapter 9. in milk, gain and gestation corresponds to the and requirements protein, P and K of described in K that incorporated in milk, used in in gestation The incorporation in milk, gain and gestation corresponds to the requirements of protein, K described in chapter gain and9. gestation corresponds to the requirements of protein, P andPKand described in Chapter 9. chapter Nitrogen 9. 12.3.1 utilisation and excretion 12.3.1 Nitrogen in utilisation excretion The N efficiency ruminantsand is low. The utilisation is higher in lactating animals than in 12.3.1 Nitrogen utilisation and excretion 12.3.1 Nitrogen utilisation excretion growing cattle andin non-producing animals. N efficiency is calculated as theanimals total amount The N efficiency ruminantsand is low. The utilisation is higher in lactating than inof N The ruminants isislow. is is higher in lactating thanthan inofin utilized incattle relation to its total intake: growing andinin non-producing animals. Nutilisation efficiency ishigher calculated as theanimals total amount N growing The NNefficiency efficiency ruminants low.The Theutilisation in lactating animals growing cattle and non-producing animals. N efficiency is calculated as the total amount of N utilized in relation to its total intake: cattle and non-producing animals. N efficiency is calculated as the total amount of N utilized in utilized intorelation Ntointake: _ its u total intake: relation its total [Eq. 12.6] ⋅ 100 N _ u _ pct = u i ⋅ 0.16 DMIN i ⋅ _CP [Eq. 12.6] ⋅ 100 N _ u _ pct = ∑ _ u i DMIN ⋅ CP ∑ i i ⋅ 0.16 ⋅ 100 [Eq. 12.6] N _ u _ pct = 12.6 i DMIi ⋅ CPi ⋅ 0.16 ∑

i is the N utilization as a proportion of the total intake, %; N_u is the amount of where N_u_pct where N_u_pct the utilization a proportion ofofthe intake, %; N_u isamount theCP amount of N N utilized, Equation 12.7; DMIi is the matter intake thetotal i=1...n'th is where N_u_pct isisthe NNutilization as as adry proportion of the total intake, %;feedstuff, N_u is thekg/d; i of utilized, Equation 12.7; DMI is the dry matter intake of the i=1...n’th feedstuff, kg/d; CP is where N_u_pct iscontent the12.7; N utilization as adry proportion of theof total intake, N_u thekg/d; amount the crude protein inDMI theiii=1...n'th feedstuff, g/kg DM; 0.16%; isfeedstuff, the Nisproportion N utilized, Equation is the matter intake theand i=1...n'th CPin is i the i of crude protein content in the i=1...n’th feedstuff, DM; and 0.16 is the N proportion in in N utilized, Equation 12.7; is the dry matter g/kg intake of the i=1...n'th kg/d; CP is protein. the protein crude protein content inDMI the feedstuff, g/kg DM; and 0.16 isfeedstuff, the N proportion ii=1...n'th ithe the crude proteinprotein content in the i=1...n'th feedstuff, g/kg DM; and 0.16 is the N proportion in N _ protein u = N _ milk + N _ gest + N _ gain the 12.7 [Eq. 12.7] N _ u = N _ milk + N _ gest + N _ gain [Eq. 12.7] N _ u = N_u N _ milk + Ntotal _ gestamount + N _ gain where the utilised g/d; N_milk, N_gest N_gain are amount the amount of N 12.7] where N_u isisthe total amount ofofutilised N,N, g/d; N_milk, N_gest andand N_gain are [Eq. the incorporated in milk, foetus and weight gain, Equation 12.8, 12.9 and 12.10 of N incorporated in milk, foetus weight gain, Equation 12,8, 12.9 and 12.10 respectively. where N_u is the total amount of and utilised N, g/d; N_milk, N_gest and N_gain arerespectively. the amount where N_u is the total amount of and utilised N, g/d; N_gest are respectively. the amount of N incorporated in milk, foetus weight gain,N_milk, Equation 12,8, and 12.9N_gain and 12.10 of incorporated milk, foetus and weight gain, Equation 12,8, 12.9 [Eq. and 12.8]12.10 respectively. N _Nmilk = MPY ⋅ 0.in 15674 12.8 [Eq. 12.8] N _ milk = MPY ⋅ 0.15674 [Eq. 12.8] N _ milk = MPYis⋅ 0the .15674 where N_milk the amount of of N N incorporated incorporated into into milk, milk, g/d; g/d; MPY MPY is the milk protein yield, Equation where is where N_milk N_milk is is the amount amount of N incorporated into milk, g/d; MPY is the the milk milk protein protein yield, yield, 183 3.4; and 0.15674 is the N content in milk protein (ISO 8968-5|IDF 020-5:2001). Equation 3.4; 3.4; and and 0.15674 0.15674 is is the the N N content content in in milk milk protein protein (ISO (ISO 8968-5|IDF 8968-5|IDF 020-5:2001). 020-5:2001). Equation 183 183 [Eq. 12.9 N _ gest = AAT _ gest ⋅ 0 . 16 ⋅ 0 . 5 N [Eq. 12.9] 12.9] N _ gest = AAT _ gest ⋅ 0.16 ⋅ 0.5 N

whereN_gest N_gestis theN incorporated into foetus, AAT is the amino required for where N_gest isisthe the NNincorporated incorporated into thethe foetus, g/d;g/d; AAT _gest is the the amino acidacid required N N_gest where into the foetus, g/d; AAT is amino acid required N_gest for gestation, Equation 9.42; 0.16 is the N proportion in the protein; and 0.5 is the utilisation gestation, Equation 9.42; 0.16 is the N proportion in the protein; and 0.5 is the utilisation for gestation, Equation 9.42; 0.16 is the N proportion in the protein; and 0.5 is the utilisation factor for factor for the the amino acids1985). (NRC, 1985). 1985). the amino acids (NRC, factor for amino acids (NRC, N_ _ gain gain = gain _ _ prot prot ⋅⋅ 00..16 16 N = gain

[Eq. 12.10] 12.10] [Eq.

12.10

where into the BW gain, g/d; gain_prot is gain, whereN_gain N_gainis theN incorporated into BW gain, gain_prot is protein the protein where N_gain isisthe the NNincorporated incorporated into thethe BW gain, g/d;g/d; gain_prot is the the protein gain,gain, Equation Equation 9.8; 0.16 is in 9.8; and 0.16 is the N proportion in the protein. Equation 9.8; and and 0.16 is the the N N proportion proportion in the the protein. protein. Total amounts amounts of of excreted excreted N N is is assumed assumed to to be be the the difference difference between between the the N N in in the the CP CP intake intake Total from from the the ration ration and and N_u: N_u: N = N_ _ excreted excreted =∑ DMI ii ⋅⋅ CP CPii ⋅⋅ 00..16 16 − −N N __ uu ∑ DMI 134 i i

12.11] [Eq. 12.11] system NorFor – The Nordic feed [Eq. evaluation

where N_excreted N_excreted is is the the total total amount amount of of N N excreted, excreted, g/day; g/day; DMI DMIi is is the the DM DM intake intake of of the the of of where

4923 4923 4919 4922 4924 4924 4920 4923 4925 4925 4921 4924 4926 4926 4922 4925 4927 4927 4923 4926

4924 4927 4928 4928 4925 4929 4929 4928 4926 4930 4930 4929 4927 4931 4931 4930 4932 4932 4931 4928 4933 4933 4929 4932 4934 4934 4933 4930 4935 4935 4934 4931 4936 4936 4935 4932

4936 4933 4937 4937 4934 4937 4938 4935 4938 4939 4936 4939 4938 4940 4940 4939 4937 4941 4941 4940 4942 4942 4938 4941 4943 4943 4939 4942 4944 4944 4943 4940 4945 4945 4944 4941 4946 4946 4945 4942 4947 4947 4946 4943 4948 4948 4944 4947 4949 4949 4945 4948 4950 4950 4946 4949 4951 4951 4947 4950 4952 4951 4948 4952 4949 4952 4953 4953 4954 4950 4955 4951 4953 4956 4952 4957 4953 4958 4959 4960 4961 4962 4963 4964 4965 4966 4967 4968 4969 4970

N _ gain =9.8; gainand _ prot ⋅ 0.16 [Eq. intake 12.10] Equation 0.16 is the is N proportion in the protein. Total Total amounts amounts of of excreted excreted N N is assumed assumed to to be be the the difference difference between between the the N N in in the the CP CP intake from the ration and N_u: from rationis and N_u: wherethe N_gain the N incorporated into to thebeBW g/d; gain_prot gain, Total amounts of excreted N is assumed thegain, difference between is thethe N protein in the CP intake

Equation 9.8;= and 0.16 is the from the ration N_u: N _ excreted DMI ⋅ CP ⋅ 0N .16proportion − N _ u in the protein. ∑and N _ excreted = ∑ DMI ii ⋅ CPii ⋅ 0.16 − N _ u

[Eq. [Eq. 12.11] 12.11]

ii Total amounts of excreted isassumed difference between theinNthe in[Eq. theintake CP intake Total amounts excreted be be thethe difference between the N CP N _ excreted = ∑of DMI ⋅ 0Nis .16 −assumed N _ u to to 12.11] i ⋅ CPi N the ration and iN_u: from the ration andisN_u: where N_excreted the total amount of N excreted, g/day; DMI is the DM intake of the of

where N_excreted is the total amount of N excreted, g/day; DMIii is the DM intake of the of content in i=1...n'th feedstuff, the i=1...n'th feedstuff, kg DM/day; CP isNthe the crude crude protein protein content in the the i=1...n'th feedstuff, the i=1...n'th feedstuff, kgtotal DM/day; CPofii is is the DM intake of the of where N_excreted is ithe amount g/day; DMI iamounts N _ excreted = ∑isDMI ⋅ CP −N _inu CP; excreted, [Eq. 12.11] g/kg DM; 0.16 the proportion N and N_u is the total of N utilized, i ⋅ 0.16 of g/kg DM; 0.16 is the proportion of N in CP; and N_u is the total amounts of N utilized, is the crude protein content in the i=1...n'th feedstuff, the i=1...n'th feedstuff, kg DM/day; CP i i Equation 12.15. Equation g/kg DM;12.15. 0.16 is the proportion of N in CP; and N_u is the total amounts of N utilized,

from

12.11

where N_excreted is the total amount of N excreted, g/day; DMI is the DM intake of the of the

i Equation 12.15. intake of the of where N_excreted is the total amount ofisN excreted, DMI i is the Excreted N in the assumed to be the difference between N and apparent i=1...n’th kgis CP the crude g/day; protein content inDM the i=1...n’th feedstuff, g/kg Excreted Nfeedstuff, infeedstuff, the faces faces isDM/day; assumed to be difference between N intake intake and apparent i isthe the i=1...n'th kg DM/day; CP the crude protein content in the i=1...n'th feedstuff, i digested N. DM; 0.16 is the proportion of N in CP; and N_u is the total amounts of N utilized, Equation 12.15. digested Excreted N0.16 in the faces is assumed to in beCP; the and difference intake and g/kg DM;N. is the proportion of N N_u isbetween the total N amounts of Napparent utilized, digested EquationN. 12.15. ⎛ ⎞ Excreted N⎛inDMI the feaces−istdassumed apparent digested N. ⎞ .16to be the difference between N intake and [Eq. N N_ _ faeces faeces = = ⎜⎜ ∑ CPii − td _ _ CP CP⎟⎟⎟ ⋅⋅ 0 0.16 ∑ DMIii ⋅⋅ CP [Eq. 12.12] 12.12] ⎠⎠⎞ to be the difference between N intake and apparent Excreted N ⎝⎜⎝⎛inii the faces is assumed ⎜ ⎟ 12.12 N _ faeces = DMI ⋅ CP − td _ CP ⋅ 0 . 16 ∑ [Eq. 12.12] i i digested N. ⎜ ⎟ ⎝ i is the amount of ⎠N excreted in the faeces, g/d; DMI is the dry matter intake where N_faeces where N_faeces is the amount of N excreted in the faeces, g/d; DMIii is the dry matter intake where N_faeces is the amount of inprotein the faeces, g/d; matter g/kg intake of the the content in the feedstuff, of the kg/d; i is the dry ⎛ feedstuff, ⎞ iiNis isexcreted the crude crude protein content in DMI theisi=1...n'th i=1...n'th feedstuff, g/kg of the i=1...n'th i=1...n'th feedstuff, kg/d;ofCP CP where N_faeces istotal amount N⋅ 0excreted in7.52; the faeces, g/d; DMI the dry matter intake ⎜⎜is∑ DMI ⎟⎟the N _ faeces =feedstuff, ⋅kg/d; CP .Equation 16 i [Eq. 12.12] i=1...n’th CP_iCP isCP, crude protein content in is the i=1...n’th feedstuff, g/kg DM; td_CP ithe i − td DM; td_CP the digested and 0.16 the proportion of N in CP. DM; td_CP is the total digested CP, Equation 7.52; and 0.16 is the proportion of N in CP. ⎝ i feedstuff, kg/d; CP ⎠ is the crude protein content in the i=1...n'th feedstuff, g/kg of the i=1...n'th

is the total digested CP, Equation i7.52; and 0.16 is the proportion of N in CP.

DM; td_CP isprotein the total CP, Equation 7.52; and 0.16 is the proportion of N in CP. Overfeeding or digested pourly ration N in urine.The difference between Overfeeding protein pourly balanced balanced rationinincreases increases N g/d; in the the urine.The difference between where N_faeces is theoramount oftheNNexcreted thethe faeces, DMI dry matter intake i is the total amount of excreted N and excreted in faeces is assumed to be the N excreted in Overfeeding protein or pourly balanced rationinincreases Nisinassumed the urine.The difference between total total amount of excreted N and the N excreted the faeces to be the N excreted in of Overfeeding the i=1...n'th protein feedstuff, or pourly kg/d; balanced CP ration increases N in the urine.The difference between is the crude protein content in the i=1...n'th feedstuff, g/kg i the urine: amount of excreted N and the N excreted in the faeces is assumed to be the N excreted in the urine: the urine: total excreted N and the excreted7.52; in theand faeces to be the DM; amount td_CP isofthe total digested CP,NEquation 0.16isisassumed the proportion of N N excreted in CP. in the urine: N _ urine = N _ excreted − N _ faeces [Eq. N _ urine = N _protein excreted N _ faeces [Eq. 12.13] 12.13] Overfeeding or−pourly balanced ration increases N in the urine.The difference between 12.13 total amount excreted the N excreted in the faeces is assumed to be the N excreted N _ urine = N _ofexcreted − NN_and faeces [Eq. 12.13]in where N_urine N_excreted total amount amount of of N N excreted, excreted, Equation where N_urine is is the the N N excretion excretion in in urine, urine, g/d; g/d; N_excreted is the the total where N_urine is the N excretion N_excreted is the urine: Equation 12.11; and N_faeces is the amount of N excreted in faeces, Equation 12.12. Equation 12.11; is the amount of N_excreted N excreted in 12.12. 12.11;N_urine and N_faeces the amount of N excreted in faeces, Equation 12.12.of where isand theN_faeces Nisexcretion in urine, g/d; is faeces, the totalEquation amount N excreted, N _ urine = N _ excreted − N _ faeces [Eq. 12.13] Equation 12.11; and N_faeces is the amount of N excreted in faeces, Equation 12.12. N _ urine N _ urine [Eq. N 12.14 [Eq. 12.14] 12.14] N __ urine urine __ pct pct = = N _ excreted ⋅⋅ 100 100 N _excreted urine where N_urine isN_the N excretion in urine, g/d; N_excreted is the total amount of [Eq. N excreted, 12.14] N _ urine _ pct = ⋅ 100 where N_urine_pct isN_faeces excreted ininurine thethe total NN excretion, where N_urine_pct isthe theNN excreted urine aproportion proportion of total excretion, Equation 12.11;Nand is the amount ofasas Na excreted in of faeces, Equation 12.12. %;%; N_urine is _ excreted N_urine the NEquation in urine, 12.13; Equation and N_excreted is the total amount of N excreted, the N in isurine, and12.13; N_excreted is the total amount of N excreted, Equation 12.11. Equation 12.11. N _ urine [Eq. 12.14] N _ urine _ pct = ⋅ 100 N _ excreted 12.3.2 Phosphorus utilisation and excretion 184 12.3.2 Phosphorus utilisation and excretion 184 The 1980). Therefor The excessary excessaryfed fedPPover overrequirement requirementisisexcreted excreted thefaeces faeces(ARC, (ARC, 1980). Therefore overfeeding 184 ininthe overfeeding P is useless. P is useless. PP efficiency efficiencyisiscalculated calculatedas: as: P _ u _ pct =

P_u

∑ DMI i ⋅ Pi i

184

⋅100

[Eq. 12.15]

12.15

where P_u_pct P_u_pctisisthe thePPutilization utilizationas asaa proportion proportionof oftotal totalPPintake, intake,%; %;P_u P_uisisthe theamount amountof ofPP utilized, where EquationEquation 12.16; DMI the iDM intake of the of of the theof i=1...n’th feedstuff, kg DM/day; and Pi is the utilized, 12.16; is the DM intake the i=1...n'th feedstuff, kg DM/day; i isDMI P content inPthe i=1...n’th DM. g/kg DM. and Pi is the content in thefeedstuff, i=1...n'thg/kg feedstuff, P _ u = (P _ milk + P _ gest + P _ gain ) ⋅ 0.7

[Eq. 12.16]

12.16

4971 4972 4973 4974 4975 4976 4977 4978

whereP_u P_uisisthe thetotal totalamount amountofofutilised utilisedP,P,g/d; g/d;P_milk, P_milk,P_gest P_gestand andP_gain P_gain where areare thethe requirement of P for milk, gestation weight gain,and Equation and 9.53 requirement of P forand milk, gestation weight9.51, gain, 9.52 Equation 9.51,respectively; 9.52 and 9.530.7 is the absorption coefficient. 0.7 is the absorption coefficient. respectively;

4979

P _ excreted = ∑ DMI − P _ uevaluation system i ⋅ Pi feed NorFor – The Nordic

4980 4981

where P_excreted is the total amount of P excreted, g/d; DMIi is the DM intake of the of the

Phosphorus excretion occurs normally in the faeces and only a minor part excretes in the urine (ARC, 1980). The excretion is estimated as the difference of total P intake and the amounts of P incorporated in milk, foetus and weight gain. i

[Eq. 12.17]

135

4970 4970 4971 4971 4972 4972 4973 4973 4974 4974 4975 4975 4976 4976 4977 4977 4978 4978

[Eq. 12.16] where where P_u P_u is is the the total total amount amount of of utilised utilised P, P, g/d; g/d; P_milk, P_milk, P_gest P_gest and and P_gain P_gain are are the the requirement of of P P for for milk, milk, gestation gestation and and weight weight gain, gain, Equation Equation 9.51, 9.51, 9.52 9.52 and and 9.53 9.53 requirement respectively; respectively; 0.7 0.7 is is the the absorption absorption coefficient. coefficient.

Phosphorus excretionoccurs occurs normally in the faeces and only a minor part excretes in the urine Phosphorus Phosphorus excretion excretion occurs normally normally in in the the faeces faeces and and only only aa minor minor part part excretes excretes in in the the (ARC, 1980). The The excretion is estimated as as thethe difference urine (ARC, 1980). The excretion is estimated estimated as the differenceofof oftotal totalPP Pintake intake and and the urine (ARC, 1980). excretion is difference total intake and the amounts of P incorporated in milk, foetus and weight amounts of P P incorporated incorporated in milk, milk, foetus gain. and weight weight gain. gain. amounts of in foetus and

4979 4979

P =∑ DMI ⋅ P − P _ u ∑ P __ excreted excreted = ∑ DMI iii ⋅ Piii − P _ u

4980 4980 4981 4981 4982 4982 4983 4983

i the DM intake of the of the where P_excreted is excreted, g/d; DMI is the DM intake the ofand the P_u is the where P_excreted is the the total total amount of P excreted, g/d;i=1...n’th DMIiii is feedstuff, kg DM/day; and Pamount is the of PP content in the feedstuff, g/kgofDM; i i=1...n'th feedstuff, feedstuff, kg kg DM/day; DM/day; and and P Pii is is the the P P content content in in the the i=1...n'th i=1...n'th feedstuff, g/kg DM; i=1...n'th i and amount of P utilised in milk, gestation gain, Equation 12.16. feedstuff, g/kg DM; and and P_u P_u is is the the amount amount of of P P utilised utilised in in milk, milk, gestation gestation and and gain, gain, Equation Equation 12.16. 12.16.

4984 4984 4985 4985 4986 4986 4987 4987 4988 4988 4989 4989 4990 4990 4991 4991 4992 4992 4993 4993

4994 4994 4995 4995 4996 4996

4997 4998 4999 5000 5001 5002 5003 5004 5005 5006 5007 5008 5009 5010 5011 5012 5013 5014 5015

[Eq. 12.17] 12.17] [Eq.

iii

12.17

where P_excreted is the total amount of P excreted, g/d; DMI is the DM intake of the of the i=1...n’th

12.3.3 Potassium utilisation utilisation andexcretion excretion 12.3.3 12.3.3 Potassium Potassium utilisation and and excretion

Cattle normally ingest more K than the requirement. Potassium utilisation is as: Cattle Potassium utilisation is calculated calculated as: as: Cattle normally normallyingest ingestmore moreKKthan thanthetherequirement. requirement. Potassium utilisation is calculated K_ _u u K = ⋅⋅ 100 100 K __ uu __ pct pct = K ∑ DMI ∑ DMIii ⋅⋅ K K ii ∑ i

iii

[Eq. 12.18] 12.18] [Eq.

i

12.18

where K_u_pct isisthe the KKutilization utilization asas proportion of of total K intake, intake, %; %; K_uK_u is the the total whereK_u_pct K_u_pctis theK utilizationas a proportion total K intake, is total the total amount of where aa proportion of total K %; K_u is amount of utilized, Equation 12.19; DMI the DM of the offeedstuff, the i=1...n'th feedstuff,and Ki is K utilized, DMI DM intake ofintake the i=1...n’th kg DM/day; ii is amount of K KEquation utilized, 12.19; Equation 12.19; DMI i is the i is the DM intake of the of the i=1...n'th feedstuff, isi=1...n’th the K K content content in the the g/kg i=1...n'th feedstuff, g/kg g/kg DM. DM. kg DM/day; and Kthe ii is theDM/day; K content inK feedstuff, DM.feedstuff, the in i=1...n'th kg and i

K_ _u u= = (K K_ _ milk milk + +K K_ _ gest gest + +K K_ _ gain gain) ⋅⋅ 0 0..9 9 K

12.19 [Eq. 12.19] 12.19] [Eq. where K_u K_u is is the the total total amount amount of of utilised utilised K, K, g/d; g/d; P_milk, P_milk, P_gest P_gest and and P_gain P_gain are are the the where requirement K milk, gestation and gain, Equation 9.66, and where K_u of is the total amount of utilised K, g/d; P_gest and P_gain requirement of K for for milk, gestation and weight weight gain,P_milk, Equation 9.66, 9.67 9.67 and 9.68 9.68are the requirement respectively; 0.9gestation is the the absorption absorption coefficient. of K for milk, and weight gain, Equation 9.66, 9.67 and 9.68 respectively; 0.9 is the respectively; 0.9 is coefficient. absorption coefficient.

185 185

The total totalamount amountofofexcreted excretedKKisiscalculated: calculated: The K _ excreted = ∑ DMI i ⋅ K i − K _ u i

[Eq. 12.20]

12.20

where K_excreted is the K excreted, g/day; DMI is the DM intake of the of the i=1...n’th feedstuff,

i DM intakeg/kg of the of and the i=1...n'th where K_excreted thethe K excreted, i is thefeedstuff, kg DM/day; and Kisi is K contentg/day; in theDMI i=1...n’th DM; K_u is the amount of feedstuff, kg DM/day; and Ki is the K content in the i=1...n'th feedstuff, g/kg DM; and K_u is K utilized in milk, gestation and gain, Equation 12.19. the amount of K utilized in milk, gestation and gain, Equation 12.19.

References 12.4 References ARC (Agricultural Reasearch Council), 1980. The nutrient requirements of ruminant livestock. Commonwealth

ARC – Agricultural Reasearch Council, 1980. The nutrient requirements of ruminant Agricultural Bureaux, Slough, UK. livestock. Slough England: Commonwealth Agricultural Bureaux.

ISO 8968-5/IDF 020-5:2001. Milk - Determination of nitrogen content. International Dairy Federation, Brussels, ISO Belgium. 8968-5/IDF 020-5:2001. Milk - Determination of nitrogen content. International Dairy Sjaunja, L.-O., L. Bævre,Belgium. I. Junkkarainen, J. Pedersen and J. Setälä, 1990. A Nordic proposition for an energy corrected Federation, Brussels, milk (ECM) formula. 26th session of the International Committee for recording the productivity of Milk Animals (ICRPMA). Sjaunja, L.-O., L. Bævre, I. Junkkarainen, J. Pedersen and J. Setälä, 1990. A Nordic

proposition for an energy corrected milk (ECM) formula. 26th session of the International Committee for recording the productivity of Milk Animals (ICRPMA).

136

NorFor – The Nordic feed evaluation system

5016 5017 5018

13.0 Standardfeed feedvalue value 13. Standard

5019 5020 5021 5022 5023 5024 5025 5026 5027

Traditional feed evaluation systems determine fixed and additive feed values of feedstuffs. The NorFor system, however,systems is a rationdetermine evaluationfixed system in principle individual Traditional feed evaluation andand additive feed each values of feedstuffs. The feedstuff has no fixed feed value because of the interactions in the digestion and metabolism NorFor system, however, is a ration evaluation system and in principle each individual feedstuff of there is aofneed compare feedstuffs for trading or in of thenutrients. hasnutrients. no fixedNevertheless, feed value because the to interactions in the digestion andpurposes metabolism optimization concentrate mixtures by linear programming in thepurposes feedstuff or industry. Nevertheless,ofthere is a need to compare feedstuffs for trading in the optimization of Therefore, we calculate feed values based on standardized conditions, where important model concentrate mixtures by linear programming in the feedstuff industry. Therefore, we calculate feed variables are fixed, e.g. fractional passage rates and efficiency in the microbial protein values based on standardized conditions, where important model variables are fixed, e.g. fractional synthesis (see Table 13.1). These fixed values represent feed rations for diets used in passage rates and efficiency inSweden. the microbial protein synthesis (see Table 13.1). These fixed values Denmark, Iceland, Norway and

5028 5029 5030 5031 5032 5033 5034 5035 5036 5037

Standard feed values for NEL, AATN and PBVN were calculated in the following way. First Standard feed values for NEL,proportion AATN andand PBV in the animal BW, DMI, concentrate content ofcalculated CP, NDF, ST andfollowing RestCHOway. wereFirst set animal N were BW, DMI,toconcentrate proportion and of CP, NDF, STthe andstandard RestCHO according Table 13.1 to standardize thecontent feed ration from which feedwere valueset is according calculated. Then for RLI, fractional rates and to Table 13.1 to values standardize thecorrNDF_fac, feed ration from which passage the standard feedr_emCP value isare calculated. Then determined usingcorrNDF_fac, Equations 13.1fractional to 13.8 (Table 13.1). values substitute the values for RLI, passage ratesThese and r_emCP are determined using Equations corresponding equations in chapter 7 whensubstitute calculating standard feed equations value of a in feedstuff 13.1 to 13.8 (Table 13.1). These values thethe corresponding Chapter 7 when with specificthe dietary characteristics. of NEL, AAT N and PBVN The standard calculating standard feed value The of a standard feedstufffeed withvalues specific dietary characteristics. (Equation 8.3, and 13.9, and respectively) receive a8.3, suffix thatand tells which standardisedreceive DMI, a suffix feed values of8.11 NEL, AAT PBV (Equation 8.11 13.9, respectively) N N 8that or 20 kgwhich DM, the calculations are based upon: NEL8the , NEL PBV and , NEL , 20, PBVN8,are N20, AAT tells standardised DMI, 8 or 20 kg DM, calculations based upon:N8NEL 8 20 AAT PBVN20,. PBV , AAT and AAT .

5038 5039

Standardised RLI, corrNDF_fac, fractional passage rates and r_emCP are calculated as:

5040

RLIstd =

5041 5042 5043 5044 5045 5046 5047 5048 5049 5050 5051 5052 5053 5054 5055 5056

M. Åkerlind and H. Volden

M. Åkerlind and H. Volden

represent feed rations for diets used in Denmark, Iceland, Norway and Sweden.

N8

N20

N8

N20

Standardised RLI, corrNDF_fac, fractional passage rates and r_emCP are calculated as: 280 DMIstd ⋅ 370

where, RLI

[Eq. 13.1]

13.1

is the rumen load index; 280 is the sum of rumen degraded ST and RestCHO in the

std where, RLIDM; rumen loadstandardized index; 280 isDM the sum of rumen RestCHO in and 370 std is the diet, g/kg DMI the intake, which degraded is either 8ST orand 20 kg DM/day; std is is the diet, g/kg DM; DMI the standardized DM intake, which is either 8 or 20 kg DM/day; std is the NDF content, g/kg DM. and 370 is the NDF content, g/kg DM.

The NDF correction factor is calculated as:

Table 13.1. Fixed values for the listed parameters used for determining standard feed values. corrNDF Parameter_ fac = e

(

− 0 .24 ⋅ RLI

)

2 .9

std

Equation

8 kg DMI[Eq. 13.2]20 kg DMI

where, is kg the correction factor for the NDF degradation rate; CurrentcorrNDF_fac body weight, 600and RLIstd is the600 rumen load index, Dry matter intake,Equation kg/d 13.1. 8 20

Concentrate proportion of the ration, % DMIstd ⋅1000 DM rCP, _ kplg/kg = 2.45 + 0.055 ⋅ + 0.0004 ⋅ 502 NDF, g/kg DM 6000.75 ST+RestCHO, g/kg DM DMI DM std ⋅ 1000 − 0.02 ⋅ 50 rrdST+rdRestCHO, _ kpc = 2.504 + 0.1375g/kg ⋅ Correction factor for NDF600 degradation rate Passage rate for liquid, %/h Passage rate for CP and ST in concentrate, %/h Passage rate for NDF in concentrate, %/h 187 Passage rate for CP and ST in roughage, %/h Passage rate for NDF in roughage, %/h Efficiency of microbial CP synthesis, g/kg rd_OM

13.2 13.3 13.4 13.5 13.6 13.7 13.8

50 50 160 160 [Eq. 13.3] 370 370 310 310 280 [Eq. 13.4]280 0.89857 0.89857 7.07943 12.5236 3.33733 6.08733 1.43505 2.61755 2.30177 4.47943 0.998465 1.64242 133.383 184.329

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_13, © Wageningen Academic Publishers 2011

137

5041 5042 5042 5043 5043 5044 5044 5045 5045 5046 5046 5047 5047 5048 5048 5049 5049 5050 5050 5051 5051 5052 5052

5053 5053 5054 5054 5055 5057 5055 5057 5056 5058 5056 5058 5057 5059 5059 5060 5058 5060 5059 5060 5061 5061

5061 5061 5062 5062 5063 5063 5064 5062 5064 5065 5062 5063 5065 5066 5064 5066 5067 5065 5067 5068 5066 5068 5069 5067 5069 5070 5068 5070 5071 5069 5071 5072 5070 5072 5071 5073

5073 5074 5072 5074 5075 5073 5075 5076 5074 5076 5077 5075 5077 5078 5076 5078 5079 5077 5079 5080 5078 5080 5081 5079 5081 5082 5080 5082 5083 5081 5083 5084 5082 5084 5085 5083 5085 5086 5084 5086 5087 5085 5087 5088 5086 5088 5089 5087 5089 5090 5088 5090 5091 5089 5091 5092 5090 5092 5093 5091 5093 5094 5092 5094 5095 5093 5095 5096 5094 5096 5097 5095 5097

where, RLIstd is the rumen load index; 280 is the sum of rumen degraded ST and RestCHO in is the rumen load index; 280 is the sum of rumen degraded ST and RestCHO in where, RLIstd the diet, g/kg DM; DMIstd is the standardized DM intake, which is either 8 or 20 kg DM/day; the diet, g/kg DM; DMIstd is the standardized DM intake, which is either 8 or 20 kg DM/day; and 370 is the NDF content, g/kg DM. and 370 is the NDF content, g/kg DM. The NDF correction factor is calculated as: The NDF correction factor is calculated as:

The NDF correction factor is calculated as:

((

− 0 .24 ⋅ RLI

− 0 .24 ⋅ RLI std corrNDF fac e − 0 .24 ⋅ RLI stdstd corrNDF __ fac == e

))

2 2 ..9 9 2 .9

[Eq. 13.2] [Eq. 13.2]

13.2

where, corrNDF_fac is the correction factor for the NDF degradation rate; and RLIstd is the where,corrNDF_fac corrNDF_facisisthe thecorrection correctionfactor factor NDF degradation rate; is the rumen the where, forfor thethe NDF degradation rate; andand RLIRLI std isstd rumen load index, Equation 13.1. rumen load index, Equation load index, Equation 13.1. 13.1. DMIstd ⋅ 1000 ⋅ 1000 + 0.0004 ⋅ 5022 r _ kpl = 2.45 + 0.055 ⋅ DMIstd r _ kpl = 2.45 + 0.055 ⋅ + 0.0004 ⋅ 50 0.75 600 0 . 600 75

[Eq. 13.3] [Eq. 13.3]

13.3

⋅ 1000 DMIstd DMI ⎞ ⎛ 0.1375 ⋅ DMI std ⋅1000 13.4 − 0.02 ⋅ 50 [Eq. 13.5] 13.4] r _ kpNDFc kpc = 2.504 .02 ⋅50⎟ ⋅ 0.43 = ⎜ 2++.504 + 0.1375⋅std ⋅ 1000 [Eq. rr __ kpc = 2.504 − 0.02−⋅050 [Eq. 13.4] DMIstd ⋅1000 ⎞⎠ ⎛⎝ 0.1375 ⋅ 600 600 600 r _ kpNDFc= ⎜ 2.504 + 0.1375⋅ − 0.02 ⋅ 50⎟ ⋅ 0.43 [Eq. 13.5] 600 ⎠ ⎝ DMIstd 1000 ⋅ ⎞ ⎛⎛ ⋅ DMI 1000 ⎞ std kpNDFc= 504 + 1375std ⋅⋅ ⋅ 1000 − [Eq. 13.5] 13.5 0..02 02 2⋅⋅ 50 50⎟⎟ ⋅⋅ 0 0..43 43 = ⎜⎜+220..504 + 00DMI −0 rrr ___ kpNDFc ..1375 kpr = 0.35 [Eq. 13.6] 600+ 0.0002 ⋅ 50187 ⎠⎠ ⎝⎝ .022 ⋅ DMIstd0.⋅75 600 1000 2 r _ kpr = 0.35 + 0.022 ⋅ 600 + 0.0002 ⋅ 50187 [Eq. 13.6] 600 0.75 DMI ⋅ 1000 DMI std std ⋅ 1000 + 13.6 rr __ kpr 0..0002 0002 ⋅⋅ 50 50 22 [Eq. 13.6] kpr = = 00..35 35 + + 00..022 022 ⋅⋅ +0 0 ..75 0 75 600 600 0.78 ⋅1.93668 r _ kpNDFr = 0.480339 + [Eq. 13.7] 0.78 ⋅1.93668 − 3.19822 ⎛ DMI std ⋅ 370 ⎞ r _ kpNDFr = 0.480339 + 13.7 [Eq. 13.7] 1+ ⎜ − 3.19822 ⎟ DMI⋅ 7std ⋅ 370 ⎠⎞ ⎛ 600 . 48383 ⎝ ⋅1.93668 ⎟ r _ kpNDFr = 0.480339 + 1 + ⎜⎝ 6000⋅.78 [Eq. 7.48383 ⎠ −− 33..19822 [Eq. 13.7] 13.7] ⋅ DMI 370 ⎛ ⎞ 19822 std std where, r_kpl is the passage 1 +rate ⎜ of the liquid ⎟ phase out of the rumen; DMIstd is the .48383 where, r_kpl r_kplDMI, thepassage rate the phase outthe ofrumen; the DMI where, isisthe rate of⋅ 7of the liquid outBW, of DMI ⎝ 600 ⎠ phase is the is standardized 8 passage or 20 kg DM; 600 isliquid the current kg; 50 isrumen; the proportion of the standardized std std

standardized 8%;or600 20 is kgthe 600 isBW, the current kg; 50 is the of in diet, %; r_kpc concentrate diet, r_kpc isDM; the passage rate protein starch inofproportion concentrates; DMI, 8 or 20inDMI, kg DM; current kg;of50 isBW, the and proportion concentrate where, r_kpl the passage rate of the liquid phase out of DMI ispassage theroughage std concentrate diet, %; r_kpc isofand the passage of protein and starch in is concentrates; r_kpNDFc isinis the passage rate NDF in concentrate; isrumen; the passage rate of where, r_kpl is the passage rate of the liquid phase outr_kpr of the the rumen; DMI is the passage rate of protein starch inrate concentrates; r_kpNDFc the rate of NDF in std standardized DMI, 8isor kg of DM; 600 isconcentrate; therate current BW, 50starch; is the proportion of r_kpNDFc the passage NDF inof r_kprkg; is the passage rate protein andisstarch; r_kpNDFr is the passage of roughage NDF; and 370 is of theroughage NDF concentrate; r_kpr the20rate passage rate roughage protein and r_kpNDFr is the passage rate concentrate inNDF; diet, %; r_kpc the passage rate andNDF; starchand in concentrates; protein and starch; theNDF passage rateof of roughage 370 is the NDF content in diet, g/kgr_kpNDFr DM. of roughage and 370 isisisthe content inprotein diet, g/kg DM. r_kpNDFc is theg/kg passage content in diet, DM. rate of NDF in concentrate; r_kpr is the passage rate of roughage protein and starch; is r_kpNDFr is from the passage rate of equation: roughage NDF; and 370 is the NDF The The fixed fixed r_emCP r_emCP iscalculated calculated fromthe thefollowing following equation: content in r_emCP diet, g/kgisDM. The fixed calculated from the following equation: ⎛1000 1000⋅⋅ DMI DMI

⎞

std ⎞ ⋅ 55.6 + 0.4166 ⋅ 310 − 0.0008868⋅31022 −54.56 ⎛ std rr __ emCP ⎟⎟ ⋅ 55.6 +from emCP ln⎜⎜ 0.4166 − 0.0008868 ⋅ 310 − 54.56 The fixed== ln r_emCP is calculated the⋅ 310 following equation: DMI ⎛⎝⎝1000⋅600 2 600 std ⎠⎞⎠

[Eq. 13.8]

13.8

r _ emCP= ln⎜ ⎟ ⋅ 55.6 + 0.4166⋅ 310 − 0.0008868⋅ 310 − 54.56 [Eq. 13.8] 600 ⎝ ⎠ ⋅ 1000 DMI ⎛ DMIefficiency 2 − 54.56 DMI is the std ⎞⎞ ⋅ 55.6 +of where, r_emCP isis⋅ the rumen microbial protein = ln _ emCP emCP ln⎛⎜⎜ 1000 ..4166 ⋅⋅ 310 − 0 ⋅⋅protein 310 ⎟⎟ ⋅ 55.6 + 0 [Eq. 13.8] std = rr _ 0of 4166 310 −microbial 0..0008868 0008868 3102synthesis; − 54 .56 where, r_emCP thestd efficiency rumen synthesis; DMI std is the standardized 600 ⎝⎝ DMI ⎠⎠kg DM; 600 where, r_emCP is the efficiency of rumen microbial protein synthesis; DMI standardized 8 or 20 600 is the standardized BW, kg;the and 310 std is is thethe of ST + RestCHO DMI 8 or 20 kg DM; 600 is the standardized BW, kg; and 310 is concentration

standardized DMI 8+ orRestCHO 20 kg DM; is theg/kg standardized BW, kg; and 310 is the concentration of ST in 600 the diet, DM. in the diet, g/kg DM. where, r_emCP the+ efficiency rumen concentration ofisST RestCHO of in the diet,microbial g/kg DM.protein synthesis; DMIstd std is the standardized DMI or 20 kgfeed DM; 600 is the8,standardized kg;AAT and N20 310uses is the The calculation of 8standard value, NEL NEL20, AATBW, the Equations N8 and The calculation of standard feedinvalue, NEL , AATN8 and AATN20 uses the Equations 7.6 8, NEL 20to concentration ofof ST + RestCHO the7.36 diet, DM. NEL AAT Equations The calculation standard value, NEL 7.6 to7.14, 7.17 to 7.23. 7.30feed to 7.34, tog/kg 7.40 7.51, 7.54, 7.55, 7.57,the 7.59, 8.1 to 8,7.38, 20, AAT N8 and N20 uses to7.14, 7.177.17 to 7.23. 7.30 to to 7.34, 7.36 to 7.38, 7.40 to 7.51, 7.54, 7.55, 7.57, 7.59, 8.1 to 8.3 and 7.6 and to7.14, to 7.23. the 7.30equations 7.34, 7.36 7.38, 7.40 to 7.51, 7.54, 7.55, 7.59, 8.1isto 8.3 8.11. However, 7.1 toto7.5 is changed to equation 13.37.57, to 13.7, 7.16 8.11. However, Equations 7.1 to 7.5 are changed to Equation 13.3 to 13.7, 7.16 is changed The calculation of standard feed value, NEL , NEL and AAT13.3 the Equations 8 20 N8 N20 8.3 and 8.11. However, to 7.5 isnutrient changed to 13.7, 7.16i,isSTi to 13.2 changed to 13.2 and 7.29the is equations changed to7.1 13.8. composition (CP CFat i,to i, NDF 8The 20, AAT N8 equation N20 uses and 7.29 is changed to 13.8. The nutrient composition (CP , CFat , NDF , ST , etc.) is the composition 7.6 to7.14, 7.17 to 7.23. 7.30 to 7.34, 7.36 to 7.38, 7.40 to 7.51, 7.54, 7.55, 7.57, 7.59, 8.1 to i i i i i changed to composition 13.2 and 7.29ofisthe changed 13.8. Thetonutrient composition (CP etc.) is the specifictofeedstuff which the feed values are calculated for i, CFat i, NDF i, ST of the feedstuffof tothe which the7.1 feed are calculated for and DMI kg.foris 8.3 and 8.11. the equations to values 7.5 is to feed equation 13.3 tocalculated 13.7, etc.) isspecific the composition specific feedstuff tochanged which the values are and DMI 1However, kg. i is 17.16 i is changed and DMIto 1 kg.and 7.29 is changed to 13.8. The nutrient composition (CPii, CFatii, NDFii, STii i is13.2 etc.) is the composition the specific feedstuff which the feed values calculated for The standardized standardized valueofofPBV PBV PBV in Equation 13.9 assumes a recirculation fixed recirculation of The value andand PBV intoEquation 13.9 assumes aare fixed N8N8 N20 N20 and DMI 1 kg. urea, based a fixed intake, totoavoid overestimating it: ii is on The standardized ofCP PBV PBV in Equation 13.9 of urea, based on avalue fixedCP intake, avoid overestimating it: assumes a fixed recirculation N8 and N20 of urea, based on a fixed CP intake, to avoid overestimating it: The and PBVN20 PBVstandardized 160 ⋅ 0.of 046PBV ) − r _N8 [Eq. 13.9] 13.9 N8mCP N20 in Equation 13.9 assumes a fixed recirculation N = rd _ CP + (value of urea, fixed CP) −intake, [Eq. 13.9] PBV rd _ CPon + (a160 ⋅ 0.046 r _ mCPto avoid overestimating it: N =based where, PBVN is the standard feed value for protein balance in the rumen, g/day; rd_CP is the (160 ) −−feed [Eq. 13.9] PBV =PBV rd ⋅⋅ 00..046 rr __ mCP PBVdegraded rd __NCP CP 160 046of mCP N is ++the the standard valueoffor balance in the rumen, 7.8; g/day; is the where, CP in rumen 1kg DM theprotein specific feedstuff, Equation 160rd_CP is a fixed N =PBV where, N is the standard feed value for protein balance in the rumen, g/day; rd_CP is the CP CP degraded rumen of 1kg DM of the specific feedstuff, 7.8; 160 a of fixed value for CP in the standardised ration (Table 13.1); 0.046 is Equation theEquation assumed proportion Nvalue that for CP degraded in the rumen of 1kg DM of the specific feedstuff, 7.8; 160 is is a fixed where, PBV is the the standard value forisprotein balance theassumed rumen, g/day; isNthe N value for CPback standardised ration (Table 13.1); 0.046 isinthe proportion that is recycled to the rumen;feed and r_mCP the rumen microbial CP yielded by rd_CP 1 kg of DM of Nin CPrecycled degraded in the rumen of 1kg DM of the specific feedstuff, Equation 7.8; by 1601 iskga DM fixedof is back to the rumen; and r_mCP is13.1). the rumen microbial CP yielded the specific feedstuff, Equation 7.30 (Table value in the standardised ration (Table 13.1); 0.046 is the assumed proportion of N that the Equation 7.30 (Table 13.1). 138specific for CPfeedstuff, NorFor – The Nordic feed evaluation system is recycledfeed back to theofrumen; and r_mCP is the rumen microbial yielded 1 kg Standard values some common feedstuffs are presented inCP table 13.2. by Note thatDM theof the specific feedstuff, Equation 7.30 than (Table 13.1). Standard feed valuesAAT of some feedstuffs presented table 13.2. Note that the lower AAT mainly due toin increased microbial standard feed value N8 is common N20 are standard feed value AAT is lower than AAT mainly due to increased microbial

in the standardised ration (Table 13.1); 0.046 is the assumed proportion of N that is recycled back to the rumen; and r_mCP is the rumen microbial CP yielded by 1 kg DM of the specific feedstuff, Equation 7.30 (Table 13.1). Standard feed values of some common feedstuffs are presented in Table 13.2. Note that the standard feed value AATN8 is lower than AATN20 mainly due to increased microbial efficiency with higher feed intake and therefore higher supply of AA from microbial protein. PBVN8 is higher than PBVN20, also due to increased microbial efficiency which incorporates more N into the microbial protein and therefore contributes less to the PBVN value. NEL8 is higher than NEL20, because increased DMI results in a higher passage rates and lowered digestibility and thus lower energy supply. Table 13.2. Standard feed values in some feedstuffs. Feedstuff

Barley Oat Triticale Brewers grain, dried Distillers grain, dried Rape seed meal Soy bean meal Vegetable fat Sugar beet pulp, dried unmolassed Grass silage very high OMD Grass silage high OMD Grass silage low OMD Maize silage high OMD Maize silage low OMD Straw, spring barley Glycerin, glycerol Urea

NorFor AATN8 PBVN8 NEL8 ID g/kg g/kg MJ/kg DM DM DM

AATN20 PBVN20 NEL20 g/kg g/kg MJ/kg DM DM DM

1-16 1-17 1-15 1-68 1-38 2-42 2-53 2-26 4-20

83 66 79 125 92 102 154 17 70

-6 20 -10 97 176 236 289 -6 -22

7.97 7.31 8.21 6.99 7.30 6.90 8.74 20.80 6.93

110 87 106 157 127 144 218 20 96

-44 -11 -50 55 127 180 210 -11 -63

7.49 6.78 7.80 6.38 6.69 6.52 8.32 18.98 6.26

6-460 6-461 6-463 6-307 6-308 6-386 12-12 13-1

60 63 62 68 66 39 82 0

92 76 49 -30 -30 -25 -120 2,922

7.73 7.50 6.64 7.36 7.05 3.44 8.12 0.00

79 80 75 87 84 44 109 0

62 49 28 -59 -56 -34 -163 2,922

7.00 6.69 5.74 6.60 6.21 2.47 7.49 0.00

NorFor – The Nordic feed evaluation system

139

5111 5112 5113 5114 5115 5116 5117 5118 5119 5120 5121 5122 5123 5124 5125 5126 5127 5128 5129 5130 5131 5132 5133 5134 5135 5136 5137 5138 5139 5140 5141 5142 5143 5144 5145 5146 5147

14.0 System evaluation H. N. I. evaluation Nielsen, M. Åkerlind and A. J. Rygh 14.Volden, System This chapterN.I. focuses on the of different aspects within the NorFor system. This H. Volden, Nielsen, M.evaluation Åkerlind and A.J. Rygh includes sensitivity analyses with respect to feed characteristics and model tests of total tract digestion, milk production, the energy system for growing cattle, RLI, feed intake and This chapter focuses on the evaluation of different aspects within the NorFor system. This includes chewing index.

sensitivity analyses with respect to feed characteristics and model tests of total tract digestion, milk production, the energy system for growing cattle, RLI, feed chewing 14.1 Sensitivity of feed degradation characteristics to intake rationand energy andindex. protein

values

14.1 Sensitivity of feed degradation characteristics to ration energy and

When the NorFor values system is used for on-farm ration optimization, forages are mainly analysed protein at commercial feed laboratories while for concentrate ingredients, values from the NorFor feed table are normally used (www.norfor.info). Within forage type, there are large variations When the NorFor system is used for on-farm ration optimization, forages are mainly analysed at in nutrient degradation characteristics, which affect (to varying degrees) the nutritional value commercial feed laboratories while for concentrate ingredients, values from the NorFor feed table of the feed ration. A sensitivity analysis is therefore useful for evaluating effects of changes in are composition normally used (www.norfor.info). forage there are large variations in nutrient the of feed ingredients on theWithin NEL and AATtype, N contents of the diet. A sensitivity degradation characteristics, affect (tothe varying degrees) the nutritional of the feed ration. analysis is also advantageouswhich for evaluating usefulness and errors related tovalue different A sensitivity analysis is therefore useful for evaluating effects of changes in the composition of feed routine analyses.

ingredients on the NEL and AATN contents of the diet. A sensitivity analysis is also advantageous

for evaluating the usefulness andiserrors to different routine in analyses. The sensitivity analysis presented basedrelated on a dairy cow simulation which a standard diet was fed and the influence of forage and concentrate degradation characteristics were tested. The sensitivity analysis presented based cow simulation in 140 which a standard Effects were tested at a DM intake ofis20 kg/don fora adairy cow of 600 kg BW and days in milk. diet was The diet the consisted of grass silageand (50% of diet DM) with an OMD of 71.4% and fed and influence of forage concentrate degradation characteristics werethe tested. Effects were concentrate mixture consisted of barley grain beet140 pulpdays (12%), maizeThe grain tested at a DM intake of 20 kg/d for a cow of(40%), 600 kgdried BW and in milk. diet consisted of (10%), soybean meal (16%), molasses (6%), rape seedmixture (5%) and mineralof barley grass silage (50% of diet DM)rapeseed with an meal OMD(9%), of 71.4% and the concentrate consisted and vitamins wasgrain (g/kg(10%), DM): CP, 179; NDF, 347; ST,rapeseed 191; CFat, grain (40%),(2%). dried The beetdiet pulpcomposition (12%), maize soybean meal (16%), meal (9%), 45; FPF, 56 and rape RestCHO, 105.and Input degradation characteristics tested sCP, iCP, molasses (6%), seed (5%) mineral and vitamins (2%). The dietwere: composition was (g/kg DM): kdCP, iNDF and347; kdNDF, and CFat, their basal, minimum maximum values arecharacteristics CP, 179; NDF, ST, 191; 45; FPF, 56 and and RestCHO, 105.parameter Input degradation presented in Table 14.1.kdCP, WheniNDF sCP and were changed, corresponding of pdCP parameter tested were: sCP, iCP, andiNDF kdNDF, and their basal, minimum values and maximum and pdNDF were changed simultaneously to maintain the same concentrations of CP and values are presented in Table 14.1. When sCP and iNDF were changed, corresponding values of NDF in the diet. The input parameter of interest was changed for one characteristic at a time pdCP and pdNDF were changed simultaneously to maintain the same concentrations of CP and and correlated responses were ignored to facilitate interpretation. For example, iNDF and NDF inare theoften diet.negatively The input correlated parameterwithin of interest changed for onetest, characteristic kdNDF feeds was but in the sensitivity they were at a time and correlated responses were to facilitate interpretation. example, iNDF and kdNDF are changed independently. Theignored sensitivity of diet energy and proteinFor values to parameter often negatively correlated within feeds but in the sensitivity test, they were changed variation was calculated as the change in the response variable relative to the change in independently. The sensitivity parameter value:of diet energy and protein values to parameter variation was calculated as the change in the response variable relative to the change in parameter value:

5148

⎛ ⎛ rmin − rmax ⎞ ⎞ ⎜ ⎜ ⎟ ⎟ ⎜ ⎜⎝ rbasal ⎟⎠ ⎟ Sensitivity = ⎜ ⎟ ⋅100 ⎜ ⎛⎜ p min − p 2 max ⎞⎟ ⎟ ⎟⎟ ⎜⎜ p basal ⎠⎠ ⎝⎝

5149 5150 5151 5152 5153

wheresensitivity sensitivityisiscalculated calculated in %, the minimum response rmaxmaximum is the maximum where in %, rmin ris theisminimum response value, value, rmax is the min response response value, pminpmin is the minimum parameter value, pmax is responsevalue, value,rbasal rbasalisisthethebasal basal response value, is the minimum parameter value, pmax is the the maximum parameter value, and pbasalisisthe thebasal basalparameter parameter value. value. maximum parameter value, and pbasal

[Eq. 14.1]

14.1

Changes in diet composition, i.e. proportions of CP, ST, NDF, SU and CFat are of major importance since they affect (inter alia) diet energy and191 metabolizable protein value (Fox et al., 2003). In the presented sensitivity analysis (Table 14.1), only the effects of variations in nutrient degradation characteristics were evaluated, since they may have an impact on the reliability of feed analyses. Both NEL and AATN are sensitive to changes in forage iNDF and kdNDF. The pdNDF from forage constitutes a large proportion of the NDF pool, and changes in both iNDF concentration and kdNDF strongly influence the energy supply for microbial growth, and hence AATN from microbial protein H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_14, © Wageningen Academic Publishers 2011

141

Table 14.1. Results of the sensitivity analysis in NorFor to changes in feed characteristics with a lactating dairy cow diet. Variables

Forage sCP1, g/kg CP iCP2, g/kg CP iNDF3, g/kg NDF kdCP4, %/h kdNDF5, %/h Concentrate sCP, g/kg CP iCP, g/kg CP iNDF, g/kg NDF kdCP, %/h kdNDF, %/h kdST6, %/h kdRestCHO7, %/h

Ranges

Sensitivity, %

Basal

Minimum

Maximum

NEL8 AATN9 MJ/kg DM g/kg DM

600 38 160 11.0 4.7

500 19 80 5.5 2.4

700 57 240 16.0 7.1

-0.1 -0.6 -5.3 -0.1 7.6

-5.1 -1.7 -3.4 -3.0 5.7

247 40 190 6.0 3.0 15.0 150

124 20 80 3.0 1.5 7.5 75

371 60 240 9.0 4.5 22.5 225

-0.1 -0.8 -2.4 -0.1 2.4 -0.9 0.3

-7.5 -3.4 -3.0 -12.8 0.5 1.5 -1.0

1

sCP = soluble crude protein. iCP = total indigestible crude protein. 3 iNDF = total indigestible NDF. 4 kdCP = fractional degradation rate of potentially degradable crude protein. 5 kdNDF = fractional degradation rate of potentially degradable NDF. 6 kdST = fractional degradation rate of potentially degradable starch. 7 kdRestCHO = fractional degradation rate of residual carbohydrates. 8 NEL = net energy lactation. 9 AAT = amino acids absorbed in the intestine. N 2

and total tract digestion of OM, which provide the basis for ME predictions. The work of Nordheim et al. (2007) demonstrated the difficulties of using NIRS to measure kdNDF. Due to the importance of kdNDF for the nutritive value of the diet and it was, therefore, decided in NorFor to calculate the kdNDF from NIRS or in vitro measurements of OMD and iNDF (see Section 5.2.1; Eriksson, 2010a). The AATN was sensitive to the proportion of sCP in the forage, although the relative difference between the selected minimum and maximum parameter values was less than for the other variables (Table 14.1). In grass silage, sCP is the dominating protein fraction, which explains why sCP has more impact on AATN than kdCP (Volden et al., 2002). Diet NEL is not sensitive to changes in forage or concentrates sCP and kdCP. This is consistent with expectations, because the segmental digestion of degradable CP has limited effects on the total tract digestion of CP. The sensitivity analysis indicates that predicted energy values are more sensitive to the chemical composition of concentrates, (e.g. ST vs. NDF) than to degradation characteristics (data not shown). However, predictions of the AATN are sensitive to kdCP, sCP and iCP. Nevertheless, this sensitivity analysis demonstrates that the NorFor system is sensitive to CHO and CP degradation characteristics in both roughage and concentrates.

142

NorFor – The Nordic feed evaluation system

14.2 Evaluation of the digestive tract sub-model Two important sub-models in NorFor are the digestive tract sub-model (which predicts the ration digestibility) and the metabolism sub-model (which predicts energy and protein supply and animal responses). When evaluating a complex model, it may be useful to evaluate several subdivisions separately, to reveal strengths and weaknesses in different parts of the system. This section summarizes the test of total tract digestion of major nutrients. Sources of data used in the evaluation of dairy cow digestion are presented in Table 14.2. The published studies were supplemented with measurements from trials conducted in connection with a student course in ruminant nutrition and physiology at the Norwegian University of Life Sciences. Total tract digestion was selected because there are minimal methodological differences between relevant studies. It is relatively simple to measure, compared to ruminal and intestinal flow measurements, which are heavily dependent on marker techniques and calculation methods used. The model was evaluated by regressing predicted values against observed values, and several statistical measures were used to assess model adequacy and behaviour, including coefficients of correlation (r) and determination (r2), mean square prediction error (MSPE) and the partitioning of MSPE (Bibby and Toutenburg, 1977). The MSPE was decomposed into errors due to: overall bias (deviation of the intercept from 0), line bias (deviation of the slope from unity) and disturbance (lack of correlation). The square root of the MSPE (RMSPE) expressed as a percentage of the observed mean was used as a measure of the prediction error. High correlations (r>0.97) between predicted and observed values were found for all the digestion variables tested (td_OM, td_CP, td_NDF and td_ST) (Figure 14.1, 14.2, 14.3, 14.4). NorFor accounted for 96% of the variation in td_OM with a mean bias of 221 g/d (1.9%). Corresponding values for td_CP, td_NDF and td_ST were 68, 68 and 106 g/d (3.3, 1.6 and 3.9%), respectively. The model slightly overpredicted td_OM, td_NDF and td_ST, while td_CP was underpredicted (Table 14.2). Except for td_ST, most of the errors in the MSPE were related to disturbance. The analysis of residuals between observed and predicted td_ST indicated that the higher general bias (0.43) was largely due to an underestimation of iST in the feedstuffs. Regression statistics (Mitchell, 1997) showed that the fitted line between predicted and observed values did not differ significantly from 1, indicating that there was no systematic deviation in digestion in the interval examined. Prediction errors were generally low, with RMSPE of 560, 167, 264 and 161 g/d for td_OM, td_CP, td_NDF and td_ST, respectively. The difference between observed and predicted values for td_CP, td_NDF and td_ST and were within 5% of the observed values for 45, 58 and 70% of the observations, respectively. Deviations were less than 10% for more than 80% of the observations. This demonstrates the ability of the NorFor model to predict ration digestion over large ranges of diet composition and feed intake. This ability is crucial for calculation of energy supply, which is based on total apparent digestion of CP, CFat and CHO in NorFor. From the mean prediction error for OM, and assuming an average ME factor of 15.5 MJ/kg digestible OM, the NEL (utilization of ME=0.6) prediction error was estimated to be 5.2 MJ, corresponding to 1.7 (5.2/3.14) kg ECM. For testing digestion in the growing cattle sub-model, a small data set from three studies (Olson and Lindberg, 1985; Olson and Lindberg, unpublished data; Wallsten, 2010) was available. The data set included results from 18 treatments, applied to both bulls (n=10) and heifers (n=8). The total apparent digestion of td_OM, td_CP, td_NDF was evaluated (Figures 14.5, 14.6 and 14.7). The coefficient of determination was high (r2>0.96) for all digestion variables tested (Table 14.3). NorFor overpredicted td_OM (6.4%) and td_NDF (9.0), while td_CP was underpredicted (13.2%). The MSPE consisted mainly overall bias, while the proportion of MSPE related to line bias was low. The regression slope between predicted and observed digestion did not differ significantly from 1. The data set was, however, too small for further analysis of the residuals.

NorFor – The Nordic feed evaluation system

143

144

NorFor – The Nordic feed evaluation system

n

11,841 1993 4235 2860

831 -134 377 24

Intercept

Observed Predicted

11,620 2061 4167 2754

Regression

Digested, g/day 0.948 1.033 0.926 1.030

Slope 0.967 0.955 0.966 0.993

r2 560 167 264 161

RMSPE 4.8 8.1 6.3 5.8

Prediction error, % 0.165 0.168 0.067 0.431

0.009 0.100 0.044 0.099

0.836 0.732 0.889 0.470

Overall bias Regression Disturbance

Proportion of MSPE

2

td_OM = apparent total tract digestion of organic matter, g/day. td_CP = apparent total tract digestion of crude protein, g/day. 3 td_NDF = apparent total tract digestion of NDF, g/day. 4 td_ST = apparent total tract digestion of starch, g/day. 5 The following trials were used: Alne, 2000; Aston et al., 1994; Bertilsson and Murphy, 2003; Eriksson et al., 2004; Kjos, unpublished data; Larsen et al. 2009b; Lund et al., 2004; Lund, unpublished data; Murphy et al., 1993; Prestløkken, 1999; Prestløkken et al., 2007; Rinne et al., 1999a, 2002; Selmer-Olsen, 1996; Tothi et al., 2003; Volden, unpublished data.

1

td_OM1 107 td_CP2 95 td_NDF3 111 td_ST4 67

Item

Table 14.2. Accuracy and precision of NorFor’s ability to predict digestion in dairy cows5.

Predicted OM digestion (g/d)

18,000

Observed - predicted

4,000

16,000 14,000 12,000 10,000 8,000 6,000 2,000 0 -2,000

0

2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 18,000

-4,000

Observed OM digestion (g/d)

Observed - predicted

Predicted CP digestion (g/d)

Figure 14.1. Predicted vs. observed organic matter (OM) digestion in dairy cows.

4,500 4,000 3,500 3,000 2,500 2,000 1,500 1000 500 0 -500

0

500

1000

1,500

2,000

2,500 3,000

3,500 4,000

4,500

-1000

Observed CP digestion (g/d)

Figure 14.2. Predicted vs. observed crude protein (CP) digestion in dairy cows.

NorFor – The Nordic feed evaluation system

145

Predicted NDF digestion (g/d) Observed - predicted

8,000 7,000 6,000 5,000 4,000 3,000 2,000 1000 0 -1000

0

1000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

-2,000 Observed NDF digestion (g/d)

Observed - predicted

Predicted ST digestion (g/d)

Figure 14.3. Predicted vs. observed NDF digestion dairy cow. 6,000 5,000 4,000 3,000 2,000 1000 0

0

1000

2,000

3,000

4,000

5,000

6,000

-1000 Observed ST digestion (g/d)

Figure 14.4. Predicted vs. observed starch (ST) digestion in dairy cows.

146

NorFor – The Nordic feed evaluation system

NorFor – The Nordic feed evaluation system

147

n

3,290 386 1,044

79 -24 100

Intercept

Observed Predicted

3,090 445 958

Regression

Digested, g/day 1.047 0.922 0.986

Slope 0.981 0.968 0.990

r2 230 64 62

RMSPE

2

td_OM = apparent total tract digestion of organic matter, g/day. td_CP = apparent total tract digestion of crude protein, g/day. 3 td_NDF = apparent total tract digestion of NDF, g/day. 4 The following trials were used: Olson and Lindberg, 1985; Olson and Lindberg, unpublished data; Wallsten, 2010.

1

td_OM1 18 td_CP2 18 td_NDF3 18

Item

Table 14.3. Accuracy and precision of NorFor´s ability to predict digestion in growing cattle4.

7.5 14.5 11.1

Prediction error, % 0.753 0.838 0.657

0.035 0.010 0.001

0.212 0.152 0.342

Overall bias Regression Disturbance

Proportion of MSPE

Predicted OM digestion (g/d)

5,000

Observed - predicted

1000

4,500 4,000 3,500 3,000 2,500 2,000 1,500 500 0

-500 0

1000

-1000

2,000

3,000

4,000

5,000

Observed OM digestion (g/d)

Predicted CP digestion (g/d)

Figure 14.5. Predicted vs. observed organic matter (OM) digestion in growing cattle.

800 700 600 500 400

Observed - predicted

300 200 100 0

0

100

200

300 400 500 600 Observed CP digestion (g/d)

700

800

Figure 14.6. Predicted vs. observed crude protein (CP) digestion in growing cattle.

148

NorFor – The Nordic feed evaluation system

Observed - predicted

Predicted NDF digestion (g/d)

3,000 2,500 2,000 1,500 1000 500 0 -500

0

500

1000

1,500

2,000

2,500

3,000

Observed NDF digestion (g/d)

Figure 14.7. Predicted vs. observed NDF digestion in growing cattle.

14.3. Evaluation of milk production in lactating dairy cows A Nordic dataset, which had not been used for model construction, was compiled. Data originated from published and unpublished trials performed in Denmark, Norway, Sweden, and Finland mainly from the 1990’s and 2000’s (Table 14.4). No selection was made among the experiments and data from all identified experiments providing information on ECM or MY, fat and protein contents, DMI and BW were included. Not all studies included information on BW or BCS changes during the trial period. Lactose content in milk was only reported in a few studies, thus a fixed content of 48 g/kg was used. The dataset included information from both long- and short-term studies, i.e. data both from continuous lactation experiments and cross-over type experiments. A further criterion was availability of data on most nutrient characteristics for each feedstuff. Experiments that only provided data on nutrient characteristics for the whole TMR were not included. Most studies did not include iNDF, sCP, kdST, kdNDF and kdCP measurements on single feedstuffs. When nutrient composition data were missing, values from the NorFor feedstuff table were used (www.norfor.info). The dataset, described in Table 14.5, consists of information from 53 trials (see Table 14.4) with five breeds (205 NRF, 77 DH, 38 SR, 5 RD and 4 JER) with a total of 329 treatments. As described in Chapter 12, ECM is predicted from the calculated total energy intake minus the estimated energy usage for maintenance, growth (only primiparous cows), pregnancy and mobilization/deposition (see Equation 12.1). Energy usage for maintenance is estimated from BW. Information on days in gestation was often not available and was therefore set to zero. Mobilization/ deposition was included via information on changes in BW or BCS. As not all studies included data on BW or BCS changes during the trial period, mobilization/deposition was assumed to be zero. Energy requirement for growth in primiparous cows was included indirectly via changes in BW or BCS. The prediction of milk protein yield is based on model estimates of AATN supply to the mammary gland and the efficiency of AATN use in synthesis of milk protein (see Equation 12.2).

NorFor – The Nordic feed evaluation system

149

Table 14.4. Studies used to evaluate the prediction of milk production, roughage intake and total dry matter intake in dairy cows. Aaes, 1991 Aaes, 1993 1,2 Bertilsson, 2007 Bertilsson, 2008 Bertilsson and Murphy, 20031 Eriksson, 2010b Eriksson et al., 20041 Hymøller et al., 20052 Ingvartsen et al., 20011, 2 Johansen, 19921 Kjos, unpublished data Kristensen, 19992 Lund, 20021 Minde and Rygh, 1997 Misciattelli et al., 20031 Mo and Randby, 19861 Mogensen and Kristensen, 20021,2

Mogensen et al., 20082 Mould, 19961 Murphy et al., 2000 Nielsen et al., 20072 Nordang and Mould, 19941 Prestløkken, 1999 Prestløkken et al., 20071 Randby, 19921 Randby, 19971 Randby, 1999a1 Randby, 1999b1 Randby, 2000a1 Randby, 2000b1 Randby, 20021 Randby, 2003 1 Randby, 20071 Randby and Mo, 19851

Randby and Selmer-Olsen, 19971 Rinne et al., 1999a1 Rinne et al., 1999b1 Sairanen, 20011 Schei et al., 20051 Shingfield et al., 2003 Steinshamn et al., 20061 Thorhauge and Weisbjerg, 20092 Thuen, 19891 Volden, 19901 Volden, 1999 Volden and Harstad, 2002 Wallsten and Martinsson, 20101 Weisbjerg, 2007 2 Weisbjerg et al., 20082 Åkerlind et al., 19992 Österman, 20031, 2

1 The

study is used for roughage intake evaluation. study is used for total dry matter intake evaluation. Studies marked with superscripts 1 and 2 included treatments with separate and TMR and feeding, respectively. 2 The

Table 14.5. Descriptive statistics of the 329 treatments from 53 Nordic trials used to evaluate prediction of ECM and milk protein yield. Item

Average

SD1

Maximum

Minimum

Days in milk ECM2, kg/d MY3, kg/d Fat, g/kg Protein, g/kg Protein yield, g/d

121 27.1 27.1 41.1 32.6 882

44 5.3 5.2 4.7 1.8 174

255 49.7 47.0 58.3 41.1 1,607

31 12.6 13.1 26.9 29,3 422

1

SD = standard deviation. ECM = energy corrected milk. 3 MY = milk yield. 2

Figures 14.8 and 14.9 illustrate the relationships between observed and predicted ECM and milk protein yields, respectively. Table 14.6 shows that across all experiments and treatments, the observed and predicted yields were similar (27.1 vs. 26.4 kg ECM/d), especially for milk protein (882 vs. 900 g protein/d). The correlation between observed and predicted milk protein values was high (r=0.97), but somewhat lower for ECM (r=0.88). Predicted ECM and milk protein values were higher than observed for 43% and 68% of the observations, respectively, with RMSE’s of 2.7 kg ECM and 48 g milk protein, corresponding to prediction errors of 9.8 and 5.5%, respectively. Predicted ECM yield was within 5% of observed values for 47% of the observations and 73% of the predicted ECM yield values were within 10% of observed values. Predicted milk protein yield was within 5% and 10% 150

NorFor – The Nordic feed evaluation system

Predicted ECM (kg/d)

50

40

30

Observed - predicted

20

10

0

0

10

20

30

40

50

-10 Observed ECM (kg/d)

Predicted milk protein yield ( g/d)

Figure 14.8. Predicted vs. observed ECM yield.

1,800 1,600 1,400 1,200 1000 800

Observed - predicted

600 400 200 0 -200

0

200

400

600

800

1000 1,200 1,400 1,600 1,800

Observed milk protein yield (g/d)

Figure 14.9. Predicted vs. observed milk protein yield.

NorFor – The Nordic feed evaluation system

151

152

NorFor – The Nordic feed evaluation system

2

26.4 900

2.6 39

Intercept

Observed Predicted

27.1 882

Regression

Production

ECM = energy corrected milk yield, kg/day. MPY = milk protein yield, g/day.

329 329

ECM1 MPY2

1

n

Item 0.88 0.98

Slope 0.78 0.94

r2 2.7 48

RMSPE

Table 14.6. Accuracy and precision of NorFor´s ability to predict ECM and milk protein yield.

9.8 5.5

Prediction error, % 0.07 0.14

0.05 0.02

0.88 0.84

Overall bias Regression Disturbance

Proportion of MSPE

of observed values for 69% and 92% of the observations, respectively. The overall and regression biases were minor for ECM and milk protein, since more than 80% of the MSPE was characterised as disturbance. The regression slopes between predicted and observed milk production variables were not significantly different from 1.

14.4 Evaluation of the energy requirement for growing cattle The estimated energy requirement for growing cattle is based on the French system (Chapter 9), modified with a 10% increase in the energy requirement for gain based on a dataset compiled from Nordic experiments with bulls. In order to perform an independent test of the NorFor energy system for growing cattle, an independent dataset containing heifer data from the Nordic countries and the US was used. Criteria for including data were heifers of dairy breeds and measurements of ADG, DMI and nutrient characteristics for each feedstuff. Hence, experiments where nutrient characteristics were available only for the whole diet were excluded. When nutrient composition data were missing, values from the NorFor feedstuff table were used (www.norfor.info). The reason for including US data was that a limited number of experiments from the Nordic countries were available. No selection was made among the experiments that were found in the literature. The dataset is described in Table 14.7 and consist of 9 trials with 54 treatments of which 47 include large breeds and 7 include Jersey. Ration energy supply was calculated using NorFor and were regarded as observed values. The energy requirements were calculated from reported BW and ADG and were regarded as predicted values. The relationship between energy supply and requirement is shown in Figure 14.10. Across all experiments and treatments, the energy supply and requirement were fairly similar (31.9 vs. 33.7 MJ NEG) with an RMSPE of 3.4 MJ NEG, corresponding to a prediction error of 10.0%. The correlation (r=0.96) was stronger than that obtained from the data used to develop the energy system for growing cattle (see Figure 9.2).

14.5. Evaluation of the rumen load index (RLI) RLI was included in the model to establish a relationship between the level of starch and sugars in the ration and their influence on digestion of NDF in the rumen (see Section 7.1.3). Therefore, RLI was evaluated against data from two studies in which rumen pH was measured (Lund, unpublished data; Larsen et al., 2009a) and where kdNDF was estimated on the basis of digestibility trials with lactating Holstein cows (Larsen et al., 2009a). The cows were fed ad libitum with grass-clover silage as the sole roughage and the concentrate consisted of barley, wheat, oat, maize, soybeans, faba beans, peas or lupins and most treatments also included soybean meal as a protein supplement in the ration.

Table 14.7. Descriptive statistics of the dataset1 used for evaluating the energy system for growing cattle, including data from nine trials with heifers and 54 treatments. Item

Average

SD2

Maximum

Minimum

DMI, kg/d BW, kg ADG, g/d Concentrates (% of DMI)

5.3 253 724 35

1.9 102 203 28

9.9 489 1,180 92

2.2 88 396 4

1

Olsson et al., 1984; Andersen et al., 1986; Ingvartsen et al., 1988; Mäntysaari, 1993; Andersen and Foldager, 1994; Van Amburgh et al., 1998; Andersen et al., 2001; Whitlock et al., 2002; Vestergaard, 2006. 2 SD = standard deviation.

NorFor – The Nordic feed evaluation system

153

70

NEG requirement (MJ/d)

60 50 40 30 20 10 0

0

10

20

30 40 NEG supply (MJ/d)

50

60

70

Figure 14.10. The relationship between ration energy supply calculated in NorFor and estimated energy requirement in heifers according to NorFor (r2=0.92; RMSPE=3.4 MJ NEG; RMSPE=10.0%). The data are from the trials described in Table 14.8. In contrast to expectations, RLI was not negatively correlated to pH in the rumen (Figure 14.11). Indeed, the data in Figure 14.11 show a positive relationship between RLI and rumen pH in the data from Lund (unpublished data) and no relationship in the data from Larsen et al. (2009a). As RLI was included in NorFor to estimate the effect of starch and sugars on the digestion of NDF in the rumen, the main interest is in evaluating RLI was as a predictor for kdNDF in the rumen, rather than as a predictor of rumen pH. This relationship is illustrated in Figure 14.12, where a negative correlation (r=-0.82) between RLI and kdNDF is evident. Thus, even though RLI was not correlated with rumen pH, increases in RLI were associated with reductions in kdNDF. This confirms that RLI is a useful term for correcting kdNDF, as described in Section 7.1.3, although this evaluation was only based on one study. However, the relationship shown in Figure 14.12 is supported by the findings of Danfær et al. (2006).

14.6. Evaluation of roughage and total dry matter intake in dairy cows Data from Nordic experiments were used to evaluate roughage and total DMI predictions. For this purpose, the data were divided in two sub-sets (Tables 14.8 and 14.9). The first (n=226) consisted of data from studies where roughage was fed ad libitum and separately, and the concentrate was fed either according to MY or as fixed amounts (Table 14.8). This dataset was used to evaluate roughage intake. In the second dataset (n=62), cows were fed total mixed rations (TMR) ad libitum and total DMI was evaluated (Table 14.9). Both datasets showed wide variations in both DMI and diet composition, and the data were in the range of those used for parameterization of the intake sub-model. Published nutrient compositions of the individual feeds were used. When nutrient composition data were missing, values from the NorFor feedstuff table were used (www.norfor.info). Predicted vs. observed roughage intake and the evaluation of the prediction errors are presented in Figure 14.13 and Table 14.10, respectively. Mean observed roughage DM intake was 11.0 kg/d 154

NorFor – The Nordic feed evaluation system

6.8 min_pH_Lund

6.6

mean_pH_Lund

Rumen pH

6.4

min_pH_Larsen mean_pH_Larsen

6.2 6.0 5.8 5.6 5.4 5.2 5.0 0.0

0.2

0.4

0.6

0.8

1.0

Rumen load index

Figure 14.11. The relationship between rumen load index calculated in NorFor and mean and minimum pH measured in the rumen. The data are from Lund (unpublished data, n=8) and Larsen et al. (2009a, n=13). 4.0 3.5

kdNDF (%/h)

3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

Rumen load index

Figure 14.12. The relationship between rumen load index calculated in NorFor and trial estimates of NDF degradation rate in the rumen (kdNDF) of lactating Holstein cows (n=13; r2=0.68; RMSPE=0.33%/h). The data are from Larsen et al. (2009a).

NorFor – The Nordic feed evaluation system

155

Table 14.8. Descriptive statistics of the dataset used for evaluating feed intake of cows, including data from 36 trials3 and 226 treatments where roughage and concentrate were fed separately. Item

Average

SD1

Maximum

Minimum

Days in milk Body weight, kg ECM2, kg/d Dry matter intake, kg/d Concentrate proportion, kg/kg DM Roughage intake, kg DM/d Roughage basis fill value, /kg DM Starch + sugars, g/kg DM Starch + sugars, kg/d

124 567 25.3 18.2 0.40 10.9 0.50 183 3.3

35 37 4.5 2.4 0.12 2.6 0.06 69 1.2

214 667 45.0 28.4 0.80 18.2 0.70 426 6.9

42 481 17.1 13.1 0.14 3.2 0.35 60 0.9

1

SD = standard deviation. ECM = energy corrected milk. 3 The trials are listed in Table 14.5. 2

Table 14.9. Descriptive statistics of the dataset used for evaluating the feed intake system for cows, including data from 12 trials3 and 62 treatments where the ration was fed as a total mixed ration. Item

Average

SD1

Maximum

Minimum

Days in milk Body weight, kg ECM2, kg/d Dry matter intake, kg/d Concentrate proportion, kg/kg DM Roughage basis fill value, /kg DM Starch + sugars, g/kg DM Starch + sugars, kg/d

96 588 33.2 21.2 0.54 0.46 208 4.4

43 48 4.6 2.1 0.1 0.04 59 1.2

200 659 49.7 25.1 0.70 0.60 419 7.8

31 441 22.4 16.6 0.27 0.42 94 1.7

1

SD = standard deviation. ECM = energy corrected milk. 3 The trials are listed in Table 14.5. 2

and mean predicted intake was 10.1 kg DM/d. The regression slope was 0.88 and not significantly different from 1. The RMSPE was 1.3 kg DM/d and 0.52 of the prediction error was related to disturbance. The overall bias accounted for 0.48 of the total MSPE, and showed that the intake model underpredicts roughage intake. Predicted roughage intake was within ±10% of observed intake for 58% of the observations. Predicted vs. observed TMR intake and results of the evaluation of the prediction error are presented in Figure 14.14 and Table 14.10, respectively. The mean observed and predicted TMR intakes were 21.2 and 21.1 kg DM/d, respectively, the regression slope was not significantly (P100 1 1 113 0.86

618 5.3 14.5 7.7 1.25 532 >100 1 1 160 0.91

625 8.9 15.3 6.9 1.10 450 32 0.96 1 120 0.87

625 10.4 17.1 7.7 1.23 450 25 0.92 1 120 0.87

350 0.3 7.8 3.8 1.08 486 >100 1 1 192 0.94

350 0.3 7.0 3.3 0.95 476 >100 1 1 366 1.12

21 37 58

27 48 75

24 46 70

24 53 77

51 70 121 21 40 61

54 71 125 28 52 80

99 130 229 20 46 66

100 148 248 20 51 71

22 39 61

21 39 60

1136

976

606

566

within ±5% of observed CTcorr for six out of eight observations. The two observations with large deviations (>15%) were from the study of Hymøller et al. (2005) and are largely responsible for a relatively low coefficient of determination (r2 = 0.21) between observed CT and predicted CI. The RMSPE was 9.2 min/kg DMr corresponding to a prediction error of 13.8%.

NorFor – The Nordic feed evaluation system

161

Predicted EI, RI and CI (min/kg DMr)

100 CT ET RT

80

60

40

20

0

0

20

40 60 80 Observed ET, RT and CT (min/kg DMr)

100

Figure 14.15. The relationship between observed eating (ET), rumination (RT) and chewing time (CT) vs. corresponding predicted indexes (EI, RI, CI). The data are from trials described in Table 14.11.

References Aaes, O. 1991. Grøncobs som erstatning for kraftfoder til malkekøer. In: Foderværdi og anvendelse af kunsttørret grønfoder til malkekøer. Statens Husdyrbrugsforsøg, Denmark. 5-19. (In Danish). Aaes, O. 1993. Total mixed rations versus separate feeding of concentrate rich rations fed restrictively or ad libitum to dairy cows. Forskningsrapport no 16 fra Statens Husdyrbrugsforsøg, Denmark. (In Danish with English summary). Åkerlind, M., K. Holtenius, J. Bertilsson and M. Emanuelson, 1999. Milk composition and feed intake in dairy cows selected for high or low milk fat percentage. Livestock Production Science 59: 1-11. Alne, J., 2000. Effects of duodenal starch and amino acid infusion on milk yield, protein content and utilization of feednitrogen in dairy cows. Master of Science Thesis. Agricultural University of Norway, Aas, Norway. (In Norwegian). Andersen, H.R. and J. Foldager, 1994. Protein requirement for replacement heifers in the period from 3 to 12 months of age. Forskningsrapport no. 26 fra Statens Husdyrbrugsforsøg, Denmark. (In Danish with English summary). Andersen, H.R., Foldager, J., Varnum, P.S. and T. Hvelplund, 1986. Forskellige proteinmængder og proteinkilder til kvieopdræt. Meddelelse no. 645 fra Statens Husdyrbrugsforsøg. (In Danish). Andersen, H.R., B.B. Andersen and H.G. Bang, 2001. Beef crossbreeding: Heifers versus bulls fed different concentrate/ roughage ratio and slaughtered at different live weight. DJF rapport no. 28. (English summary). Aston, K, C. Thomas, R. Daley, J.D. Sutton and M.S. Dhanoa, 1994. Milk production from grass silage diets: effects of silage characteristics and the amount of supplementary concentrate. Animal Production 59: 31-41. Bertilsson, J., 2007. Agrodrank som foder till mjölkkor. Kungsängendagarna 2007. Rapport nr 267. Swedish University of Agricultural Sciences, Uppsala, Sweden. pp. 21-23. (In Swedish). Bertilsson, J., 2008. Effects of chop length of silage on chewing time, feed consumption and milk production. In: Hopkins, A., T. Gustafsson, J. Bertilsson, G. Dalin, N. Nilsdotter-Linde and E. Spörndly (eds.). Biodiversity and animal feed: Future challenges for grassland production. Grassland Science in Europe volume 13, Uppsala, Sweden, pp. 783-785. Bertilsson, J. and M. Murphy. 2003. Effects of feeding clover silages on feed intake, milk production and digestion in dairy cows. Grass and Forage Science 58: 309-322.

162

NorFor – The Nordic feed evaluation system

Danfær, A., P. Huhtanen, P. Udén, J. Sveinbjörnsson and H. Volden, 2006. Karoline – A Nordic model for feed evaluation. Model description. In: Kebreab, E., J. Dijkstra, A. Bannink, J. Gerrits and J. France (eds.). Nutrient digestion and utilisation in farm animals: Modelling approaches. CAB International, Wallingford, UK, pp. 383-406. Eriksson, T. 2010a. In vitro methods for indigestible Neutral Detergent Fiber (iNDF). Proceedings of the 1th Nordic Feed Science Conference, Report no. 274, Swedish University of Agricultural Sciences, Uppsala, Sweden, pp. 41-45. Eriksson, T. 2010b. Nitrogen metabolism in dairy cows fed restricted amounts of grass–clover silage supplemented with seeds from narrow-leafed lupin or pea. Livestock Science 131: 39-44. Eriksson, T., M. Murphy, P. Ciszuk and E. Burstedt, 2004. Nitrogen balance, microbial protein production. and milk production in dairy cows fed fodder beets and potatoes or barley. Journal of Dairy Science 87: 1057-1070. Fox, D., G.L.O. Tedeschi, T.P. Tylutki, J.B. Russell, M.E. Van Amburgh, L.E. Chase, A.N. Pell and T.R. Overton. 2004. The Cornell Net Carbohydrate and Protein System model for evaluating herd nutrition and nutrient excretion. Animal Feed Science and Technology 112: 29-78. Garmo, T.H., A.T. Randby, M. Eknaes, E. Prestløkken and P. Nørgaard, 2008. Effect of grass silage chop length on chewing activity and digestibility. In: Hopkins, A., T. Gustafsson, J. Bertilsson, G. Dalin, N. Nilsdotter-Linde and E. Spörndly (eds.). Biodiversity and animal feed: Future challenges for grassland production. Grassland Science in Europe, volume 13, pp. 810-812. Hymøller, L., M.R. Weisbjerg, P. Nørgaard, C.F. Børsting and N.B. Kristensen, 2005. Majsensilage til malkekøer. DJF rapport nr. 65. Danmarks Jordbrugsforskning. Foulum. Denmark. (In Danish). Ingvartsen, K.L., O. Aaes and J.B. Andersen, 2001. Effects of pattern of concentrate allocation in the dry period and early lactation on feed intake and lactational performance in dairy cows. Livestock Production Science 71: 207-221. Ingvartsen, K.L., J. Foldager, J.B. Larsen and V. Østergaard, 1988. Growth and milk yield by Jersey reared at different planes of nutrition. Beretning no. 645 fra Statens Husdyrbrugsforsøg. (In Danish). Johansen, A., 1992. A comparison between meadow fescue and timothy silage. Doctor Scientarium Thesis. Agricultural University of Norway, Aas, Norway. Kristensen, V.F., 1999. Grovfoderkildens betydning for malkekoens produktion og fodereffektivitet. Temamøde Malkekøernes og kviernes ernæring April 1999. Statens Husdyrbrugsforsøg rapport 118, pp. 18-33. (In Danish). Larsen, M., N.B. Kristensen and T. Hvelplund, 2009a. Stivelsen til malkekøer – perspektivering af resultaterne fra stivelsesprojektet. In: Fodringsdag - Temadag om aktuelle fodringsspørgsmål. 1 September 2009. Dansk Kvæg, Århus, Denmark, pp. 54-61. (In Danish). Larsen, M., P. Lund, M.R. Weisbjerg and T. Hvelplund, 2009b. Digestion site of starch from cereals and legumes in lactating dairy cows. Animal Feed Science and Technology 153: 236-248. Lund, P., M.R. Weisbjerg and T. Kristensen, 2004. The effect of heat treatment on degradability and microbial synthesis of protein in the rumen. Journal of Animal Feed Science 13: 143-146. Lund, P., 2002. The effect of forage type on passage kinetics and digestibility of fibre in dairy cows. Dissertation. The Royal Veterinary and Agricultural University, Department of Animal Science and Animal Health, Copenhagen, Denmark. 171 pp. Minde, A. and A.J. Rygh, 1997. Metoder for å bestemme nedbrytingshastighet, passasjehastighet og vomfordøyelighet av NDF hos mjølkeku. Masteroppgave. Norges Landbrukshøgskole. 118 pp. (In Norwegian). Misciattelli, L., V.F. Kristensen, M. Vestergaard, M.R. Weisbjerg, K. Sejrsen and T. Hvelplund, 2003. Milk production, nutrient utilization and endocrine responses to increased postruminal lysine and methionine supply in dairy cows. Journal of Dairy Science 86: 275-286. Mitchell, P.L. 1997. Misuse of regression for empirical validation of models. Agricultural Systems 54: 313-326. Mo, M. and Å.T. Randby, 1986. Wilted silage for lactating dairy cows. Proceedings of 11th General Meeting of European Grassland Federation. Troia, Portugal, 4-9 May. Mogensen, L. and T. Kristensen, 2002. Effect of barley or rape seed cake as supplement to silage for high-yielding organic dairy cows. Acta Agriculturae Scandinavica. Section A - Animal Science 53: 243-252. Mogensen, L., P. Lund, T. Kristensen and M.R. Weisbjerg, 2008. Effects of toasting blue lupins, soybeans or barley as supplement for high-yielding organic dairy cows fed grass-clover silage ad libitum. Livestock Science 115: 249-257. Mould, F., 1996. Fôropptak og produksjon hos melkekyr ved et lavt PBV i rasjonen. Husdyrforsøksmøtet, pp. 121127. (In Norwegian). Murphy, M., H. Khalili and P. Huhtanen, 1993. The substitution of barley by other carbohydrates in grass silage based diets to dairy cows. Animal Feed Science and Technology 41: 279-296.

NorFor – The Nordic feed evaluation system

163

Murphy, M., M. Åkerlind and K. Holtenius, 2000. Rumen fermentation in lactating cows selected for milk fat content fed two forage to concentrate ratios with hay or silage. Journal of Dairy Science 83: 756-764. Mäntysaari, P. 1993. The effects of feeding level and protein source of the diet on growth and development at slaughter of prepurbertal heifers. Acta Agriculturae Scandinavica Section A - Animal Science 43: 44-51. Nielsen, N.I., N.C. Friggens, T. Larsen, J.B. Andersen, M.O. Nielsen and K.L. Ingvartsen, 2007. Effect of changes in diet energy density on feed intake, milk yield and metabolic parameters in dairy cows in early lactation. Animal 1: 335-346. Nordang, L. and F. Mould, 1994. Surfôr med ulikt innhold av PBV til mjølkekyr. Husdyrforsøksmøtet, pp. 99-103. (In Norwegian). Nørgaard, P., E. Nadeau and A. Randby, 2010. A new Nordic structure evaluation system for diets fed to dairy cows: a meta analysis. In: Sauvant, D., J. Van Milgen, P. Faverdin and N. Friggens (eds.). Modelling nutrient digestion and utilization in farm animals. Wageningen Academic Publishers, Wageningen, the Netherlands, pp. 112-121. Olsson, K., I. Olsson and L. Lindell, 1984. Effect of the amount of roughage fed to heifers during the rearing period on feed consumption and milk production during the first lactation. Report no. 131. Swedish University of Agricultural Sciences Österman, S., 2003. Extended calving interval and increased milking frequency in dairy cows. Acta Universitatis Agriculturae Suecia. Agria 383. Swedish University of Agricultural Sciences, Uppsala, Sweden. Prestløkken, E., 1999. Protein value of expander-treated barley and oats for ruminants. Doctor Scientarium Thesis, Agricultural University of Norway, Aas, Norway. 137 pp. Prestløkken, E., Å.T. Randby, M. Eknæs and T. Garmo, 2007. Tidlig og normalt høstet gras. Opptak av surfôr og produksjon hos mjølkekyr. Husdyrforsøket, Thon Hotel Arena 14-15 February 2007, pp. 25-28. (In Norwegian). Randby, Å.T., 1992. Grass-clover silage for dairy cows. Proceedings of 14th General Meeting, European Grassland Federation, Lahti, Finland, 8-11 June, pp. 272-275. Randby, Å.T., 1997. Feeding of silage effluent to dairy cows. Acta Agriculturae Scandinavica Section A - Animal Science 47: 20-30. Randby, Å.T., 1999a. Grass silage treated with two different acid-based additives for dairy cows. The Fifth International Symposium on the Nutrition of Herbivores. San Antonio, Texas, USA. 8 pp. Randby, Å.T., 1999b. Effect of increasing levels of a formic acid based additive on ad libitum intake of grass silage and the production of meat and milk. Proceeding of the XII International Silage Conference, Uppsala, Sweden, pp. 205-206. Randby, Å.T., 2000a. Round bale grass silage treated with an additive containing both lactic acid bacteria and formate for dairy cows and steers. Sixth British Grassland Society Research Conference, SAC, Craibstone, Aberdeen, Scotland, pp. 121-122. Randby,Å.T., 2000b. Effekten av økende dosering med et maursyreholdig ensileringsmiddel på næringsverdien av surfôret i produksjonsforsøk med kyr og kastrater. Hellerud Forsøksgård, Rapport nr. 25. 37 pp. (In Norwegian). Randby, Å.T., 2002. Silage fermentation by concentrate type interaction. Proceedings of the XIII International Silage Conference, Auchincruive, Scotland, pp. 312-313. Randby, Å.T., 2003. Høstetid og fôrkvalitet. Grønn kunnskap 7: 27-43. (In Norwegian). Randby, Å.T. 2007. Effect of propanol and dimethylsulphide in grass silage on organoleptic milk quality. Journal of Animal and Feed Science 16, Supplement 1: 102-107. Randby, Å. and M. Mo, 1985. Surfôr tilsatt foraform i sammenlikning med maursyresurfôr til melkekyr, Hellerud høsten 1984. M-141. Rapporter fra forsøksvirksomheten ved Institutt for husdyrernæring, pp. 75-78. (In Norwegian). Randby, Å.T. and I. Selmer-Olsen, 1997. Formic acid treated or untreated round bale grass silage for dairy cows. 8th International Symposium of Forage Conservation. Brno, Czech Republic, pp. 160-161. Randby, Å.T., I. Selmer-Olsen and L. Baevre, 1999. Effect of ethanol in feed on milk flavor and chemical composition. Journal of Dairy Science 82: 420-428. Rinne M., P. Huhtanen and S. Jaakkola. 1997. Grass maturity effects on cattle fed silage-based diets. 2. Cell wall digestibility, digestion and passage kinetics. Animal Feed Science and Technology 67: 19-35. Rinne M., S. Jaakkola, K. Kaustell, T. Heikkilä and P. Huhtanen, 1999a. Silage harvested at different stages of grass growth versus concentrate foods as energy and protein sources in milk production. Animal Science 69: 251-263. Rinne, M., S. Jaakkola, T. Varvikko and P. Huhtanen, 1999b. Effects of type and amount of rapeseed feed on milk production. Acta Agriculturae Scandinavica, Section A - Animal Science, 49: 137-148.

164

NorFor – The Nordic feed evaluation system

Rinne M., P. Huhtanen and S. Jaakkola. 2002. Digestive processes of dairy cows fed silages harvested at four stages of maturity. Journal of Animal Science 80: 1986-1998. Rustas, B.O., P. Nørgaard, A.R. Jalali and E. Nadeau, 2010. Effects of physical form and stage of maturity at harvest of whole-crop barley silage on intake, chewing activity, diet selection and faecal particle size of dairy steers. Animal 4: 67-75. Sairanen, A., J.I. Nousiainen and H. Khalili, 1999. Korkean väkirehumäärän vaikutus maitotuotokseen ja tuotannon kannattavuuteen. In: Mitä Suomi syö - ja millä hinnalla? Agro-Food ‘99, Helsinki. pp. 7. (In Finnish). Selmer-Olsen, I., 1996. Microbial protein synthesis in the rumen of cows fed extensively or restricted fermented grass silage. Proceedings of the XIth International Silage Conference, Aberystwyth, Wales, pp. 226-227. Schei, I., H. Volden and L. Bævre, 2005. Effects of energy balance and metabolizable protein level on tissue mobilization and milk performance of dairy cows in early lactation. Livestock Production Science 95: 35-47. Shingfield, K.J., A. Vanhatalo and P. Huhtanen, 2003. Comparison of heat-treated rapeseed expeller and solvent-extracted soya-bean meal as protein supplements for dairy cows given grass silage-based diets. Animal Science 77: 305-317. Steinshamn, H., L.T. Mydland, H. Volden and E. Thuen. 2006. The effect of red clover proportion in timothy silage on nitrogen utilization in lactating dairy cows. In: Wachendorf, M, Á. Helgadóttir and G. Parente (eds.). Sward dynamics. N-flows and forage utilisation in legume-based systems. Grado. Italy. 10-12 November. 2005. ESAAzienda regionale per lo sviluppo rurale, Italy, pp. 293-297. Thorhauge, M. and M.R. Weisbjerg, 2009. Effekt af fodring med mættet eller umættet fedt hos SDM-DH, RDM og Jersey. Kvæginfo no. 1938. Dansk Landbrugsrådgivning, Aarhus, Denmark. (In Danish). Thuen, E., 1989. Influence of source and level of dietary protein on milk yield, milk composition and some rumen and blood components in dairy cows. Doctor Scientarium Thesis. Agricultural University of Norway, as, Norway. 151 pp. Tothi, R, P. Lund, M.R. Weisbjerg and T. Hvelplund, 2003. Effect of expander processing on fractional rate of maize and barley starch degradation in the rumen of dairy cows estimated using rumen evacuation and in situ techniques. Animal Feed Science and Technology 104: 71-94. Van Amburgh, M.E., D.G. Fox, D.M. Galton, D.E. Bauman and L.E. Chase, 1998. Evaluation of National Research Council and Cornell net carbohydrate and protein systems for predicting requirements of Holstein heifers. Journal of Dairy Science 81: 509-526. Vestergaard, M., K.F. Jørgensen, I. Fisker, H.R. Andersen and K. Sejrsen, 2006. Forhøjet proteintildeling til store kvier - Resultater fra forsøg på Kvægbrugets ForsøgsCenter. In: Fodringsdag - Temadag om aktuelle fodringsspørgsmål. 29 august. Dansk Kvæg, pp. 41-45. (In Danish). Volden, H. 1990. Response of dietary AAT density on milk yield - Review of production trials in Norway. Proceedings of the seminar on fat and protein feeding to the dairy cow, 15-16 October, Eskilstuna, Sweden, pp. 81-89. Volden, H., 1999. Effects of level of feeding and ruminally undegraded protein on ruminal bacterial protein synthesis, escape of dietary protein, intestinal amino acid profile, and performance of dairy cows. Journal of Animal Science 77: 1905-1918. Volden, H. and O.M. Harstad, 2002. Effects of duodenal amino acid and starch infusion on milk production and nitrogen balance in dairy cows. Journal of Animal Science 80, Supplement 1. 321 pp. Volden, H.L., T. Mydland, V. Olaisen, 2002. Apparent ruminal degradation and rumen escape of soluble nitrogen fractions in grass, and grass silage administered intraruminally to lactating dairy cows. Journal of Animal Science 80: 2704-2716. Wallsten, J., J. Bertilsson, E. Nadeau and K. Martinsson, 2010. Digestibility of whole-crop barley and oat silages in dairy heifers. Animal 4-3: 432-438. Weisbjerg, M.R. and K. Soegaard, 2008. Feeding value of legumes and grasses at different harvest times. In: Hopkins, A., T. Gustafsson, J. Bertilsson, G. Dalin, N. Nilsdotter-Linde and E. Sporndly (eds.). Biodiversity and animal feed: Future challenges for grassland production. Grassland Science in Europe, volume 13, pp. 513-515. Weisbjerg, M.R., 2007. Lucerneensilage til malkekøer – resultater fra nye forsøg. In: Fodringsdag - Temadag om aktuelle fodringsspørgsmål. 5-12. Dansk Kvæg, Aarhus, Denmark. (In Danish). Weisbjerg, M.R., L. Wiking, N.B. Kristensen and P. Lund, 2008. Effects of supplemental dietary fatty acids on milk yield and fatty acid composition in high and medium yielding cows. Journal of Dairy Research 75: 142-152. Whitlock, B.K., M.J. Vandehaar, L.F.P. Silva and H.A.Tucker, 2002. Effect of dietary protein on prepurbertal mammary development in rapidly growing dairy heifers. Journal of Dairy Science 85: 1516-1525.

NorFor – The Nordic feed evaluation system

165

15. The NorFor IT system, ration formulation and optimisation H. Volden, H.D. Rokkjær, A. Göran and M. Åkerlind This chapter describes how we have converted a whole-animal model into a commercial computer tool to be used by advisors and farmers to formulate and optimise diets for dairy cows and growing cattle. The NorFor system is a semi-mechanistic model that has many theoretical advantages for the formulation and evaluation of feed rations. The system represents a significant development in models of nutritional formulation. It incorporates a detailed description of the protein and carbohydrate composition of the ration, which is used in the algorithms to predict degradation, digestion and metabolism in the gastro-intestinal tract and also the supply of metabolizable energy and protein. It can be used not only to estimate nutritional requirements but also provides substantial benefits for predicting nutrient utilization and the environmental impact from different diets. A complex model with semi-mechanistic algorithms and non-linear functions requires the development of a powerful, user-friendly computer tool before it can be used in practical ration formulation and optimisation. Feed is one of the major costs in modern cattle production, so it is important to consider both nutritional and economic aspects when formulating optimal rations. A key objective in modern, computerised ration formulation is auto-balancing (Boston et al., 2000), i.e. identification of the least expensive combination of feed ingredients that provides all considered nutrients within specified ranges. Use of an optimisation (auto-balancing) approach in the formulation of feed rations for cattle requires detailed nutritional and economic knowledge, including information regarding available feeds, the animals and management practices. Some feeds are home-grown while others are purchased as individual ingredients or as concentrate mixtures, and knowledge of their characteristics, obtained from either analyses or tables are needed. In addition, information on feed prices is required. Important animal information includes lactation stage, lactation number, BW, gestation day and age. Management factors include housing (tied up or loose housed) and feeding system (total mixed ration or separate feeds; pasture or indoor). All of these factors affect feed requirements and it is therefore essential that all relevant information is readily available. For these reasons, NorFor has put substantial efforts and resources into the development of an effective, economic IT platform and computer tool that can be used in practical ration evaluation and optimisation.

15.1 IT platforms NorFor is used by nutrition advisors in Denmark, Iceland, Norway and Sweden. An objective in the development of the computer tools was to use an IT platform that allows operators to access the latest version of the model. Hence, all users had to be able to access it via the same server system. Two IT platforms were developed: one web-based and one in which operators use an offline client, synchronized with the NorFor sever before starting the ration formulation. Denmark, Iceland and Norway have chosen to use an online solution, while Sweden is using an offline solution. Although, all countries use the common NorFor severs, each country, except Iceland, has developed their own national client and interface, to enable the NorFor system to be integrated with other national advisory tools. Figure 15.1 presents an overview of the common IT system, which consists of a server application, a homepage, an administration tool and Web services for dairy management software (National herd recording system), laboratories and feed companies. The NorFor servers host four web services: the feed analysis system (FAS), the feedstuff table (FST), the feed ration calculator (FRC), and the one-day feeding control (OFC) system.

15.2 The feed analysis system (FAS) NorFor has developed a web-client that is used by linked laboratories to report results from herd level feed analyses (Figure 15.2). The laboratories connect to the system and upload analytical results, H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_15, © Wageningen Academic Publishers 2011

167

Figure 15.1. An architectural overview of the NorFor IT system.

Figure 15.2. An architectural overview of the feed analysis system (FAS). and the system calculates feed characteristics (using the FST Calculator) from other analytical data, as described in Chapter 6. Based on the NorFor feed code system, tabulated values are added to the feed samples if absent. If the feed sample is from a herd registered in the national herd recording system (Figure 15.3), it is assigned a herd-specific id, making it automatically available for use by the FRC for ration optimisation or by the OFC for calculating herd feed efficiency. In the NorFor server, each herd has its own feedstuff table, in which the analysed feeds and those copied from the FST are saved. When roughage is analysed at the laboratory, the farmer or advisor reports additional information about the feed sample. This information included date of harvest, cutting number, botanical composition, feed additives, storage system for silage and theoretical cutting length. These additional values are stored, together with the feed characteristics, in a common feed database, which can be used for 168

NorFor – The Nordic feed evaluation system

Figure 15.3. An architectural overview of the link between NorFor and the national client. statistical evaluation of factors such as the annual roughage quality and the relationships between feed additives and silage fermentation quality. Approximately 25,000 roughage samples are analysed according to NorFor specifications annually.

15.3 Feedstuff table (FST) NorFor has developed a comprehensive feedstuff table (FST) that is continuously updated. Updating and monitoring the FST is done by use of the NorFor administration tool. From the NorFor sever, the FST is available to each national client and when working on herd-specific feed plans, feedstuffs are copied from the FST into the herd feedstuff table for further use. For external users, the FST is available at www.norfor.info. Figure 15.4 shows an overview of the NorFor FST, which is organized in a hierarchal directory system, in which the highest level is region, dividing the feedstuffs into country specific and common NorFor categories. The feed directory is further divided into feed groups, by 19 sub-directories, and the feed groups are divided into feed types. For example, the feed group ‘grains, is divided into ‘grains’, ‘dry grain by-products’ and ‘wet grain by-products’. The ‘forages and roughage’ group is divided into seven feed types: ‘pasture grass and clover-grass’, ‘grass and clover-grass’, ‘whole crop’, ‘grass and clover-grass silage’, ‘whole crop silage’, ‘hay and ‘straw’, and ‘grass pellets’. Figure 15.4 also shows an example of choosing a parameter setting (‘NDF characteristics’). These settings are used to generate reports on variables such as NDF, starch, protein, amino acids fatty acids, fermentation products, minerals, vitamins or total characteristics. A part of the feed table system is also information on commercial compound feeds including prices, reported by the feed industry (Figure 15.5). When feed rations are formulated, the operators copy NorFor – The Nordic feed evaluation system

169

Figure 15.4. Overview of the NorFor feedstuff table and example of NDF characteristics in feedstuffs from the feed group forages and roughage.

Figure 15.5. An architectural overview of the feed company web-service in NorFor. 170

NorFor – The Nordic feed evaluation system

actual company products into the herd specific feedstuff table. NorFor has developed a web-service to help the feed industry report these details to the FST.

15.4 Dairy management system Each Nordic country has its own national herd recording system, in which both records for individual cows (e.g. lactation number, age, days in milk, milk yield, milk composition, BCS, fertility and health) and herd information (e.g. 305-d milk yield, milk composition, slaughter weight and slaughter age) are reported. When formulating diets in the NorFor system, herd information is available (Figure 15.3), which means that rations can be formulated for either individual cows or groups of cows in the herd.

15.5 Feed ration calculator (FRC) A requirement for the NorFor IT system was that it should allow both the calculation and optimisation of rations. The chosen solution was three calculator add-inns and a black-box optimisation approach (Figure 15.6). The FRC Calculator is used to evaluate feed rations and to predict production responses when the feed input is known. The FAS/FST Calculator is used to calculate standard feed values for a single feedstuff (see Chapter 13) by combining values obtained from the FRC calculator and information from the FAS and FST. The SNOPT™ Optimizer is used to optimise (auto-balance) a feed ration i.e. produce a ration that provides all the required nutrients at the lowest possible cost. This means that when using the optimiser individual feed prices must be reported through the herd feedstuff table. The SNOPT Optimizer is a package of software and algorithms, developed by researchers (Gill et al., 2005) at Stanford University and the University of California (http://www.sbsi-sol-optimize.com), for solving largescale optimisation problems (linear and nonlinear). It is especially effective for nonlinear problems whose functions and gradients are expensive to evaluate. SNOPT was implemented in NorFor with guidance from the TOMLAB company (http://www.tomopt.com/tomlab).

SNOPT optimizer

OFC calculator

Animal data

FRC calculator

Herd feedstuffs

FAS/FST calculator

Figure 15.6. The add-inns used in the NorFor FRC calculation and optimisation.

NorFor – The Nordic feed evaluation system

171

6046 6046 6046 6047 6047 6047 6048 6048 6048 6049 6049 6049 6050 6050 6050 6051 6051 6051 6052 6052 6052 6053 6053 6053 6054 6054 6054 6055 6055 6055 6056 6056 6056 6057 6057 6057 6058 6058 6058 6059 6059 6059 6060 6060 6060 6061 6061 6061 6062 6062 6062 6063 6063 6063 6064 6064 6064 6065 6065 6065 6066 6066 6066 6067 6067 6067 6068 6068 6068 6069 6069 6069 6070 6070 6070 6071 6071 6071 6072 6072 6072 6073 6073 6073 6074 6074 6074

The optimisation software module consists of a high-level interface written in C# targeting the Microsoft®.NET platform, allowing optimisation problems to be specified, and a low-level SNOPT 7 solver library built using Fortran 77 code. The functions optimized are constructed using C#-like syntax and compiled in a library called the high-level code. For nonlinear problems such as those posed in NorFor, SNOPT employs a sparse SQP algorithm, described by Gill et al. (2005). The nonlinear solver requires knowledge of the first derivatives of all (nonlinear) functions, and if some are unknown, they are estimated numerically. Initially, the NorFor system used numerical gradients for all nonlinear functions. However, the calculation engine applies automatic differentiation (AD). This eliminates the round-off errors normally associated with numerical estimation and yields exact derivatives down to machine precision. Tests made during the development of the AD package showed an increasedmethod robustness without significantly reducing performance by usefloating of AD. This is operator-overloading method applied might might theoretically theoretically be be slower slower than than using using regular regular floating operator-overloading applied important, since the operator-overloading method applied might theoretically be slower than using operator-overloading method applied might theoretically be slower than using regular floating point expressions. point expressions. point expressions. regular floating point expressions. The question of whether the optimum for each been The question of whether SNOPT SNOPT finds finds the global global optimum for each problem problem has has been raised. raised. The question of finds global optimum for has Thethe question ofwhether whetherSNOPT SNOPT findsthe thethe global optimum foreach eachproblem problem hasbeen beenraised. raised. For the For types of problem,, this depends on nature of the nonlinear constraints. For the the types types of of problem,, problem,, this this depends depends on on the the nature nature of of the the nonlinear nonlinear constraints. constraints. If If the the For If the types of problem, this depends on the nature of the nonlinear constraints. If the constraint constraint functions functions are are known known explicitly, explicitly, aa person person might might be be able able to deduce deduce whether whether or or not not functions constraint constraint functions are known explicitly, adoing person might be able to toor deduce whether ora not are known explicitly, a person might be able to deduce whether not they define they define a convex region, but although this automatically using a computer they define define aa convex convex region, region, but but although although doing doing this this automatically automatically using using aa computer computer convex region, they algorithm (without symbolic description) might be possible, possible, it would would be far far from from trivial. description) It is is but although doingaathis automatically using a computer algorithm (without a symbolic algorithm (without symbolic description) might be it be trivial. algorithm (without a symbolic–description) might be possible, itstarting would be far from trivial. It It is easier non-convexity it suffices to find different points that give might to beprove possible, it would be–far from trivial. It two is easier to prove non-convexity – it suffices to find easier to prove non-convexity it suffices to find two different starting points that give easier to prove non-convexity – it suffices towhich find two different starting points that give significantly different local optimal points, can be done with little knowledge of the two different starting points that give significantly different local optimal points, which can be done significantly different different local local optimal optimal points, points, which which can can be be done done with with little little knowledge knowledge of of the the significantly nature of the functions involved. Nevertheless, SNOPT is an industry standard nonlinear with little knowledge of the nature of the functions involved. Nevertheless, SNOPT is an industry nature of the functions involved. Nevertheless, SNOPT is an industry standard nonlinear nature of the solver functions involved. Nevertheless, SNOPT is an industry standard nonlinearfind optimisation that even on problems. It standard nonlinear solver well that often well even on non-convex problems. It may optimisation solver optimisation that often often performs performs well even performs on non-convex non-convex problems. It may may well well find optimisation solver that often performs well even on non-convex problems. It may well find the global optimum by chance,bydepending depending on the the given given starting point. However, method to to the optimum chance, on point. However, aaa method wellglobal find the global by optimum chance, depending on starting the given starting point. However, a method to the global optimum by chance, depending on the given starting point. However, method to increase run more than one optimisation, increase the odds of identifying global optimality is isto totosimply simply run more than oneone optimisation, increasethe theodds oddsof ofidentifying identifyingglobal globaloptimality optimalityis simply run more than optimisation, with increase the odds of identifying global optimality is to simply run more than one optimisation, with different starting points. If all runs give the same optimum, it is likely, although not with different starting points. If runs all runs runs give the same optimum, it is likely,although althoughnot not100% certain, different starting points. If allIf givegive thethe same optimum, it it is is likely, with different starting points. all same optimum, likely, although not 100% certain, that the global optimum has been found. A multi-start algorithm has been 100% certain, that the global optimum has been found. A multi-start algorithm has been that thecertain, globalthat optimum has been found. Abeen multi-start algorithm hasalgorithm been implemented in the NorFor 100% the global optimum has found. A multi-start has been implemented in NorFor software system, which provides an automated method for implemented in the thewhich NorFor software system, which provides an automatedmultiple method starting for implemented in the NorFor software system, which provides an automated method for software system, provides an automated method for generating generating multiple multiple starting starting points points that that can can be be used used for for repeated repeated local local optimisation optimisation runs. runs. It It points that generating generating multiple starting points that can be used for repeated local optimisation runs. It according can be used for repeated local optimisation runs. It automatically detects similar solutions automatically detects similar solutions according to criteria and can used automatically detects and similar solutions according to tolerance tolerance criteria and can be be used to to problem is automatically detects similar solutions according to tolerance criteria and can be used to to tolerance criteria can be used to obtain good indications of whether a particular obtain good good indications indications of of whether whether aa particular particular problem problem is is convex convex or or not. not. In In aa test test of of 20 20 actual actual obtain obtain indications whether athe particular problem is convex or not. a test of 20found actualto have a convexgood or not. In aevery test of of 20 actual NorFor problems, every inInthe set was NorFor problems, problem in set was found to have aaproblem unique optimal point, NorFor problems, every problem in the set was found to have unique optimal point, NorFor problems, every problem in the set was found to have a unique optimal point, unique optimal point, although for a subset of the problems certain starting points (typically one or although although for for aaa subset subset of of the the problems problems certain certain starting starting points points (typically (typically one one or or two two of of 20-50 20-50 although for subset of the problems certain starting points (typically one or two of 20-50 two of 20-50 randomly chosen points) resulted in infeasible solutions. This is not a cause for concern, randomly chosen points) resulted in infeasible solutions. This is not a cause for concern, since randomly chosen points) resulted in infeasible solutions. This is cause for concern, since randomly chosen points)be resulted innot infeasible solutions. This is not not aaand cause concern, can since sincewill therealways will always points do not all constraints, the for multi-start can there be do all constraints, and algorithm there will always be points points that that do that not fulfil fulfil allfulfil constraints, and the the multi-start multi-start algorithmalgorithm can there will always be points that do not fulfil all constraints, and the multi-start algorithm can be applied ininattempts attempts totosolve solve problems that seem to to be be infeasible. be applied appliedin attemptsto solveproblems problems that seem infeasible. be that seem to be infeasible. be applied in attempts to solve problems that seem to be infeasible. The in the NorFor system is the The principal principaloptimisation optimisationproblem problem NorFor system to identify the cheapest combination The principal optimisation problem in in thethe NorFor system is to toisidentify identify the cheapest cheapest The principal optimisation problem in the NorFor system is to identify the cheapest combination of feed ingredients that meets nutritional requirements. The linear cost is function of feed ingredients that meets nutritional requirements. The linear cost function simply the total combination of feed ingredients that meets nutritional requirements. The linear combination of feed ingredients that meets nutritional requirements. The linear cost cost function function is simply the total cost of a ration, expressed: cost of a ration, is simply the total totalexpressed: cost of of aa ration, ration, expressed: expressed: is simply the cost nn

6075 6075 6075

n pi x min ff ((xx )) = min = ∑ ∑p pi x ii min x f (x ) = i∑ =1 i x i

15.1

6076 6076 6076 6077 6077 6077

≤ Ax Ax ≤ ≤b bU bbb L ≤ L L ≤ Ax ≤ b U U cc L ≤ cc((xx )) ≤ cc U ≤ ≤ L ≤ c(x ) ≤ c U U cL

15.2

15.3

15.4

6078 6078 6078 6079 6079 6079 6080 6080 6080 6081 6081 6081 6082 6082 6082 6083 6083 6083 6084 6084 6084 6085 6085 6085 6086 6086 6086 6087 6087 6087 6088

xx

ii = =11

n,x ≥ 0 n pp ,, x ∈R n ≥0 0 p, x x∈ ∈R R n ,, x x≥

m1 m1 × ×n bU ∈ ∈R Rm A∈ ∈R Rm m111 ,, A m 111 .×nn n bbb L ,, b L L ,bU U ∈R ,A∈R m m 22 cc L ,, cc U ,, cc(( xx )) ∈ 2 ∈R Rm 2 cL L , cU U , c( x ) ∈ R where, p is the price unit of feedstuff i; xi is the amount of feedstuff ii used in the ration. i is the price per per of i; of the where, p i nn where, pii isthe theset price per unit unit of feedstuff feedstuff i; xxiii is is the then;amount amount of feedstuff feedstuff i used used in inelements; the ration. ration. R implies of real numbers with dimension A is a matrix with constant n n R implies implies the the set set of of real real numbers numbers with with dimension dimension n; n; A A is is aa matrix matrix with with constant constant elements; elements; bbbLLL R L are linear constraints expressed on vector form; c and c are non-linear constraints. and b U L U and and bbUUU are are linear linear constraints constraints expressed expressed on on vector vector form; form; ccLLL and and ccUUU are are non-linear non-linear constraints. constraints.

15.5 15.6

The set of constraints constraints expresses expresses nutritional nutritional demands, demands,NorFor which include include DMI, energy, energy, AATNN,, The DMI, AAT 172set of – The Nordic feed evaluation system The set of constraints characteristics expresses nutritionalrumen demands, which which include DMI, energy, AATNN, PBV PBVNNN and and nutritional nutritional characteristics characteristics (e.g., (e.g., rumen rumen load load index, index, fatty fatty acids, acids, NDF NDF and and minerals). minerals). PBV and nutritional (e.g., load index, fatty acids, NDF and minerals). N The The constraints constraints are are divided divided into into m m111 linear linear and and m m222 nonlinear nonlinear constraints, constraints, aaa special special case case being being linear and m nonlinear constraints, case being The constraints are divided into m 1are 2nonlinear, the so-called ratio constraints that originally being quotientsspecial of linear sums:

where, pi is the price per unit of feedstuff i; xi is the amount of feedstuff i used in the ration. Rn implies the set of real numbers with dimension n; A is a matrix with constant elements; bL and bU are linear constraints expressed on vector form; cL and cU are non-linear constraints. The set of constraints expresses nutritional demands, which include DMI, energy, AATN, PBVN and nutritional characteristics (e.g. rumen load index, fatty acids, NDF and minerals). The constraints are divided into m1 linear and m2 nonlinear constraints, a special case being the so-called ratio constraints that are originally nonlinear, being quotients of linear sums:

∑ ∑

n i =1 i i n i =1 i

wx

6089

r (x ) =

6090 6091 6092 6093 6094 6095 6096 6097 6098 6099 6100 6101 6102 6103 6104 6105 6106 6107 6108 6109 6110 6111 6112 6113 6114 6115 6116 6117 6118 6119 6120 6121 6122 6123 6124 6125 6126 6127 6128 6129 6130 6131 6132 6133 6134 6135 6136

where, wi are constant real-valued weighting values. These constraints are easily transformed into at most two linear constraints. When optimising for more than one animal where, wi is constant real-valued These constraints easily transformed into at simultaneously, two runs are madeweighting — first forvalues. the individual animals andare then for the entire most two constraints. When optimising for problem more than onetoanimal group withlinear additional constraints. The multi-animal leads a blocksimultaneously, structure in the two runs constraint 15.7 illustrates a hypothetical with four with animals (four constraints. are made: matrix. first forFigure the individual animals and then forsituation the entire group additional blocks), 10 feeds and five constraints which link different animals together), theillustrates The multi-animal problem leads to a(three blockof structure in the constraint matrix. Figureand 15.7 total usage of feedstuffs 1, 2 and is limited. a hypothetical situation with four3animals (four blocks), 10 feeds and five constraints (three of which

x

15.7

link different animals together), and the total usage of feedstuffs 1, 2 and 3 is limited.

Figure 15.7

One of the main requirements when selecting the optimisation platform was that the execution time

One of the mainArequirements selecting the optimisation platform client was that the execution should be fast. performancewhen test was conducted with the Norwegian involving simultaneous time should be fast. A performance test was conducted with the Norwegian client involving optimisation for 30 cows, nine constraints and five feedstuffs. The obtained optimisation time was simultaneous optimisation for 30 cows, nine constraints and five feedstuffs. The optimisation 0.23 second per cow. time was 0.23 s per cow. The goal of optimisation is to find the least costly combination of feeds that meets a set of constraints (minimum and maximum) ensuring that amounts of nutrients supplied are within specified ranges. Thus, a set of constraints is needed for both feed ingredients and nutrients. In a practical situation, the amount of home grown feed is known and feed constraints are often based on their availability. Therefore, the amount of a feed in the ration may often be adjusted by the operator to take into account a minimum or maximum that must be used from an inventory perspective. When feeds are purchased an optimisation approach is useful when the feed budget and purchasing of ingredients are planned. Nutritional constraints are based on either their direct effects on production responses, i.e., the requirements (NEL and AATN) of animals to meet target performance criteria or nutrient amounts and concentrations in the rations, e.g., PBV, RLI, NDF, Starch, CP, fatty acids and minerals. It is possible to optimise 84 nutritional variables in NorFor but not all of them have recommended constraints at present. When rations are optimised in the national clients, we start with a recommended default setting for eight nutritional constraints (Table 15.1). Table 15.1

Table 15.2 illustrates the process of ration optimisation with a set of diets formulated with variations in the constraints of RLI (0.6 vs. 0.45) and AAT/NEL (15 vs. 16 g/MJ). TMR rations were formulated for a target of 35 kg ECM and in addition to the standard optimisation constraints, several other nutritional variables were used to evaluate and characterise the rations. Reducing the RLI (diet 1 vs. diet 2) reduced the proportion of wheat grain in the optimal diet by 7.5 percent and increased the proportion of cold-pressed rapeseed by 5.8 percent. Thesparsity change pattern in RLI affected the starch content, which decreased by 24 % when Figure 15.7. The for a problem with four animals, ten feedstuffs, five constraints the was reduced. Increasing the minimum requirement of AAT/NEL from 15 to 16 g/MJ andRLI three constraints limiting total usage of feedstuffs. (Diet 2 vs. Diet 3) increased the proportion of extracted soybean meal and decreased the proportion of rapeseed in the optimal diet by 3.7 and 4.6 percent, respectively. This change resulted in an increased CP concentration in the diet and minor changes in the AA composition. The simple examples presented in Table 15.2 demonstrate how the combination of auto-balancing and nutritional constraints change the optimal diet composition. They also NorFor – The Nordic feed evaluation system 173 show the sensitivity of the optimisation procedure to changes in the constraints and highlight

The goal of optimisation is to find the least costly combination of feeds that meets a set of constraints (minimum and maximum) ensuring that amounts of nutrients supplied are within specified ranges. Thus, a set of constraints is needed for both feed ingredients and nutrients. In a practical situation, the amount of home grown feed is known and feed constraints are often based on their availability. Therefore, the amount of a feed in the ration may often be adjusted by the operator to take into account a minimum or maximum that must be used from an inventory perspective. When feeds are purchased an optimisation approach is useful when the feed budget and purchasing of ingredients are planned. Nutritional constraints are based on either their direct effects on production responses, i.e. the requirements (NEL and AATN) of animals to meet target performance criteria or nutrient amounts and concentrations in the rations, e.g. PBV, RLI, NDF, Starch, CP, fatty acids and minerals. It is possible to optimise from 84 nutritional variables in NorFor but not all of them have recommended constraints at present. When rations are optimised in the national clients, we start with a recommended default setting for eight nutritional constraints (Table 15.1). Table 15.2 illustrates the process of ration optimisation with a set of diets formulated with variations in the constraints of RLI (0.6 vs. 0.45) and AAT/NEL (15 vs. 16 g/MJ). TMR rations were formulated for a target of 35 kg ECM and in addition to the standard optimisation constraints, several other nutritional variables were used to evaluate and characterise the rations. Reducing the RLI (diet 1 vs. diet 2) reduced the proportion of wheat grain in the optimal diet by 7.5% and increased the proportion of cold-pressed rapeseed by 5.8%. The change in RLI affected the starch content, which decreased by 24% when the RLI was reduced. Increasing the minimum requirement of AAT/NEL from 15 to 16 g/MJ (Diet 2 vs. Diet 3) increased the proportion of extracted soybean meal and decreased the proportion of rapeseed in the optimal diet by 3.7 and 4.6%, respectively. This change resulted in an increased CP concentration in the diet and minor changes in the AA composition. The simple examples presented in Table 15.2 demonstrate how the combination of auto-balancing and nutritional constraints changes the optimal diet composition. They also show the sensitivity of the optimisation procedure to changes in the constraints and highlight the importance of ensuring that feed rations obtained by optimisation are assessed by a trained nutritionist before they are implemented. Table 15.1. Default optimisation constraint settings in NorFor for dairy cows. Item

Ration fill value Energy balance, % AATN/NEL2, g/MJ AATN response, % PBVN3, g/kg DM RLI4, g/g Fatty acids, g/kg DM Chewing index, min/kg DM

Constraint Minimum

Maximum

IC·0.97 99.5 15 95 10

IC1 100.5

20 32

40 0.6 45

1

IC = animal intake capacity. AATN/NEL = ratio between amino acids absorbed in the small intestine (AATN) and net energy lactation (NEL) available for milk production. 3 PBV = protein balance in the rumen. 4 RLI = rumen load index. 2

174

NorFor – The Nordic feed evaluation system

Table 15.2. Ration optimisation with varied constraints for rumen load index and AAT/NEL for a 600 kg dairy cow producing 35 kg ECM. Ration Diet 1

Diet 2

Diet 3

Ingredients, % of DM Grass silage, medium OMD1 Grass silage, high OMD Maize silage, high OMD Wheat grain Wheat, brewers grain Soybean meal Rapeseed, cold pressed Salt Mineral and vitamin mix

16.5 15.8 24.8 26.6 6.3 3.4 5.6 0.23 0.68

16.1 20.2 23.3 19.1 6.3 2.8 11.4 0.22 0.67

16.2 19.5 24.8 19.0 6.3 6.5 6.8 0.23 0.68

Total dry matter, kg/d Cost, NOK

22.2 29.86

22.3 29.64

22.2 30.10

8.45 100 15 95 10 0.6 42 41 57.2 154 340 265 33 6.54 2.21 2.49 32 47 28

8.45 100 15 95 20 0.45 50 42 59.6 165 366 202 38 6.61 2.29 2.53 37 51 30

8.45 100 16.0 98 20 0.45 44 42 60.4 170 363 205 36 5.59 2.22 2.50 35 48 29

Ration fill value, FV/kg DM Energy balance, % AATN/NEL, g/MJ2 AATN response, % PBVN, g/kg DM3 RLI, g/g4 Crude fat, g/kg DM Chewing index, min/kg DM Roughage proportion, % of DM Crude protein, g/kg DM NDF, g/kg DM Starch, g/kg DM Sugar, g/kg DM Lysine, % of AATN Methionine, % of AATN Histidine, % of AATN Calcium, g/kg DM Phosphorus, g/kg DM Magnesium, g/kg DM 1

OMD = organic matter digestibility. AATN/NEL = ratio between amino acids absorbed in the small intestine (AATN) and net energy lactation (NEL) available for milk production. 3 PBV = protein balance in the rumen. N 4 RLI = rumen load index. 2

NorFor – The Nordic feed evaluation system

175

In the system development, testing and implementation process, NorFor has developed a PC program called the NorFor Development Tool (NDT). The NDT runs off-line, incorporates all the NorFor equations and compiles the three calculator add-inns (Figure 15.8). This means that it is possible both to calculate and optimize rations with NDT. The program is used by the personnel who develop and maintain the system. It is easy to edit existing equations and to include new equations in the program (Figure 15.9). It is also possible to both import and export data between NDT and Microsoft Excel, which makes it easy to test and evaluate experimental data. Import of feedstuff specifications into the NDT from the online NorFor feedstuff table is implemented. An essential feature of the NDT is the unit test, which allows multiple tests to be run simultaneously to ensure that changes in the equation sets do not have unexpected effects on different parts of the model. Before a new version of the NorFor system becomes available for the national clients, it is extensively evaluated in the NDT. After approval, the new/edited equations are transferred to the operative NorFor servers. This procedure ensures that the system is updated safely.

15.6 One-day feeding control (OFC) The OFC is a ration evaluator that is used to calculate herd feed efficiency (mainly in terms of energy and nitrogen) for a specific day or period. It compiles data from the national herd recording system and from the FST and FRC. The number of animals in each category (cows, calves, heifers and bulls) and the status of individual animals (e.g. MY, milk chemical, gestation day and age) are available from the national herd recording system for the specific test day. The feed intake is measured for each animal category or groups of animals and, in combination with information from the herd feedstuff table, the FRC is used to calculate ration composition, production responses and feed efficiency. The ration evaluation and OFC are then used for optimising/adjusting the ration for the next feeding period.

Figure 15.8. The architectural structure of the NorFor Development tool.

176

NorFor – The Nordic feed evaluation system

Figure 15.9. Example of a dialogue box in the NorFor Development Tool used to edit and test new equations.

References Boston, R.C., D.G. Fox, C. Sniffen, E. Janczewski, R. Munson and W. Chalupa, 2000. The conversion of a scientific model describing dairy cow nutrition and production to an industry tool: the CPM dairy project. In: McNamara, J., J. France and D.E. Beever (eds.). Modelling nutrient utilization in farm animals. CABI Publishing, Wallingford, UK, pp. 361-377. Gill, P.E., W. Murray and M.A. Saunders, 2005. SNOPT: An SQP algorithm for large-scale constrained optimisation. SIAM Review 47: 99-131.

NorFor – The Nordic feed evaluation system

177

Keyword index A AATN 83 –– available for growth 98 –– available for milk production 95 –– balance 96 –– efficiency 95, 100 administration tool 169 amino acids 70, 71, 72, 73 ammonia 44, 62 –– correction 37 animal production level 82 apparent digestibility 75, 77 ATP 66, 69 B black-box optimisation body condition score body protein

171 27 101

C carbohydrate cation-anion difference cellulolytic bacteria chewing index –– eating index –– particle length –– recommendation –– rumination index concentrates crude protein

35 57, 108 64 127 128 128, 129 104 129 66 34, 45, 46, 62

D daily gain degradation rate –– in feed mixtures Development Tool digestion methods drying temperature

30 59 57 176 42 42

E effective rumen degradability 66, 67 endogenous protein 83 energy balance 91 energy corrected milk 28 energy requirement 85, 86, 87, 90 enzymatic digestion 71 F fatty acids feed analysis

73, 104 41

feed ration calculator feedstuff table fermentation products –– acids –– alcohols G gestation global optimum gross energy growth function I input data in sacco –– mobile bag in vitro in vivo IT platforms IT system J Jersey

167, 171 167, 169 38 38 38 27 172 81 31 27, 29 35, 48, 51, 63 34 45, 55 45 167 167 27

L lactation curves 28 large intestine 74 –– degradation of neutral detergent fibre in feedstuff 75 –– degradation of starch 75 –– microbial protein synthesis 75 linear cost function 172 Lucas principle 35, 55 M mature body weight 28, 31 metabolizable energy 81 –– efficiency 81 methane 81 microbial –– efficiency 67 –– growth 62, 66 –– protein 83 –– protein synthesis 67 milk protein maximum production 96 mineral recommendations 105, 108, 113 minerals 38 mobile nylon bag 52

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9, © Wageningen Academic Publishers 2011

179

N national client national herd recording system net energy lactation neutral detergent fibre in feedstuff –– degradation –– indigestible –– potentially degradable nitrogen utilisation non-linear optimization code O one-compartment –– degradation model –– digestion model one-day feeding control optimisation organic matter –– digestibility

169 171 82 35 55, 65 35, 51 35 134 115

74 65 167, 176 174 33 45

P partial efficiency of metabolizable energy 82 –– kg 82 –– km 82 –– kmg 82 particle length 33, 47 passage rate 59, 60, 61 PBVN 70 PBVN recommendation 102 phosphorus utilisation 135 potassium utilisation 136 prediction of milk yield 133 prediction of weight gain 133 protein 71, 73, 93, 94 –– production 28 protein requirement –– gestation 102 –– growth 97 –– mobilization and deposition 101 R RestCHO rumen evacuation technique rumen load index S sensitivity analysis silage index small intestine SNOPT Optimizer standard feed value –– AATN8 180

37, 63 60 64, 104 141 38, 56 71, 73 171 137 138

–– AATN20 138 –– NEL8 138 –– NEL20 138 –– PBVN8 138 –– PBVN20 138 starch 35, 36, 63, 73 substitution rate 113 system evaluation 141 –– chewing index 159 –– digestive tract sub-model 143 –– energy requirement for growing cattle 153 –– milk production 149 –– roughage intake 154 –– rumen load index 153 –– sensitivity analysis 142 –– total dry matter intake 154 T theoretical chopping length triacylglycerols two-compartment digestion model U urea correction

128 66 65 37

V vitamin recommendations 108 –– vitamin A 109 –– vitamin D 109 –– vitamin E 109 vitamins 39 voluntary feed intake 113 –– capacity of bulls 119 –– capacity of Jersey 121 –– capacity of lactating dairy cows 115 –– capacity of steers and heifers 120 –– fill value 56 –– metabolic correction factor 118, 123 –– metabolic regulation factor 117 –– recommendations 104 –– substitution correction factor 122 –– substitution rate factor 117 W water-free amino acids web services

72 167

NorFor – The Nordic feed evaluation system

EAAP – European Federation of Animal Science

NorFor Nordic feed evaluation system

The European Association for Animal Production wishes to express its appreciation to the Ministero per le Politiche Agricole e Forestali and the Associazione Italiana Allevatori for their valuable support of its activities

NorFor – The Nordic feed evaluation system

EAAP publication No. 130

edited by: Harald Volden

Wageningen Academic P u b l i s h e r s

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned. Nothing from this publication may be translated, reproduced, stored in a computerised system or published in any form or in any manner, including electronic, mechanical, reprographic or photographic, without prior written permission from the publisher: Wageningen Academic Publishers P.O. Box 220 6700 AE Wageningen The Netherlands www.WageningenAcademic.com [email protected]

ISBN: 978-90-8686-162-0 e-ISBN: 978-90-8686-718-9 DOI: 10.3920/978-90-8686-718-9 ISSN 0071-2477

Photographer: Jens Tønnesen from Dansk Landbrugs Medier

First published, 2011

©Wageningen Academic Publishers The Netherlands, 2011

The individual contributions in this publication and any liabilities arising from them remain the responsibility of the authors. The designations employed and the presentation of material in this publication do not imply the expression of any opinion whatsoever on the part of the European Association for Animal Production concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The publisher is not responsible for possible damages, which could be a result of content derived from this publication.

Table of contents Preface Anders H. Gustafsson

9

Acknowledgements

11

Glossary

13

1. Introduction H. Volden and A.H. Gustafsson

21

2. Overall model description H. Volden

23

3. Animal input characteristics. M. Åkerlind, N.I. Nielsen and H. Volden

27

4. Feed fraction characteristics H. Volden

33

5. Feed analyses and digestion methods M. Åkerlind, M. Weisbjerg, T. Eriksson, R. Tøgersen, P. Udén, B.L. Ólafsson, O.M. Harstad and H. Volden

41

6. Feed calculations in NorFor H. Volden

55

7. Digestion and metabolism in the gastrointestinal tract H. Volden and M. Larsen

59

8. Energy and metabolizable protein supply H. Volden and N.I. Nielsen

81

9. Animal requirements and recommendations N.I. Nielsen and H. Volden

85

10. Prediction of voluntary feed intake H. Volden, N.I Nielsen, M. Åkerlind, M. Larsen, Ø. Havrevoll and A.J. Rygh

113

11. Chewing index system for predicting physical structure of the diet P. Nørgaard, E. Nadeau Å. Randby and H. Volden

127

12. Prediction of milk yield, weight gain and utilisation of N, P and K M. Åkerlind and H. Volden

133

13. Standard feed value M. Åkerlind and H. Volden

137

NorFor – The Nordic feed evaluation system

7

14. System evaluation H. Volden, N.I. Nielsen, M. Åkerlind and A.J. Rygh

141

15. The NorFor IT system, ration formulation and optimisation H. Volden, H.D. Rokkjær, A. Göran and M. Åkerlind

167

Keyword index

179

8

NorFor – The Nordic feed evaluation system

Preface The present volume is the first comprehensive, published description of the Nordic Feed Evaluation System, NorFor. It includes the results of extensive work initiated as a project in 2001 by the farmers’ dairy cooperatives in Denmark, Norway, Iceland and Sweden. The overall aim was to create an identical, common feed evaluation system for all four countries to facilitate communication between farmers, consultants and feed industry representatives. Additional goals were to make dissemination of relevant research findings faster and more efficient, both within the four countries and internationally. Towards that end, a group of experts in ruminant nutrition was commissioned to develop such a system. Since we regarded this as a request to overhaul the current national systems comprehensively, we concluded that it provided an opportunity to develop a completely new feed evaluation system based robustly on current, published knowledge. NorFor was a development project that ran from 2002 to 2006 and included comparisons of various feed evaluation systems applied in western countries. The development also included validation of the models based on available research data from the Nordic countries. Significant parts of the development work involved close cooperation with scientists at the agricultural universities and research institutes in the Nordic countries. In addition, we have shared information and have had constructive discussions with experts in the feed industry and feed laboratories. We greatly appreciate the past and ongoing cooperation. From 2007 onwards, NorFor has been jointly funded and managed by the four organizations mentioned above through a cooperative organization (NorFor F.M.B.A.). The present description of NorFor includes all parts of the model. References to the most important sources of information are also given. We thereby provide a description of the biological basis for NorFor. The NorFor project group hopes this volume will be useful for dairy farmers, advisors, scientists and all other actors in the dairy industry in the four NorFor countries. We also recommend the reader to use the IT tools, since many of the nutritional interactions in the model are easier to use and understand if they are applied than if the texts, tables and graphs are read. NorFor Anders H. Gustafsson Leader of NorFor 2002-2009

NorFor – The Nordic feed evaluation system

9

Acknowledgements This publication summarizes the entire work of NorFor, including development and documentation. Numerous people have contributed to the compilation of this written document. NorFor has been developed over several years, thus the nature of their contributions has differed greatly. However, despite these differences, every contribution was equally appreciated. Thus, NorFor F.M.B.A. and the authors would like to sincerely thank everyone who has contributed to the project, and to this first version of the NorFor scientific publication. The main author of the present publication was Harald Volden, of TINE and the Norwegian University of Life Sciences, who has been the leader of the NorFor scientific development group. In addition, various people have played important roles in preparing the publication, including: Nicolaj I. Nielsen of AgroTech and the Danish Cattle Association; Maria Åkerlind of the Swedish Dairy Association; and Peder Nørgaard of the University of Copenhagen. Other persons, listed in the relevant chapters in this publication, have also contributed as authors. In addition, the following persons (in alphabetical order of family names) have contributed in different working groups during the development and/or documentation of NorFor: • Feed evaluation and feeding standards: Ole Aaes, Jan Bertilsson, Anders H. Gustafsson, Øystein Havrevoll, Mårten Hetta, Mogens Larsen, Maria Mehlqvist, Elisabeth Nadeau, Nicolaj I. Nielsen, Peder Nørgaard, Åshild Randby, Arnt Johan Rygh, Ingunn Schei, Harald Volden and Maria Åkerlind • Feed analysis and tables: Lars Bævre, Julia Beck, Margareta Emanuelson, Torsten Eriksson, Odd Magne Harstad, Erica Lindberg, Maria Mehlqvist, Jens Møller, Bragi Lindal Olafsson, Rudolf Thøgersen, Peter Udén, Harald Volden, Martin Riis Weisbjerg and Maria Åkerlind • IT-system: Johannes Frandsen, Henrik D. Rokkjær, Anders Göran (Tomlab), Niels Jafner, Ågot Ligaarden, Aud Elin Rivedal and Harald Volden • Communication: Marie Liljeholm, Maria Mehlqvist and Arnt Johan Rygh • National advisory tools: Åse Marit Flittie Anderssen, Henrik Martinussen, Patrik Nordgren, Berglind Ósk Óðinsdóttir and Maria Åkerlind. We would also like to thank all scientists who have contributed to fruitful discussions and made valuable comments on the model and documents. The present documentation was reviewed by Torben Hvelplund, University of Aarhus, Denmark, and Peter Udén, Swedish University of Agricultural Sciences, Sweden and they are greatly acknowledged for their work. Responsibility for the final content of the model and this present EAAP publication of the NorFor systems rests entirely with the authoring group and NorFor F.M.B.A.

NorFor – The Nordic feed evaluation system

11

Glossary Abbreviation AA AA-N AAj AAj_AATN AATN AATN_bal AATN_dep AATN_Eff AATN_gain AATN_gest AATN_maint AATN_milk AATN_mob AATN_NEG AATN_NEG_Min AATN_NEL AATN20 AATN8 ACF ADG ADG_calc Age Age_end Age_start ALF APL Avail_AAT b_car BCS BCS_change BCS_kg BUF BW BW_birth BW_calc BW_end BW_mat BW_start C12:0 C14:0 C16:0 C18:0 C18:1 C18:2 C18:3 C20:5

Unit g N/100 g N g/d % g/d % g/d g/d g/d g/d g/d g/MJ g/MJ g/MJ g/kg DM g/kg DM g/kg DM g/d g/d days days days g/kg DM g/d mg/kg DM BCS/d kg/BCS g/kg DM kg kg kg kg kg kg g/100 g FA g/100 g FA g/100 g FA g/100 g FA g/100 g FA g/100 g FA g/100 g FA g/100 g FA

Parameter name Amino acids Amino acid nitrogen in feedstuff Individual amino acid Individual amino acid absorbed in small intestine Amino acids absorbed in the small intestine AAT balance AAT for deposition in cows AAT efficiency of utilisation AAT requirement for weight gain in primiparous cows and growing cattle AAT requirement for gestation in cows and heifers AAT requirement for maintenance AAT requirement for milk production AAT for mobilisation in cows Available AAT to energy ratio in growing cattle AAT/NEG minimum recommendation Available AAT to energy ratio in cows Standard feed value for AAT at 20 kg DMI Standard feed value for AAT at 8 kg DMI Acetic acid in feedstuff Average daily body weight gain Estimated daily weight gain Current age Age at calving or sale Age at the start of a feeding period Alcohol in feedstuff Animal production level Available AAT for milk production Beta-carotene in feedstuff Body condition score, from 1 to 5 Daily change in body condition score Weight per unit body condition score Butyric acid in feedstuff Current body weight Body weight at birth Calculated body weight at a given age Body weight at calving or sale Body weight for a mature animal Body weight at the start of a feeding period Lauric acid in feedstuff Myristic acid in feedstuff Palmitic acid in feedstuff Stearic acid in feedstuff Oleic acid in feedstuff Linoleic acid in feedstuff Linolenic acid in feedstuff EPA, Eicosapentaenoic acid in feedstuff

NorFor – The Nordic feed evaluation system

13

Abbreviation

Unit

Parameter name

C22:6 Ca Ca_gain

g/100 g FA g/kg DM g/d

Ca_gest Ca_intake_Min Ca_main Ca_milk CAD CFat CFatD CHO CHOD CI Cl Cl_gain

g/d g/d g/d g/d mEq/kg DM g/kg DM %

Cl_gest Cl_intake_Min Cl_main Cl_milk Co conc_share corrNDF_fac CP CP_intake CPcorr CPD CTo Cu DIM DH DM DM1 DM2 DMcorr DMuncorr DMI DMIc DMIstd

g/d g/d g/d g/d mg/kg DM % of DM

EB EBW EBWG ECM ECM_response ECMherd e_Comp eCP ED

% kg g/d kg/d kg/d kg/d g/100g g/d

DHA, Docosahexaenoic acid in feedstuff Calcium in feedstuff Calcium requirement for weight gain in growing cattle and primiparous cows Calcium requirement for gestation Calcium minimum recommendation Calcium requirement for maintenance Calcium requirement for milk production Cation anion difference in feedstuff Crude fat in feedstuff Apparent total digestibility of crude fat Carbohydrates Apparent total digestibility of carbohydrates Chewing index of feedstuff Chloride in feedstuff Chloride requirement for weight gain in growing cattle and primiparous cows Chloride requirement for gestation Chloride minimum recommendation Chloride requirement for maintenance Chloride requirement for milk production Cobalt in feedstuff Concentrate share in ration Correction factor for kdNDF Crude protein in feedstuff Daily intake of crude protein Crude protein corrected in feedstuff Apparent total digestibility of crude protein Observed chewing time Copper in feedstuff Days in milk Danish Holstein Dry matter in feedstuff Dry matter, first step Dry matter, second step Dry matter, corrected for volatile losses Dry matter, not corrected for volatile losses Dry matter intake Dry matter intake of concentrate Dry matter intake when calculating standard feed values Energy balance Empty body weight Empty body weight gain Energy corrected milk Predicted ECM response Average daily ECM yield per cow in the herd Amino acid composition in endogenous amino acids Endogenous crude protein Efficient degradability

14

% min/kg DM g/kg DM g/d

g/kg DM g/d g/kg DM % min/kg NDF mg/kg DM days g/kg g/kg g/kg DM1 g/kg g/kg kg DM/d kg DM/d 8 or 20 kg DMI

NorFor – The Nordic feed evaluation system

Abbreviation

Unit

Parameter name

EFOS

% of OM

EI Ep erd erd_CP erd_NDF erd_RestCHO erd_ST ETo f_milk FA FA6mm from pool 1 to 2 Rumen passage rate of NDF particles>6mm from pool 2 Rumen passage rate of protein and starch particles6mm Microbial synthesis of amino acids in the rumen Microbial crude fat synthesis in the rumen Microbial protein synthesis in the rumen Microbial synthesis of organic matter in the rumen Microbial starch synthesis in the rumen Passage of organic matter to small intestine Red Danish Crude fat degraded in the rumen Crude protein degraded in the rumen Corrected crude protein degraded in the rumen Fermentation products from feed degraded in the rumen NDF degraded in the rumen NH3-crude protein degraded in the rumen Rumen degraded organic matter

18

g/kg rd OM

g/d g/d g/d g/d

NorFor – The Nordic feed evaluation system

Abbreviation

Unit

Parameter name

rd_pdCP

g/d

rd_pdNDFc

g/d

rd_pdNDFr

g/d

rd_pdST rd_RestCHO rd_sCP rd_sST rd_ST RestCHO RestCHOcorr RFA RI RLI

g/d g/d g/d g/d g/d g/kg DM g/kg DM g/kg CFat min/kg DM g/g NDF

Rough_share roughage_appetite RTo RUP S s+pdCP sCP Se si_outOM sid_AA

% of DM % min/kg NDF

Potentially degradable crude protein degraded in the rumen Potentially degradable NDF in concentrate degraded in the rumen Potentially degradable NDF in roughage degraded in the rumen Potentially degradable starch degraded in the rumen Rest CHO fraction degraded in the rumen Soluble crude protein degraded in the rumen Soluble starch degraded in the rumen Total starch degraded in the rumen Rest fraction in feedstuff Rest fraction corrected in feedstuff Residual fatty acids Rumination index Rumen load index, rumen degraded starch and rest fraction per unit of NDF Roughage percentage in the diet Roughage appetite proportion Observed ruminating time Rumen undegradable protein Sulphur in feedstuff Sum of sCP and pdCP in feedstuff Soluble crude protein in feedstuff Selenium in feedstuff Passage of organic matter to the large intestine Amino acids from feed digested in the small intestine Individual amino acids from feed digested in the small intestine Crude fat from feed digested in the small intestine Crude protein from feed digested in the small intestine Endogenous amino acids digested in the small intestine Microbial amino acids digested in the small intestine Microbial crude protein digested in the small intestine Microbial fatty acids digested in the small intestine Microbial starch digested in the small intestine Starch from feed digested in the small intestine Swedish Holstein Swedish Red Soluble starch in feedstuff Starch in feedstuff Content of starch and sugar in the diet Total intake of starch and sugar Apparent total digestibility of starch Sugar in feedstuff

g/kg DM g/kg CP g/kg CP mg/kg DM g/d g/d

sid_AAj sid_CFat sid_CP

g/d g/d

sid_eAA

g/d

sid_mAA sid_mCP

g/d g/d

sid_mFA sid_mST sid_ST SH SR sST ST ST_SU_DM ST_SU_intake STD SU

g/d g/d g/d g/kg ST g/kg DM g/kg DM g/d % g/kg DM

NorFor – The Nordic feed evaluation system

19

Abbreviation

Unit

Parameter name

SubR TAF TCL td_CFat td_CHO td_CP td_CPcorr td_OM TMR uNDF uNDS uOM Urea N VFA VitA VitD VitE VOS

none g/kg DM mm g/d g/d g/d g/d g/d g/kg DM g/kg DM g/kg DM g/g

YHerd Zn WPC

kg ECM/year mg/kg DM weeks

Substitution rate Total acids in feedstuff Theoretical cutting length Total apparently digested crude fat Total apparently digested carbohydrates Total apparently digested crude protein Total apparently digested crude protein corrected Total apparently digested organic matter Total mixed ration Undegraded NDF Undegraded neutral detergent solids Undegraded organic matter Urea nitrogen Volatile fatty acids Vitamin A in feedstuff Vitamin D in feedstuff Vitamin E in feedstuff Organic matter digestibility of feedstuff in vitro (VOS method) Yield level in the herd Zinc in feedstuff Weeks post calving

20

1000 IU/kg DM 1000 IU/kg DM IU/kg DM % of OM

NorFor – The Nordic feed evaluation system

1. Introduction H. Volden and A.H. Gustafsson Feed is one of the major expenses in modern cattle production. In addition to feed prices, its overall costs are affected by the efficiency of feed utilization and the output of animal products to be marketed. Hence, there is a clear need to evaluate feed quality in order to maximise profitability. This requires information on both animal requirements and nutrient supply, since formulation of an appropriate ration involves balancing available feeds in proportions that match the amounts of nutrients supplied to the animals’ nutrient requirements as closely as possible. There are two principal methods used to describe animal nutrition: those based on mechanistic approaches, which describe responses to nutrients from chemical and physiological processes in the gastrointestinal tract and intermediary, and empirical approaches describing simple relationships between nutrient intakes and production responses. The challenge in the development of new feed evaluation systems is to accurately predict responses to nutrients so that any difference in product income and feed costs can be maximized, while improving overall feed efficiency. Feed efficiency is also of great importance due to its impact on enteric methane emission to the atmosphere and nitrogen and phosphorus passing into the environment. Feed evaluation for cattle has long traditions. Important milestones are introductions of the Weende analysis by Henneberg and Stohmann (1864), and the starch equivalent system by Kellner (1912). In the Nordic countries the introduction of the milk production value by Hansson (1913) and the Scandinavian feed unit by Møllgaard (1929) were of great magnitude for modern cattle production. Extensive research in the 1950’s and 60’s further improved our knowledge in feed evaluation, and in the period from 1970 to 1990 most countries introduced new systems for energy based either on metabolizable energy (ME) or net energy for lactation (NEL) or growth (NEG) (Van der Honing and Alderman, 1988). Also the knowledge of protein evaluation for cattle have increased considerably during the last 40 years, from use of total or digestible crude protein to the protein evaluation systems that predict the host animal amino acid (AA) supply from dietary protein escaped ruminal degradation and from microbial protein synthesized in the rumen (Madsen, 1985; NRC, 1985; Vérité et al., 1987; AFRC, 1992; Tamminga et al., 1994). Traditional feed evaluation systems are additive and generally do not take into account interactions in digestion and nutrient metabolism. When interactions and non-linear relationships are considered, in attempts to describe feed metabolism, individual feeds will no longer have fixed values. Hence, the value of a given feedstuff will depend on how the feed is used. This means that the feed value cannot be determined until we know which feed ration will be used and the production situation in which it will be applied. Therefore, the development of a new feed evaluation system must consist of a ration evaluation system which can be used to optimize nutrient supply and production responses, rather than a system that can predict individual feed values. Development of new feed evaluation systems (Russell et al., 1992; Sniffen et al., 1992; Fox et al., 1992; NRC, 2001) and mechanistic models (Baldwin, 1995; Danfær et al., 2006), which describe nutrient supply and the nutritional requirements of cattle, are important for understanding ruminant nutrition and constitute an important step towards implementing more sophisticated nutritional strategies to optimize production responses in cattle. The use of non-linear semi-mechanistic feed evaluation systems for ration optimization requires powerful computing tools. Recent improvements in non-linear optimization tools and algorithms enable the development of more complex feed evaluation systems that can also be used in practice. The aim of this book is to provide a detailed description of the semi-mechanistic feed evaluation system NorFor, which has already been implemented by advisory services in four Nordic countries, i.e. Denmark, Iceland, Norway and Sweden.

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_1, © Wageningen Academic Publishers 2011

21

References AFRC (Agricultural and Food Research Council), 1992. Nutritive requirements of ruminant animals: protein. AFRC Technical Committee on Response to Nutrients. Report No. 9. Nutrition Abstracts and Reviews: Series B 62: 787-835. Baldwin, R.L., 1995. Modelling Ruminant Digestion and Metabolism. Chapman and Hall, London, UK. 592 pp. Danfær, A., P. Huhtanen, P. Udén, J. Sveinbjörnsson and H. Volden, 2006. The Nordic dairy cow model, Karoline – Description. In: Kebreab, E., J. Dikstra, A. Bannink, J. Gerrits and J. France (eds.). Nutrient Digestion and Utilisation in Farm Animals: Modelling Approaches. CAB International, Wallingford, UK, pp. 383-406. Fox, D.G., C.J. Sniffen, J.D. O’Connor, J.B. Russell and P.J. Van Soest, 1992. A net carbohydrate and protein system for evaluating cattle diets. III. Cattle requirements and diet adequacy. Journal of Animal Science 70: 3578-3596. Hansson, N., 1913. En ny metod för beräkning av fodermedlens produktionsvärde vid utfodring af mjölkkor. Meddelande Nr. 85 från centralanstalten för försöksväsendet på jordbruksområdet. Ivar Hæggströms Boktryckeri Aktiebolag, Stockholm, Sweden. (In Swedish). Henneberg, W. and F. Stohmann, 1864. Begründung einer rationellen fütterung der wiederkauer. Schwetske and Söhne, Braunschweig, Germany. (In German). Kellner, O., 1912. Die Ernährung der Landwirtschaftlichen nutztiere. Raul Parey, Berlin, Germany. (In German). Madsen, J., 1985. The basis for the proposed Nordic protein evaluation system for ruminants. The AAT-PBV system. Acta Agriculturae Scandinavica Supplement 25: 9-20. Møllgaard, H., 1929. Fütterungslehre des milchviehs. Verlag von M. and H. Schalper, Hannover, Germany. (In German). NRC (National Research Council), 1985. Ruminal nitrogen usage. National Academy Press, Washington, DC, USA. 138 pp. NRC, 2001. Nutrient requirements of dairy cattle. Seventh revised edition, National Academy Press, Washington, DC, USA. 408 pp. Russell, J.B., J.D. O’Connor, D.G. Fox, P.J. Van Soest and C.J. Sniffen, 1992. A net carbohydrate and protein system for evaluating cattle diets. I. Ruminal fermentation. Journal of Animal Science 70: 3551-3561. Sniffen, C.J., J.D. O’Connor, P.J. Van Soest, D.G. Fox and J.B. Russell, 1992. A net carbohydrate and protein system for evaluating cattle diets. II. Carbohydrate and protein availability. Journal of Animal Science 70: 3562-3577. Tamminga, S., W.M. Van Straalen, A.P.J. Subnel, R.G.M. Meijer, A. Steg, C.J.G. Wever and M.C. Block, 1994. The Dutch protein evaluation system: the DVE/OEB-system. Livestock Production Science 40: 139-155. Van der Honing, Y. and G. Alderman, 1988. Feed evaluation and nutritional requirements. III. Ruminants. Livestock Production Science 19: 217-278. Vérité, R., P. Chapoutot, B. Michalet-Doreau, J.L. Peyraud and C. Poncet, 1987. Révision du système des protéines digestibles dans l’intestin (PDI). Bulletin Technique C.R.Z.V. Thei1, INRA 70: 19-34.

22

NorFor – The Nordic feed evaluation system

2. Overall model description H. Volden The NorFor system is a semi-mechanistic, static and science-based model, which predicts nutrient supply and requirements for maintenance, milk production, growth and pregnancy in cattle. The model can be divided into five parts: (1) an input section describing characteristics of the animal and feeds available; (2) a module simulating processes in the digestive tract and the intermediary metabolism, termed the feed ration calculator (FRC); (3) a module predicting feed intake; (4) a module predicting the physical structure of the diet; and (5) an output section describing nutrient supply, nutrient balances and production responses (Figure 2.1). The input variables for the model are animal and feed characteristics. For dairy cows, the main input variables are body weight (BW), stage of lactation, pregnancy day and planned or potential daily milk production. For growing animals (bulls, steers and heifers) input variables are BW and average daily weight gain (ADG). The feed dry matter (DM) is separated into ash, crude protein (CP), crude fat (CFat), neutral detergent fibre (NDF), starch (ST), sugar (SU), fermentations products (FPF) such as organic acids and alcohols, and a residual fraction (RestCHO). The CP is divided into soluble (sCP), potentially degradable (pdCP), indigestible (iCP) and ammonia (NH3N). The NDF is divided into a total indigestible (iNDF) and a potentially degradable (pdNDF) fraction. The ST is divided into soluble (sST), potentially degradable (pdST) and indigestible (iST) fractions. The FPF are separated into lactic acid (LAF), volatile fatty acids (VFA) and alcohols. Fractional degradation rates (kd) of the soluble and potentially degradable feed fractions are also required for the model. The FRC consists of four sections: (1) the rumen, (2) the small intestine, (3) the large intestine and (4) metabolism (Figure 2.2). Feed organic matter (OM) entering the rumen is either fermented and used for microbial production, or it escapes from the rumen for further digestion in the lower digestive tract. Ruminal degradation of CP, ST and NDF in concentrate feeds are assumed to follow first-order single-compartment kinetics, while degradation of NDF in roughage is modelled as a Input Feed characteristics (chemical composition and particle length) Animal characteristics (body weight, breed, stage of lactation)

Physical structure (chewing time)

Gastro-intestinal tract and intermediary metabolism Nutrient calculator Nutrient digestion and metabolism

Feed intake

Output Nutrient supply Ration energy and protein value. Predicted milk yield, protein production, Nutrient balances in the rumen, etc.

Figure 2.1. Overview of the NorFor model. H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_2, © Wageningen Academic Publishers 2011

23

FEED

24

RUMEN

SMALL INTESTINE

HIND GUT

iNDF

pdNDF

pdNDF

pdNDF

F ib r e

iNDF

iST

iST

iST

pdST

pdST

pdST

St a r ch

sST

sST

Rest CHO

Rest CHO

R est FPF

FA

FA

iCP

Glycerol , galactose

CFat

Lipids iCP

sCP

sCP

mCFat

mCP

Microbial growth

Energy and nitrogen for microbial growth

pdCP

pdCP

P r ot ein

Energy and nitrogen for microbial growth

mST

mST

mCFat

mCP

mCP

Microbial growth

NH 3

eCP

eCP

PBV N

Net energy

Energy Energy

Energy

of CP_intake

UREA 4.6%

Milk or gain

AA in milk or gain

AAT N to milk or gain Utilization

AAT N

dep. / mob.

AAT N for hair and skin

Endogenous N

Endogenous urinary-N

INTERMEDIARY METABOLISM

Figure 2.2. Flow diagram of the gastrointestinal tract and intermediary compartments of the NorFor feed evaluation system. For all abbreviations check the abbreviation list.

FAECES

NorFor – The Nordic feed evaluation system

two-compartment system, with a non-escapable and an escapable pool. The nutrients available for microbial growth come from ruminally degraded NDF, ST, RestCHO, glycerol, CP and LAF. The efficiency of microbial synthesis depends on the level of feed intake and diet composition. The input to the small intestine consists of OM from microbes, unfermented feed fractions escaping from the rumen and endogenous secretions. These components are partly digested in the small intestine and are either metabolised or enter the large intestine. The OM passing into the large intestine is subjected to microbial fermentation, and the digested OM not used for microbial synthesis is absorbed and metabolised. Faecal excretion consists of OM from microbes synthesised in the large intestine, feed that has escaped previous digestion and undigested rumen microbial material. The intermediary metabolism section yields ME calculated from total tract digestible OM. Net energies for maintenance, lactation, growth and pregnancy are predicted from the ME. Different coefficients are used to calculate NE for maintenance, lactation, and growth. Net energy for lactation is used for dairy cows, while NEG is used as the energy measurement for growing cattle. The nitrogen (N) fractions entering the intermediary metabolism consist of NH3N absorbed from the rumen, dietary, microbial and endogenous amino acids (AA) absorbed from the small intestine, and NH3N from the large intestine. The absorbed AAs are utilized for maintenance, growth, pregnancy and milk production. The efficiency of AA utilization is specific for each production/process. The metabolizable protein available for animal production is assigned as amino acids absorbed from the small intestine (AATN). The N which is not used for maintenance or production is excreted in the urine. Predicting nutritive value is only one part of ration formulation as formulation involves both the selection of feed ingredients and the prediction of feed intake. Therefore, the NorFor system contains a module to predict the intake of feeds. For prediction of feed intake dietary fill values (FV) and animal intake capacity (IC) are applied. In roughages, FV is calculated from OM digestibility (OMD) and NDF content, and in ensiled forages the basic FV is also corrected for content of VFA, LAF and NH3N. Animal IC is dependent on BW, milk yield, stage of lactation, lactation number, ADG and physical activity. The model uses a combination of dietary physical effects and metabolic factors to impact feed intake, and the effect of easily fermentable carbohydrates on roughage intake is accounted for by using a substitution rate factor (SubR). A minimum amount of large particles is essential for optimal rumen function. Hence, a module to evaluate the physical structure of the diet is included in NorFor. The dietary physical effect is described by a total chewing index (CI), which is calculated as the sum of an eating (EI) and ruminating (RI) index for each individual feed. The EI value reflects the associated chewing activity as feed is consumed and is calculated from the particle length and NDF content of the feed. The RI value is calculated from particle length, NDF content and a hardness factor, which is dependent on the iNDF content of the feed. The hardness factor reflects the lignification of the structural fibre of the feed and the associated physical force required for the comminution of large particles. The output from a model calculation in NorFor describes the intake of the individual feeds in the total ration. It consists of variables describing efficiencies of digestion and nutrient utilization, production (milk, ADG), N excretion, nutrient balances, energy (NEL or NEG) and protein (AATN) values of the ration.

NorFor – The Nordic feed evaluation system

25

3. Animal input characteristics. M. Åkerlind, N.I. Nielsen and H. Volden Animal characteristics are needed for calculation of nutrient requirements and IC for both cows and growing cattle. Information on animal parameters, such as BW, is also needed when calculating ruminal variables, such as passage rates of feed fractions out of the rumen and efficiency of microbial protein synthesis. Moreover, the physical activity of the animal (simply classified according to whether it is loose or tied up) will affect the IC and energy requirements.

3.1 Input data for cows All required input data for cows are listed in Table 3.1 and default values for different breeds are compiled in Table 3.2. The information needed for cows in the system is breed, lactation number, BW, stage of lactation (days in milk) and milk production. Pregnancy and whether the cow is dry or lactating are also factors required for determining IC and requirements for gestation. When calculating energy and protein requirements for gestation, information on mature body weight (BW_mat) is needed since this variable is breed-specific (see Table 3.2 and Sections 9.1.3 and 9.2.5). BW_mat is also needed to determine protein requirements for growth in primiparous cows (Section 9.2.3). Animal weight changes can be estimated as changes in body condition score (BCS), where the BW per condition scores depends on the breed. The BCS in NorFor uses a scale from 1 to 5, where 1 is thin and 5 is fat (Gillund et al., 1999). One unit of BCS corresponds to 60 kg BW except for Jersey which is set to 45 kg BW; these values are based on data from several previous studies (e.g. Enevoldsen and Kristensen,1997; Gillund et al., 1999; Nielsen et al., 2003; Bossen, 2008). Table 3.1 Input data for cows in NorFor. Input data

Unit

Dry cow Lactation number Days in milk Daily milk yield Protein content in milk Fat content in milk Lactose content in milk Yield level of the herd Body weight, current Body weight, mature2 Weight gain in primiparous cows Days of gestation Daily change in body condition score Weight per unit body condition score2 Activity Breed3

none1 No. days kg/day g/kg g/kg g/kg kg ECM kg kg kg/day days BCS/day kg/BCS none1 none1

1

No unit. Default values can be taken from Table 3.2. 3 Abbreviations for breeds are explained in Table 3.2. 2

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_3, © Wageningen Academic Publishers 2011

27

429 428 430 429 431 430 432 431 433 432 434 433 435 434 435 436 436 437

compiled in Table 3.2. The information needed for cows in default the system is breed, lactation number, All required input data for cows are listed in Table 3.1 and values for different breeds are BW, stageinofTable lactation in milk) andneeded milk production. whether the cow is dry compiled 3.2. (days The information for cows inPregnancy the systemand is breed, lactation number, or lactating arelactation also factors IC andPregnancy requirements gestation. When BW, stage of (daysrequired in milk)for anddetermining milk production. andfor whether the cow is dry calculating protein requirements for gestation, information on mature body weight or lactatingenergy are alsoand factors required for determining IC and requirements for gestation. When (BW_mat) needed since this requirements variable is breed-specific Table 3.2on andmature Sections 9.1.3 and calculating isenergy and protein for gestation,(see information body weight 9.2.5). BW_mat is also needed to determine protein requirements for3.2 growth in primiparous cows (BW_mat) is needed since this variable is breed-specific (see Table and Sections 9.1.3 and Table 3.2. Default values for mature body weight (BW_mat) and the weights corresponding to a (Section 9.2.3). is also needed to determine protein requirements for growth in primiparous cows 9.2.5). BW_mat body condition score unit (BCS_kg) for dairy cows of different breeds. (Section 9.2.3). Table Abbr.3.1 3.1 Table 3.2

Dairy breed

BW_mat kg

BCS_kg kg/BCS

437 438 439 438 439 440 441 440 442 441 443 442 444 443 445 444 445 446

DH 3.2 Danish Holstein 640 60 Table IB Icelandic breed 470 60 where the Animal weight changes can be estimated as changes in body condition score (BCS), JER Jersey 440 45 BW per weight condition scorescan depends on the breed. The BCS in NorFor usesscore a scale fromwhere 1 to 5,the Animal changes be estimated as changes in body condition (BCS), NR Norwegian Red 600 where 1 is thin and 5 is fat (Gillund et al., 1999). One unit of BCS corresponds to 60 kg BW BW per condition scores depends on the breed. The BCS in NorFor uses a scale60 from 1 to 5, RD Danish Red 660 60 except for Jersey which is set to 45 kg BW; these values are based on data from several previous where 1 is thin and 5 is fat (Gillund et al., 1999). One unit of BCS corresponds to 60 kg BW studies (e.g.Jersey Enevoldsen and Kristensen,1997; Gillund et are al., based 1999; on Nielsen et al., 2003; previous Bossen, SH for Swedish 640 60 except which is setHolstein to 45 kg BW; these values data from several 2008). SR Swedish Red 620 60 studies (e.g. Enevoldsen and Kristensen,1997; Gillund et al., 1999; Nielsen et al., 2003; Bossen, 2008).

446 447 448 447 449 448 449 450

The energy requirement for milk production is based on the production of energy corrected milk The energy requirement for milk production is equations based on (Sjaunja the production1990) of energy corrected (ECM), which can be calculated either of depending on milk The energy requirement for milkfrom production is two based on the productionetofal., energy corrected milk (ECM), which can be calculated from either of two equations (Sjaunja et al., 1990) depending on whether information milk lactose is available: (ECM), which can beon calculated fromcontent either of two equations (Sjaunja et al., 1990) depending on whether information on milk lactose content is available: whether information on fmilk lactose content _ milk p _ milkis available: l _ milk ⎞ ⎛ [Eq. 3.1]

450 451 451 452 453 452 454 453 454

ECM = MY ⋅ ⎜ 0.01 + 0 .122 ⋅ + 0 .077 ⋅ + 0.053 ⋅ ⎟ ⎝⎛ f _10 milk p _10 milk l _10 milk ⎠⎞ ECM = MY ⋅ ⎜ 0.01 + 0 .122 ⋅ + 0 .077 ⋅ + 0.053 ⋅ ⎟ 10 10 10 ⎠ ⎝

f _ milk p _ milk ⎞ ⎛ ECM = MY ⋅ ⎜ 0 .25 + 0 .122 ⋅ + 0 .077 ⋅ ⎟ ⎝⎛ f _10 milk p _10 milk ⎠⎞ ECM = MY ⋅ ⎜ 0 .25 + 0 .122 ⋅ + 0 .077 ⋅ ⎟ 10 ⎠ where ECM⎝ is the energy corrected milk,10 kg/day;

[Eq. 3.1]

3.1

[Eq. 3.2] [Eq. 3.2]

3.2

MY is the daily milk yield, kg/day; and f_milk,

where the energy corrected kg/day; MYand is the daily milk yield, in kg/day; f_milk, g/kg. p_milkECM and isl_milk are the contentsmilk, of fat, protein anhydrous lactose milk, and respectively, p_milk and l_milk are the corrected contents ofmilk, fat, protein in milk, respectively, where ECM is the energy kg/day; and MYanhydrous is the dailylactose milk yield, kg/day; and f_milk, g/kg. p_milk and l_milk are therations contents fat, protein and the anhydrous lactose in milk, respectively, When formulating feed forofgroups of cows, daily ECM yield (ECMherd) can be estimated g/kg. from breed-specific lactation curves: 23 ECMherd = a + b ∙ YHerd – c ∙ DIM + ln(DIM) ∙ d

23

3.3

where ECMherd is the daily estimated ECM yield, kg/d; YHerd is the herd’s average ECM yield per cow, kg/305 d; DIM is days in milk; a, b, c and d are regression coefficients presented in Table 3.3 for primiparous and multiparous cows of different breeds. Standard lactation curves are available for different dairy breeds and lactation numbers. The ECMHerd value refers to the herd’s 305-d lactation yield, which is based on national herd recording data. Danish cow recording data (Danish Cattle Association) have been used to parameterize the standard lactation curves for the dairy breeds Jersey, Danish Holstein and Danish Red. Icelandic, Norwegian and Swedish national herd recording data are the basis for the standard lactation curves for the Icelandic breed, Norwegian Red, Swedish Holstein and Swedish Red cattle, respectively (Farmers Association of Iceland; Tine Dairies in Norway; Swedish Dairy Association). Daily milk protein yield is calculated from milk yield and milk protein content according to Equation 3.4. Protein production is essential for calculating the AA requirement for milk production: MPY = MY ∙ p_milk

3.4

where MPY is the milk protein yield, g/day; MY is the milk yield, kg/day; and p_milk is the milk protein content, g/kg.

28

NorFor – The Nordic feed evaluation system

Table 3.3. The multiple regression coefficients a, b, c and d is used to predict daily ECM yield from standardised lactations curve in Equation 3.3. Breed1

Lactation

a

b

c

d

DH DH IB IB JER JER NR NR RD RD SH SH SR SR

primiparous multiparous primiparous multiparous primiparous multiparous primiparous multiparous primiparous multiparous primiparous multiparous primiparous multiparous

-3.10 3.17 0.88 6.34 -8.15 -2.98 -7.09 -0.59 -6.20 1.45 -4.05 2.93 -4.46 -0.21

0.00325 0.00338 0.00288 0.00269 0.00324 0.00335 0.00314 0.00310 0.00335 0.00330 0.00299 0.00299 0.00304 0.00317

0.01685 0.06040 0.04009 0.06234 0.02810 0.05630 0.06440 0.09100 0.02138 0.06973 0.03560 0.06100 0.03970 0.06920

1.140 1.025 1.627 1.569 2.654 2.305 3.741 3.378 1.750 1.844 2.591 1.997 2.677 2.520

1 The

abbreviations of the different breeds are shown in Table 3.2.

3.2 Input data for growing cattle When calculating nutrient requirements and the IC for growing cattle, information on BW, ADG, sex, breed and activity are needed, and for heifers the days of gestation are also required (Table 3.4). Table 3.4. Input data for growing cattle. Parameter

Unit

Sex Breed2 Activity Body weight3 Average daily gain3 Days of gestation (heifers) Body weight, birth4 Body weight, start4 Body weight, end4 Body weight, mature4 Age, start4 Age, current4 Age, end4

none1 none1 none1 kg g/day days kg kg kg kg days days days

1

No unit. classifications of different breeds are shown in Table 3.5. 3 Body weight and daily gain are required parameters and can be estimated from the parameters that are marked by 4. 4 These parameters are required if current body weight and average daily gain data are not available. 2 The

NorFor – The Nordic feed evaluation system

29

485 486

sex, breed and activity are needed, and for heifers the days of gestation are also required (Table 3.4).

487 488 489

Table 3.4

490

3.2.1 Estimation of body weight and daily weight gain from a growth function

491 492 493 494 497 495 498 496 499 497 500 498 501 499 502 500 503 501 504 502 505 503 506 504 507 505 508 506 509 510 507 508 509 510 511

Whenplanning planningfeed feedrations rations animals of different ages, of BW_mat, ageBW andatBW When forfor animals of different ages, datadata of BW_mat, age and the at the start of the period andand planned ageage and BW at atthetheend start ofrearing the rearing period planned and BW endofofthe therearing rearingperiod period are are needed. needed. Expected Expected BW and can be estimated from the following logistic growth equation based BW and ADG canADG be estimated from the following logistic growth equation based on: on:

512 511 512 513 514 513 515 516 514 517 515 518 516 519 517 520 518 521 519 522 520 523 521 524 522 525 523 526 524 527 525 528 526 529 527 530 528 531 529 532 530 533 531 534 532 533 535 534 535

3.2.1 Estimation of body weight and daily weight gain from a growth function

(Age−for )))) age, kg; BW_start is the BW[Eq. ( A⋅(1−e( −B⋅BW Agethe _ start where BW_calc the⋅ eestimated current at the start, kg; 3.5] BW _ calc = BW _is start 3.5 A is described in Equation 3.6; B is described in Equation 3.8; Age is the current age, days; Age_start is the age at start, days. IfBW Age_start iscurrent 0 the BW_start the same is asthe BW_birth. whereBW_calc BW_calc theestimated estimated kg; isBW_start the kg; start, kg; A where isisthe BW forfor thethe current age,age, kg; BW_start is the BW atBW the at start,

is is described 3.7; in1)))) Equation3.8; 3.8;Age Age the current age, days; Age_start A describedininEquation Equation 3.6; B B is described described in Equation isisthe current age, days; ⋅(1−e(−B⋅(Age−Age _ start+24 A⋅(1−e(−B⋅(Age−Age _ start)))) ⋅1000 ADG calcis =atthe BW _ start e (AIf − BW _ is start ⋅ e (is Age_start age at ⋅start, days. If Age_start 0 the BW_start the same as BW_birth. is the_ age start, days. Age_start is 0 theisBW_start the same as BW_birth. [Eq. 3.6]

) ( ADG _ calc = (BW _ start ⋅ e (A⋅(1−e(−B⋅(Age−Age _ start+1)))) − BW _ start ⋅ e (A⋅(1−e(−B⋅(Age−Age _ start)))) )⋅1000 where ADG_calc is the estimated average daily gain for the current age, g/day; BW_start is the

3.6

[Eq. 3.6] BW at the start, kg; A is described in equation 3.7; B is described in equation 3.8; Age is the where ADG_calc is the estimated average daily gain for the current age, g/day; BW_start is the BW current age, days; Age_start is the age at start, days. If Age_start is 0 the BW_start is the same as where ADG_calc is is thedescribed estimatedin average daily3.7; gainBfor the current in age, g/day; BW_start at the start, kg; A Equation is described Equation 3.8; Ageisisthe the current BW_birth. BW at the start, kg; A is described in equation 3.7; B is described in equation 3.8; Age is the

age, days; Age_start is the age at start, days. If Age_start is 0 the BW_start is the same as BW_birth.

current age, days; Age_start is the age at start, days. If Age_start is 0 the BW_start is the same as Factor A and B in Equations 3.5 and 3.6 are calculated as: BW_birth. Factor A and B in Equations 3.5 and 3.6 are calculated as: _ mat .1 ⎞ Factor⎛ABW and B in⋅ 1Equations 3.5 and 3.6 are calculated as:

A = ln⎜⎜ [Eq. 3.7] 3.7 ⎟⎟ ⎝ BW _ start ⎠ ⎛ BW _ mat ⋅ 1.1 ⎞ ⎟ A = ln⎜⎜ [Eq. 3.7] BW__start mat ⋅ 1⎟⎠.1 / BW _ start ) ⎞ ⎛⎝ ln(BW ⎟⎟ ln ⎜⎜ ln( BW _ mat ⋅ 1.1 / BW _ end ) ⎠ [Eq. 3.8] 3.8 B= ⎝ _ start ⎛ ln(Age BW __ end mat −⋅ 1Age .1 / BW _ start ) ⎞ ⎟⎟ ln ⎜⎜ ln(and BW _Bmat .1 / BW _used end ) ⎠in Equations 3.5 and 3.6; BW_mat is the mature body weight, kg; where A are⋅ 1factors [Eq. 3.8] B = ⎝A where and B are −factors used in Equations 3.5 and 3.6; BW_mat is the mature body weight, kg; end Ageweight _ start BW_startAge is _the body at start or at birth, kg; BW_end is the body weight at the end of the

BW_start is the body weight at start or at birth, kg; BW_end is the body weight at the end of the feedingperiod; period;Age_start Age_start at start, days; Age_end theatage theofend feeding is is thethe ageage at start, days; and and Age_end is theisage theat end the of the feeding where and BIfare factors used Equations 3.5 and 3.6; BW_mat the mature body weight, kg; period,Aperiod, days. Age_start is 0in days, the BW_start is the same asis BW_birth. feeding days. BW_start is the body weight at start or at birth, kg; BW_end is the body weight at the end of the

feeding period; Age_start is the age at start, days; and Age_end is the age at the end of the Anexample exampleofofthe thelogistic logistic growth function of ADG during the rearing is shown in An growth function of ADG and and BW BW during the rearing periodperiod is shown feeding period, days. in Figure3.1. 3.1. Figure An example of the logistic growth function of ADG and BW during the rearing period is shown Figure 3.1of the feeding period depends on whether the animal will be sold, slaughtered or used for The end in Figure 3.1.

replacement. Default values for birth weights (BW_birth) and BW_mat that can be used for estimating

The of the feeding period depends on whether the animal will be sold, slaughtered or used for BWend and Figure 3.1ADG for different breeds and gender are shown in Table 3.5. BW_birth values for beef replacement. Default values for birth weights (BW_birth) and BW_mat that can be used for breeds were collected from Danish, Norwegian and Swedish national recording data compiled by estimating BW and ADG for different breeds and gender are shown in table 3.5. BW_birth values The end of the feeding period depends on whether animal will Research be sold, slaughtered used for thebeef Danish Cattle Norwegian Meattheand Poultry andordata Cattle statistics for breeds wereAssociation, collected from Danish, Norwegian and Swedish nationalCentre recording replacement. Default values for birth weights (BW_birth) andDanish, BW_mat that can beand used for (2007), respectively. Values for BW_mat were taken from Norwegian Swedish slaughter compiled by the Danish Cattle Association, Norwegian Meat and Poultry Research Centre and estimating BW and ADG for different breeds and gender are supplied shown inby table 3.5. BW_birth data for animals over 4respectively. years of ageValues over the 10 years the Danish Cattlevalues Association; Cattle statistics (2007), forlast BW_mat were taken from Danish, Norwegian for beef breeds were collected from Danish, Norwegian and Swedish national recording data Norwegian and data Poultry Research Centre and Dairy respectively. and Swedish Meat slaughter for animals over 4 years of Swedish age over the lastAssociation, 10 years supplied by the compiled by the Danish Cattle Association, Norwegian Meat and Poultry Research Centre and Danish Cattle Association; Norwegian Meat and Poultry Research Centre and Swedish Dairy Cattle statistics (2007), respectively. Values for BW_mat were taken from Danish, Norwegian Association, respectively. and Swedish slaughter data for animals over 4 years of age over the last 10 years supplied by the Danish Cattle Association; Norwegian Meat and Poultry Research Centre and Swedish Dairy Association, Table 3.5 respectively. Table 3.5 25 25

30

NorFor – The Nordic feed evaluation system

ADG_calc, g/day BW_calc, kg

1000 900 800 700 600 500 400 300 200 100 0

ADG, 24mo ADG, 27mo BW, 24mo BW, 27mo

0

5

10 15 Age, months

20

25

30

Figure 3.1. A logistic growth function predicting BW (BW_calc) and estimated weight gain (ADG_ calc) during the rearing of a heifer which is scheduled to calve at a live weight of 560 kg at either 24 or 27 months of age. To achieve a final weight of 560 kg, the weight gain is faster for a heifer finished at 24 months of age (ADG, 24mo; BW, 24mo) than for a heifer finished at 27 months (ADG, 27mo; BW, 27mo). Table 3.5. Default values for birth weight (BW_birth) and mature body weight (BW_mat) for heifers, bulls and steers of different breeds. Breeds Early maturing dairy breeds Danish Holstein Danish Red Icelandic breed Jersey Norwegian Red Swedish Holstein Swedish Red Early maturing beef breeds Aberdeen Angus Dexter Galloway Hereford Highland cattle Tiroler Grauvieh Late maturing beef breeds Belgian Blue Blonde d’Aquitaine Charolais Chianina Limosin Piemontese Saler Simmenthal Early·late maturing breed Crossbred1 1

BW_birth BW_birth BW_mat BW_mat BW_mat for heifer kg for bulls kg for heifer kg for bulls kg for steers kg 40 40 33 28 39 39 39

41 41 33 30 41 41 41

640 660 470 440 600 640 620

950 950 800 650 950 950 950

750 750 700 550 750 750 750

36 21 34 40 29 39

38 24 35 42 30 42

700 340 550 700 500 700

950 450 850 950 700 950

750 400 750 750 600 750

44 44 46 50 41 41 39 44

47 47 49 55 43 43 41 46

850 850 750 850 750 750 750 750

1,200 1,200 1,200 1,200 1,200 1,200 1,200 1,200

1,050 1,050 1,050 1,050 1,050 1,050 1,050 1,050

42

44

750

1,050

950

Crossbreeds of early and a late maturing breeds are assigned specific factors in some of the equations in Chapter 9.

NorFor – The Nordic feed evaluation system

31

References Bossen, D., 2008. Feeding strategies for dairy cows. Individual feed allocation using automatic live weight registrations as management parameter. PhD thesis. University of Copenhagen, Faculty of Life Sciences, Denmark. 159 pp. Enevoldsen, C. and T. Kristensen, 1997. Estimation of body weight from body size measurements and body condition scores in dairy cows. Journal of Dairy Science 80: 1988-1995. Gillund, P., O. Reksen, K. Karlberg, Å.T. Randby, I. Engeland and B. Lutnaes, 1999. Testing of a body condition score method in Norwegian cattle. Norsk Veterinærtidsskrift 111: 623-632. (In Norwegian with English summary). Nielsen, H.M., N.C. Friggens, P. Løvendahl, J. Jensen and K.L. Ingvartsen, 2003. Influence of breed, parity, and stage of lactation on lactational performance and relationship between body fatness and live weight. Livestock Production Science 79: 119-133. Sjaunja, L.-O., L. Bævre, I. Junkkarainen, J. Pedersen and J. Setälä, 1990. A Nordic proposition for an energy corrected milk (ECM) formula. 26th session of the International Committee for recording the productivity of Milk Animals (ICRPMA).

32

NorFor – The Nordic feed evaluation system

4. Feed fraction characteristics H. Volden Absorbed nutrients are provided from fermentation and digestion in the gastrointestinal tract, and the predictions are sensitive to the nutrient profiles of the feed. NorFor has an extensive feed table (www.norfor.info) that lists chemical composition and digestion characteristics of typical Nordic feedstuffs. The feed table values are continuously updated as new information becomes available.

4.1 Definition of roughage and concentrates Roughage and concentrates are feedstuff classes that are generally used and several criteria characterize both groups of feedstuffs, e.g. fibre content, energy density, moisture content and particle length. In NorFor it is necessary to identify a feedstuff either as ‘roughage’ or ‘concentrate’. This is decided from information on particle length; feedstuffs with particle lengths greater or less than 6 mm are characterized as roughages and concentrates, respectively. The identification of feedstuff class is necessary when describing the feed FV and in the calculation of the CI. This information is also necessary to predict the fractional passage rate of a feed fraction out of the rumen. Feedstuffs are further classified as liquid, fine, coarse, rolled, chopped, and unchopped, since this is used for identification of most frequent particle length (PL) which affects the calculated CI values.

4.2 Division of organic matter The DM is separated into OM and ash. The OM is further divided into the following components: CP, NDF, ST, CFat, LAF, VFA, alcohols and a calculated RestCHO fraction including SU. The main chemical components are further divided into sub-fractions (Figure 4.1) that have uniform rates of kd in the rumen.

Soluble Potentially degradable Total indigestible

Protein NDF

‘Soluble’ Potentially degradable Total indigestible

Amino acids Potentially degradable Total indigestible

Starch

Fermentation pr. Sugar Residual CHO Crude fat

Fatty acids

Ash

Figure 4.1. The feed fractionation scheme in NorFor.

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_4, © Wageningen Academic Publishers 2011

33

4.3 Crude protein fractions and amino acids The composition and properties of dietary proteins strongly influence protein metabolism in the gastrointestinal tract of ruminants, hence information on these variables is essential for robust feed evaluation. Feed CP (N·6.25) consists primarily of AA and smaller amounts of non-amino nitrogen, which is composed of NH3N, urea and nucleic acids. In NorFor CP is partitioned into three fractions: (1) soluble (sCP), (2) potentially degradable in the rumen (pdCP) and (3) total tract-indigestible (iCP). As example, Table 4.1 presents average CP fractions for a limited number of common feedstuffs. The sCP fraction contains soluble proteins, peptides, free AA, and non-amino nitrogen. When calculating ruminal degradation of the sCP fraction instantaneous degradation of NH3N and urea is assumed and a fractional degradation rate of 150%/h is used for the soluble AA fractions (Russel et al., 1992; Volden et al., 1998, 2002). The pdCP fraction is assumed to be insoluble, but degradable by microbes in the rumen. Individual rates of degradation (kdCP) for the pdCP fraction are estimated from ruminal degradation profiles using the in sacco technique (see Section 5.2.2 for a description of the method), assuming that the degradation profile follows first-order kinetics. Intestinal digestibility of the undegraded rumen feed protein (RUP) may vary considerably between feeds (Hvelplund, 1985; Van Straalen and Tamminga, 1990; Volden and Harstad, 1995). The in sacco mobile bag technique (MBT) provides an easy and fast method to determine intestinal digestibility of protein (Hvelplund et al., 1992, 1994) (see Section 5.2.4 for a description of the method). In NorFor, the MBT technique is used to determine feed iCP. The iCP represents a protein fraction that is completely resistant to both microbial degradation in the rumen and digestion in the small intestine. However, ruminal pre-incubation has been shown to have important effects on the intestinal digestion of protein in several feeds (Volden and Harstad, 1995;Vanhatalo et al., 1996). Consequently, iCP is determined in feed residues after 16 and 24 h of ruminal incubation for concentrates and roughages, respectively. Volden and Harstad, 1995 observed that pre-incubation in the rumen increased intestinal digestion and thus reduced the iCP. This implies that the feed protein contains a fraction that is not degradable Table 4.1. Crude protein (CP) fractions in selected feeds and the corresponding fractional degradation rate of the potential degradable protein fraction. Feedstuff

Barley Wheat Maize Peas Rape seed Rape seed meal Soybean meal Maize silage, high digestibility Maize silage, low digestibility Grass silage, high digestibility Grass silage, low digestibility

NorFor feed code

CP g/kg DM

Protein fractions, g/kg CP sCP1

pdCP2

iCP3

001-0016 001-0020 001-0014 003-0006 002-0007 002-0042 002-0053 006-0307 006-0309 006-0461 006-0463

113 131 96 239 218 388 516 78 75 173 144

290 220 114 711 280 216 160 464 437 668 628

670 750 886 289 660 734 840 432 459 267 228

36 29 51 28 85 58 11 140 140 38 49

kdCP4 %/h 11.3 14.3 2.5 9.1 9.5 9.5 7.9 4.6 4.6 12.1 8.0

1

sCP=soluble crude protein fraction. pdCP=potentially degradable crude protein. 3 iCP=total indigestible crude protein. 4 kdCP=fractional degradation rate of pdCP. 2

34

NorFor – The Nordic feed evaluation system

in the rumen, but is digestible in the small intestine, explaining why the sum of sCP + pdCP + iCP in most feedstuffs is not 1000 g/kg CP. The remaining CP fraction is added to the protein fraction digested in the small intestine. The amount of RUP is one of the primary factors determining the amount of dietary AAs entering the small intestine. NorFor predicts individual AAs absorbed from the small intestine. Thus, information regarding the AA profile of the RUP is required. A variable fraction of the feed protein is degraded in the rumen and the AA profile of the RUP may differ from the AA profile of the intact feed protein. Several authors have compared the AA profiles of intact feed protein and in sacco residues after ruminal incubation (Skórko-Sajko et al., 1994; Skiba et al., 1996; Weisbjerg et al., 1996; Zebrowska et al., 1997; Prestløkken, 1999; Harstad and Prestløkken, 2000; Harstad and Prestløkken, 2001). Although a limited number of feeds were evaluated, the acquired data suggest that there are only minor differences between the AA profiles of intact feed protein and RUP. Therefore, the AA profile and the sum of AA nitrogen (AA-N) in RUP are assumed to be the same as those of the original feedstuff. The NorFor feedstuff table (www.norfor.info) has information on the AA profiles of individual feeds.

4.4 Carbohydrate fractions Carbohydrates (CHO) are heterogeneous feed constituents that collectively comprise the largest component of cattle diets, and make the major contribution to supporting ruminal microbial growth. CHO digestion in the gastrointestinal tract varies considerably and the end-products from ruminal fermentation and intestinal digestion are the major nutrient supply for animal production. Moreover, the rate of CHO degradation in the rumen is a major factor affecting ruminal pH and thus the rumen environment and feed utilization. A good balance between the different CHO fractions is essential to maintain normal rumen function and optimize rumen fermentation. Carbohydrates can be divided into two main fractions, structural and non-structural The structural CHO originates from plant cell wall material (consisting of cellulose, hemicellulose and lignin), which in NorFor is defined as NDF (Van Soest, 1994). The NDF fraction is further divided into a potentially degradable fraction (pdNDF) and a total indigestible fraction (iNDF). The iNDF is determined from ruminal in sacco incubation for 288 h (see Section 5.2.3 for a description of the method), and the pdNDF fraction is estimated as total NDF minus iNDF. In concentrates, the rate of pdNDF degradation (kdNDF) is estimated by in sacco degradation, assuming first-order degradation kinetics (see Section 5.2.2 for a description of the method). In roughage, kdNDF is predicted from the combination of in vivo OMD and iNDF and using the Lucas principle applied to the excretion of non-fibre matter (Weisbjerg et al., 2004a, b). A complete description of the method and calculations is presented in Section 6.1. Examples of NDF content and degradation characteristics in commonly used feedstuffs are presented in Table 4.2. The non-structural carbohydrates consist of carbohydrate-based cell contents, including ST, SU, β-glucans and some of the pectins. In NorFor ST and SU are determined analytically, while β-glucans and pectins are part of the RestCHO fraction. Starch is an important feed constituent for high-yielding cattle due to its high digestibility and importance as a source of substrates for propionic acid fermentation in the rumen. Inadequate ST intake may depress microbial activity and thus reduce microbial protein synthesis. However, excessive levels of ST may depress fibre digestibility and roughage utilization, as well as causing ruminal acidosis and abnormalities in the rumen tissue (Owens et al., 1998). Starch digestibility in the rumen depends on grain type, grain processing and particle length (Mills et al., 1999; Offner and Sauvant, 2004; Larsen et al., 2009). In NorFor, ST is partitioned into three fractions (Table 4.3): ‘soluble’ (sST), potentially degradable (pdST) and indigestible (iST). The ST fractions are calculated from in sacco degradation profiles (see Section 5.2.2). The sST fraction represents small starch particles NorFor – The Nordic feed evaluation system

35

Table 4.2. NDF fractions in selected feeds and the corresponding fractional degradation rate of the potential degradable NDF fraction. Feedstuff

Barley Wheat Maize Peas Rape seed Rape seed meal Soybean meal Maize silage, high digestibility Maize silage, low digestibility Grass silage, high digestibility Grass silage, low digestibility

NorFor feed code

NDF g/kg DM

NDF fractions, g/kg NDF pdNDF1

iNDF2

001-0016 001-0020 001-0014 003-0006 002-0007 002-0042 002-0053 006-0307 006-0309 006-0461 006-0463

198 127 111 126 188 290 133 338 397 501 597

836 831 913 984 686 500 939 828 803 888 793

164 169 87 16 314 500 61 172 197 112 207

kdNDF3 %/h 3.2 3.5 3.0 10.4 13.3 6.7 5.0 3.4 3.1 5.0 3.7

1

pdNDF=potentially degradable NDF. iNDF=total indigestible NDF. 3 kdNDF=fractional degradation rate of pdNDF. 2

Table 4.3. Starch fractions in selected feeds and the corresponding fractional degradation rate of the potential degradable starch fraction. Feedstuff

Barley Wheat Maize Peas Maize silage, high digestibility Maize silage, low digestibility

NorFor feed code

Starch g/kg DM

Starch fractions, g/kg ST sST1

pdST2

iST3

001-0016 001-0020 001-0014 003-0006 006-0307 006-0309

615 667 712 511 347 299

350 399 230 230 500 500

650 601 770 770 500 500

3 5 30 30 10 10

kdST4 %/h 39.2 59.5 9.0 9.0 40.0 40.0

1

sST=small starch granules (‘water soluble starch fraction’). pdST=potentially degradable starch. 3 iST=total indigestible starch. 4 kdST=fractional degradation rate of pdST. 2

that are lost from nylon bags washed in cold water. It is assumed that the sST fraction follows the ruminal liquid pool and the fractional degradation rate is set to 150%/h, irrespective of feed source. The pdST fraction is calculated as the difference between the asymptotic value obtained from in sacco incubation and the sST value. The iST fraction is assumed to be completely resistant to digestion in the small intestine and is determined by the MBT (Norberg et al., 2007). The fractional degradation rates (kdST) of pdST are feed-specific and highly variable, ranging from 9%/h (maize) to 79%/h (oats) (Table 4.3). The kdST is estimated from in sacco degradation profiles.

36

NorFor – The Nordic feed evaluation system

704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 720 720 719 720 721 721 721 722 722 722 723 723 723 724 724 724 725 725 725 726 726 726 727 727 727 728 728 728 729 729 729 730 730 730 731 731 731 732 732 732 733 733 733 734 734 734 735 735 735 736 736 736 737 737 737 738 738 738 739 739 739 740 740 740 741 741 741 742 742 742 743 743 743 744 744 744 745 745 745 746 746 746 747 747 747 748 748 748 749 749 749 750 750 750 751 751 751 752 752 752 753 753 753 754 754 754 755 755 755 756 756 756 757 757 757 758 758 758 759 759 759

MBT (Norberg et al., 2007). The fractional degradation rates (kdST) of pdST are feed-specific and highly variable, ranging from 9%/h (maize) to 79%/h (oats) (Table 4.3). The kdST is estimated from in sacco degradation profiles. Table 4.3

The RestCHO fraction includes β-glucans, pectic substances and SU. Although the SU fraction

The RestCHO fraction includes β-glucans, pectic substances and SU. Although the SU fraction maybe beanalytically analyticallydetermined, determined, it categorized is categorized a part of RestCHO the RestCHO fraction. This is because may it is as aaspart of the fraction. This is SU is not analyzed in all in samples and tabulated values areare normally used. The because SUroutinely is not routinely analyzed all samples and tabulated values normally used. TheRestCHO is estimated by: RestCHO is estimated by: Re stCHO = 1000 − Ash − CP − CFat − NDF − ST − FPF

[Eq. 4.1]

4.1

where residual carbohydrate fraction in the g/kg DM; the ash content whereRestCHO RestCHOisisthethe residual carbohydrate fraction infeed, the feed, g/kg Ash DM;isAsh is the ash content in the feed, g/kg DM; CPCP is the crude protein content in theinfeed, g/kg DM; CFat is CFat the crude fatcrude fat in the feed, g/kg DM; is the crude protein content the feed, g/kg DM; is the content in the feed, g/kg DM; and is sum of products in content in in the the feed, feed, g/kg g/kg DM; DM; NDF and FPF FPF is the the sum of the theinfermentation fermentation products in isthe thethefeed, feed, content is the NDF content the feed, g/kg DM; ST starch content inthe the feed, g/kg DM; NDF is the NDF content in the feed, g/kg DM; ST is the starch content content Equation Equationin4.7. 4.7. feed, g/kg DM; and FPF is the sum of the fermentation products in the feed, in the feed, the fermentation products in the feed, Equation 4.7. Equation 4.7.g/kg DM; and FPF is the sum of 34 If N, the the RestCHO RestCHO If the the feedstuff feedstuff contains contains low low molecular molecular weight weight N N fractions, fractions, such such as as urea urea and and NH NH33N, the3RestCHO If contains low molecular weight Nformulas: fractions, such such as urea If the thefeedstuff feedstuff contains low molecular weight N fractions, asand ureaNH and N, the RestCHO is corrected (RestCHOcorr) using the 3N,NH is corrected (RestCHOcorr) using the following following formulas: is using thethe following formulas: is corrected corrected(RestCHOcorr) (RestCHOcorr) using following formulas: For For urea urea correction: correction: For For urea ureacorrection: correction: UreaN UreaN CFat − NDF − ST − FPF Re Re stCHOcorr stCHOcorr = = 1000 1000 − − Ash Ash − − CPcorr CPcorr − − CP CP ⋅⋅ UreaN − − CFat − NDF − ST − FPF 2915 − CFat − NDF − ST − FPF Re stCHOcorr = 1000 − Ash − CPcorr − CP ⋅ 2915 2915

[Eq. [Eq. 4.2] 4.2] [Eq. 4.2]

CP NH 3 N CFat − NDF − ST − FPF Re Re stCHOcorr stCHOcorr = = 1000 1000 − − Ash Ash − − CPcorr CPcorr − − CP ⋅⋅ NH 33 N − − CFat − NDF − ST − FPF 6 1000 − CFat − NDF − ST − FPF Re stCHOcorr = 1000 − Ash − CPcorr − .25 ⋅ 1000 6.25 1000

[Eq. [Eq. 4.3] 4.3] [Eq. 4.3]

⎛⎛ NH NH3N N ⎞⎞ CPcorr CPcorr = CP = ⎜⎛⎜11 − − NH33N ⎟⎟⎞ ⋅⋅ CP 1000 ⎠⎟⎠ ⋅ CP CPcorr = ⎝⎜⎝1 − 1000 1000 ⎠ ⎝

[Eq. [Eq. 4.4] 4.4] [Eq. 4.4]

4.2

For ammonia correction: For For ammonia ammoniacorrection: correction: For ammonia correction:

4.3

whereRestCHOcorr RestCHOcorr is the residual carbohydrate fraction corrected for low molecular where is residual carbohydrate fraction corrected for nitrogen where RestCHOcorr is the the residual carbohydrate fraction corrected for low low molecular molecular nitrogen nitrogen fractions in the feed, g/kg DM; Ash is the ash content in the feed, g/kg DM; CP is the crude protein where RestCHOcorr is the residual carbohydrate fraction corrected for low molecular nitrogen fractions in the feed, g/kg DM; Ash is the ash content in the feed, g/kg DM; CP is the crude fractions in the feed, g/kg DM; Ash is the ash content in the feed, g/kg DM; CP is the crude fractions the feed, feed, the content infat thecontent feed, g/kg DM; CPg/kg is the crude contentcontent ininthe DM; CFat theash fat content in the g/kg DM; NDF is the protein in feed, g/kg DM; CFat is the crude in the feed, DM; NDF is protein content in the theg/kg feed,DM; g/kgAsh DM;isis CFat iscrude the crude fat content in feed, the feed, g/kg DM; NDF is neutral protein content incontent thefibre feed, g/kg DM; CFat isDM; the crude incontent the content feed, DM;feed, NDF isDM; FPF the neutral detergent in feed, g/kg DM; ST is the in the g/kg detergent fibre incontent the feed, g/kg ST isfat the starch ing/kg the g/kg the neutral detergent fibre content in the the feed, g/kg DM; STcontent is the starch starch content in feed, the feed, g/kg the neutral fibre content in products theinfeed, g/kg ST is the starch content inCPcorr the feed, g/kg protein DM; FPF the sum fermentation the feed, Equation 4.7, DM; is the sumis of products the in feed, Equation 4.7, g/kgg/kg DM; CPcorr isis DM; FPF isdetergent thefermentation sum of of fermentation products in theDM; feed, Equation 4.7, g/kg DM; CPcorr iscrude DM; FPF the sum of products themolecular feed, Equation 4.7, g/kg CPcorr crude protein content in the corrected for low N g/kg DM; NH the Nisis isammonia the crude protein content infermentation the feed feedfor corrected for in low molecular N fractions, fractions, g/kgDM; DM; NH3N content inisthe feed corrected low molecular N fractions, g/kg DM; NH N 3N is 3the crude protein content in the feed corrected for low molecular N fractions, g/kg DM; NH ammonia N content in feed, g/kg N; UreaN is the urea N content in feed, g/kg N; 2915 3N is the ammonia N content in feed, g/kg N; UreaN is the urea N content in feed, g/kg N; 2915 content in feed, g/kg N; UreaN is the urea N content in feed, g/kg N; 2915 corresponds to the CP ammonia N content in content feed, g/kg N; UreaN is the urea N content in feed, g/kg N; 2915 corresponds to of kg corresponds to the the CP content of one one kg of of urea. urea. content of one kgCP of urea. corresponds to the CP content of one kg of urea. The CPcorr is calculated as: The calculated The CPcorr CPcorris calculatedas: The CPcorr isiscalculated as:as: 4.4

where CPcorr is the crude protein content in the feed corrected for low molecular weight N fractions,

where the crude N where CPcorr CPcorr is crude protein protein content content in in the the feed feed corrected corrected for for low low molecular molecular weight N g/kg DM; CPisisthe theCP crude protein content in the g/kg DM; and NH N isweight the or urea where CPcorr the crude protein content the feedfeed, molecular fractions, g/kg is crude protein in the g/kg DM; and N is ammonia theN 3 fractions, g/kgisDM; DM; CP is the the crude proteinincontent content incorrected the feed, feed, for g/kglow DM; and3NH NHweight 3N is the N content in the feed, g/kg N. fractions, g/kg DM; CP is the crude protein content in the feed, g/kg DM; and NH N is the ammonia or urea N content in the feed, g/kg N. 3 ammonia or urea N content in the feed, g/kg N. ammonia or urea N content in the feed, g/kg N. The RestCHO RestCHOfraction fractionrepresents representsthe therapidly rapidlyfermented fermentedwater-soluble water-soluble CHO fractions; The CHO fractions; aa a heterogeneous The RestCHO fraction represents the rapidly fermented water-soluble CHO fractions; The RestCHO fraction rapidly CHO fractions; arate (Weisbjerg et heterogeneous group carbohydrates, ranging from SU, has ruminal degradation group of carbohydrates, rangingthe from SU, fermented which awhich high ruminal degradation heterogeneous group of ofrepresents carbohydrates, ranging from has SU,water-soluble which has aa high high ruminal degradation heterogeneous group of carbohydrates, ranging from SU, which has a high ruminal degradation rate (Weisbjerg et al., 1998) to soluble fibre fractions, which have moderate degradation al., 1998) to soluble fibre fractions, which have moderate degradation rates (Sniffen et al., 1992; rate (Weisbjerg et al., 1998) to soluble fibre fractions, which have moderate degradation rates rates rate (Weisbjerg et al.,Lanzas 1998) to fibre which have rates (Sniffen al., 1992; et al., In NorFor the fractional degradation of (Sniffen etal., al.,2007). 1992; Lanzas et soluble al., 2007). In fractions, NorFor the fractional degradation rate of RestCHO RestCHO Lanzas etet In NorFor the2007). fractional degradation rate ofmoderate RestCHOdegradation inrate roughage (kdRestCHO) (Sniffen et al., 1992; Lanzas al., 2007). in (kdRestCHO) is calculated as: in roughage (kdRestCHO) is et calculated as:In NorFor the fractional degradation rate of RestCHO is roughage calculated as: in roughage (kdRestCHO) is calculated as: SU ⎛⎛ SU ⎞⎞⎟ ⋅ 150 SU ⎞⎞⎟ ⋅ 10 + ⎛⎛⎜ SU kd = ⎜⎜⎛11 − − kd Re Re stCHO stCHO = SU ⎟⎠⎞ ⋅ 150 SU ⎟⎠⎞ ⋅ 10 + ⎜⎝⎛ Re stCHO Re stCHO ⎝ Re stCHO Re stCHO ⎠⎟ ⋅ 10 + ⎝⎜ ⎠⎟ ⋅ 150 kd Re stCHO = ⎝⎜1 − ⎝ Re stCHO ⎠ ⎝ Re stCHO ⎠

[Eq. [Eq. 4.5] 4.5] [Eq. 4.5]

4.5

where kdRestCHO is the degradation rate of the rest carbohydrate fraction, %/h; SU is the sugar

where kdRestCHO is the rate of fraction, %/h; SU is where kdRestCHO is the degradation degradation ratecarbohydrate of the the rest rest carbohydrate carbohydrate fraction,4.1; %/h; is the the sugar sugar content, g/kg DM;RestCHO RestCHO is therest rest fraction, Equation theSU fractional degradation where kdRestCHO is the degradation of the rest carbohydrate fraction, SU is the sugar content, g/kg is carbohydrate fraction, 4.1; the content, g/kg DM; DM; RestCHO is the the restrate carbohydrate fraction, Equation Equation 4.1;%/h; the fractional fractional rate of SU is 150%/h; and the fractional degradation rate of the non-sugar part of the RestCHO content, g/kg DM; RestCHO is the rest carbohydrate fraction, Equation 4.1; the fractional

fraction is 10%/h. In concentrate feeds the kdRestCHO is set to 150%/h. 35 35 35

NorFor – The Nordic feed evaluation system

37

760 761 762 763

degradation ratefat of SU is 150 %/h;acids and the fractional degradation rate of the non-sugar part of the 4.5 Crude and fatty RestCHO fraction is 10 %/h.

764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781

richCrude than CPfatorand CHO, although 4.5 fatty acids the energy value is dependent on the proportion of fatty acids (FA)

782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808

Fatconcentrate normally feeds constitutes a low proportion of cattle diets. However, fat is often used as an energy In the kdRestCHO is set to 150%/h. source for high productivity cattle. On a weight basis, fat is approximately 2.2 times more energy-

in the CFat. constitutes However, arumen microorganisms cannot toleratefat high levels of as fatan(Jenkins, Fat normally low proportion of cattle diets. However, is often used energy 1993). Therefore, bothproductivity the level and type FA are basis, important to consider when optimizing cattle source for high cattle. Onofa weight fat is factors approximately 2.2 times more diets. The NorFor (www.norfor.info) includes information on FA profiles of individual energy-rich than CPfeedstuff or CHO,table although the energy value is dependent on the proportion of fatty feedstuffs. FA metabolism in the cannot gastro tolerate intestinal tract is not and is, acids (FA) inHowever, the CFat.individual However, rumen microorganisms high levels of expressed fat (Jenkins, Therefore, both and typemodule. of FA areConsequently, important factors consider when therefore,1993). not implemented in the the level metabolism onlytoinformation on dietary includes onlevel FA of the optimizing cattle diets. NorFor feedstuff (www.norfor.info) FA composition can beThe evaluated, which istable important when focusing on theinformation optimal FA profiles individual feedstuffs. However, individual FA metabolism intestinal tract feeds. diet, on of milk fat quality and milk off-flavours. The proportion of FAininthe thegastro CFat varies between is notmain expressed andlipids is, therefore, implemented in are the metabolism module. Consequently, The storage in plantnot seeds and grains triacylglycerols, which consist of three fatty only information on dietary FA composition can be evaluated, which is important acids attached to glycerol. In these feeds the proportion of FA in CFat is 80 towhen 85%focusing (Danfær et al., on the optimal FA level of the diet, on milk fat quality and milk off-flavours. The proportion of 2006). In contrast, forage lipids are mainly in the form of galactolipids, which contain galactose in FA in the CFat varies between feeds. The main storage lipids in plant seeds and grains are addition to glycerol and unsaturated Thus, their In FAthese proportions triacylglycerols, which consist of threefatty fatty acid acidsmoieties. attached to glycerol. feeds theare substantially lower: approximately 45% (Danfær et al., 2006). When the rumen OM fermentation is predicted in proportion of FA in CFat is 80 to 85% (Danfær et al., 2006). In contrast, forage lipids are mainly NorFor, theofglycerol and galactose are added to the feed fractions, and serve as energy in the form galactolipids, which contain galactose in fermentable addition to glycerol and unsaturated fatty sources for microbial growth. acid moieties. Thus, their FA proportions are substantially lower: approximately 45% (Danfær et al., 2006). When the rumen OM fermentation is predicted in NorFor, the glycerol and galactose are to the fermentable feed fractions, and serve as energy sources for microbial growth. 4.6added Fermentation products

4.6 Fermentation products Ensiled forages dominate as winter roughages in the Nordic countries. The feedstuffs have highly

Ensiled dominate as winter in the Nordic The feedstuffs highly acids) variableforages FPF (i.e. VFA, LAF and roughages alcohols) contents. Thecountries. VFA (acetic, propionic have and butyric variable FPF (i.e. account VFA, LAF alcohols) contents. Themajor VFA (acetic, can collectively forand up to 70 g, and LAF (the organicpropionic acid) for and 40 tobutyric 150 g per kg DM acids) collectively account to 70 g, and lactic acid (the major organic for 40 tosources for of the can silage. The VFA, whichfor areupend-products of ruminal fermentation, areacid) not energy 150 g per kg DM of the silage. The VFA, which are end-products rumen microorganisms. Lactic acid represents a minor sourceofofruminal energy fermentation, for microbialare growth, and not sources forlactic rumenacid microorganisms. a minor energy the energy ATP yield from metabolism isLactic only acid 25%represents of the yield fromsource CHOof fermentation (Van for microbial growth, and the ATP yield from lactic acid metabolism is only 25% of the yield Soest, 1994). A high content of fermentation products in silages has a negative effect on forage intake from CHO fermentation (Van Soest, 1994). A high content of fermentation products in silages (Huhtanen et effect al., 2002, 2007). Therefore, NorFor relative NorFor silage index (Huhtanen has a negative on forage intake (Huhtanen et al.,incorporates 2002; 2007).the Therefore, et al., 2002)the when calculating feed(Huhtanen intake (seeet Section 6.2). The silage index information incorporates relative silage index al., 2002) when calculating feed requires intake (see on fermentation acids (TAF), calculated as: section 6.2). The silage index requires information on fermentation acids (TAF), calculated as: TAF = LAF + ACF + PRF + BUF

[Eq. 4.6]

4.6

where of of total fermentation acidsacids in theinfeed, g/kg DM; the content whereTAF TAFisisthe thecontent content total fermentation the feed, g/kgLAF DM;isLAF is the of content of lactic g/kg DM; ACF is the acid content in the in feed, PRF is the lacticacid acidininthe thefeed, feed, g/kg DM; ACF is acetic the acetic acid content theg/kg feed,DM; g/kg DM; PRF is the propionic feed, g/kg DM; BUF is the acid acid content in thein feed, propionicacid acidcontent contentininthethe feed, g/kg DM; BUF is butyric the butyric content the g/kg feed, g/kg DM. DM

Ensiled forage also contains variable amounts of alcohol (Randby et al., 1999). This is taken into

Ensiled forage also contains variable amounts of alcohol (Randby et al., 1999). This is taken into accountwhen whenFPF FPFisiscalculated: calculated: account FPF = TAF + FOF + ALF

[Eq. 4.7]

4.7

36 where content of fermentation products in thein feed, TAF is the content where FPF FPFisisthethe content of fermentation products theg/kg feed,DM; g/kg DM; TAF is theofcontent of fermentation acids in the feed, Equation 4.6, g/kg DM; FOF is the content of formic acid in the fermentation acids in the feed, Equation 4.6, g/kg DM; FOF is the content of formic acid in the feed, feed, g/kg DM; the content of alcohol feed, g/kgDM. DM. g/kg DM; ALFALF is theis content of alcohol in in thethe feed, g/kg

809 810 811 812 813 814 815 816

4.7 Minerals

817

4.8 Vitamins

4.7 Minerals

Mineral elements may adversely affect production, animal health and reproduction (NRC, 2001). Several elements are also important for maintaining normal rumen function. NorFor predicts Mineral elements may adversely affect production, animal health and reproduction (NRC, 2001). mineral supply and incorporates requirements for the following elements: Calcium (Ca), Several elements are also important for maintaining normal rumen function. NorFor(S), predicts Phosphorus (P), Magnesium (Mg), Potassium (K), Sodium (Na), Chlorine (Cl), Sulfur Iron mineral (Fe), Manganese (Mn), Zinc (Zn), Copper (Cu), Cobalt (Co), Selenium (Se), Molybdenum (Mo) and Iodine (I). Animal mineral requirements are described in Chapter 10. The NorFor feedstuff 38 (www.norfor.info) includes information on the mineral NorForcontent – The of Nordic feedfeedstuffs. evaluation system individual table

supply and incorporates requirements for the following elements: Calcium (Ca), Phosphorus (P), Magnesium (Mg), Potassium (K), Sodium (Na), Chlorine (Cl), Sulfur (S), Iron (Fe), Manganese (Mn), Zinc (Zn), Copper (Cu), Cobalt (Co), Selenium (Se), Molybdenum (Mo) and Iodine (I). Animal mineral requirements are described in Chapter 9. The NorFor feedstuff table (www.norfor. info) includes information on the mineral content of individual feedstuffs.

4.8 Vitamins Rumen microbes synthesize adequate amounts of vitamin K to meet the requirements of most cattle, except in young calves. However, cattle require a dietary source of vitamins A and E. Vitamin D is also normally included in the diet. NorFor incorporates requirements for vitamins A, E and D (see Chapter 9) and the NorFor feedstuff table (www.norfor.info) provides information on the vitamin A, E and D contents of individual feedstuffs.

References Danfær, A., P. Huhtanen, P. Udén, J. Sveinbjörnsson and H. Volden, 2006. The Nordic dairy cow model, Karoline – Description. In: Kebreab, E., J. Dijkstra, A. Bannink, J. Gerrits and J. France (eds.). Nutrient digestion and utilisation in farm animals: Modelling approaches. CAB International, Wallingford, UK, pp. 383-406. Harstad, O.M. and E. Prestløkken, 2000. Effective degradability and intestinal digestibility of individual amino acids in solvent-extracted soybean meal (SBM) and xylose-treated SBM (SoyPass®) determined in sacco. Animal Feed Science and Technology 83: 31-47. Harstad, O.M. and E. Prestløkken, 2001. Rumen degradability and intestinal digestibility of individual amino acids in corn gluten meal, canola meal and fish meal determined in sacco. Animal Feed Science and Technology 94: 127-135. Huhtanen, P., H. Khalili, J.I. Nousiainen, M. Rinne, S. Jaakkola, T. Heikkilä and J. Nousiainen, 2002. Prediction of the relative intake potential of grass silage by dairy cows. Livestock Production Science 73: 111-130. Huhtanen, P., M. Rinne and J. Nousiainen, 2007. Evaluation of the factors affecting silage intake of dairy cows: a revision of the relative silage dry-matter index. Animal 1: 758-770. Hvelplund, T., 1985. Digestibility of rumen microbial protein and undegraded dietary protein estimated in the small intestine of sheep and by the in sacco procedure. Acta Agriculturae Scandinavica Supplement 25: 132-144. Hvelplund, T., M.R. Weisbjerg and L.S. Andersen, 1992. Estimation of the true digestibility of rumen undegraded protein in the small intestine of ruminants by the mobile bag technique. Acta Agriculturae Scandinavica, section A: Animal Science 42: 34-39. Hvelplund, T., F.D. DeB. Hovell, E.R. Ørskov and D.J. Kyle, 1994. True intestinal digestibility of protein estimated with sheep on intragastric infusion and with the mobile nylon bag technique. Proceedings of the Society of Nutrition Physiology, 3. pp. 64. Jenkins, T.C., 1993. Lipid metabolism in the rumen. Journal of Dairy Science 76: 3851-3863. Lanzas, C., C.J. Sniffen, S. Seo, L.O. Tedeschi and D.G. Fox, 2007. A revised CNCPS feed carbohydrate fraction scheme for formulating rations for ruminants. Animal Feed Science and Technology 136: 167-190. Larsen, M., P. Lund, M.R. Weisbjerg and T. Hvelplund, 2009. Digestion site of starch from cereals and legumes in lactating dairy cows. Animal Feed Science and Technology 153: 236-248. Mills, J.A.N., J. France and J. Dijkstra, 1999. A review of starch digestion in the lactating dairy cow and proposals for a mechanistic model: 1. Dietary starch characterization and ruminal starch digestion. Journal of Animal and Feed Sciences 8: 291-340. Norberg, E., O.M. Harstad and H. Volden, 2007. Assessment of recovery site of mobile nylon bags for measuring ileal digestibility in dairy cows. Journal of Dairy Science 90: 418-421. NRC (National Research Council), 2001. Nutrient requirements of dairy cattle. Seventh revised edition, National Academy press, Washington, DC, USA. Offner, A. and D. Sauvant, 2004. Prediction of in vivo starch digestion in cattle from in sacco data. Animal Feed Science and Technology 111: 41-56. Owens, F.N., D.S. Secrist, W.J. Hill and D.R. Gill, 1998. Acidosis in cattle – a review. Journal of Animal Science 76: 1275-1286.

NorFor – The Nordic feed evaluation system

39

Prestløkken, E., 1999. Protein value of expander-treated barley and oats for ruminants. Ph.D. Dissertation No. 5. Department of Animal Science, Agricultural University of Norway, Aas, Norway. Randby, Å.T., I. Selmer-Olsen and L. Bævre, 1999. Effect of ethanol in feed on milk flavour and chemical composition. Journal of Dairy Science 82: 420-428. Russell, J.B., J.D. O’Connor, D.G. Fox, P.J. Van Soest and C.J. Sniffen, 1992. A net carbohydrate and protein system for evaluating cattle diets. I. Ruminal fermentation. Journal of Animal Science 70: 3551-3561. Skórko-Sajko, H., T. Hvelplund and M.R. Weisbjerg, 1994. Rumen degradation and intestinal digestibility of amino acids in different roughages estimated by nylon bag techniques. Journal of Animal and Feed Sciences 3: 1-10. Skiba, B., M.R. Weisbjerg and T. Hvelplund, 1996. Effective degradability in the rumen and total tract digestibility of amino acids in different roughages determined in sacco. Journal of Animal and Feed Sciences 5: 347-363. Sniffen, C.J., J.D. O’Connor, P.J. Van Soest, D.G. Fox and J.B. Russell, 1992. A net carbohydrate and protein system for evaluating cattle diets. II. Carbohydrate and protein availability. Journal of Animal Science 70: 3562-3577. Vanhatalo, A., P. Dakowski and P. Huhtanen, 1996. Effects of stage of growth and duration of rumen incubation time on intestinal digestibility of rumen-undegradable nitrogen of grass by mobile-bag method in cows. Acta Agriculturae Scandinavica, Section A - Animal Science 46: 1-10. Van Soest, P.J, 1994. Nutritional Ecology of the Ruminant. (2nd ed.) Cornell University press, Ithaca, NY, USA. 476 pp. Van Straalen, W.M. and S. Tamminga, 1990. Protein degradation of ruminant diets. In: Wiseman, J. and D.J.A. Cole (eds.). Feedstuff Evaluation. Butterworths, London, pp. 55-72. Volden, H. and O.M. Harstad, 1995. Effect of rumen incubation on the true indigestibility of feed protein in the digestive tract determined by the nylon bag techniques. Acta Agriculturae Scandinavica, Section A - Animal Science 45: 106-115. Volden, H., W. Velle, O.M. Harstad, A. Aulie and Ø. Sjaastad, 1998. Apparent ruminal degradation and rumen escape of lysine, methionine and threonine - administered intraruminally in mixtures to high yielding cows. Journal of Animal Science 76: 1232-1240. Volden, H.L., T. Mydland, V. Olaisen, 2002. Apparent ruminal degradation and rumen escape of soluble nitrogen fractions in grass, and grass silage administered intraruminally to lactating dairy cows. Journal of Animal Science 80: 2704-2716. Weisbjerg, M.R., T. Hvelplund, S. Hellberg, S. Olsson and S. Sanne, 1996. Effective rumen degradability and intestinal digestibility of individual amino acids in different concentrates determined in sacco. Animal Feed Science and Technology 62: 179-188. Weisbjerg, M.R., T. Hvelplund and B.M. Bibby, 1998. Hydrolysis and fermentation rate of glucose, sucrose and lactose in the rumen. Acta Agriculturae Scandinavica, Section A - Animal Science 48: 12-18. Weisbjerg, M.R., T. Hvelplund and K. Søegaard, 2004a. Prediction of digestibility of neutral detergent solubles using the Lucas principle. Journal of Animal and Feed Sciences 13, Supplement 1: 239-242. Weisbjerg, M.R., T. Hvelplund and K. Søegaard, 2004b. Prediction of NDF digestibility based on assumptions about true digestibility and endogenous loss of NDS. Journal of Animal and Feed Sciences, 13, Supplement 1: 243-246. Zebrowska, T., Z. Dlugolecka, J.J. Pajak and W. Korczynski, 1997. Rumen degradability of concentrate protein, amino acids and starch, and their digestibility in the small intestine of cows. Journal of Animal and Feed Sciences 6: 451-470.

40

NorFor – The Nordic feed evaluation system

5. Feed analyses and digestion methods M. Åkerlind, M. Weisbjerg, T. Eriksson, R. Tøgersen, P. Udén, B.L. Ólafsson, O.M. Harstad and H. Volden Feed characteristics are determined via chemical analyses and digestion methods. Specific NorFor methods for determining parameters such as DM, sCP, iNDF and the in sacco methods are fully described in this chapter. Tables 5.1 and 5.2 present an overview of the feed analysis and digestion methods, respectively. Table 5.1. Recommended NorFor feed analysis methods. Parameter

Abbrev.

Unit

Reference method

Dry matter in concentrate Dry matter in roughage Ash Crude protein

DM

g/kg

EC No. 152/2009

DM

g/kg

Ash CP

g/kg DM g/kg DM

Soluble CP

sCP

g/kg CP

Ammonia nitrogen

NH3N

g N/kg N

Individual AA1 Crude fat Individual FA2

AAj CFat FAj

Neutral detergent fibre Starch

NDF

Free choice of MgO-method or Autoanalyzer (Broderick and Kang, 1980) g/100g CP EC No. 152/2009 g/kg DM EC No. 152/2009 g/100 g FA CEN ISO/TS 17764-1:2007 CEN ISO/TS 17764-2:2007 g/kg DM ISO 16472:2006 IDT

ST

g/kg DM

Lactic, propionic, butyric, formic acids and alcohol (ethanol) Sugar Calcium Phosphorus Magnesium Potassium Sodium Chloride Sulphur

LAF, PRF, g/kg DM BUF, FOF, ALF SU Ca P Mg K Na Cl S

g/kg DM g/kg DM g/kg DM g/kg DM g/kg DM g/kg DM g/kg DM g/kg DM

EC No. 152/2009 EC No. 152/2009 or Dumas

NorFor method

See Section 5.1.1 See Section 5.1.2 See Section 5.1.3

Spectrophotometric method or the plate count method described by Bach Knudsen (1997) and Bach Knudsen et al. (1987) HPLC or GC

EC No. 152/2009 ICP or free choice of method EC No. 152/2009 or ICP ICP or free choice of method ICP or free choice of method ICP or free choice of method EC No. 152/2009 or ICP ICP or free choice of method

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_5, © Wageningen Academic Publishers 2011

41

Table 5.1. Continued. Parameter

Abbrev.

Unit

Reference method

Iron, Copper, Manganese and Zinc Other micro minerals Vitamin A β-carotene Vitamin D Vitamin E

Fe, Cu, Mn, Zn

g/kg DM

EC No. 152/2009 or ICP

g/kg DM IU/kg DM IU/kg DM IU/kg DM IU/kg DM

Free choice of method EC No. 152/2009 or Jensen et al. (1998) EC No. 152/2009 or Jensen et al. (1998) Any appropriate method is acceptable EC No. 152/2009 or Jensen et al. (1999)

VitA b-car VitD VitE

NorFor method

1 The amino acids that can be reported in the NorFor feed tables are: alanine, arginine, asparagine, cysteine, glutamine,

glycine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine and valine. 2 The fatty acids that can be reported in the NorFor feed tables are FA50 MJ NEG/d) and vice versa for small animals. However, it has not been possible to improve this relationship. Table 9.2. NE requirement (MJ/d) for growth in primiparous cows related to BW and daily gain.

Figure 9.2

BW, kg

Daily weight gain, g/d

The utilisation coefficient of ME to NE for growth, corrected for the proportion of energy that originates from protein retention, is calculated according to the INRA (1989) equation:

250

500

750

k g _ corr = 0.35 + 0.25 ⋅ (1 − Ep )2

[Eq. 9.6] 400 4.4 7.5 10.6 500 4.5 7.6 10.7 for growth, corrected for 10.9 the proportion of where 600 kg_corr is the utilisation 4.7coefficient of ME to NE7.8 energy that originates from protein retention; and 700 in the total amount of 4.8energy retained per day 7.9 11.1 Ep is the daily NE requirement for protein retention relative to the total daily NE retention, Equation 9.7.

NorFor – The Nordic feed evaluation system 5.48 ⋅ gain _ prot

Ep =

5.48 ⋅ gain _ prot + 9.39 ⋅ gain _ fat

[Eq. 9.7]

87

⎛ 80_ prot + 9.39 ⋅ gain _ fat ) ⋅ 4.184 ⋅ k mg NEG _ gain = ⎜ (5.48 ⋅ gain ⎜⎛ 41000 .184 k gk_mg corr NEG _ gain = ⎝⎜ (5.48 ⋅ gain 70_ prot + 9.39 ⋅ gain _ fat ) ⋅ 1000 ⋅ k ⎜ g _ corr ⎝

NEG requirement (MJ/d)

2834 2834 2835 2835 2836 2837 2836 2838 2837 2839 2838 2840 2839 2841 2840 2842 2841 2843 2842 2844 2843 2845 2844 2846 2845 2847 2846 2848 2847 2849 2848 2850 2849 2851 2850 2852 2851 2853 2852 2854 2853 2855 2854 2856 2855 2857 2856 2858 2857 2859 2858 2860 2859 2861 2860 2862 2861 2863 2862 2863 2864 2864 2865 2866 2865 2867 2866 2868 2867 2869 2868 2870 2869 2870 2871 2871

⎞ ⎟ ⋅ 1.10 ⎟⎞ ⎠⎟ ⋅ 1.10 ⎟ ⎠

[Eq. 9.5] [Eq. 9.5]

where NEG_gain is the60 daily NE requirement for gain in growing cattle, MJ NEG/d; 5.48 and 9.39 the energyiscontent inNE protein and fat, respectively, kcal/g;cattle, gain_prot is the daily whereare NEG_gain the50 daily requirement for gain in growing MJ NEG/d; 5.48protein and retention, Equation 9.8; gain_fat is theand daily retention , Equation 9.9; the factor 9.39 are the energy content in protein fat,fatrespectively, kcal/g; gain_prot is the 4.184/1000 daily protein converts kcal to MJ;9.8; kmg is the combined utilisation coefficient of ME to the NE factor for maintenance retention, Equation is the daily fat retention , Equation 9.9; 4.184/1000and 40gain_fat growth, Equation 8.5kand is the utilisation coefficient of ME to NE for for growth, corrected g_corr converts kcal to MJ; combined utilisation coefficient of ME to NE maintenance and mg iskthe 30 k thatisoriginates for the proportion of energy from coefficient protein retention daily energy the utilisation of MEintothe NEtotal for growth, corrected growth, Equation 8.5 and g_corr retention, Equationof9.6. for the proportion energy 20 that originates from protein retention in the total daily energy retention, Equation 9.6. 10 kg_corr takes into account the dependence of the utilization of ME for The utilization coefficient growth on the proportion of energy retained as protein. Thus, the ratio between kmg and kg_corr of ME foris The utilization coefficient 0 kg_corr takes into account the dependence of the utilization used to on adjust NE requirement for gain toasaccount for variations in the proportion of kenergy growth the the proportion of energy retained protein. Thus, the ratio between k and is mg g_corr 0 10 20 30 40 50 60 70 80 that from retention theto total daily for energy retention (INRA, 1989).ofThe 10% usedoriginates to adjust the NEprotein requirement forin gain account variations in the proportion energy adjustment (factor in Equation 9.5) fordaily growth was added because a dataset NEG supply (MJ/d) that originates from1.1 protein retention inin theNE total energy retention (INRA, 1989).compiled The 10% from 11 Norwegian andinsix Swedish9.5) experiments bulls involving 120 treatments adjustment (factor 1.1 Equation in NE for with growth was added because a dataset(Homb, compiled 1974; Hole, 1985; Olsson Lindberg, 1985; 1987; Johansen, 1992; Skaara, 1994; from 11 Norwegian and sixand Swedish experiments with bulls 120 treatments (Homb, Figure 9.2. The relationship between energyOlsson, supply andinvolving energy requirement after a modification of Randby, 2000a; Randby, 2000b; Berg aand Volden, 2004; Berg, 2005; Berg, 1974; Hole, 1985; Olsson and Lindberg, 1985; Olsson, 1987; 1992; Skaara, 1994; in order to the original French system where 10% increase inVolden NEJohansen, forand growth was implemented unpublished; Havrevoll, unpublished; unpublished) showed underprediction of line shows Y=X Randby, Randby, 2000b; Berg Olsson, and Volden, 2004;and Volden andan Berg, 2005; Berg, obtain a2000a; closer relationship between energy supply energy requirements. Solid RMSE = 4.4; see Figure 9.2). Although, this modification was made, requirements Havrevoll, (r2 = 0.90 and unpublished; unpublished; Olsson, unpublished) showed an underprediction of 2=0.90 and and dotted line shows relationship between supply and requirement (r RMSE=4.4). The 2 that there is also a slope effect indicating that the estimated energy supply Figure 9.2 shows is too = 0.90 and RMSE = 4.4; see Figure 9.2). Although, this modification was made, requirements (r data are from the trials described in Table 9.11. high compared to that the energy larger animals that (>50the MJestimated NEG/d) and vicesupply versa for Figure 9.2 shows there isrequirement also a slope in effect indicating energy is too smallcompared animals. However, it has not been possible improve(>50 this MJ relationship. high to the energy requirement in largertoanimals NEG/d) and vice versa for small animals. However, it has not been possible to improve this relationship. requirement Figure 9.2 in larger animals (>50 MJ NEG/d) and vice versa for small animals. However, we were not able Figure 9.2to improve this relationship when using the described approach. The utilisation coefficient of ME to NE for growth, corrected for the proportion of energy that originates fromcoefficient protein retention, is NE calculated according tocorrected theforINRA (1989) Theutilisation utilisation coefficient of ME to for NEgrowth, for growth, for the equation: proportion The of ME to corrected the proportion of energy of thatenergy that originates retention, is calculated according to thetoINRA (1989) equation: originatesfrom fromprotein protein retention, is calculated according the INRA (1989) equation: [Eq. 9.6] k g _ corr = 0.35 + 0.25 ⋅ (1 − Ep )2 [Eq. 9.6] k g _ corr = 0.35 + 0.25 ⋅ (1 − Ep )2 9.6 where kg_corr is the utilisation coefficient of ME to NE for growth, corrected for the proportion of energy theistotal amount of coefficient energy retained perto day originates fromcorrected protein andof wherekin kg_corr is utilisation coefficient of ME NEgrowth, for growth, for the proportion of where thethe utilisation of ME NEtothat for corrected for theretention; proportion g_corr Ep is the NE amount requirement for protein retention relative to the total daily NE retention, energy in the retained perper day that originates from protein retention; and and Ep is energy indaily thetotal total amountofofenergy energy retained day that originates from protein retention; Equation 9.7. Ep the daily NE requirement for protein retention relative thetotal totaldaily dailyNE NEretention, retention, Equation 9.7. theisdaily NE requirement for protein retention relative totothe Equation 9.7. 5.48 ⋅ gain _ prot [Eq. 9.7] 9.7 Ep = + 9_.39 ⋅ gain _ fat 5.48 ⋅ gain5._48prot ⋅ gain prot [Eq. 9.7] Ep = 5.48 ⋅ gain _ prot + 9.39 ⋅ gain _ fat

where Ep is the daily NE requirement for protein retention relative to the total daily NE retention;

2872 2873 2872 2874 2873 2875 2874 2876 2875 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886

where Ep is the daily NE requirement for protein retention relative to the total daily NE retention; 5.48 and9.39 9.39are arethethe energy content in protein andrespectively, fat, respectively, kcal/g; gain_prot is the daily 5.48 content infor protein andretention fat, gain_prot the daily whereand Ep is the daily energy NE requirement protein relative tokcal/g; the total daily NEis retention; protein retention, Equation 9.8; and gain_fat is the daily fat retention, Equation 9.9. protein and gain_fat is and the daily fat retention,kcal/g; Equation 9.9. is the daily 5.48 andretention, 9.39 are Equation the energy9.8; content in protein fat, respectively, gain_prot protein retention, Equation 9.8; and gain_fat is the daily fat retention, Equation 9.9. The following followingequations equations used to calculate the daily retention of protein and fat (Robelin and The areare used to calculate the daily retention of protein and fat (Robelin & Daenicke,1980) 1980)and and examples of and fat and protein retention in growing cattle Daenicke, examples of fat protein retention in growing cattle are givenare in given Tablesin Tables 110 9.3 9.3 and and9.4. 9.4. 110

2887 2888 2889 2890

where to factor_5 are dependent on the type of system 88 gain_fat is the daily fat retention, g/d; factor_3 NorFor – The Nordic feed evaluation growing cattle (Table 9.5); EBW is the empty body weight, Equation 9.12; Fat_mass is the fat content in the EBW, Equation 9.13 and EBWG is the empty body weight gain, Equation 9.10.

gain _ prot = factor _ 1 ⋅ 1.06 ⋅ (EBWG − gain _ fat) ⋅ FFM0.06

[Eq. 9.8]

9.8

where daily protein retention, kg/d;kg/d; factor_1 is dependent on the on typethe of type growing wheregain_prot gain_protisisthethe daily protein retention, factor_1 is dependent of growing cattle is the empty bodybody weight gain, gain, Equation 9.10; gain_fat is the daily fat daily fat cattle(Table (Table9.5); 9.5);EBWG EBWG is the empty weight Equation 9.10; gain_fat is the retention, FFM is fat freefree mass, Equation 9.11.9.11. retention,Equation Equation9.9; 9.9;and and FFM is fat mass, Equation ⎛ ⎛ ⎛ 1000 3 + 2 ⋅_factor _ 4 ⋅ ln( EBW _) )5⋅ factor _ 5 ⎞⎞ ⋅ Fat _ mass⎛ ⎞ ⎛⎜ (factor ⎞⎞ _ 3 + 2_⋅ factor 4 ⋅ ln( EBW ) ) ⋅ factor [Eq. ⎟ ⎟ ⋅ ( EBWG⎟)⎟1,⋅78 gain __fat fat==⎜ ⎛⎜⎜1000 ( EBWG )1.78 ⋅ ⎜ ⋅ Fat _ mass ⎞⎟ ⋅ ⎜ (⎟factor gain ⎟⎟ ⎜ . 1 78 ⎜ ⎝⎜ ⎝ ⎟ ⎟ EBW ⎠ ⎜ ⎠ 1,78 _ 5 EBW Factor Factor _ 5 ⎠⎠ ⎝ ⎠⎠ ⎝ ⎝⎝

9.9]

9.9

Table 9.3. Estimated fat and protein retention in heifers of dairy breeds related to BW and ADG. BW, kg

200 300 400 500

800 g ADG

600 g ADG

Fat, g/d

Protein, g/d

Fat, g/d

Protein, g/d

142 207 270 333

134 125 114 102

85 124 162 199

105 101 95 88

Table 9.4. Estimated fat and protein retention in bulls of dairy breeds related to BW and ADG. BW, kg

ADG, g/d

Fat, g/d

Protein, g/d

100 200 300 400 500 100 200 300 400 500

1,200 1,500 1,400 1,200 1,000 1,000 1,300 1,200 1000 800

86 200 297 309 290 62 175 226 223 195

200 238 211 172 139 168 211 187 151 120

2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886

The following equations are used to calculate the daily retention of protein and fat (Robelin & Table 9.5.1980) Coefficients used for fatretention and protein retention in growing Daenicke, and examples of calculating fat and protein in growing cattle are given cattle. in Tables 9.3 and 9.4.

2887 2888 2889 2890

where onon thethe type ofof growing where gain_fat gain_fatisisthe thedaily dailyfat fatretention, retention,g/d; g/d;factor_3 factor_3totofactor_5 factor_5are aredependent dependent type growing cattle9.5); (Table 9.5);isEBW is the empty body weight, Equation Fat_mass is fat the content fat cattle (Table EBW the empty body weight, Equation 9.12;9.12; Fat_mass is the in the content in the EBW, 9.13 and EBWG is body the empty body weight gain, Equation EBW, Equation 9.13Equation and EBWG is the empty weight gain, Equation 9.10. 9.10.

2891

EBWG =

2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902

Animal

Factor_1

Factor_2

Factor_3

gain _ prot = factor _ 1 ⋅ 1.06 ⋅ (EBWG − gain _ fat) ⋅ FFM0.06

Heifer or steer 0.1616

-6.311

1.811

Factor_4 0

Factor_5 0.8

Factor_6 [Eq. 9.8]

1.046

Factor_7 -0.3939

1 Bull EM -1.68retention, 0.0189 0.1609 1.0 on the 1.023 -0.2855 where gain_prot is0.1541 the daily protein kg/d; factor_1 is dependent type of growing 1 9.5); 0.1541 cattle (Table EBWG is the-5.433 empty body1.5352 weight gain, is the daily fat Bull LM 0 Equation 9.10; 1.2 gain_fat1.024 -0.2704 retention, 9.9; and FFM is fat free1.3708 mass, Equation 9.11. 1.0 Bull CB1Equation0.1541 -5.7541 0.0442 1.023 0.27795 ⎛ ⎛ 1000 ⋅ Fat _ mass ⎞ ⎛ (factor _ 3 + 2 ⋅ factor _ 4 ⋅ ln( EBW ) ) ⋅ factor _ 5 ⎞ ⎞ 1 EM: early maturing; CB: cross breed. ⎜ ⎟ ⎟ ⋅ ( EBWG )1,78 _ fat = ⎜ ⎜ maturing; LM:⎟ ⋅late gain ⎜⎝ ⎟⎟ EBW ⎠ ⎜⎝ Factor _ 51,78 ⎠⎠ ⎝

EBW ⋅ factor _ 6 ⋅ ADG BW

[Eq. 9.9]

[Eq. 9.10]

9.10

whereEBWG EBWGisisthe thedaily daily empty body weight gain, the empty body weight, Equation where empty body weight gain, g/d; g/d; EBWEBW is theisempty body weight, 9.12; BW is the weight of the animal, kg. However, BW for primiparous cows is corrected with the Equation 9.12; BW is the weight of the animal, kg. However, BW for primiparous cows is followingwith termthe600/BW_mat; is dependent onisthe type of on growing (Table 9.5); and corrected following termfactor_6 ·600/BW_mat; factor_6 dependent the typecattle of growing ADG(Table is the average live weight, g/d.of live weight, g/d. cattle 9.5); anddaily ADGgain is theofaverage daily gain FFM = EBW − Fat _ mass

[Eq. 9.11]

where FFM is the fat free mass of the EBW, kg; EBW is the empty body weight, Equation 9.12; Fat_mass the fat content in the EBW,system Equation 9.13. and NorFor – TheisNordic feed evaluation EBW = e (factor _ 7 + factor _ 6⋅ln(BW ))

[Eq. 9.12]

9.11 89

2894 2894 2895 2895 2896 2896 2897 2897 2898 2898 2899 2899 2900 2900 2901 2901 2902 2902 2903 2903 2904 2904 2905 2905 2906 2906 2907 2907 2908 2908 2909 2909 2910 2910 2911 2911 2915 2915 2912 2912 2916 2916 2913 2915 2917 2913 2917 2916 2914 2918 2918 2914 2917 2919 2919 2918 2920 2920 2919 2921 2921 2920 2922 2922 2921 2922 2923 2923 2924 2924 2923 2925 2925 2924 2926 2926 2925 2927 2927 2926 2928 2928 2927 2929 2929 2928 2930 2930 2929 2931 2931 2930 2932 2932 2931 2933 2933 2932 2934 2934 2933 2935 2935 2934 2936 2936 2935 2937 2937 2936 2938 2938 2937 2939 2939 2938 2940 2940 2939 2941 2941 2940 2942 2942 2941 2943 2943 2942 2944 2944 2943 2945 2945 2944 2946 2946 2945 2947 2947 2946 2948 2948 2947 2949 2949 2948 2950 2950 2949 2951 2951 2950 2952 2952 2951 2953 2953 2952 2954 2954 2953 2955 2955 2954 2956 2956 2955 2957 2957 2956 2958 2958 2957 2959 2959 2958 2959

corrected corrected with with the the following following term term ·600/BW_mat; ·600/BW_mat; factor_6 factor_6 is is dependent dependent on on the the type type of of growing growing cattle cattle (Table (Table 9.5); 9.5); and and ADG ADG is is the the average average daily daily gain gain of of live live weight, weight, g/d. g/d. [Eq. [Eq. 9.11] 9.11]

FFM = EBW − Fat _ mass FFM = EBW − Fat _ mass

9.12; where kg; EBW is empty body weight, Equation whereFFM FFMis thefat fatfree freemass massof theEBW, EBW, EBW is the empty body weight, Equation where FFM isisthe the fat free mass ofofthe the EBW, kg;kg; EBW is the the empty body weight, Equation 9.12;9.12; and and Fat_mass is content the EBW, Equation 9.13 Fat_mass is the fatfat content inin the Fat_mass is the the fat content in theEBW, EBW,Equation Equation9.13. 9.13.. and _ 7 + factor _ 6⋅ ln(BW )) EBW = e ((factor EBW = e factor _ 7 + factor _ 6⋅ln(BW ))

[Eq. [Eq. 9.12] 9.12]

9.12

where body weight, kg; and are on of whereEBW EBWis theempty empty body weight, kg; factor_6 7 are dependent on the type of growing where EBW isisthe the empty body weight, kg; factor_6 factor_6 and 77and are dependent dependent on the the type type of growing growing cattle (Table 9.5); and BW is is thethe weight of the the animal, kg. However, However, BW for for primiparous cows is iscows is cattle is the weight of animal, kg. BW primiparous cows cattle(Table (Table9.5); 9.5);and andBW BW weight of the animal, kg. However, BW for primiparous corrected term ·600/BW_mat. BW_mat is BW 3.5). corrected with the following term ·600/BW_mat. BW_mat is the the maturity BW (Table (Table 3.5). 3.5). correctedwith withthe thefollowing following term 600/BW_mat. BW_mat is maturity the maturity BW (Table 2 + factor_ 3⋅ ln( EBW ) + factor _ 4 ⋅ (ln( EBW ) 22 )) Fat _ mass = ee (factor_ Fat _ mass = (factor_ 2 + factor_ 3⋅ ln( EBW ) + factor _ 4 ⋅ (ln( EBW ) )) 2

[Eq. [Eq. 9.13] 9.13]

9.13

where content in EBW, kg; EBW is body weight, Equation whereFat_mass Fat_massis thefat content in the EBW, EBW is empty the empty weight, Equation 9.12; where Fat_mass isisthe the fatfat content in the the EBW, kg; kg; EBW is the the empty bodybody weight, Equation 9.12; factor_2 to factor_4 are dependent on the type of growing cattle (see Table 9.5). 9.12; factor_2 to factor_4 are dependent on the type of growing cattle (see Table 9.5). factor_2 to factor_4 are dependent on the type of growing cattle (see Table 9.5). where Protein_mass is the protein content in the EBW, kg; factor_1 is dependent on the type where Protein_mass is the protein content in the EBW, kg; factor_1 is dependent on the type of of 9.11.. growing cattle cattle (see (see Table Table 9.5); 9.5);1.and and FFM is the fat free mass of the EBW, Equation 9.11 growing 06 FFM is the fat free mass of the EBW, Equation 1 . 06 [Eq. 9.14] Pr otein _ mass = factor _ 1 ⋅ FFM where protein content on the type of in the EBW, kg; factor_1 is dependent [Eq. 9.14] 9.14 Pr oteinProtein_mass _ mass = factoris_the 1 ⋅ FFM growing Table 9.3 9.3cattle (see Table 9.5); and FFM is the fat free mass of the EBW, Equation 9.11. Table

where Protein_mass is the protein content in the EBW, kg; factor_1 is dependent on the type of

Table 9.3 9.4 cattle (see Table 9.5); and FFM is the fat free mass of the EBW, Equation 9.11. Table 9.4 growing Table 9.5 Table 9.4 9.5

111 111

9.1.5 Mobilization and deposition Table 9.5 9.1.5 Mobilization Mobilization and deposition deposition 9.1.5 and NorFor considers mobilization deposition BCSsection3.1) (see Section 3.1)primifor both NorFor considers considers mobilization and and deposition usingusing BCS (see (see for both both and primi- and NorFor mobilization and deposition using BCS section3.1) for primi- and 9.1.5 Mobilization and deposition multiparous cows.The The amount of NE deposited when aiscow is fed its above its requirement energy requirement multiparous cows. amount of NE deposited when a cow fed above energy multiparous cows. The amount of NE deposited when a cow is fed above its energy requirement andhas hasconsiders positive energy balance (EB>100%) is BCS calculated as: NorFor mobilization and(EB>100%) deposition using (see for both primi- and and has positive energy balance (EB>100%) is calculated calculated as:section3.1) and aaapositive energy balance is as: multiparous cows. The amount of NE deposited when a cow is fed above its energy requirement and a positive balance (EB>100%) NELhas _ dep dep = NEL NEL − −energy NE __ ma ma int − − NEL NEL gain − − NE NEis__calculated gest − − NEL NELas: _ milk milk [Eq. NEL _ = NE int __ gain gest _ [Eq. 9.15] 9.15]

9.15

NELamount _ dep = of NELNE − NE _ ma intwhen − NEL gain − NE gestEB90%) values. The ratio of 500 ratio of were omitted from the dataset due to unrealistically high AATN_Eff (>90%) values. The500 AATN/NEG for growth was calculated from the amount of AATN available for growth and the AATN/NEG for growth was calculated from the amount of AATN available for growth and the NEG for growth. The relationships AAT efficiency and AAT _NEG, BW 500 requirement 181 between NEG requirement for106 growth. The relationships between AATNN efficiency and AATNN_NEG, BW and ADG are shown in Figures 9.7, 9.8 and 9.9. and Figures 9.7, 9.8 and 9.9. 600ADG are shown in94 170 113 197

700 9.11 Table Table 9.11 Figure 9.7 Figure 9.7

98

Figure 9.8 Figure 9.8

86

158

101

183

NorFor – The Nordic feed evaluation system

3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233

ADG ⎞ ⎛ ⎜1.88 − 0.03821⋅ AATN _ NEG − 0.00176 ⋅ BW − 0.2283 ⋅ ⎟ 1000 ⎟ ⋅100 AATN _ Eff = ⎜ [Eq. 9.34] 1 Table 9.11. Descriptive statistics of +the dataset ⋅used for developing the equation for predicting ⎜ + 0.0000019014 ⎟ 1.83 5 ⋅ BW 0 . 000000016 AAT _ NEG N ⎝ ⎠

the AATN efficiency for growth in growing cattle. The dataset included trials with Norwegian and Swedish bulls involving 120 treatments.

where AATN_Eff is the efficiency with which AATN is used for growth, %; BW is the weight of the animal, kg; ADG is the averageAverage daily live weightMinimum gain, g/d and AAT N_NEG is the ratio Maximum SD between AATN and energy available for growth, Equation 9.35.

DMI (kg/d) 6.4 2.8 9.9 1.8 BW (kg) 322 113 568 120 ADG (g/d) AAT − AAT _ ma int 1,144 1,567 196 − AATN _ gest 548 N N = AAT [Eq. 9.35] N _ NEG Protein retention (g/d) NEG_ gain173 114 226 26 PBVN (g/kg DM) 22 -29 77 22 AAT /NEG (g/MJ) 16.9 9.7 29.8 is the whereNAATN_NEG is the ratio between AATN and energy available for growth, g/MJ; AATN4.3 AATNNsupplied efficiency (%) 50.7 8.11; AAT 1.4 99.4 23.0 AAT from the feed ration, Equation N_maint is the AATN required for The ratio between MP and energy available for growth is calculated as:

maintenance, Equation 9.22; AATN_gest is the AATN required for gestation in heifers, Equation 1 (Homb, 1974; Hole, 1985; Olsson and Lindberg, 1985; Olsson, 1987; Johansen, 1992; Skaara, 1994; Randby, 2000a; 9.42; and NEG_gain is the amount of energy used for growth, Equation 9.5.

Randby, 2000b; Berg and Volden, 2004; Volden and Berg, 2005; Berg, unpublished data; Havrevoll, unpublished

The of AATN data). utilization for growth was calculated by dividing the amount of protein data;efficiency Olsson, unpublished retained by the amount of AATN available for growth. The amount of protein retained was estimated using Equation 9.8 (see section 9.1.4). The AATN available for growth was calculated as the totalAAT AATN_NEG supply is minus the AAT N required for maintenance and gestation. Some observations where the ratio between AAT and energy available for growth, g/MJ; AATN is were omittedNfrom the dataset due to unrealisticallyNhigh AATN_Eff (>90%) values. The ratio of the AATN supplied from the feed ration, Equation 8.11; AATN_maint is the AATN required for AATN/NEG for growth was calculated from the amount of AATN available for growth and the maintenance, Equation 9.22; AAT the AATAAT for gestation in heifers, Equation N_gest is between N required NEG requirement for growth. The relationships N efficiency and AATN_NEG, BW 9.42; andare NEG_gain the amount energy and ADG shown in is Figures 9.7, 9.8of and 9.9. used for growth, Equation 9.5.

The efficiency of AATN utilization for growth was calculated by dividing the amount of protein Table 9.11

retained by the amount of AATN available for growth. The amount of protein retained was estimated

Figure 9.7 using Equation 9.8 (see Section 9.1.4). The AATN available for growth was calculated as the total

AATN supply minus the AATN required for maintenance and gestation. Some observations were

Figure omitted9.8 from the dataset due to unrealistically high AATN_Eff (>90%) values. The ratio of AATN/NEG

for growth was calculated from the amount of AATN available for growth and the NEG requirement for growth. The relationships between AATN efficiency and AATN_NEG, BW and ADG are shown in Figures 9.7, 9.8 and 9.9. Based on the AAT supply from the feed ration and the AAT _Eff, the amount of AAT retained Figure 9.9

N

N

for growth can be calculated as:

N

Based on the AATN supply from the feed ration and the AATN_Eff, the amount of AATN retained for growth can be calculated as: ADG ⎞ ⎛ ⎜1.88 − 0.03821 ⋅ AATN _ NEG − 0.00176 ⋅ BW − 0.2283 ⋅ ⎟ 1000 ⎟ AATNretained = AATN ⋅ ⎜ ⎜ + 0.0000019014 ⋅ BW1.83 + 0.000000016 ⋅ AAT _ NEG 5 ⎟ N ⎝ ⎠

9.36

[Eq. 9.36] where AATNretained is the AATN retained for growth, g/d; AATN is AATN supplied from the feed ration, Equation 8.11; AATN_NEG is the ratio between AATN and energy available for growth; Equation 9.35; BW is the weight of the animal, kg; and ADG is the average daily live weight gain, g/d. 119

The maximum retention of AATN in growth can be calculated by putting the derivative of this equation equal to 0, which results in: AAT N _ NEG _ Min =

(1.88 − 0.00176 ⋅ BW − 0.2283 ⋅ ADG / 1000 + 0.0000019014 ⋅ BW 1.83 ) 0 .03821 ⋅ 2 − 0 .000000016 ⋅ 5 ⋅ AAT N _ NEG 4

9.37

However, the above equation is not suited for calculating the AATN requirement for growing cattle because it is not possible to calculate the minimum AATN requirement (AATN_NEG _Min) using AATN_NEG as an input, since AATN_NEG is a ration variable. Therefore, AATN_NEG was replaced NorFor – The Nordic feed evaluation system

99

AATN efficiency (%)

120 100 80 60 40 20 0

8

10

12

14

16

18

20

22

24

26

28

30

AATN (g/MJ NEG)

Figure 9.7. AATN efficiency (%) in growing cattle in relation to AATN supply. AATN supply is defined as the AATN available for growth per MJ NEG available for growth. A multiple regression with AATN supply, BW (see Figure 9.8) and ADG (see Figure 9.9) as significant explanatory variables explained the variation in AATN efficiency with high accuracy (r2=0.97 and RMSE=2.9). The data are from the trials described in Table 9.11.

AATN efficiency (%)

120 100 80

3234 3235 3236 3237 3238

60the AATN retained for growth, g/d; AATN is AATN supplied from the feed where AATNretained is ration, Equation 8.11; AATN_NEG is the ratio between AATN and energy available for growth; Equation 9.35; BW40 is the weight of the animal, kg; and ADG is the average daily live weight gain, g/d. 20

3239 3240

The maximum retention 0 of AATN in growth can be calculated by putting the derivative of this equation equal to 0, which results in: 50 100 150 200 250 300 350 400 450 500 550 600

3241

AAT N _ NEG _ Min =

3242

(1.88 − 0.00176 ⋅ BW − 0.2283 ⋅ ADG 1000 + 0 .0000019014 ⋅ BW 1 .83 ) BW /(kg) 0 .03821 ⋅ 2 − 0 .000000016 * 5 * AAT N _ NEG 4

Figure 9.8. AATN efficiency (%) in growing cattle in relation to BW based on the trials described in [Eq. 9.37] Table 9.11. A multiple regression with BW, AATN supply (see Figure 9.7) and ADG (see Figure 9.9) However, the above equation is not suited for calculating the AAT requirement for growing cattle N AAT efficiency as significant explanatory variables explained the variation in with high accuracy N because it is not possible to calculate the minimum AAT N requirement (AATN_NEG _Min) using 2 (r =0.97 and RMSE=2.9).

3243 3244 3245 3246 3247 3248 3249 3250

AATN_NEG as an input, since AATN_NEG is a ration variable. Therefore, AATN_NEG was replaced with a value of 14, which corresponds to the requirement of an animal with a BW of 300 with value corresponds to theoptimum requirement ofthe an economic animal with a BWthe of 300 kg and kg anda an ADGofof14, 800which g/d. Since the biological is rarely optimum, an ADG of 800 g/d. Since the biological optimum is rarely the economic optimum, the biological biological optimum (=maximum retention of AATN) was lowered by 1 g AATN/NEG. Therefore, optimum (=maximum of AAT ) was lowered 1 g AAT /NEG. Therefore, the following the following equation isretention a modification of NEquation 9.37, inby which 1 is subtracted and 14 replaces N equation is a modification of Equation 9.37, in which 1 is subtracted and 14 replaces AATN/NEG: AAT N/NEG:

3251

AAT N _ NEG _ Min =

3252

[Eq.is9.38] where AATN_NEG_Min is the minimum AATN requirement (g/MJ NEG); BW the weight of the animal, kg; and ADG is the average daily live weight gain, g/d.

3253 3254 3255 3256 3257 3258

(1.88 − 0.00176 ⋅ BW − 0.2283 ⋅ ADG / 1000 + 0.0000019014 ⋅ BW 1.83 ) − 1 0 .03821 ⋅ 2 − 0 .000000016 ⋅ 5 ⋅ 14 4

9.38

where AATN_NEG_Min is the minimum AATN requirement (g/MJ NEG); BW is the weight of the animal, kg; and ADG is the average daily live weight gain, g/d.

100 NorFor – The Nordic feed evaluation system In contrast to the AATN balance for dairy cows, the AAT N balance for growing cattle is calculated more simply: AATN _ NEG

3234 3235 3236 3237 3238

where AATNretained is the AATN retained for growth, g/d; AATN is AATN supplied from the feed ration, Equation 8.11; AATN_NEG is the ratio between AATN and energy available for growth; Equation 9.35; BW is the weight of the animal, kg; and ADG is the average daily live weight gain, g/d.

120 The maximum retention of AATN in growth can be calculated by putting the derivative of this equation equal to 0,100 which results in:

3241

AAT N _ NEG _ Min =

3242 3243 3244 3245 3246 3247 3248 3249 3250

[Eq. 9.37] 60 However, the above equation is not suited for calculating the AATN requirement for growing cattle because it is not possible 40 to calculate the minimum AATN requirement (AATN_NEG _Min) using AATN_NEG as an input, since AATN_NEG is a ration variable. Therefore, AATN_NEG was replaced with a value20of 14, which corresponds to the requirement of an animal with a BW of 300 kg and an ADG of 800 g/d. Since the biological optimum is rarely the economic optimum, the biological optimum (=maximum retention of AATN) was lowered by 1 g AATN/NEG. Therefore, 0 the following equation is a modification of Equation 9.37, in which 1 is subtracted and 14 replaces 500 600 700 800 900 1000 1,100 1,200 1,300 1,400 1,500 1,600 1,700 1,800 AATN/NEG: ADG (g/d)

3251

AAT N _ NEG _ Min =

3252 3253 3254 3255 3256

AATN efficiency (%)

3239 3240

(1.88 − 0.00176 ⋅ BW − 0.2283 ⋅ ADG / 1000 + 0.0000019014 ⋅ BW 1.83 )

80

0 .03821 ⋅ 2 − 0 .000000016 * 5 * AAT N _ NEG 4

(1.88 − 0.00176 ⋅ BW − 0.2283 ⋅ ADG / 1000 + 0.0000019014 ⋅ BW 1.83 ) − 1 4

0 .03821cattle ⋅ 2 − 0 .in 000000016 5 ⋅ 14 Figure 9.9. AATN efficiency (%) in growing relation⋅to ADG based on the trials described in [Eq. 9.38] Table 9.11. A multiple regression with ADG, AATN supply (see Figure 9.7) and BW (see Figure 9.8) as significant explanatory variables explained the variation in AATN efficiency with high accuracy (r2=0.97 and RMSE=2.9). where AAT N_NEG_Min is the minimum AATN requirement (g/MJ NEG); BW is the weight of

the animal, kg; and ADG is the average daily live weight gain, g/d.

3257 3258

In contrast contrasttotothe theAAT AAT dairy cows, the AAT for growing is calculated N balance N balance for for dairy cows, the AAT for growing cattle iscattle calculated In N balance N balance more simply: more simply:

3259

AATN _ bal =

3260 3261 3262 3263

N N N N where AATfor N_bal is the AATN balance, %; AATN_NEG is the ratio between AATN and energy available growth, Equation 9.35; and AATN_NEG_Min is the minimum AATN requirement, available for growth, Equation 9.35; and AATN_NEG_Min is the minimum AATN requirement, Equation 9.38. Equation 9.38.

AATN _ NEG ⋅100 AATN _ NEG _ Min

[Eq. 9.39]

9.39

where AAT _bal is the AAT balance, %; AAT _NEG is the ratio between AAT and energy

9.2.4 Mobilization Mobilizationand anddeposition deposition 9.2.4

3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275

In dairy cows, mobilized or deposited body protein will affect the amount of AATN available for In dairy cows, mobilizedInorNorFor deposited bodyN protein will affect amount of AATN available for milk mobilization (AATthe milk protein production. the AAT N_mob) or AATN deposition protein production. In to NorFor the AAT (AAT AAT deposition (AATN_dep) (AAT is related the animal's energy balance. This means that or when a Ncow is in negative N_dep) N mobilization N_mob) is related to theprotein animal’s balance.It This meansthat thatbody when a cowreserves is in negative energy balance willenergy be mobilized. is assumed energy containenergy 135 g balance protein/kg (NRC, 1985) and,Itbased on values from INRA (1989), 24.8 MJ energy is supplied per (NRC, protein will be mobilized. is assumed that body energy reserves contain 135 g protein/kg kg mobilized body on mass, and the energy requirement for MJ deposition MJ/kg. The 1985) and, based values from INRA (1989), 24.8 energyisis31.0 supplied per kgutilization mobilized body i.e. The 80%utilization (Tammingaofetmobilized al., 1994),protein of mobilized milk production is assumedisto31.0 be high, mass, and theprotein energyforrequirement for deposition MJ/kg. whereas utilizationisof AATN for deposition in80% adult(Tamminga animals has et been shown towhereas be low the utilization for milkthe production assumed to be high, i.e. al., 1994), (INRA, NRC, 2001); it is assumed to be 120 50%. The amount of AATN originating from of AAT1989; N for deposition in adult animals has been shown to be low (INRA, 1989; NRC, 2001); it mobilized protein is calculated as: is assumed to be 50%. The amount of AAT originating from mobilized protein is calculated as:

3276

AATN _ mob =

3277 3278 3279 3280 3281 3282 3283

from body mass, g/d; NEL_mob is the where N mobilized whereAAT AATNN_mob _mobisisthe theamount amountofofAAT AAT N mobilized from body mass, g/d; NEL_mob is the amount amount of energy mobilized from body Equation 9.18; is the AATcontent perkg kgbody mass; N contentper of energy mobilized from body mass,mass, Equation 9.18; 135135 is the AAT N body mass; 24.8 is the energyper supplied per kg mobilized body MJ NEL/kg; and 0.8torefers 24.8 is the energy supplied kg mobilized body mass, MJmass, NEL/kg; and 0.8 refers the efficiency to the efficiency of mobilized AATN for milk production.

3284

AATN _ dep =

3285 3286 3287 3288

N

135 ⋅ 0.8 ⋅ NEL _ mob 24.8

[Eq. 9.40]

9.40

of mobilized AATN for milk production.

When cows are in positive energy balance the AATN requirement for protein deposition is When cows calculated as:are in positive energy balance the AATN requirement for protein deposition is calculated

as:

135 ⋅ 2 ⋅ NEL _ dep 31.0

[Eq. 9.41]

is the feed amount of AATN required where AAT N_dep NorFor – The Nordic evaluation systemfor protein deposition, g/d; NEL_dep is the amount of energy deposited, Equation 9.17; 135 is the AATN content per kg body mass; 31.0 is the energy content per kg body mass, MJ NEL/kg; and two refers to an efficiency of 50% for AATN deposition.

101

3280 3281 3282 3283

When cows are in positive energy balance the AATN requirement for protein deposition is calculated as:

3284

AATN _ dep =

3285 3286 3287 3288 3289 3290

where g/d; NEL_dep is the whereAAT AATNN_dep _depisisthe theamount amountofofAAT AATNNrequired requiredfor forprotein proteindeposition, deposition, g/d; NEL_dep is the amount amount of energy deposited, Equation the AAT per kg body mass; 31.0isisthe energy N content of energy deposited, Equation 9.17; 9.17; 135 is135 theisAAT content per kg body mass; 31.0 N refers to an efficiency of 50% for the energy content per kg body mass, MJ NEL/kg; and two content per kg body mass, MJ NEL/kg; and 2 refers to an efficiency of 50% for AATN deposition. AATN deposition.

3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310

to the efficiency of mobilized AATN for milk production.

135 ⋅ 2 ⋅ NEL _ dep 31.0

[Eq. 9.41]

9.41

The AATN requirements for deposition and mobilization are illustrated in Table 9.12.

The AATN requirements for deposition and mobilization are illustrated in Table 9.12.

9.2.5 Gestation

Table 9.12

The AATN requirement for gestation in heifers and cows depends on the stage of pregnancy and includes the AATN required for the development and maintenance of all conceptus tissues, including 9.2.5 Gestation the foetus. The AATN requirement equation is modified from on NRC uses mature The AATN requirement for gestation in heifers and cows depends the(1985) stage ofand pregnancy and BW as a scaling factor instead of birth weight. includes the AATN required for the development and maintenance of all conceptus tissues,

including the foetus. The AATN requirement equation is modified from NRC (1985) and uses -0.00262 ∙ gest_day 8.5357 – (13.1201 ) - 0.00262 ∙ gest_day mature BW asBW_mat a scaling factor instead birth weight.∙ e ∙ 34.375 ∙ eof 600 −0.00262⋅ gest _ day AATN_gest = BW _ mat 8 . 5357 − ( 13 . 1201 ⋅ e ) − 0 .00262⋅gest _ day ⋅ 34.375 ⋅ e 0.5 600 [Eq. 9.42] AATN _ gest =

9.42

5 gestation, g/d; BW_mat is the mature BW, kg (Table where AATN_gest is the daily AAT required0.for 3.2 and 3.5); and gest_day is the actual day of gestation; 0.5 is the assumed AATN utilization for where AATN_gest is the daily AAT required for gestation, g/d; BW_mat is the mature BW, kg gestation. (Table 3.2 and 3.5); and gest_day is the actual day of gestation; 0.5 is the assumed AAT N

utilization for gestation.

The AATN requirement for gestation is illustrated in Figure 9.10.

The AATN requirement for gestation is illustrated in Figure 9.10.

9.3 PBVN

Figure 9.10

The minimum PBVN recommendation for dairy cows is related to milk yield according to the following equation, which is illustrated in Figure 9.11.

9.3 PBVN

-0.14 ∙ ECM) for dairy cows is related to milk yield according to the The PBVN recommendation PBVminimum N_DM_Min = 10 ∙ (1 – e following equation, which is illustrated in Figure 9.11.

9.43

where PBVN_Min is the minimum recommended level of PBVN, g/kg DM; and ECM is the daily 121 energy-corrected milk yield, Equation 3.1 and 3.2. Table 9.12. Weight change and AATN requirement for deposition and mobilization in cows depending on breed and change in BCS. Daily change in BCS1

Weight change, kg/d

AATN requirement, g/d

Large breed

Jersey

Large breed

Jersey

-0.02 -0.01 0 0.01 0.02

-1.2 -0.6 0 0.6 1.2

-0.9 -0.45 0 0.45 0.9

-154 -77 0 192 384

-115 -58 0 144 288

1

One BCS=60 kg for large breeds and 45 kg for Jersey, i.e. 0.01 BCS corresponds to 600 g BW for large breeds and 450 g BW for Jersey.

102

NorFor – The Nordic feed evaluation system

AATN requirement (g/d)

350

DH Jersey

300 250 200 150 100 50 0

0

50

100

150

200

250

300

Days in gestation

Figure 9.10. Estimated AATN requirement (g/d) for pregnancy depending on days of gestation and mature BW. Jersey has a mature BW of 440 kg; Holstein has a mature BW of 640 kg. Figure 9.11 shows that for cows producing more than 20 kg ECM per day, the minimum PBVN recommendation is 10 g/kg DM, while for cows producing less than 20 kg the minimum PBVN is reduced to zero when the milk production is 0, i.e. for dry cows. If recirculation of urea to the rumen is taken into account, then the minimum PBVN should theoretically be 0, implying that the N available for microbial growth is similar to the N required for microbial growth. The NorFor PBVN recommendation for lactating dairy cows is based on a former Danish recommendation, which was set to 0 g/d (Strudsholm et al., 1999) on the basis of several production trials (Kristensen, 1997; Madsen et al., 2003), but raised to 10 g/kg, for two reasons. Firstly, unlike the former Danish system, NorFor takes into account the recirculation of urea to the rumen in the calculation of PBVN (see Equation 7.35), and secondly the feed passage rate is lower than in the former AAT/PBV system (Madsen et al., 1985), which will result in a higher ruminal degradation of protein. A maximum of 40 g PBVN per kg DM is recommended for both lactating and dry cows

PBVN (g/kg DM)

12 10 8 6 4 2 0

0

10

20

30 40 ECM (kg/d)

50

60

Figure 9.11. PBVN recommendation for dairy cows as a function of ECM yield. NorFor – The Nordic feed evaluation system

103

to avoid increased energy expenditure for the excretion of N via urine (Danfær, 1983; Reynolds, 2000) and decreased fertility (Rajala-Schultz, 2001). For growing cattle the minimum PBVN recommendation is 0 g/kg DM. This recommendation is not based on production trials but on a theoretical approach, i.e. that there is a balance between CP entering the rumen and CP leaving the rumen. The maximum PBVN recommendation is set to 55 g/kg DM based on Olsson (1987). Olsson (1987) conducted several trials with intensively reared bulls and concluded a maximum recommendation of 4 g PBV/MJ ME. Assuming a 60% ME to NEG conversion efficiency, and an average content of 7.5 MJ NEG/kg DM, the maximum recommendation is 50 g PBVN/kg DM (4/0.6=6.7 g PBVN/MJ NEG·7.5). Since the energy content of a ration may be higher than 7.5 MJ NEG/kg DM, it was decided that NorFor has a maximum recommendation of 55 g PBVN per kg DM.

9.4 Rumen load index The maximum recommended RLI (see Chapter 7 and Equation 7.15 for calculation) is set to 0.6, which gives a moderate reduction of NDF degradation rate. A RLI of 0.6 corresponds to a diet with 240-280 g starch per kg DM depending on starch source. The sugar and starch content of a diet will often be 290-320 g/kg DM. These maximum levels of starch and sugar are in accordance with previous Danish recommendations (Strudsholm et al., 1999). The NRC (2001) recommends diets with a non-fibre carbohydrates content ranging from 360-440 g/kg DM, depending on the NDF content of the ration; 360 g/kg DM for diets with a low NDF content (250 g/kg DM), and 440 g/kg DM for diets with a high NDF content (330 g/kg DM). No health problems were reported in Danish trials with dairy cows fed diets containing 330-340 g starch/kg DM using maize silage and barley as starch sources (Hymøller et al., 2005).

9.5 Chewing time Chewing time index (CI) is used for cows and growing cattle in NorFor to ensure good rumen function, see further descriptions in Chapter 11. The minimum recommendation for CI is 32 minutes per kg DM for large breeds and 30 minutes per kg DM for Jersey cows. These recommendations are based on assessments in relation to the former Danish chewing index (Nørgaard, 1986). However, in the future the recommendations will be validated against data where e.g. rumen pH and milk fat content has been measured.

9.6 Intake capacity Calculations of the ration FV and animal IC are described in Chapter 10. The total FV of the ration should not exceed the IC of the animal. When optimising the feed ration FV should be as close to IC as possible in order to ensure the animals satiety. However for practical reasons NorFor has different minimum recommendations for IC for lactating cows, dry cows and growing cattle.

9.7 Fatty acids The type of fat, i.e. chain length, degree of saturation and physical form, is of importance for the response in milk yield (Børsting et al., 2003). Too high fat content in dairy cattle diets will reduce feed intake and fibre degradation in the rumen (NRC, 2001). The NRC (2001) recommends that the CFat content in a ration should not exceed 6-7% of DM, and the Danish maximum recommendation for growing cattle is 50-60 g/kg DM (Strudsholm et al., 1999). A relationship was established between ECM yield and FA/kg DM, based on Danish trials with fat supplementation suggesting that the maximal ECM yield is obtained with 46 g FA/kg DM for fat with a degree of saturation of approximately 50 (Børsting et al., 2003). If the fat supplementation consists of FAs with a high 104

NorFor – The Nordic feed evaluation system

3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381

The type of fat, i.e. chain length, degree of saturation and physical form, is of importance for the response in milk yield (Børsting et al., 2003). Too high fat content in dairy cattle diets will reduce feed intake and fibre degradation in the rumen (NRC, 2001). The NRC (2001) recommends that the CFat content in a ration should not exceed 6-7% of DM, and the Danish maximum recommendation for growing cattle is 50-60 g/kg DM CFat of DMI (Strudsholm et al., degreeAofrelationship saturation was thenestablished the maximum ECM yield can beFA/kg achieved 75 gonFA/kg DM (Børsting 1999). between ECM yield and DM,atbased Danish trials with supplementation suggesting that the ECM yield is obtained g FA/kg et al.,fat2003). NorFor has implemented a maximal maximum recommendation ofwith 45 g46FA/kg DM for both DM fat growing with a degree of saturation of approximately 50 content (Børstingofet64 al.,g/kg 2003). If the fat cowsforand cattle. This corresponds to a CFat DM, assuming a 50/50 supplementation consists of FAs with degree ofinsaturation the and maximum ECM ratio of roughage to concentrate, anda high FA contents the CFat then of 650 750 g/kg in yield roughage and can be achievedrespectively. at 75 g FA/kg DM (Børsting et al., 2003). NorFor has implemented a maximum concentrates, recommendation of 45 g FA/kg DM for both cows and growing cattle. This corresponds to a CFat content of 64 g/kg DM, assuming a 50/50 ratio of roughage to concentrate, and FA contents 9.8 Minerals in the CFat of 650 and 750 g/kg in roughage and concentrates, respectively.

3382 3383 3384 3385 3386 3387 3388

Mineral requirements are generally based on a factorial approach in which the requirements for 9.8 Minerals

3389 3390 3391 3392 3394 3394 3394 3393 3394 3394 3394 3394 3395 3394 3395 3394 3395 3395 3395 3395 3395 3395 3396 3395 3396 3396 3396 3396 3396 3396 3396 3396 3397 3397 3397 3397 3397 3397 3397 3397 3397 3398 3398 3398 3398 3398 3398 3398 3399 3398 3399 3398 3399 3399 3399 3399 3399 3400 3399 3400 3399 3400 3400 3400 3400 3400 3401 3400 3401 3400 3401 3401 3401 3401 3401 3401 3401 3402 3402 3402 3402 3402 3402 3402 3402 3402 3403 3403 3403 3403 3403 3403 3403 3404 3403 3404 3403 3404 3404 3404 3404 3404 3405 3405 3404 3405 3404 3405 3405 3405 3405 3406 3406 3405 3406 3405 3406 3406 3406 3406 3407 3406 3407 3406 3407 3407 3407 3407 3407 3408 3408 3407 3408 3407 3408 3408 3408 3408 3409

9.8.1 Macro minerals Recommendations for calcium, phosphorus, magnesium, sodium, potassium and chlorine for dairy Recommendations forcattle calcium, sodium, potassium and chlorine for dairy cows and growing arephosphorus, calculatedmagnesium, using the following equations. Note that the absorption cows and growing calculated using the following equations. the absorption coefficients rangecattle fromare16% for magnesium to 90% for sodiumNote andthat potassium:

maintenance, milk production, growth and gestation are considered. The requirements also include a

Mineral requirements are generally based on a factorial approach in which the requirements for safety margin, although this margin notgestation quantified, based onThe therequirements lowest absorption maintenance, milk production, growthisand are but considered. also coefficients from the literature. Furthermore, these low absorption coefficients also apply for mineral supplements, include a safety margin, although this margin is not quantified, but based on the lowest although they have higher true absorption coefficients (NRC, 2001). absorption coefficients from the literature. Furthermore, these low absorption coefficients also apply for mineral supplements, although they have higher true absorption coefficients (NRC, 9.8.1 Macro minerals 2001).

coefficients range_from Ca_maint = factor 1 ⋅ BW16% 0.50for magnesium to 90% for sodium and potassium: [Eq. 9.45] Ca_maint = factor _1 ⋅ BW 00..50 [Eq. 9.45] Ca_maint = factor 1 [Eq. 9.45] Ca_intake_ _ ma0int 9.44] Ca_maint = Min factor=_ _Ca 1 ⋅⋅ BW BW .50 50+ Ca _ milk + Ca _ gain + Ca _ gest [Eq. 9.45] Ca_maint = factor _ 1 ⋅ BW 0.50 [Eq. 9.45] Ca_maint = factor _ 1 ⋅ BW 0 . 50 [Eq. Ca_maint = factor _ 1 ⋅ BW 0.50 [Eq. 9.45] 9.45] Ca _ milk= =factor 1 . 22_⋅1ECM 0 . 50 [Eq. 9.46] Ca_maint ⋅⋅ BW 9.45] Ca _ milk 1 .. 22 50 [Eq. 9.46] Ca_maint = =factor _⋅1ECM BW 000....50 50 9.45] Ca [Eq. 9.46] Ca _ _ milk milk = =1 1 . 22 22 ⋅⋅ ECM ECM 0 0 . 50 50 [Eq. 9.46] Ca _ milk = 1 . 22 ⋅ ECM 0 . 50 [Eq. 9.46] Ca .. 50 [Eq. 9.46] 9.46] Ca _ _ milk milk = =1 1 .. 22 22 ⋅⋅ ECM ECM 0 00.22 50 [Eq. − 0.22 ⋅ ADG 123 Ca = 1 ⋅⋅ ECM .. 50 [Eq. 9.46] Ca_gain ⋅ BW_mat 0.50 9.47] [Eq. − 0.22 Ca _ _ milk milk= =9.83 1 .. 22 22 ECM 0 00.22 50 ⋅⋅ BW ADG 9.46] 0.22 − 0.22 1000 ADG Ca_gain = 9.83 ⋅ BW_mat BW ⋅ 0.50 [Eq. 9.47] 0.22 ⋅⋅ BW − 0.22 ⋅⋅ ADG1000 0.50 Ca_gain = 9.83 ⋅⋅ BW_mat [Eq. 9.47] Ca_gain = 9.83 BW_mat BW 0.50 [Eq. 9.47] 0.22 − 0.22 1000 ADG Ca_gain = 9.83 ⋅ BW_mat ⋅ BW ⋅ 0.50 1000 [Eq. 9.47] 0.22 − 0.22 ADG 0.22 − 0.22 Ca_gain = = 9.83 9.83 ⋅⋅ BW_mat BW_mat BW ⋅ ADG1000 0.50 0.50 [Eq. 9.47] 9.47] Ca_gain ⋅⋅ 0BW [Eq. .00007 ⋅gest⋅⋅)ADG ⋅gest 1000 0.22 − 1000 0.50 ⎛ 9.83 ⎞ 0.22 − − 0.22 0.22 Ca_gain = ⋅⋅ BW_mat ⋅⋅ 00BW 0.02456 ⋅ e(((000...05581 [Eq. 9.47] 05581 − ..00007 ⋅⋅gest ))ADG ⋅⋅gest ⎛ ⎞ Ca_gain = 9.83 BW_mat BW ⋅ 0.50 05581 − 00007 gest gest [Eq. 9.47] ⎜ ⎟ 1000 ⎛ ⎞ 0.02456 e ⋅ Ca_gest = ⎜⎛⎜ 0.02456⋅ e(0.05581− 0.00007⋅gest)⋅gest 1000 .50 [Eq. 9.48] ⎟⎞⎟⎞⎟ 0 0.02456 e ⋅ ( 0 . 05581 − 0 . 00007 ⋅ gest ) ⋅ gest = Ca_gest 0 . 50 ( 0 . 05581 − 0 . 00007 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) ⎜ ⎛ [Eq. 9.48] Ca_gest 0 . 50 = [Eq. 9.48] 0.02456 e ⋅ ( 0 . 05581 − 0 . 00007 ⋅ gest ) ⋅ gest 0.02456 e − ⋅ ⎜ ⎟ ⎛ ⎞ Ca_gest − 0− .00007 ⋅gest )⋅gest 00..05581 00..00007 ⋅⋅((gest − 1)⋅(gest − 1) ⎠⎟⎞ 0..50 ⎝⎜⎛ −0.02456 [Eq. 9.48] ⋅⋅ ee⋅⋅ (ee0(((.05581 − − 0.02456 0.02456 Ca_gest = = [Eq. 0.05581 05581 −.00007 0.00007 00007 ⋅(gest gest −11))⋅⋅((gest gest − −11)) ⎟ 0 0.02456 − ⋅⋅gest ))⋅⋅gest Ca_gest = ⎝⎜⎜⎝⎜⎛⎝⎛ − 0..50 50 [Eq. 9.48] 9.48] 0.02456 0.05581 0.00007 ⋅(gest −1)⋅(gest −1) ⎠⎟⎠⎟⎞⎠⎟⎞ 0 Ca_gest = 50 05581 − 00− .00007 gest gest 0.02456 ⋅⋅ ee⋅⋅ ((ee00((..05581 [Eq. 9.48] ⎜⎜⎝ − 0.02456 − 0 . 05581 − 0 . 00007 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) 0.02456 ⎟ (0.05581− 0.00007⋅(gest −1)⋅(gest −1) ⎠⎟ 0.50 Ca_gest = ⋅⋅ eeint [Eq. 9.49] 9.48] 0.02456 −in0.02456 ⎠ 0.50 = ⎜⎝⎜⎝ − Ca_gest P_intake_M = P _ ma + P _ milk + P _ gain + P _ gest [Eq. 9.48] ⎟ ⎠ ( 0 . 05581 − 0 . 00007 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) ⎜⎝ −in (0.05581 0.00007 ⋅_ (gest −1+ )⋅(P gest −1) ⎟⎠ 0.02456 ⋅⋅ eeint P_intake_M = P _ ma + P __ −milk + P gain __ gest [Eq. 9.49] P_intake_M in = P _ ma int + P milk + P _ gain + P gest [Eq. 9.49] − 0.02456 P_intake_M in = P _ ma int + P _ milk + P _ gain + P _ gest ⎝ ⎠ [Eq. 9.49] P_intake_M in [Eq. P_intake_M in = =P P_ _ ma ma int int + +P P ___ milk milk + +P P ___ gain gain + +P P ___ gest gest [Eq. 9.49] 9.49] P_intake_M in = P _ ma int + P milk + P gain + P gest [Eq. 9.49] P_maint = 1.0in⋅ DMI 0ma .70int + P _ milk + P _ gain + P _ gest [Eq. P_intake_M = P _ [Eq. 9.50] 9.49] P_maint = 1 . 0 ⋅ DMI 0 . 70 P_intake_M in = P _ ma int + P _ milk + P _ gain + P _ gest [Eq. 9.50] 9.49] P_maint = = 11..00 ⋅⋅ DMI DMI 0 0..70 70 [Eq. 9.50] 9.50] P_maint [Eq. P_maint = 1 . 0 ⋅ DMI 0 . 70 [Eq. 9.50] 9.50] [Eq. P_maint = 1 . 0 ⋅ DMI 0 . 70 P_maint = 1.0 ⋅ DMI 0.70 [Eq. 9.50] P_milk==factor _ 2 ⋅ MY 0.70 [Eq. 9.50] 9.51] P_maint 1 . 0 ⋅ DMI 0 . 70 [Eq. P_milk = factor _ 2 ⋅ MY 0 . 70 P_maint = 1.0 ⋅ DMI 0.700.70 9.51] [Eq. 9.50] P_milk _ 2 ⋅ MY [Eq. P_milk= = factor factor MY 70 [Eq. 9.51] 9.51] P_milk = factor___ 222 ⋅⋅⋅ MY MY 000...70 70 [Eq. 9.51] P_milk = factor [Eq. P_milk= factor_ 2 ⋅ MY 0.70 0.22 − 0.22 ADG [Eq. 9.51] 9.51] P _ gain =factor 1.2 + ⋅ BW_mat ⋅ BW −− 0.22 ⋅ ADG1000 0.70 [Eq. 9.52] 0.22 P_milk = __ 4.635 22 ⋅⋅ MY 00..70 [Eq. 0.22 P _ gain 1.2 + 4.635 0.70 9.52] P_milk == factor MY⋅⋅ BW_mat 70 [Eq. 9.51] 9.51] 0.22 ⋅⋅⋅ BW − 0.22 0.22 ⋅⋅⋅ ADG P _ gain = 1.2 + 4.635 BW 0.70 9.52] ADG 1000 P _ gain = 1.2 + 4.635 ⋅⋅ BW_mat BW_mat BW [Eq. 9.52] 0.22 − 0.22 1000 0.70 ADG 0.22 − 0.22 P _ gain = 1.2 + 4.635 BW_mat ⋅ BW ⋅ 0.70 1000 [Eq. 9.52] ADG 0.22 − 0.22 [Eq. 9.52] ADG P _ gain = 1.2 + 4.635 ⋅ BW_mat ⋅ BW ⋅ 0.70 1000 0.70 P _ gain = 1.2 + 4.635(⋅0BW_mat ⋅ BW ⋅ [Eq. 9.52] 1000 0.22 − 0.22 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest 1000 ⎛ ⎞ ADG 0.22 − 0.22 P _ gain = 1.2 + 4.635 ⋅ BW_mat ⋅ BW ⋅ 0.70 0.02743 e ⋅ [Eq. 9.52] ADG ( 0 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest ⎛ ⎜ ⎟ ⎞ P _ gain==⎛ 1.2 + 4.635 ⋅(00BW_mat ⋅ BW ⋅ 0.70 [Eq. 9.52] ( . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest 1000 ⎞ ⋅ 0.02743 e P_gest 0 . 70 − 0.000075⋅gest)⋅gest ⋅⋅ ee (0..05527 1000⎟⎟⎞ ⎜⎜⎛⎜⎛ 0.02743 [Eq. 9.53] 0.02743 = 05527 − 0 . 000075 ⋅ gest ) ⋅ gest P_gest 0 . 70 ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) ⎟ ⎞ [Eq. 9.53] P_gest 0 . 70 = [Eq. 9.53] e ⋅ ( 0 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest ⎟ ⎛⎜⎛ −0.02743 ⎞ P_gest 0 . 70 = ⋅ 0.02743 e ( 0 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) ⎝ ⎠ [Eq. 9.53] ⎜ ⎟ ⎞ 0.70 0.02743 e ⋅ ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) − ⋅ 0.02743 e ⋅ P_gest = 0.02743 e ⎜ ⎟ [Eq. 9.53] ⎝⎝⎜⎜⎛ − ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) 0.02743 e ⋅ ⎠ ⎟ ( 0 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest P_gest 0 . 70 = ⎞ ⎠ ⎟ [Eq. 9.53] 0.02743 e ⋅ ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) P_gest = ⎝⎜⎛ − 0 . 70 ( 0 . 05527 − 0 . 000075 ⋅ gest ) ⋅ gest 0.02743 e ⋅ [Eq. 9.53] ⎞ ⎠ ⎟ 0.02743 e − ⋅ ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) 0.02743 e ⋅ ⎜ ⎟ ⎝ ⎠ ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) ⎜− ⎟ 0 P_gest ..70 = 0.02743 e ⋅ [Eq. 0.02743 e − ⋅ ⎝ ⎠ P_gest 0 70 = [Eq. 9.53] 9.53] ⎝⎜⎜ Min = Mg _ ma ⎠⎟⎟ Mg_intake_ int + Mg _ milk + Mg _ gain + Mg _ gest 9.54] ( 0 . 05527 − 0 . 000075 ⋅ ( gest − 1 ) ⋅ ( gest − 1 ) Mg_intake_ (0.int 05527 − 0_ ⋅Mg (gest −1)⋅(gest −1_ 0.02743 ⋅⋅__eema = Mg + Mg milk + __ gain + Mg [Eq. 9.54] Mg_intake_ Min = Mg int + Mg __.000075 milk + Mg + Mg __) gest gest ⎝⎝ − ⎠⎠ [Eq. 9.54] 0.02743 −Min Mg_intake_ Min = Mg __ ma ma int + Mg milk + Mg __ gain gain + Mg gest [Eq. 9.54] Mg_intake_ Min = Mg ma int + Mg _ milk + Mg gain + Mg _ gest [Eq. 9.54] Mg_intake_ Min = Mg _ ma int + Mg _ milk + Mg _ gain + Mg _ gest [Eq. Mg_intake_Min = Mg _ ma int+ Mg _ milk + Mg _ gain + Mg _ gest [Eq. 9.54] 9.54] [Eq. 9.55] Mg_maint = ⋅ 0 . 003 BW 0 . 16 Mg_intake_ Min = Mg ma 9.54] Mg_maint = 0 ..003 [Eq. 9.55] Mg_intake_ Min =⋅⋅ BW Mg 00__..16 ma int int+ + Mg Mg __ milk milk + + Mg Mg __ gain gain + + Mg Mg __ gest gest 9.54] [Eq. 9.55] Mg_maint = 0 003 BW 16 [Eq. 9.55] Mg_maint= 0.003⋅ BW 0.16 [Eq. 9.55] Mg_maint= 0.003⋅ BW 0.16 [Eq. 9.55] Mg_maint = 0 . 003 ⋅ BW 0 . 16 [Eq. 9.55] Mg_maint = 0..15 003 ⋅ BW 00..16 9.56] Mg_milk = ⋅⋅ ECM 16 9.56] Mg_milk ==0 00...15 0...16 16 [Eq. Mg_maint 003 ⋅⋅ BW 9.56] Mg_milk ⋅⋅ ECM ECM 16 [Eq. 9.55] 9.55] Mg_maint 003 BW 000 9.56] Mg_milk= ==0 00..15 15 ECM 0..16 16 [Eq. 9.56] Mg_milk= 0.15⋅ ECM 0.16 [Eq. 9.56] Mg_milk = 0 . 15 ⋅ ECM 0 . 16 9.56] Mg_milk 0.16 0.16 [Eq. 9.57] Mg_gain == 00.15 .45⋅ ECM ⋅ ADG ADG 1000 [Eq. 9.57] 9.56] Mg_milk = 0 . 15 ⋅ ECM 0 . 16 ADG Mg_gain = 0 . 45 ⋅ 0 . 16 [Eq. 9.57] Mg_gain = 0 . 45 ⋅ 0 . 16 9.56] Mg_milk 01000 .16 0.16 [Eq. 9.57] Mg_gain == 00.15 .45⋅ ECM ⋅ ADG 1000 [Eq. 9.57] 1000 0.16 Mg_gain = 0.45 ⋅ ADG ADG [Eq. 9.57] 1000 ADG Mg_gain = 0 . 45 ⋅ 0 . 16 [Eq. 9.58] 9.57] Mg_gain .450⋅ .16 1000 0.16 Mg_gest == 00.33 [Eq. [Eq. 9.58] 9.57] Mg_gain ..45 ⋅ .ADG 0.16 Mg_gest == 000..33 16 1000 [Eq. ADG 9.57] Mg_gest 0 [Eq. 9.58] Mg_gain 450 Mg_gest = == 0 0.33 33 0⋅ ..16 16 1000 0.16 [Eq. 9.58] Mg_gest = 0.33 0.16 1000 [Eq. 9.58] Mg_gest = 0 . 33 0 . 16 [Eq. 9.58] NorFor – The Nordic feed evaluation system Mg_gest = 0Min .33 = 0.16 [Eq. 9.58] Na_intake_ Na _ ma int+ Na _ milk + Na _ gain + Na _ gest 9.59] [Eq. Na_intake_ Min = Na _ ma int+ Na _ milk + Na _ gain + Na _ gest [Eq. 9.59] Mg_gest = ..33 0 Na_intake_ Na [Eq. 9.58] 9.59] Mg_gest =0 0Min 33 = 0..16 16 [Eq. 9.58] Na_intake_ Min = Na _ _ ma ma int int+ + Na Na _ _ milk milk + + Na Na _ _ gain gain + + Na Na _ _ gest gest 9.59] Na_intake_Min = Na _ ma int+ Na _ milk + Na _ gain + Na _ gest [Eq. Na_intake_Min Min = = Na Na _ _ ma ma int int+ + Na Na _ _ milk milk + + Na Na _ _ gain gain + + Na Na _ _ gest gest [Eq. 9.59] 9.59] Na_intake_ [Eq. 9.59] Na_maint= factor_ 3 ⋅ BW 0.90 [Eq. 9.60]

(( ((( ( (( ((( ( ( (((( (

(( ((( (( (( ( (( (( (( ( (( (( ( ((

(( ((( (

(( ((( (

)) (( ))) ((( )(

(( ((( ( ( (((( (

)) ))) )

)) (( ))) ((( )(

(( ((( ( ( (((( (

)) )) ) ))

)) ))) )

)) ))) ) )) )) ))

)) ))) )

)) ))) )

)) ))) )

)) ))) )

9.44 9.45 9.46 9.47 9.48 9.49 9.50 9.51 9.52 9.53 9.54 9.55 9.56 9.57

105

3405 3405 3405

Mg_milk =00...15 15 ECM ..16 Mg_milk ECM 16 Mg_milk Mg_milk== 15⋅⋅⋅⋅⋅ECM ECM0000 0...16 16 Mg_milk ==000..15 15 ECM 16

3406 3406 3406

ADG ADG Mg_gain === 000..45 .45⋅⋅⋅ADG 016 .16 Mg_gain ADG1000 Mg_gain 1000 0000...16 Mg_gain == 00..45 45⋅⋅ADG .16 Mg_gain 45 1000 1000 16 1000

[Eq. 9.57] [Eq. 9.57] [Eq. 9.57] [Eq. 9.57] [Eq. 9.57]

3407 3407 3407

Mg_gest Mg_gest===000..33 .33 33000..16 .16 16 Mg_gest Mg_gest == 00..33 33 00..16 16 Mg_gest

[Eq. 9.58] [Eq. 9.58] [Eq. 9.58] [Eq. 9.58] [Eq. 9.58]

9.58

3408 3408

Na_intake_ Min = Na Na _ ma ma int + Na Na _ milk milk + Na Na _ gain gain Na gest Na_intake_ Min Na ma int Na milk Na gain Na _ gest Na_intake_ ++++ ___gest Na_intake_Min Min=== = Na Na___ _ma maint int+++ +Na Na___ _milk milk+++ +Na Na___ _gain gain +Na Na _gest gest Na_intake_ Min int Na

[Eq. 9.59] [Eq. 9.59] [Eq. 9.59] [Eq. 9.59] [Eq. 9.59]

9.59

3409 3409 3409

Na_maint = factor factor _ 3 ⋅ BW 0.90 Na_maint factor BW ..90 90 Na_maint Na_maint BW Na_maint=== = factor factor____3333⋅⋅⋅⋅BW BW0000..90 90

[Eq. 9.60] [Eq. 9.60] [Eq. 9.60] [Eq. 9.60] [Eq. 9.60]

9.60

3410 3410 3410

Na_milk MY 90 Na_milk Na_milk== =000..63 .63 63⋅⋅⋅⋅MY MY0000...90 .90 Na_milk Na_milk== 00..63 63⋅MY MY 0.90 90

[Eq. 9.61] [Eq. 9.61] [Eq. 9.61] [Eq. [Eq.9.61] 9.61]

9.61

3411 3411 3412 3411 3412 3412 3412 3412 3412 3412 3412 3413 3413 3413 3413 3413 3413 3413 3414 3414 3414 3414 3414 3414 3414 3415 3415 3415 3415 3415 3415 3415 3416 3416 3416 3416 3416 3416 3416 3416 3417 3417 3417 3417 3417 3417 3417 3418 3418 3418 3418 3418 3418 3418 3419 3419 3419 3419 3419 3419 3419 3420 3420 3420 3420 3420 3420 3420 3421 3421 3421 3421 3421 3421 3422 3422 3422 3422 3422 3422 3422 3423 3423 3423 3423 3424 3423 3424 3423 3424 3424 3425 3423 3424 3425 3424 3425 3425 3426 3424 3425 3426 3425 3426 3426 3427 3425 3426 3427 3426 3427 3427 3428 3426 3427 3428 3427 3428 3428 3428 3429 3427 3428 3429 3428 3429 3429 3429 3430 3428 3429 3430 3429 3430 3430 3430 3431 3429 3430 3431 3431 3431 3432 3430 3431 3432 3432 3432 3433 3431 3432 3433 3433 3433 3434 3432 3433 3434 3434 3434 3435 3433 3434 3435 3435 3436 3434 3435 3436 3436 3437 3435 3436 3437 3437 3438 3436 3437 3438 3438 3439 3437 3438 3439 3439 3439 3440 3438 3440 3440 3441 3439 3440 3441 3441 3441 3442 3440 3442 3442 3442 3443 3441 3442 3443 3443 3443 3442 3443 3443

ADG Na_gain 90 ADG Na_gain 000..90 Na_gest == = 1111...4.444⋅⋅⋅0⋅ADG .90 1000 Na_gain .90 1000 Na_gain = Na_gain =111..44.400⋅ADG 90 1000 00..90 Na_gest = ..ADG 90 Na_gest = 90 1000 1000 Na_gest = 1 . 4 0 . 90 Na_gest ==11..44 00..90 90 Na_gest Na_gest = 1 . 4 0 . 90 124 124 Na_gest = 1 . 4 0 . 90 Na_gest = 1.4in0.=90K _ ma int + K _ milk + K _ gain 124 K_intake_M + K _ gest 124 K_intake_M = _ int + _ + _ + K K_intake_Min in == K K __ ma ma int ++ K K __milk milk ++K K __gain gain +124 K_ _ gest gest K_intake_M in K ma int K milk K gain gest K_intake_Min in == K K__ma maint int++KK__milk milk++KK__gain gain+++KKK___gest gest K_intake_M K_intake_M in = K _ ma int + K _ milk + K _ gain + K _ K_intake_M gest K_intake_Min in ==KK__ma maint int++KK__milk milk++KK__gain gain++KK_ _gest gest

(((

[Eq. 9.56] [Eq. 9.56] [Eq. 9.56] [Eq. 9.56] [Eq. 9.56]

))))

(((

))))

( (((

) )))) )

( ((( (

) )))) )

[Eq. 9.62] 9.62 [Eq. 9.62] [Eq. 9.63] 9.62] [Eq. [Eq.9.62] 9.62] [Eq. 9.63] [Eq. 9.63] [Eq. 9.63] [Eq.9.63] 9.63] [Eq. [Eq. 9.63] 9.63 [Eq. 9.63] [Eq. 9.63] 9.64] [Eq. 9.64] [Eq. 9.64] [Eq. 9.64] [Eq. 9.64] [Eq. 9.64] [Eq. 9.64] [Eq. 9.64] 9.64 [Eq. 9.64] K_maint= (0.038⋅ BW + factor_ 4 ⋅ DMI) 0.90 [Eq. 9.65] ( ) K_maint = 0 ..038 ⋅⋅ BW + factor _ 4 ⋅⋅ DMI 0 ..90 [Eq. 9.65] K_maint = 0 038 BW + factor _ 4 DMI 0 90 [Eq. 9.65] ( ) K_maint== (00..038 factor___444⋅⋅DMI ⋅DMI DMI 90 [Eq. 9.65] 038⋅⋅⋅BW BW+ factor ..90 [Eq. 9.65] ))))000..090 K_maint +++factor [Eq. 9.65] K_maint BW factor _ 4 ⋅ DMI = (0.038 9.65 [Eq. 9.65] K_maint ) )0.090 90 [Eq. 9.65] K_maint 038 factor__44⋅ DMI ⋅ DMI .90 [Eq. 9.65] K_milk===1(.(500.⋅.038 MY⋅⋅BW 0BW .90++factor [Eq. 9.66] K_milk = 1 . 5 ⋅ MY 0 . 90 [Eq. 9.66] K_milk= MY 90 [Eq. 9.66] K_milk 90 [Eq. 9.66] =11 1...55 5⋅⋅⋅MY MY000 0....90 90 [Eq. 9.66] K_milk = [Eq. 9.66] K_milk 90 = [Eq. 9.66] 9.66 K_milk ==111...555⋅⋅⋅MY MY 0 . 90 [Eq. 9.66] K_milk MY 0 . 90 [Eq. 9.66] ADG K_gain = 1.6 ⋅ ADG 0.90 [Eq. 9.67] 1000 ADG K_gain = 1 . 6 ⋅ 0 . 90 [Eq. 9.67] K_gain = 1.6 ⋅ADG 00..90 [Eq. 9.67] ADG1000 K_gain 90 [Eq. 9.67] = 11..66⋅⋅ADG 90 1000 [Eq. 9.67] K_gain = [Eq. 9.67] 9.67 K_gain = 90 [Eq. 9.67] 1000 0000....90 1000 ADG1000 K_gain == 111...666⋅⋅⋅ADG 90 [Eq. 9.67] 1000 K_gain 0 . 90 [Eq. K_gest = 1.0 0.90 1000 [Eq. 9.67] 9.68] 1000 K_gest [Eq. 9.68] K_gest = = 11..00 00..90 90 [Eq. 9.68] 9.68 K_gest [Eq. 9.68] =111...000 000...90 90 [Eq. 9.68] K_gest = [Eq. 9.68] K_gest = 90 [Eq. 9.68] K_gest ==11..00 00..90 [Eq. 9.68] K_gest 90 [Eq. 9.68] Cl_intake_ Min = Cl _ ma int + Cl _ milk + Cl _ gain + Cl _ gest 9.69] Cl_intake_ [Eq. 9.69] Cl_intake_Min Min = = Cl Cl _ _ ma ma int int + + Cl Cl _ _ milk milk + + Cl Cl _ _ gain gain + + Cl Cl _ _ gest gest [Eq. 9.69] Min = = Cl Cl___ma ma int ++Cl Cl milk ++Cl Cl gain ++Cl Cl gest Cl_intake_Min maint int+ Cl____milk milk+ Cl____gain gain+ Cl____gest gest 9.69 [Eq. [Eq. 9.69] 9.69] Cl_intake_ [Eq. 9.69] Cl_intake_ Min = Cl ma int + Cl milk + Cl gain + Cl gest [Eq. 9.69] Cl_intake_ Min = Cl _ ma int + Cl _ milk + Cl _ gain + Cl _ gest [Eq. 9.69] Cl_intake_Min = Cl _ ma int + Cl _ milk + Cl _ gain + Cl _ gest [Eq. 9.69] Cl_maint = 00..0225 0225⋅⋅ BW BW 00..90 90 [Eq.9.70] 9.70] Cl_maint= = [Eq. 9.70 Cl_maint 0225 BW [Eq. 9.70] = 0000....0225 0225⋅⋅⋅⋅BW BW00000....90 90 Cl_maint = [Eq. 9.70] Cl_maint .90 90 [Eq. 9.70] [Eq. 9.70] Cl_maint = 0225 BW 90 [Eq. 9.70] Cl_maint ==00..0225 ⋅⋅BW 00.90 [Eq. 9.70] Cl_maint 0225 BW . 90 [Eq. 9.70] Cl_milk = 11..15 15⋅⋅ MY MY 00..90 90 [Eq.9.71] 9.71] Cl_milk= = [Eq. 9.71 Cl_milk 15 MY 90 [Eq. 9.71] Cl_milk =1111....15 15⋅⋅⋅⋅MY MY0000....90 90 [Eq. 9.71] [Eq. 9.71] Cl_milk= 90 [Eq. 9.71] Cl_milk = 15 MY 90 [Eq. 9.71] Cl_milk= 1.15⋅ MY 0.90 [Eq. 9.71] Cl_milk= 1.15⋅ ADG MY 0.90 [Eq. 9.71] Cl gain = = 11..00 ⋅⋅ ADG 90 [Eq.9.72] 9.72] ADG1000 00..90 9.72 Cl ___ gain gain [Eq. Cl = 1 . 0 ⋅ 0 . 90 [Eq. 9.72] ADG ADG 1000 Cl __ gain 0000...90 [Eq. 9.72] gain = = 111...000⋅⋅⋅ADG 90 1000 [Eq. 9.72] . 90 [Eq. 9.72] Cl gain = 90 [Eq. 9.72] ADG 1000 1000 0.90 Cl _ gain = 1.0 ⋅ ADG1000 [Eq. 9.72] 1000 Cl _ gain==11.0.00⋅ .90 1000 0.90 [Eq. 9.73] 9.72] Cl_gest [Eq. 1000 9.73 Cl_gest 90 [Eq. 9.73] Cl_gest == =111...000 000...90 90 [Eq. 9.73] Cl_gest [Eq. 9.73] =11..00 00..90 90 [Eq. 9.73] Cl_gest = Cl_gest [Eq. [Eq. 9.73] 9.73] Cl_gest = 1.0 0.90 [Eq. 9.73] Cl_gestCa_intake_Min = 1.0 0.90 [Eq. 9.73] where is the the calcium recommendation fordairy dairycows, cows, g/d;Ca_maint, Ca_milk, where Ca_intake_Min is calcium the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, where Ca_intake_Min is recommendation for g/d;Ca_maint, Ca_milk, where Ca_intake_Min is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, where Ca_intake_Min is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, Ca_intake_Min is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, where Ca_intake_Min is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, Ca_gain, Ca_gest is a cows requirement of Ca for maintenance, milk production, gain and gestation, where is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, Ca_gain, Ca_gest is a cows requirement offor Camaintenance, forfor maintenance, milk production, gain and gestation, Ca_gain, Ca_gest is aa cows requirement of Ca milk production, gain and gestation, where Ca_intake_Min is the calcium recommendation dairy cows, g/d;Ca_maint, Ca_milk, Ca_gain, Ca_gest is cows requirement of Ca for maintenance, milk production, gain and gestation, Ca_gain, Ca_gest is requirement of milk production, gain and gestation, Ca_gain, Ca_gest is aa cows cows requirement ofCa Ca for maintenance, milk production, gain and gestation, Ca_gain, Ca_gest is requirement of for maintenance, milk production, gain and gestation, respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are where Ca_intake_Min is the calcium recommendation for dairy cows, g/d;Ca_maint, Ca_milk, Ca_gain, Ca_gest cows requirement Cafor formaintenance, maintenance, milk production, gain and gestation, respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are Ca_gain, Ca_gest isisaaacows cows requirement ofofCa Ca for maintenance, milk production, gain and gestation, respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and Ca_gain, Ca_gest is a cows requirement of Ca for maintenance, milk production, gain and gestation, respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and respectively; P_intake_Min, Mg_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min are recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and recommendations of phosphorus, magnesium, sodium, potassium and for dairy cows and recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and growing cattle, g/d; BW is the the weight of the the animal, animal, kg; BW_mat ischlorine the mature weight ofthe the respectively; P_intake_Min, Na_intake_Min, K_intake_Min and Cl_intake_Min areanimal, recommendations ofBW phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and growing cattle, g/d; is weight of kg; BW_mat is the weight of recommendations of phosphorus, magnesium, sodium, potassium and chlorine dairy cows and growing cattle, g/d; BW is the weight of the animal, kg; BW_mat is the mature weight of the growingcattle, cattle, g/d; BW isMg_intake_Min, the weight ofanimal, the animal, kg; BW_mat ismature thefor mature weight of the growing g/d; BW is the weight of the kg; BW_mat is the mature weight of the growing cattle, g/d; BW is the weight of the animal, kg; BW_mat is the mature weight of the growing cattle, g/d; BW is the weight of the animal, kg; BW_mat is the mature weight of the animal, kg (Table 3.2 and Table 3.5); ECM is the energy-corrected milk yield, Equation 3.1 and 3.2; recommendations of phosphorus, magnesium, sodium, potassium and chlorine for dairy cows and3.2; growing cattle, g/d; BW is the weight of the animal, kg; BW_mat is the mature weight of the animal, kg (Table 3.2 and Table 3.5); ECM is the energy-corrected milk yield, Equation 3.1 and 3.2; growing cattle, g/d; BW is the weight of the animal, kg; BW_mat is the mature weight of the animal, kg (Table 3.2 and Table 3.5); ECM is the energy-corrected milk yield, Equation 3.1 and kg (Table 3.2 and Table 3.5); ECM is theisisweight energy-corrected milk yield, Equation 3.1and and 3.2; MY is animal, kg (Table 3.2 and 3.5); milk Equation 3.1 3.2; animal, kg (Table 3.2 and Table 3.5); ECM the energy-corrected yield, Equation 3.1 and 3.2; animal, kg (Table 3.2 and Table 3.5); ECM the energy-corrected milk yield, Equation 3.1 and 3.2; MY is the milk yield, kg/d; ADG is theECM daily gain ofgrowing growing cattle including primiparous growing cattle, g/d; BW is Table the weight of theis animal, kg;of BW_mat ismilk theyield, mature weight of3.1 the animal, kg (Table 3.2 and Table 3.5); ECM isthe theenergy-corrected energy-corrected milk yield, Equation and 3.2; MY is the milk yield, kg/d; ADG is the daily gain cattle including primiparous animal, kg (Table 3.2 and Table 3.5); ECM isweight the energy-corrected milk yield, Equation 3.1 and 3.2; MY is the the milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous MY is milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous the milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous the milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous MY is the milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous cows, g/d; gest is the actual day in gestation from day 190 of gestation until calving; The animal, kg (Table 3.2 and Table 3.5); ECM is the energy-corrected milk yield, Equation 3.1 and MY isg/d; the milk milk yield, kg/d;ADG ADG isgestation thedaily dailyfrom weight gain of growing cattle including primiparous 3.2; cows, cows, gest is the actual day in day 190 gestation until calving; The MY is the yield, kg/d; is the weight gain of growing cattle including primiparous cows, g/d; gest is the actual day in gestation from day 190 of gestation until calving; The cows, g/d; gest is the actual day in gestation from day 190 ofof gestation until calving; The g/d; isgest is the actual day in gestation from day190 190 of gestation until calving; Theand denominators g/d; gest is the actual day in gestation from day 190 gestation until calving; The cows, g/d; is the actual day in gestation from day of until calving; The denominators equations are absorption coefficients; factor_1 is0.031 0.031 for lactating cows and MY the gest milk yield, kg/d; ADG is the daily weight gain of growing cattle including primiparous g/d; gest is theequations actual day in gestation from day 190 ofgestation gestation until calving; The in the are absorption coefficients; is for lactating cows cows, g/d; gest is the actual day in gestation from day 190 offactor_1 gestation until calving; The denominators in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows and denominators in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows and denominators in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows denominators in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows and in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows and 0.0154 for non-lactating cows, g Ca/kg BW; factor_2 is 0.98 for large breeds and 1.17 for Jersey, cows, g/d; gest is the actual day in gestation from day 190 of gestation until calving; The denominators inthe theequations equations are absorption coefficients; factor_1 0.031 for lactating cows cows,are gCa/kg Ca/kg BW;factor_2 factor_2 is0.98 0.98 forlarge large breeds and 1.17for for Jersey, gg0.0154 for denominators in absorption coefficients; factor_1 is is 0.031 for lactating cows andand 0.0154 for non-lactating cows, Ca/kg BW; factor_2 is 0.98 for large breeds and 1.17 for Jersey, g 0.0154 for non-lactating cows, ggg BW; is for breeds and 1.17 Jersey, g 0.0154 for non-lactating cows, Ca/kg BW; factor_2 is for large breeds and 1.17 for Jersey, g 0.0154 for non-lactating cows, gglactating Ca/kg BW; factor_2 isis0.98 0.98 for large breeds and 1.17 for Jersey, ggP/kg non-lactating cows, g Ca/kg BW; factor_2 is 0.98 for large breeds and 1.17 for Jersey, g P/kg milk; 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, g milk; denominators in the equations are absorption coefficients; factor_1 is 0.031 for lactating cows and 0.0154 for non-lactating cows, Ca/kg BW; factor_2 0.98 for large breeds and 1.17 for Jersey, 0.0154 for non-lactating cows, g Ca/kg BW; factor_2 is 0.98 for large breeds and 1.17 for Jersey, g factor_3 is 0.038 for cows and 0.015 for non-lactating cows and growing cattle, g P/kg milk; milk; factor_3 is 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, g P/kg factor_3 is 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, g P/kg milk; factor_3 is 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, g milk; factor_3 is 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, ggg Na/kg factor_3 is 0.038 for lactating cows and 0.015 for non-lactating cows and growing cattle, Na/kg BW; factor_4 is 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, g 0.0154 for factor_3 non-lactating cows, glactating Ca/kg cows BW; factor_2 is for 0.98 for large cows breeds and 1.17 forcattle, Jersey, milk; factor_3 isis0.038 0.038 forlactating cows and 0.015 fornon-lactating non-lactating cows and growing cattle, P/kg milk; is for lactating 0.015 cows and growing cattle, g 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, g Na/kg BW; factor_4 6.1 for and 2.6 for non-lactating and growing g Na/kg BW; factor_4 is 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, g Na/kg BW; factor_4 is 6.1 for lactating cows and for non-lactating cows growing cattle, g BW; factor_4 is 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, gnonK/kg is dry matter intake which is2.6 set to 3.3% and1.67% 1.67% ofand BW for lactating and P/kg factor_3 0.038 for lactating cows and 0.015 for and non-lactating cows and growing g DM; Na/kg BW;and factor_4 isfor 6.1lactating for lactating cows and 2.6 for non-lactating cows and growing cattle, ggnonNa/kg BW; factor_4 is 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, gcattle, BW;milk; factor_4 is 6.1is cows and 2.6 for non-lactating cows and growing cattle, K/kg DM; and DMI dry matter intake which is set to 3.3% of BW for lactating and DM; and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonK/kg DM; DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonK/kg DM; and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonDM; and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonlactating cows, respectively. For growing cattle, DMI is calculated according to Equations 9.74 and Na/kg BW; factor_4 is 6.1 for lactating cows and 2.6 for non-lactating cows and growing cattle, g DM; and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonK/kg DM; and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and nonrespectively. For growing cattle, DMI is calculated according to Equations 9.74 and and DMI is dry matter intake which is set to 3.3% and 1.67% of BW for lactating and non-lactating lactating cows, respectively. For growing cattle, DMI is calculated according to Equations 9.74 and lactating cows, respectively. For growing cattle, DMI isiscalculated calculated according totoEquations Equations 9.74 and lactating cows, For cattle, DMI is according 9.74 and lactating cows, respectively. For growing cattle, calculated according Equations 9.74 and 9.75. K/kg and respectively. DMI isFor dry growing matter intake which isDMI set toiscalculated 3.3% andaccording 1.67% of to BW for lactating and non-9.75. lactating cows, respectively. For growing cattle, DMI according to Equations 9.74 lactating cows, respectively. Forgrowing growing cattle, DMI calculated according Equations 9.74 and cows,DM; respectively. cattle, DMI is is calculated toto Equations 9.74 and 9.75. 9.75. 9.75. lactating cows, respectively. For growing cattle, DMI is calculated according to Equations 9.74 and

The recommendations formacro macrominerals mineralsare arefrom fromNRC NRC(2001) (2001)with with modifications for phosphorus 9.75.recommendations for macro minerals are from NRC (2001) with modifications for phosphorus recommendations for macro minerals are from NRC (2001) with modifications for phosphorus The for modifications for phosphorus Therecommendations recommendations for macro minerals areNRC from NRC (2001) with modifications for phosphorus The for macro minerals are from (2001) with modifications for phosphorus recommendations for macro minerals are from NRC (2001) with modifications for phosphorus The for minerals are (2001) with modifications for phosphorus sincerecommendations content inmilk milk isdifferentiated differentiated between large breeds and Jersey based on the phosphorus content in milk is differentiated between large breeds and Jersey based on recommendations formacro macro minerals arefrom fromNRC NRC (2001) with modifications for phosphorus the phosphorus content in milk is differentiated between large breeds and Jersey based on since the phosphorus content in is between large breeds and Jersey based on since the phosphorus content in milk is differentiated between large breeds and Jersey based on since the phosphorus content in milk is differentiated between large breeds and Jersey based on the phosphorus content in milk is differentiated between large breeds and Jersey based on since the phosphorus content ininmilk isisSwedish differentiated between large breeds and Jersey based on Danish Aaes, 2004) and Swedish data(Lindmark-Månsson (Lindmark-Månsson et al.,.,2003). .,2003). 2003). In NRC The recommendations for macro minerals are from NRC (2001) with modifications forbased phosphorus et al In NRC (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson the phosphorus content milk differentiated between large breeds and Jersey on et al 2003). In NRC Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson Danish (Sehested and Aaes, 2004) and data et al ., In NRC Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson et al ., 2003). In NRC Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson et al ., 2003). In NRC Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson et al., 2003). In NRC (2001) Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson et al ., 2003). In NRC the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but inin (2001)the forabsorption absorption ofSwedish calciumand andphosphorus phosphorus depend on the but since phosphorus in milk is differentiated between large breeds and Jersey based on Danish (Sehested andcontent Aaes, 2004) and data (Lindmark-Månsson et al .,feedstuff, 2003). In NRC the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but in (2001) the coefficients for of calcium depend on the feedstuff, but in (2001) the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but in the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but inin the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but in NorFor (2001) the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but in these are fixed at 50% and 70%, respectively, based on values from NRC (2001). DMI is NorFor 50% and 70%, respectively, based on values from NRC (2001). DMI is Danish (Sehested and Aaes, 2004) and Swedish data (Lindmark-Månsson et al ., 2003). In NRC (2001) the coefficients for absorption of calcium and phosphorus depend on the feedstuff, but NorFor these are fixed at 50% and 70%, respectively, based on values from NRC (2001). DMI NorFor these are fixed at 50% and 70%, respectively, based on values from NRC (2001). DMI isis NorFor these are at 50% and 70%, respectively, based from NRC (2001). DMI is NorFor these are fixed fixed at 50% and 70%, respectively, basedon onvalues values from NRC (2001). DMI is is required these are fixed at 50% and 70%, respectively, on values from NRC (2001). DMI NorFor these are at 50% and based from NRC DMI isin (2001) the coefficients for absorption ofrespectively, calcium andbased phosphorus depend on the(2001). feedstuff, but NorFor these arefixed fixed at 50% and70%, 70%, respectively, basedon onvalues values from NRC (2001). DMI is 125 125 125 125 NorFor these are fixed at 50% and 70%, respectively, based on values from NRC (2001). DMI is 125 125 125 125 125 NorFor – The Nordic feed evaluation system 106

3444 3444 3445 3445 3446 3446 3447 3447 3448 3448 3449 3449 3450 3450 3451 3451 3452 3452 3453 3453 3454 3454 3455 3455 3456 3456 3457 3457 3458 3458 3459 3459 3460 3460 3461 3461 3462 3462 3463 3463 3464 3464 3465 3465 3466 3466 3467 3467 3468 3468 3469 3469 3470 3470 3471 3471 3472 3472 3473 3473 3474 3474 3475 3475 3476 3476 3477 3477 3478 3478 3479 3479 3480 3480 3481 3481 3482 3482 3483 3483 3484 3484

required to estimate the maintenance requirement of phosphorus and potassium and was set to 3.3% to BW estimate thecows maintenance requirement of NorFor phosphorus and potassium and wasset settofor to3.3% 3.3% of BW required todairy estimate theaccording maintenance requirement of phosphorus andfor potassium and was of for to NRC (2001). equations DMI were developed of for dairybased cows according toofNRC (2001). NorFor equations for DMI were developed for for growing forBW dairy cows according to NRC NorFor equations for DMIsystem. were developed growing cattle, on a subset the(2001). data used to develop the feed intake growing cattle,on based on a subset the data to develop feedintake intakesystem. system. cattle, based a subset of theofdata usedused to develop thethefeed DMIHeifer Steers = BW / 100 ⋅ (0.000004 ⋅ BW 22 − 0.0049 ⋅ BW + 3.1033) DMIHeifer Steers = BW / 100 ⋅ (0.000004 ⋅ BW − 0.0049 ⋅ BW + 3.1033)

[Eq. 9.74] [Eq. 9.74]

DMIBulls = BW/100⋅ (2.9 − 0.002⋅ BW) DMIBulls = BW/100⋅ (2.9 − 0.002⋅ BW)

[Eq. 9.75] [Eq. 9.75]

9.74 9.75

where DMIHeifersSteers and DMIBulls are estimated DMI for heifers, steers and bulls, kg DM/d; BW where DMIHeifersSteers and DMIBulls are estimated DMI for heifers, steers and bulls, kg DM/d; BW theweight weight theanimal, animal, where DMI and DMI HeifersSteers Bulls are estimated DMI for heifers, steers and bulls, kg DM/d; BW isisthe ofofthe kg.kg. is the weight of the animal, kg. Thecalcium calciumrecommendation recommendation growing cattle calculated from tabulated in Pehrson The forfor growing cattle was was calculated from tabulated values values in Pehrson The recommendation forusing growing cattle was calculated etal al. (1975)and andisiscalculated calculated following equation: from tabulated values in Pehrson et .calcium (1975) using thethe following equation: et al. (1975) and is calculated using the following equation: ⎛ 4.744 + 0.0361⋅ BW + 0.00000565⋅ BW2 ⎞ Ca _ int ake _ Min = ⎜⎛ 4.744 + 0.0361⋅ BW + 0.00000565⋅ BW2 ⎟⎞ Ca _ int ake _ Min = ⎜ + 0.0106 ⋅ ADG + 0.00000622⋅ ADG2 ⎟ ⎝⎜ 2 ⎠⎟ ⎠ ⎝ + 0.0106⋅ ADG + 0.00000622⋅ ADG

9.76

[Eq. 9.76] [Eq. 9.76]

where is the calcium recommendation for growing cattle, g/d; BW isg/d; the BW weight whereCa_intake_Min Ca_intake_Min is the calcium recommendation for growing cattle, is of the weight where Ca_intake_Min is the recommendation for growing cattle, g/d; BW the weight of the animal, kg; and is calcium theisaverage daily live weight gain of gain the animal, g/d.isHeifers late of the animal, kg; ADG and ADG the average daily live weight of the animal, g/d.inHeifers in late the animal, kg; and ADG is theCa average daily live weight gain of 9.48. the animal, g/d. Heifers in late gestation require anan additional supply for for gestation, Equation gestation require additional Ca supply gestation, Equation 9.48. gestation require an additional Ca supply for gestation, Equation 9.48. The recommendation for sulfur is set to 2 g/kg DM for dairy cows and growing cattle (NRC, 2001). The recommendation sulfur isto set2 to 2 g/kg DMdairy for cows dairyand cows and growing cattle2001). (NRC, 2001). The forfor sulfur is set g/kg DM cattleon(NRC, The recommendation macro mineral recommendations for dairy cowsfor and growing cattlegrowing that depend the BW and The macro mineral recommendations for dairy cows and growing cattle that depend on the BW and The macro mineral recommendations dairy ECM or ADG are illustrated in Tablesfor 9.13 andcows 9.14.and growing cattle that depend on the BW and ECMororADG ADGareare illustrated in Tables and 9.14. ECM illustrated in Tables 9.13 9.13 and 9.14. Table 9.13 Table 9.8.2 9.13 Micro minerals Table 9.14 Table Micro9.14 mineral recommendations for dairy cows and growing cattle are shown in Table 9.15. The

9.8.2 Micro copper (Cu)minerals requirement in the ration is dependent on the molybdenum and sulfur contents; the 9.8.2 Micro minerals

Micro mineral recommendations cows(Table and growing cattle areifshown in Table 9.15. The recommendation of 10 mg Cufor perdairy kg DM 9.15) is valid the ration contains less than 1 mg Micro mineral recommendations for2dairy cowsper and cattle1989). are shown Table 9.15. of The copper (Cu) requirement in the ration dependent on the(NRC, molybdenum andinsulfur contents; the molybdenum (Mo) and less than g is sulfur kggrowing DM Higher amounts molybdenum copper (Cu) requirement inCu theper ration is dependent on the molybdenum andcontains sulfur contents; recommendation of 10 mg kgcoefficient DM (Table 9.15) is valid ifbetween the ration less thanthe 1 be below and sulfur decrease the absorption of Cu. The ratio Cu and Mo should not recommendation of 10 mg Cu per kg2DM (Table 9.15) is valid if the ration contains less than 1 mg molybdenum (Mo) and less than g sulfur per kg DM (NRC, 1989). Higher amounts of 4:1molybdenum (NRC, 1989).(Mo) Synthesis ofthan the hormone the thyroid gland is inhibited by goitrogenic mg and less 2absorption g sulfurthyroxine per kg DMin(NRC, 1989). Higher amounts of molybdenum and sulfur decrease the coefficient of Cu. The ratio between Cu and Mo substances, which are common in rape seed products thatofcontain (Hermansen molybdenum sulfur the absorption Cu.thyroxine Theglucosinolates ratioinbetween Cu and Mois et al., should not be and below 4:1 decrease (NRC, 1989). Synthesiscoefficient of the hormone the thyroid gland 1995). Since rape seed products are commonly used in feed rations for dairy cows in Nordic should notbybegoitrogenic below 4:1substances, (NRC, 1989). Synthesis of theinhormone thecontain thyroid gland countries, is inhibited which are common rape seedthyroxine products in that the recommended iodine level was set to 1.0 mg/kg DM for dairy cows, although NRC (2001) only inhibited by goitrogenic substances, which are common in rape seed products that contain glucosinolates (Hermansen et al., 1995). Since rape seed products are commonly used in feed rations recommends 0.6Nordic mg/kgcountries, DM. the substantial amounts ofmg/kg goitrogenic substrates, the glucosinolates et al.,If1995). Sincecontains rape seed products commonly used inDM feedforrations for dairy cows (Hermansen in the ration recommended iodine level are was set to 1.0 dairy for dairy cows inNRC Nordic countries, theincreased recommended iodine level wasration set tocontains 1.0 mg/kg DM1999). for dairy iodine recommendation should be 2.0 mg/kg (Strudsholm etsubstantial al., cows, although (2001) only recommends 0.6tomg/kg DM. IfDM the cows, although NRC (2001) only recommends 0.6 mg/kg DM.should If the ration containstosubstantial amounts of goitrogenic substrates, the iodine recommendation be increased 2.0 mg/kg DM amounts of goitrogenic substrates, the iodine recommendation should be increased to 2.0 mg/kg DM (Strudsholm et al., 1999). al., 1999). (Strudsholm Table 9.13.etRecommendations (g/day) for macro minerals for growing cattle gaining 750 g/day1.

The recommendation includes requirements for maintenance and gain. 126

BW, kg

Calcium

Phosphorus126 Magnesium

Sodium

Potassium

200 kg 400 kg 600 kg

24 32 40

12 15 18

5 8 11

22 37 52

6 10 13

1 Phosphorus, magnesium, sodium and potassium recommendations are from NRC (2001). The calcium recommendation

is from Pehrson et al. (1975).

NorFor – The Nordic feed evaluation system

107

Table 9.14. Recommendations (g/day) for macro minerals for dairy cows depending on BW and milk yield (MY) or ECM1. BW, kg

MY or ECM, kg/d

Calcium

Phosphorus2 Magnesium Sodium

Potassium

430 430 430 600 600 600

17/22 24/31 31/40 25/25 35/35 45/45

80 102 124 109 136 163

49 60 72 63 77 91

151 166 181 201 218 235

29 37 46 35 44 53

34 40 46 43 50 57

1 The recommendations are from NRC (2001) with a slight modification for phosphorus, for which the milk content is

from Lindmark-Månsson et al. (2003) and Sehested and Aaes (2004). 2 The P recommendation is dependent on breed. Jersey cows (BW=430 kg) have a content of 1.17 g P/kg milk whereas 0.98 g P/kg milk is used for large breed cows (BW=600 kg). MY and ECM yield were assumed to be equal for large breeds but for Jersey cows MY was multiplied by 1.3 to get ECM yield.

Table 9.15. Recommendations for micro minerals (mg/kg DM) for dairy cows1 and growing cattle2.

Dairy cows Growing cattle

Cobalt

Copper3

Iodine4

Iron

Manganese Selenium5 Zinc

0.1 0.1

10 10

1.0 0.5

50 50

40 20

0.2 0.1

50 30

1 The

recommendations are based on NRC (2001) with the exception of selenium. recommendations are based on NRC (2000). 3 The recommendation is dependent on the content of molybdenum and sulfur in the ration. 4 The recommendation considers some content of goitrogenic substrates, which are sometimes present in rape seed products (Hermansen et al., 1995). 5 The recommendation is from former Swedish recommendations (Spörndly, 2003). 2 The

9.8.3 Cation anion difference In order to reduce the risk of milk fever, the CAD has been introduced as a nutritional variable in feeding, especially for dry cows. The NorFor recommendation for CAD is -150 to 0 mEq/kg DM in the dry period, or at least for the last 3 to 4 weeks of the dry period (Horst et al., 1997; Moore et al., 2000; Kristensen, 2005). For lactating cows the recommendation is 200 to 450 mEq/kg DM since Borucki Castro et al. (2004) found that increasing CAD from 140 to 450 mEq/kg DM had no effect on DMI or milk yield. The CAD should never exceed 450 mEq/kg DM and this is important when adding buffer to the ration (Kristensen, 2005). Research results in the literature are equivocal about whether a reduction in dietary K and a moderate CAD in the dry period are sufficient to avert milk fever (Overton and Waldron, 2004). Potassium is therefore included in the calculation of CAD (see Chapter 6) and NorFor does not provide any recommendations for K in relation to CAD.

9.9 Vitamins Cattle require vitamins A, D, E and K. However, the dietary requirements are only absolute for vitamins A and E, since vitamin K is synthesized by ruminal and intestinal bacteria, and vitamin D is synthesized by the action of ultraviolet radiation in the skin (NRC, 2001). NorFor has 108

NorFor – The Nordic feed evaluation system

recommendations for the supply of fat soluble vitamins (A, D and E) to dairy cows and growing cattle, which are from NRC (2001) and shown in Table 9.16. The recommendations from NRC (2001) are based on supplemental vitamins and, thus, do not take into account the natural content of vitamins in feedstuffs. NorFor includes the natural content of vitamins A, D and E when calculating the supply of these vitamins from a feed ration and recommendations are therefore a total recommendation. The beta-carotene in feedstuffs is converted to vitamin A such that 1 mg beta-carotene corresponds to 400 IU of vitamin A (Strudsholm et al., 1999). Table 9.16. Recommendations for vitamins A, D and E supplies (IU/kg BW) to dairy cows and growing cattle are based on NRC (2001). Requirements are expressed as total amounts, i.e. supplemental plus vitamins provided by feedstuffs.

Dairy cows Growing cattle

Vitamin A1, 2

Vitamin D3

Vitamin E4

110 80/110

30 10

0.8/1.6 0.3/1.6

1

Beta-carotene can be converted to vitamin A and the total requirement can be met from beta-carotene. 1 mg betacarotene corresponds to 400 IU vitamin A (Strudsholm et al., 1999). 2 The requirement of 80 IU/kg BW is for bulls, steers and non-pregnant heifers, while the requirement of 110 IU/kg BW is for pregnant heifers (NRC, 2001). 3 The requirement is from Strudsholm et al. (1999). 4 The requirement of 0.8 IU/kg BW is for lactating cows, while the requirement of 1.6 IU/kg BW is for dry cows and heifers in late gestation. The requirement for growing cattle is modified from NRC (2001), which suggests a range from 0.26 to 0.8 IU/kg BW, where the lowest value refers to grazing animals and the highest value to animals fed stored forages.

References Berg, J. and T. Matre, 2001. Produksion av storfekjøtt. Landbruktforlaget. Norway. (In Norgwegian). Berg, J. and H. Volden, 2004. Økologisk storfekjøtt. Buskap nr 4. 16-17. (In Norwegian). Borucki Castro, S.I., L.E. Phillip, V. Girard and A. Tremblay, 2004. Altering dietary cation-anion difference in lactating dairy cows to reduce phosphorus excretion to the environment. Journal of Dairy Science 87: 1751-1757. Børsting, C.F., J.E. Hermansen and M.R. Weisbjerg, 2003. Fedtforsyningens betydning for mælkeproduktionen. In: Strudsholm, F. and K. Sejrsen (eds.). Kvægets ernæring og fysiologi. DJF-rapport nr. 54, Danmarks Jordbrugsforskning, Denmark, pp. 133-152. (In Danish). Danfær, A., 1983. Næringsstoffernes intermediære omsætning. In: Østergaard, V. and A. Neimann-Sørensen (eds.). Optimale foderrationer til malkekoen. Foderværdi, foderoptagelse, omsætning og produktion. Beretning nr. 551 fra Statens Husdyrbrugsforsøg, Denmark 12.1-12.65. (In Danish). Garcia, F., J. Agabriel and D. Micol, 2007. Alimentation des bovins en croissance et à l’engrais. In: INRA (ed.). Alimentation des bovin, ovins et caprin. Besoin des animaux - Valeurs des aliments. Edition Quæ, Versaille, France, pp. 89-120. (In French). Hermansen, J., O. Aaes, S. Ostersen and M. Vestergaard, 1995. Rapsprodukter til malkekøer – mælkeydelse og mælkekvalitet. Forskningsrapport nr. 29 fra Statens Husdyrbrugsforsøg, Denmark. (In Danish). Hole, J.R, 1985. The nutritive value of silage made from Poa pratensis ssp. Alpigena and Phlenum pratense. III. Growing bulls fed silage made from the regrowth of Poa pratensis or Phlenum pratense. Report No. 232. Department of Animal Nutrition, Agricultural University of Norway, Aas, Norway. 17 pp. Homb, T., 1974. Grovfôr-kraftfôr-forholdet ved intensivt oppdrett av 14 måneder gamle slakteokser. Institutt for husdyrernæring og fôringslære, Norges landbrukshøgskole. Melding nr. 164, Norway. 18 pp. (In Norwegian). Horst, R.L., J.P. Goff, T.A. Reinhardt and D.R. Buxton, 1997. Strategies for preventing milk fever in dairy cattle. Journal of Dairy Science 80: 1269-1280.

NorFor – The Nordic feed evaluation system

109

Hvelplund, T. and J. Madsen, 1990. A study of the quantitative nitrogen metabolism in the gastro-intestinal tract, and the resultant new protein evaluation system for ruminants. The AAT-PBV system. The Royal Veterinarian and Agricultural University, Copenhagen, Denmark. Hymøller, L., M.R. Weisbjerg, P. Nørgaard, C.F. Børsting, and N.B. Kristensen, 2005. Majsensilage til malkekøer. DJF rapport nr. 65. Danmarks Jordbrugsforskning, Denmark. (In Danish). INRA (Institut National de la Recherche Agronomique), 1989. Ruminant nutrition: Recommended allowances and feed tables. John Libbey and Co Ltd, London, UK. 389 pp. Johansen, A., 1992. A comparison between meadow fescue and timothy silage. Doctor Scientarium Thesis. Agricultural University of Norway, Aas, Norway. 108 pp. Kristensen, N.B., 2005. Brug CAB til optimering af foderrationer til malkekøer. I: Fodringsdag - Temadag om aktuelle fodringsspørgsmål. 30 august. Dansk Kvæg, Denmark, pp. 23-42. (In Danish). Kristensen, V.F., 1997. Optimal proteinforsyning. In: Malkekøernes ernæring. Intern rapport nr. 88. Statens Husdyrbrugsforsøg, Denmark, pp. 46-55. (In Danish). Kristensen, V.F., 1995. Forudsigelse af foderoptagelsen hos malkekøer. Intern rapport nr. 61. Statens Husdyrbrugsforsøg, Denmark. 28 pp. (In Danish). Kristensen, V.F., 1983. Styring af foderoptagelsen ved hjælp af foderrationens sammensætning og valg af fodringsprincip. In: Østergaard, V. and A. Neimann-Sørensen (eds.). Optimale foderrationer til malkekoen. Foderværdi, foderoptagelse, omsætning og produktion. Beretning nr. 551 fra Statens Husdyrbrugsforsøg, Denmark. 7.1-7.35. (In Danish). Lindmark-Månsson, H., R. Fondén and H.E. Pettersson, 2003. Composition of Swedish dairy milk. International Dairy Journal 13: 409-425. MacRae, J.C., P.J. Buttery and D.E. Beever, 1988. Nutrient interactions in the dairy cow. In: P.C. Garnsworthy (ed.). Nutrition and lactation in the dairy cow. Butterworths, London, UK, pp. 55-75. Madsen, J., 1985. The basis for the proposed Nordic protein evaluation system for ruminants. The AAT-PBV system. Protein Evaluation for Ruminants. Acta Agriculturae Scandinavica Supplement 25: 13-19. Madsen, J., L. Misciatteli, V.F. Kristensen and T. Hvelplund, 2003. Proteinforsyning til malkekøer. In: Strudsholm, F. and K. Sejrsen (eds.). Kvægets ernæring og fysiologi. DJF-rapport nr. 54. Danmarks Jordbrugsforskning, Denmark, pp. 113-121. (In Danish). Minde, A. and A.J. Rygh, 1997. Metoder for å bestemme nedbrytingshastighet, passasjehastighet og vomfordøyelighet av NDF hos mjølkeku. Master thesis. Norges landbrukshøgskole, Norway. 118 pp. (In Norwegian). Moore, S.J., M.J. Van de Haar, B.K. Sharma, T.E. Pilbeam, D.K. Beede, H.F. Bucholtz, J.S. Liesman, R.L. Horst and J.P. Goff, 2000. Effects of altering dietary cation-anion difference on calcium and energy metabolism in peripartum cows. Journal of Dairy Science 83: 2095-2104. Mydland, L.T., 2005, Molecular and chemical characterization of the rumen microbiota and quantification of rumen microbial protein synthesis. Ph.D. thesis. Norwegian University of Life Sciences, Aas, Norway. 155 pp. Nielsen, H.M., N.C. Friggens, P. Løvendahl, J. Jensen and K.L. Ingvartsen, 2003. Influence of breed, parity, and stage of lactation on lactational performance and relationship between body fatness and live weight. Livestock Production Science 79: 119-133. NRC (National Research Council), 1985. Nutrient Requirement of Dairy Cattle. Fifth revised edition. National Academy Press, Washington, DC, USA. NRC, 1989. Nutrient Requirement of Dairy Cattle. Sixth revised edition. National Academy Press, Washington, DC, USA. NRC, 2000. Nutrient Requirement of Beef Cattle. Seventh revised edition. National Academy Press, Washington, DC, USA. NRC, 2001. Nutrient Requirement of Dairy Cattle. Seventh revised edition. National Academy Press, Washington, DC, USA. Nørgaard, P., 1986. Physical structure of feeds for dairy cows - A new system for evaluation of the physical structure in feedstuffs and rations for dairy cows. In: Neimann-Sørensen A. (ed.). New developments and future perspectives in research of rumen function. Brussels, Belgium, pp. 85-107. Olsson, I., 1987. Effect of protein supply on the performance of intensively reared bulls - evaluation of the DCP and the Nordic AAT-PBV protein evaluation system. Dissertation. Report 164. Swedish University of Agricultural Sciences, Uppsala, Sweden. Olsson, I. and J.E. Lindberg, 1985. Performance of bull calves fed rations with different calculated amino acid flow to the duodenum. Acta Agriculturae Scandinavica Supplement 25.

110

NorFor – The Nordic feed evaluation system

Overton, T.R. and M.R. Waldron, 2004. Nutritional management of transition dairy cows: Strategies to optimize metabolic health. Journal of Dairy Science 87: E105-E119. Pehrson, B., V. Kossila, J.T. Hansen and N.E. Hansen, 1975. Förslag til normer för makro- och mikromineraler till nötkreatur och svin. Foderjournalen, pp. 55-106. (In Swedish). Prestløkken, E., 1999. Protein value of expander-treated barley and oats for ruminants. Ph.D. thesis. Agricultural University of Norway, Aas, Norway. 137 pp. Rajala-Schultz, P.J., W.J.A. Saville, G.S. Frazer and T.E. Wittum, 2001. Association between milk urea nitrogen and fertility in Ohio dairy cows. Journal of Dairy Science 84: 482-489. Randby, Å.T., 2000a. Ensilert eller propionsyrekonservert bygg til mjølkekyr og kastrater. Husdyrforsøksmøtet 2000. Institutt for husdyrfag, NLH, Aas, Norway, pp. 409-412. (In: Norwegian). Randby, Å.T., 2000b. Effekten av økende dosering med et maursyreholdig ensileringsmiddel på næringsverdien av surfôret i produksjonsforsøk med kyr og kastrater. Hellerud rapport nr 25, Aas, Norway. 37 pp. (In Norwegian). Randby, Å.T., 2003. Høstetid og fôrkvalitet. Grønn kunnskap. 7: 27-43. (In Norwegian). Reynolds, C.K., 2000. Forage evaluation using measurements of energy metabolism. In: Givens, D.I., E. Owen, R.F.E. Axford and H.M. Omed (eds.). Forage evaluation in ruminant nutrition. CAB International, Wallingford, UK, pp. 95-111. Robelin, J. and R. Daenicke, 1980. Variations of net requirements for cattle growth with live weight, live weight gain, breed and sex. Annales de Zootechnie 29: 99-118. Sehested, J. and O. Aaes, 2004. Fosfor til kvæg. DJF-rapport nr. 60. Danmarks Jordbrugsforskning, Denmark. (In Danish). Schei, I., H. Volden and L. Bævre, 2005. Effects of energy balance and metabolizable protein level on tissue mobilization and milk performance of dairy cows in early lactation. Livestock Production Science 95: 35-47. Sjaunja, L.O., L. Bævre, L. Junkkarainen, J. Pedersen and J. Setälä, 1990. A Nordic proposition for an energy corrected milk (ECM) formula. 26th session of the International Committee for Recording the Productivity of Milk Animals (ICRPMA). Skaara, E., 1994. The effect of N-fertilization on the protein value of silage assist by the AAT/PBV system. Doctor Scientarium Thesis. Agricultural University of Norway, Aas, Norway. 95 pp. Spörndly, R., 2003. Fodermedelstabeller för idisslare. Rapport nr. 257. Department of Animal Nutrition and Management, Swedish University of Agricultural Sciences, Uppsala, Sweden. 96 pp. (In Swedish). Strudsholm, F., O. Aaes, J. Madsen, V.F. Kristensen, H.R. Andersen, T. Hvelplund and S. Østergaard, 1999. Danske fodernormer til kvæg. Rapport nr. 84. Landbrugets Rådgivningscenter, Århus, Denmark. (In Danish). Subnel, A.P.J., R.G.M. Meijer, W.M. Van Straalen and S. Tamminga, 1994. Efficiency of milk protein production in the DVE protein evaluation system. Livestock Production Science 40: 215-224. Tamminga, S., W.M. van Straalen, A.P.J. Subnel, R.G.M Meijer, C. Steg, J.G. Wever and M.C. Block, 1994. The Dutch protein evaluation system: the DVE/OEB-system. Livestock Production Science 40: 139-155. Thuen, E., 1989. Influence of source and level of dietary protein on milk yield, milk composition and some rumen and blood components in dairy cows. Doctor Scientarium Thesis. Agricultural University of Norway, Norway. 151 pp. Van Es, A.J.H., 1975. Feed evaluation for dairy cows. Livestock Production Science 2: 95-107. Van Es, A.J.H., 1978. Feed evaluation for ruminants. I. The system in use from May 1978 onwards in the Netherlands. Livestock Production Science 5: 331-345. Van Straalen, W.M., 1994. Modelling of nitrogen flow and excretion in dairy cows. Ph.D. thesis, Department of Animal Nutrition, Wageningen Agricultural University, Wageningen, the Netherlands. 205 pp. Volden, H., 1990. Response of dietary AAT density on milk yield - Review of production trials in Norway. Proceedings of the seminar on fat and protein feeding to the dairy cow. October 15-16, Eskilstuna, Sweden, pp. 81-89. Volden, H., 1999. Effects of level of feeding and ruminally undegraded protein on ruminal bacterial protein synthesis, escape of dietary protein, intestinal amino acid profile, and performance of dairy cows. Journal of Animal Science 77: 1905-1918. Volden, H., 2001. Utvikling av et mekanistisk system for vurdering av fôr til drøvtyggere, AAT-modellen. In: Fôropptak og fôrmiddelvurdering hos drøvtyggere. Fagseminar 18.-19. september 2001. Quality Hotell Halvorsbøle, Jevnaker, Norway. 30 pp. (In Norwegian). Volden, H. and O.M. Harstad, 2002. Effects of duodenal amino acid and starch infusion on milk production and nitrogen balance in dairy cows. Journal of Animal Science, Supplement 1 80: 321. Volden, H. and J. Berg, 2005. Nytt fôrvurderingssystem. Buskap nr 5: 12-14. (In Norwegian).

NorFor – The Nordic feed evaluation system

111

3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955

10. Prediction of voluntary feed intake 10.0 Prediction of voluntary feed intake

H. Volden, Volden,N. N.II Nielsen, Nielsen,M. M.Åkerlind, Åkerlind,M.M. Larsen, Havrevoll and A.J. Rygh H. Larsen, Ø. Ø. Havrevoll, A. J. Rygh Describingthe thenutrient nutrient variables in aand diettheir andinteractions their interactions is an important part of ration Describing variables in a diet is an important part of ration formulationasasproductivity productivity of the dairy is sensitive the profile of the nutrients formulation of the dairy cow cow is sensitive to the to profile of the nutrients absorbed.absorbed. However, ofof feed intake is probably the most important determinant of production. However,the theprediction prediction feed intake is probably the most important determinant of production. Feed by by animal andand feedfeed characteristics. The most Feedintake intakeisisprimarily primarilyinfluenced influenced animal characteristics. The important most important animal animal characteristics areand BWphysiological and physiological including stage lactation, milkproduction, stage characteristics are BW state,state, including stage of of lactation, milk production, stage of gestation, live weight gain and body condition score. Feed characteristics of gestation, live weight gain and body condition score. Feed characteristics such as digestibility such as digestibility and fiber content have both a strong influence on rumen fill and, hence, feed and fibre content have both a strong influence on rumen fill and, hence, feed intake (Kristensen, intake (Kristensen, 1983). However, several studies (Rinne et al., 2002; Garmo et al., 2007) have 1983). However, several studies (Rinne et al., 2002; Garmo et al., 2007) have shown that cows may shown that cows may stop eating before the fill capacity of the rumen is reached. This has been stop eating before theregulation fill capacity of which the rumen reached. This hastobeen attributed attributed to metabolic (MR), is alsoisan important factor consider when to metabolic regulationfeed (MR), which is alsointake an important factor to consider when predicting intake. predicting intake. Physical regulation is related to the ruminal NDF poolfeed (Bosch et Physical intake regulation ruminal pool (Bosch al., 1992; Rinne et of al.,the 2002) and is al ., 1992; Rinne et is al.,related 2002) to andthe is partly an NDF indirect effect of theetenergy concentration partly an indirect effect ofinthe energy concentration the diet because DMI declines in a curvilinear diet because DMI declines a curvilinear manner withofincreasing energy density of the diet manner with increasing energy density of the diet (Mertens, 1994). (Mertens, 1994). Several toto predict DMI forfor practical ration formulation. These Severalsystems systemshave havebeen beendeveloped developed predict DMI practical ration formulation. These systems systems vary both complexity in their complexity and of choice of variables included (animal, nutritional and vary both in their and choice variables included (animal, nutritional and environmental). environmental). In the Danish1983) (Kristensen, 1983)(Jarrige, and French (Jarrige, 1986) systems, intake from In the Danish (Kristensen, and French 1986) systems, feed intakefeed is predicted is predicted from dietary fill values and animal fill capacity. In the system developed by Mertens dietary fill values and animal fill capacity. In the system developed by Mertens (1987), intake is (1987), intake is regulated both physically by the diet (NDF) and metabolically by the energy regulated both physically by the diet (NDF) and metabolically by the energy demand of the animal. demand of the animal. A similar approach is also used to predict DMI in NorFor. Each individual A similar approach is also used(FV; to predict Each individual is capacity assigned a basic feed is assigned a basic fill value sectionDMI 6.2) in andNorFor. the animal is assigned an feed intake fill value (FV; Section 6.2) and the animal is assigned an intake capacity (IC) expressed in the same (IC) expressed in the same units as the feed. When predicting DMI, the following general units as the When predicting DMI, the following general equation must be fulfilled: equation mustfeed. be fulfilled: IC= FV_ intake

[Eq.10.1]

10.1

whereIC ICisisthe theanimal animalintake intake capacity FV_intake is total the total intake expressed where capacity andand FV_intake is the feedfeed intake expressed in fill in fill units, whichwhich is calculated as described in Section 6.2. units, is calculated as described in section 6.2. Despite thethe feed FVFV andand thusthus the FV_intake is notisconstant; it varies Despitethis thisgeneral generalrelationship, relationship, feed the FV_intake not constant; it varies with with the concentrate substitution rate(SubR), (SubR),which which affects affects roughage if the the concentrate substitution rate roughageintake. intake.Moreover, Moreover, if the energy energy concentration the ration is high, relative the animal’s requirements, above concentration of the of ration is high, relative to thetoanimal’s requirements, the the above general equation general equation will overestimate feed intake. An MR factor should therefore be included in the will overestimate feed intake. An MR factor should therefore be included in the equation when equation when formulating rations to meet specific animal production levels (e.g. weight gain or formulating rations to meet specific animal production levels (e.g. weight gain or milk yield). milk yield).

In NorFor, NorFor,the theICICofofeach each animal category estimated feeding experiments in which In animal category has has beenbeen estimated from from feeding experiments in which animals were fed total mixed rations ad libitum or fed only roughage ad libitum with fixed animals were fed total mixed rations ad libitum or fed only roughage ad libitum with fixed levelslevels of concentrate. tried to toevaluate evaluatehow howdifferent different types of roughage of concentrate.These Theseexperiments experiments have tried types of roughage are are utilized and also the interaction between roughage and differentcategories categories utilized and also the interaction between roughage andconcentrate concentrate by by different of of animals. From each experiment, the FVs of the different feeds were calculated assuming that the FV_intake represented the IC of the animals.

10.1. Dairy cows

149

A database of 183 dietary treatments was compiled from Norwegian, Danish, Swedish and Icelandic production experiments. (Table 10.1) in which grass or maize silages constituted the sole or dominant forage. The breeds represented in the database were the dominant Nordic dairy breeds, i.e. DH, SR, SH, NR and IB (Table 3.2). Animal and feed characteristics for the large dairy breeds, JER and IB are presented in Table 10.2, 10.3 and 10.4, respectively. The total DMI varied from 9.7 to 32.3 kg/d H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_10, © Wageningen Academic Publishers 2011

113

Table 10.1. Studies used to develop the feed intake module for dairy cows in NorFor. Bertilsson and Murphy, 2003 Bossen et al., 2010a Bossen et al., 2010b Hetta et al., 2007 Hymøller et al., 2005 Johansen, 1992 Kristensen, 1999 Kristensen et al., 2003 Kristensen et al., unpublished data Misciattelli et al., 2003 Mo and Randby, 1986 Mould, 1996 Nordang and Mould, 1994 Randby, 1992

Randby, 1997 Randby, 1999a Randby, 1999b Randby, 2000 Randby, 2002 Randby, 2007 Randby and Mo, 1985 Randby and Selmer-Olsen, 1997a Randby et al., 1999 Rikarðsson, 1997 Rikarðsson, 2002 Schei et al., 2005 Sveinbjörnsson and Harðarsson, 2008 Weisbjerg et al., unpublished data

Table 10.2. Descriptive statistics of data used to develop the equation for predicting intake capacity of dairy cows: large dairy breeds. Item

Average

SD1

Maximum

Minimum

Days in milk Body weight, kg ECM2, kg/d Dry matter intake, kg/d Concentrate proportion, kg/kg DM Roughage intake, kg DM/d Roughage basis fill value, FV/kg DM Starch + sugars, kg/d

123 577 29.9 19.7 0.42 11.2 0.51 2.7

65 59 5.8 3.4 0.101 2.3 0.031 0.7

328 775 48.9 32.3 0.84 19.4 0.76 5.0

7 437 15.1 9.7 0.14 3.7 0.42 1.5

1 SD=standard

2 ECM=energy

deviation. corrected milk.

Table 10.3. Descriptive statistics of data used to develop the equation for predicting intake capacity of dairy cows: Jersey cows. Item

Average

SD1

Maximum

Minimum

Days in milk Body weight, kg ECM2, kg/d Dry matter intake, kg/d Concentrate proportion, kg/kg DM Roughage intake, kg DM/d Roughage basis fill value, FV/kg DM Starch + sugars, kg/d

119 458 25.8 16.0 0.40 9.6 0.44 3.6

77 43 6.0 2.8 0.09 2.3 0.020 0.7

294 598 47.2 23.8 0.59 18.7 0.48 5.9

7 308 4.0 5.3 0.11 3.1 0.41 1.2

1SD=standard

2ECM=energy

114

deviation. corrected milk.

NorFor – The Nordic feed evaluation system

Table 10.4. Descriptive statistics of data used to develop the equation for predicting intake capacity of dairy cows: Icelandic cows. Data from Baldursdóttir (2010). 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000

1

Item From each experiment, the FVs of the different Averagefeeds were SD calculatedMaximum animals. assuming thatMinimum the FV_intake represented the IC of the animals.

Days in milk 58 34 147 7 Body weight, kg 439 47 619 300 10.1. ECM2Dairy , kg/d cows 20.4 4.7 41.0 7.6 Dry matterofintake, kg/d treatments was compiled15.1 2.1 Danish, Swedish 25.3 and 7.1 A database 183 dietary from Norwegian, Icelandic production experiments. (Table 10.1) in which maize silages constituted the 0.28 Concentrate proportion, kg/kg DM 0.41 grass or0.09 0.49 sole or dominant forage. The breeds represented in the Roughage intake, kg DM/d 9.0 database were 1.8 the dominant 15.6Nordic dairy4.0 breeds, i.e., basis DH, SR, NRFV/kg and IBDM (Table 3.2). Animal characteristics0.62 for the large 0.48 Roughage fill SH, value, 0.53 and feed 0.025 dairy breeds, JER kg/d and IB are presented in Table 10.2, The total DMI Starch + sugars, 2.510.3 and 10.4, 0.5 respectively.3.1 0.5

varied from 9.7 to 32.3 kg/d and the concentrate intake from 2.7 to 19.8 kg DM/d. The roughage FV ranged from 0.42 to 0.76 per kg DM. A multiple regression approach was used to derive the 1SD=standard deviation. equation for predicting animal IC. 2 ECM=energy corrected milk.

Table 10.1 Table 10.2 and the concentrate intake from 2.7 to 19.8 kg DM/d. The roughage FV ranged from 0.42 to 0.76 Table 10.3 per kg10.4 DM. A multiple regression approach was used to derive the equation for predicting animal IC. Table

Based 1995), thethe following modified multiple regression Basedon onthe thework workofofKristensen Kristensen(1983, (1983, 1995), following modified multiple regression equation equation used to describe IC: was usedwas to describe IC: IC _ cow = ( a ⋅ DIM b ⋅ e c ⋅ DIM − DIM − d + e ⋅ ECM + ( BW − f ) ⋅ g )

10.2

[Eq. 10.2]

where intake capacity of lactating dairy dairy cows; cows; DIM isDIM days is in milk; ECM is the whereIC_cow IC_cowisisthethe intake capacity of lactating days in milk; ECM is the energy is is thethe animal body weight, kg; and a, b, a, c, b, d, c, e, d, f, ge,are energycorrected correctedmilk, milk,kg/d; kg/d;BW BW animal body weight, kg; and f, g are regression regression coefficients Table 10.3). coefficients (see Table(see 10.5). ® The et et al.,al., 1998) in Microsoft Excel, whichwhich employs a generalized reduced reduced ® Excel, The Solver Solvertool tool(Fylstra (Fylstra 1998) in Microsoft employs a generalized gradient non-linear optimization code (Lasdon et al., 1978; Sveinbjörnsson et al., 2006), was gradient non-linear optimization code (Lasdon et al., 1978; Sveinbjörnsson et al., 2006), was used used to parameterize the equation and fit regression coefficients to the above model by to parameterize themean equation and fit regression coefficients the above model by minimizing the minimizing the root square prediction error (RMSPE) for to animal IC. The regression root mean square prediction error (RMSPE) for animal IC. The regression coefficients for predicting coefficients for predicting the IC of lactating dairy cows are presented in Table 10.5.

the IC of lactating dairy cows are presented in Table 10.5.

Table 10.5

Figure 10.1 illustrates how the IC coefficients of large dairyused breeds varies with lactation period, with a 305-d Table 10.5. Multiple regression to predict dairy cow intake capacity (IC). yield potential of between 6 000 and 10 000 kg ECM. For Icelandic cows the same approach as 1 in Table 10.5 is from described above was used to parameterize the IC,regression and the parameter values Cow category Multiple coefficients the work of Baldursdóttir (2010).

a

b

c

d

e

f

-0.0006 -0.0006 -0.0004 -0.0004 -0.0014 -0.0013

0.55 0.55 0.25 0.15 0.65 0.14

0.091 500 0.091 575 [Eq. 10.3] 0.110 360 0.110 405 0.015 400 0.035 523

The DMI is also affected by animal activity, and when cows are on pasture, or confined in a loose-house system, an IC exercise value is added (Kristensen, 1995):

Primiparous large dairy breeds Multiparous large breeds ) ⋅ factorR IC = (IC _ animal + ICdairy _ exercize Jersey primiparous cows Jersey multiparous cows Icelandic primiparous cows Icelandic multiparous cows

1

2.59 2.82 2.25 2.40 4.07 4.77

0.134 0.134 0.134 0.134 0.087 150 0.071

g 0.006 0.006 0.006 0.006 0.002 0.0013

IC=(a·DIMb·ec·DIM – DIM–d + e·ECM + (BW – f)·g).

NorFor – The Nordic feed evaluation system

115

4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043

used to parameterize the equation and fit regression coefficients to the above model by minimizing the root mean square prediction error (RMSPE) for animal IC. The regression coefficients for predicting the IC of lactating dairy cows are presented in Table 10.5. Table 10.5

Figure 10.1 illustrates how the IC of large dairy breeds varies with lactation period, with a 305-d

Figure 10.1 illustrates how the6,000 IC ofand large10,000 dairy breeds varies with lactationcows period, a 305-d yield potential of between kg ECM. For Icelandic thewith same approach as yield potential of between 6 000 and 10 000 kgthe ECM. For Icelandic cows the sameinapproach as is from described above was used to parameterize IC, and the parameter values Table 10.5 described above was used to parameterize the IC, and the parameter values in Table 10.5 is from the work of Baldursdóttir (2010). the work of Baldursdóttir (2010).

The DMI DMIisisalso alsoaffected affected by ⋅animal when arepasture, or confined (0.45activity, ) ⎞or confined − activity, 0.09 ⋅ pH +when ACF ⋅ (1.cows 5− 0.223 ⋅on pHpasture, ⎛ LAF The and)and cows are on in a in a loose⎟ ⎜ by animal house system, an ICanexercise value is added (Kristensen, 1995): loose-house system, IC exercise value is added (Kristensen, 1995): [Eq. 5.5] DM corr = DM uncorr + ⎜ + PRF ⋅ (1.4 − 0.182 ⋅ pH) + BUF ⋅ (1.9 − 0.272 ⋅ pH)⎟ ⎟ ⎜ ⎟ ⎜ ) ⋅ factorR IC = (IC _ animal + is ICeither _ exercize [Eq. 10.3] ⋅ 0.6in Equation 10.2 or IC_dry described 10.3 ALF + NH 3N where IC_animal described in Equation ⎠ ⎝ +IC_cow 10.5 or 10.6; IC_exercize is 0 if the cows are tied up and 0.15 if the cows are in a loose housing where IC_animal is either IC_cow in Equation 10.2 or IC_dryindescribed Equation 10.5 system or on pasture; factorR is an addescribed libitum correction factor as described Equationin 10.4. or 10.6; IC_exercize is 0 if the cows are tied150 up and 0.15 if the cows are in a loose housing system Figure 10.1 or on pasture; factorR is an ad libitum correction factor as described in Equation 10.4. ⎛ ACF ⋅ (1.5 − 0.223 ⋅ pH ) + PRF ⋅ (1.4 − 0.182 ⋅ pH )⎞ ⎟ 5.6] = DMfed +⎜ DM corrare If libitum, e.g.e.g. , in in situations where roughage intake is not optimal, due to due to[Eq. ⎜ + BUF If cows cows arenot notuncorr fedadad libitum, situations not optimal, feed )where ⋅ (1.4do − 0not ⋅ pHmaximal + ALFroughage +roughage .182 NH 3 N ⋅ intake 0intake, .6 ⎟⎠ is ⎝ feed shortage or feeding systems that allow an adjustment factor shortage or feeding systems that do not allow maximal roughage intake, an adjustment factor is is introduced: introduced: ⎛ roughage _ appetite ⎞ − 13 ⎟ + 0.8502 factorR = 0.0214 ⋅ ⎜ 5 ⎝ ⎠

[Eq.10.4]

sid _ mAA = r _ mAA ⋅ 0.85

10.4

[Eq. 7.43]

where factorR is an adjustment factor for the intake capacity; roughage_appetite is the appetite

where factorR is an adjustment factor for the intake capacity; roughage_appetite is the appetite proportion,%%ofofexpected expected libitum intake. adad libitum intake. proportion,

In non-lactating non-lactatingcows, cowsone an of equation on the work of based Kristensen used to predict IC: In the twobased alternative equations, on the (1995) work ofisKristensen (1995) are used to predict IC: IC _ dry = (5 + ((BW − f ) ⋅ 0.006)) ⋅ factor _ h IC _ dry = (5 + ((BW − 575) ⋅ 0.006)) ⋅ factor _ h

[Eq.10.5]

[Eq.10.5] 10.5

where IC_dry is the intake capacity of non-lactating cows, FV/d; BW is the body weight, kg; f is the [Eq.10.6] Jersey=0.83). IC _ dry = (factor _ i f+in (YHerd 6500)factor_h ⋅ 0.0003 + (is BW f ) ⋅ 0.006 ) ⋅ factor _ hfactor (large breeds=1; regression factor Table−10.5; the−breed correction DM _ int ake ⎛ ST _ SUof_ non-lactating ⎞ ⎛ ST _isSU ⎞ where IC_dry the+ intake factor_i the age and−gestation = 0is.97 − 0.2119 cows, ⋅⎜ FV _ SubR 0.562 ⋅ capacity 0.1932 5.122 ⎟ ⋅ 0.05 ⎜ ⎟ ⋅ 0.1 −FV/d; 1000 1000 ⎝ ⎠ ⎝ ⎠ stage factor (less than 260 days of gestation for primiparous cows = 5.3; multiparous cows = 5.8; 10.0 >260 days of gestation: primiparous cows = 4.8; multiparous cows = 5.3);IC BW is the body 10,000 kg ECM weight, kg; f 9.0 is the regression factor f in table 10.5; factor_h is the breed correction factor (large IC 6,000 kg ECM breeds = 1; Jersey = 0.83). YHerd is the yield level in the herd, 305-d average. [Eq.10.8] 8.0 When DMI is calculated, the IC is set equal to the FV_intake. Several factors affect the 7.0 FV_intake and it isPage generally accepted that increases in concentrate supplementation decrease the Equations on 85 in the proof: roughage intake 6.0 (Thomas, 1987). This is not only due to the filling effect of additional concentre but also by the accompanying reduction in ruminal NDF digestibility (Khalili and Huhtanen, 1991; Huhtanen 5.0 and Jaakkola, 1993; Stensig et al., 1998), which results in a higher roughage FV. 4.184FVk mg However, the concentrate substitution rate is not a constant (INRA, 1989); the true roughage ⋅ which for cows factor_1= 0.29256, for heifers and steers factor _ 1 = 90 ⋅ feed but 4.0 size varies with ration and composition. Moreover, MR is not related to each individual 1000 k m , for bulls instead to the entire ration. Bosch et al. (1992), Rinne et al. (2002) and Eriksen and Ness (2007) 3.0 k mgsize, indicating that the DMI is .184 pool have shown that cows do not eat to a constant ruminal4NDF factor _ 1concentration = 91⋅ ⋅ be used metabolically regulated and that dietary NDF can an indirect of of dairy breeds or crossbreed bulls ofmeasure beef breeds 2.0 1000 k m andasfor MR (Mertens, 1987; The same that was the development of animal 0 1994). 25 50 75 dataset 100 125 150 used 175in200 225 250 275 300 IC was k mg also used to establish a relationship between FV_intake and the effects of SubR and MR. When 4.184 Days factorinto _ 1 =account 100 ⋅ SubR⋅and MR, ; the 4.184/1000 converts to MJ;equation: the km is used to taking the factor FV_intake can in bemilk calculated fromkcal the general 1000 k m

Intake capacity (IC)

3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4001 4000 4002 4003 4004 4005 4006 4007 4008 4009 4010

Figure 10.1. throughout a 305-d lactation period at two production levels calculate FV_ intake = ∑Feed DMIi ⋅intake FVi + ∑capacity DMIj ⋅ FV(IC) [Eq.10.7] j ⋅FV_ SubR + FV_ MR (6,000 and 10,000 kg ECM). i j

4044 151

116

NorFor – The Nordic feed evaluation system

4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4045 4042 4046 4043 4047 4048 4044 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086

where IC_dry is the intake capacity of non-lactating cows, FV/d; factor_i is the age and gestation stage factor (less than 260 days of gestation for primiparous cows = 5.3; multiparous cows = 5.8; >260 days of gestation: primiparous cows = 4.8; multiparous cows = 5.3); BW is the body weight, kg; f is the regression factor f in table 10.5; factor_h is the breed correction factor (large breeds = 1; Jersey = 0.83). YHerd is the yield level in the herd, 305-d average.

When DMI is calculated, the IC is set equal to the FV_intake. Several factors affect the FV_intake

and it DMI is generally accepted increases supplementation decrease the roughage When is calculated, the ICthat is set equal to in theconcentrate FV_intake. Several factors affect the intake (Thomas, Thisaccepted is not only to the in filling effect ofsupplementation additional concentrate FV_intake and it is1987). generally thatdue increases concentrate decrease but the also by roughage intake (Thomas, 1987). is not onlydigestibility due to the filling effect of Huhtanen, additional concentre the accompanying reduction in This ruminal NDF (Khalili and 1991; Huhtanen but by the accompanying in ruminal NDF digestibility (Khalili and Huhtanen, andalso Jaakkola, 1993; Stensigreduction et al., 1998), which results in a higher roughage FV. However, the 1991; Huhtanen and Jaakkola, 1993; et al ., 1998), whichthe results a higherFV roughage concentrate substitution rate is not aStensig constant (INRA, 1989); true in roughage varies FV. with ration However, the concentrate substitution rate notrelated a constant (INRA, 1989); the FVthe entire size and composition. Moreover, MR is is not to each individual feedtrue butroughage instead to varies with ration size and composition. Moreover, MR is not related to each individual feed but ration. Bosch et al. (1992), Rinne et al. (2002) and Eriksen and Ness (2007) have shown that cows instead to the entire ration. Bosch et al. (1992), Rinne et al. (2002) and Eriksen and Ness (2007) do not eat to a constant ruminal NDF pool size, indicating that the DMI is metabolically regulated have shown that cows do not eat to a constant ruminal NDF pool size, indicating that the DMI is and that dietary NDF concentration be⋅ (used measure of MR (Mertens, 1987; ) + ACF ) ⎞ ⋅ (0.45 − 0.09 ⋅ pHcan 1.5 − 0.as 223an ⋅ pHindirect ⎛ LAF metabolically regulated ⎜ and that dietary NDF concentration can⎟be used as an indirect measure of 1994). The same dataset that was used in the development of animal IC wasofalso used establish a [Eq. 5.5] MR (Mertens, 1987; 1994). The same dataset that was used in the development animal ICtowas ⎟ ⎜ ( ) ( ) DM corr = DM uncorr + + PRF ⋅ 1.4 − 0.182 ⋅ pH + BUF ⋅ 1.9 − 0.272 ⋅ pH ⎟ ⎜ relationship between FV_intake and the effects of SubR and MR. When taking into account SubR also used to establish⎜ a relationship between FV_intake and the ⎟effects of SubR and MR. When + NH3 N ⋅ 0.6 ⎠equation: ⎝ + ALF and MR, FV_intake can be calculated fromcan thebe general taking intothe account SubR and MR, the FV_intake calculated from the general equation: where FV_intake is the feed intake expressed in fill units; FVi is the fill value of the i’th concentrate FV/kg DM; FVj is the basic fill value of the j’th roughage, Equation 6.10 and FV_ intake =feed, ∑ DMI [Eq.10.7] 10.6 i ⋅ FVi + ∑ DMIj ⋅ FVj ⋅FV_ SubR + FV_ MR 6.11; FV_SubR rate factor, 0 to1, Equation 10.8 and 10.15; FV_MR is the the i is the substitution j metabolic regulation ⎛factor, 10.9 and ⋅10.16. ) + PRF (1.4 − 0.182 ⋅ pH )⎞ is the fill value of the i’th concentrate ACF ⋅ (1Equations .5 − 0.223 ⋅ pH where ⎟i [Eq. 5.6] = DM uncorr is + ⎜the feed intake expressed in fill units; FV DM corrFV_intake ⎜ + BUF ⋅ (1.4 − 0.182 ⋅ pH ) + ALF + NH N ⋅ 0.6 ⎟

feed, FV/kg DM; the related basis fill value of the j’th Equation 6.10 and 6.11; FV_SubR 3 roughage, ⎝ j is ⎠ Traditionally, SubRFV has been to the effect 151of concentrate per se. However, it may be is the substitution rate factor, 0 to1, Equation and 10.14; FV_MR is the metabolic difficult to define a feedstuff as strictly ‘roughage’10.7 or ‘concentrate’. In NorFor, variations in theregulation factor, Equations 10.8 and 10.15. SubR are explained by changes in the NDF digestion in the rumen and the effect of rapidly degradable CHO’s on digestion in the rumen. This is similar to the effect used to explain the [Eq. sid _ mAA =ofr ruminal _ mAA .85 been correction NDF digestibility, RLI of (Equation 7.15)per in Chapter 7. Therefore, Traditionally, SubR⋅ 0has related toi.e. thethe effect concentrate se. However, it 7.43] may the be difficult suitability includingas dietary ST‘roughage’ and SU in the of theInsubstitution factor (FV_SubR) to define aoffeedstuff strictly or estimation ‘concentrate’. NorFor, variations in the SubR are was assessed, ExcelinSolver tool to develop a SubR function based on CHO’s explained by employing changes inthe theMicrosoft® NDF digestion the rumen and the effect of rapidly degradable the and SU supply and its effect on roughage FV:to explain the correction of ruminal NDF on dietary ruminalSTdigestion. This is similar to the effect used

digestibility, i.e. the RLI (Equation 7.15) in Chapter 7. Therefore, the suitability of including dietary _ SU _ DM_ h ⎞ ⎛ ST _ SU _ int ake ⎞ IC __dry + ((+in BW − f )⋅ ⎛⎜⋅estimation 0ST .006 )) ⋅ factor FV SR ==(SU 05.97 0.562 .2119 − 5.122 05 [Eq.10.5] ⎟ ⋅ 0.1 − 0.1932 ⋅factor ⎜ ⎟ ⋅ 0.assessed, ST and the of− 0the substitution (FV_SubR) was employing the 1000 1000 ⎝ ⎠ ⎝ ⎠ ® Microsoft Excel Solver tool to develop a SubR function based on the dietary ST and SU supply [Eq.10:8] and its effect on roughage FV:

where FV_SubR is the ⎛roughage substitution⎞ correction factor, 0 to 1; ST_SU_DM is the ST _ SU _ DM ⎛ ST _ SU _ int ake ⎞ 5.122 + 0.562 ⋅ ⎜ sugars in the−diet, ⋅⎜ −the FV _ SubR =of 0.97 0.2119 .1 − 0.1932 ⎟ ⋅ 0DM; sugar ⎟ ⋅ 0.05and proportion starch and g/g and ST_SU_intake is starch 1000 1000 ⎝ ⎠ ⎝ ⎠ intake, g/d.

10.7

where FV_SubR is the roughage substitution correction factor, 0 to 1; ST_SU_DM is the proportion

Figure 10.2 illustrates how FV_SubR is affected by the amount and starch proportion ST +intake, SU in g/d. of starch and sugars in thethe diet, g/g DM; and ST_SU_intake is the and of sugar [Eq.10.8] the ration; an increased proportion of SU + ST in the diet has a negative effect on the roughage FV and 10.2 this effect is alsohow related to the SU +isST intake.by Thethe latter effectand may explain why Figure illustrates theproof: FV_SubR affected amount proportion of ST + SU in the Equations on Pageconcentrate 85 in the replacing a starchy with a soluble fibre-based concentrate has been have a FV and ration; an increased proportion of SU + ST in the diet has a negative effect found on thetoroughage highly variable effects on forage intake (Huhtanen, 1993; Van Vuuren et al., 1993; Petit and this effect is also related to the SU + ST intake. The latter effect may explain why replacing a starchy Tremblay, 1995; Huhtanen et al., 1995; Hymøller et al., 2005). concentrate with a soluble fibre-based concentrate has been found to 4have variable effects .184 ak highly mg factorPetit _ 1 = and 90 ⋅ Tremblay, ⋅ which for cows factor_1= 0.29256, for heifers andetsteers on forage intake (Huhtanen, 1993; Van Vuuren al., 1993; 1995; Huhtanen et , for bulls 1000 k Figure 10.2 m

al., 1995; Hymøller et al., 2005).

4.184 k mg factor _ 1 = 91 ⋅ causing ⋅ In MR isordefined as a regulatory factor cows to stop eating before reaching their of NorFor, dairy breeds crossbreed 1000 k m and for bulls of beef breeds In NorFor, is defined as a regulatory causingrelated cows factor to stop eating before reaching their full ruminal MR FV capacity. Physiologically, thisfactor is an animal rather than a direct k therefore 4.capacity. 184 mg Physiologically, ruminal effect, and it is expected to influence animal IC. However, in NorFor it is full ruminal FV this is an animal related factor rather than a direct factor _ 1 = 100 ⋅ ⋅ ; the factor 4.184/1000 converts kcal to MJ; the km is used to addedruminal to the feed of the equation whentocalculating forIC. computational k m expected effect, andside it is 1000 therefore influence DMI animal However, inreasons. NorForThe it isFV_MR added toisthe feed

calculated side of theas: equation when calculating DMI for computational reasons. The FV_MR is calculated as: calculate ⎛ ⎜ 2.530 FV _ MR = ⎜1.453 − ⎜ ⎛ (0.466 − FV _ r ) ⎞ ⎜ ⎜ 0.065 ⎟⎠ 1 + e⎝ ⎝

⎞ ⎟ ⎟ ⋅ IC ⎟ 8 ⎟ ⎠

[Eq.10.9]

where FV_MR is the roughage metabolic correction factor; FV_r is the mean basis roughage fill value in the diet, FV/kg DM, Equation 6.10; IC is animal intake capacity, Equation 10.3; and the IC/8 ratio is the adjustment factor for animal IC level.

NorFor – The Nordic feed evaluation system 152

10.8

117

1.2

1.1

1 FV_SubRr

Starch + sugar (g/g DM)

0.05

0.40 0.35 0.30 0.25 0.20 0.15 0.10

0.9

1

2

0.8 8 7 5 6 4 3 Starch + sugar (kg/d)

Figure 10.2. A principle figure describing the effect of starch + sugar intake and their proportion in the diet on roughage substitution rate (FV_SubR) factor. where FV_MR is the roughage metabolic correction factor; FV_r is the mean basis roughage fill value in the diet, FV/kg DM, Equation 6.10; IC is animal intake capacity, Equation 10.3; and the IC/8 ratio is the adjustment factor for animal IC level. Dividing IC by 8 adjusts the standard IC so that FV_MR can be used across dairy breeds. Mertens (1994) demonstrated that the threshold between an energy regulated intake and a fill limited intake was dependent on the milk yield. Maximum fill and minimum energy concentration was achieved at a higher NDF diet concentration (440 g/kg DM versus 320 g NDF/kg DM) for low-yielding cows (20 kg ECM) than for high yielding cows (40 kg ECM). Figure 10.3 shows the relationship between FV_MR and FV_r at two levels of IC (6 and 8, which represent different milk yields). At an FV_r of 0.48 the FV_MR is zero, and with a higher and increasing FV_r a low yielding cow will reach its IC relative faster than a high yielding cow. However, at low FV_r values an opposite effect is achieved, which probably is explained by a higher diet energy density and thus a metabolic regulation of the intake (Mertens, 1994). Parameterization of Equation 10.6 showed that FV_MR makes a highly significant contribution to the prediction of roughage intake. The RMSPE was reduced by 19% when the FV_MR was introduced into the FV_intake equation. The need for this correction factor demonstrates that ruminal fill is not the only factor limiting roughage intake, as there are other important metabolic regulations (Forbes, 1995). Since FV_MR is related to roughage FV, and thus to the concentration of NDF in roughage, the rate of rumen particle size reduction may also partly explain this correction factor, since physical particle breakdown is not accounted for when calculating roughage FV. The introduction of FV_SubR and FV_MR in the prediction of DMI demonstrates that roughage FV is not a constant. Ration formulation is therefore performed by an iteration process in NorFor, using a computer program capable of solving non-linear optimization problems (see Chapter 15). The intake sub-model is further evaluated in Chapter 14.

118

NorFor – The Nordic feed evaluation system

Metabolic regulation factor (FV_MR)

1.5 1.3

IC=8 IC=6

1.0 0.8 0.5 0.3 0.0 0.25 0.30 -0.3

0.35

0.40

0.45

0.50

0.55

0.60

0.65

0.70

0.75

-0.5 -0.8 -1.0 -1.3

Roughage basis fill value (FV_r)

Figure 10.3. Relationship between roughage basis fill value (FV_r) and metabolic regulation factor (FV_MR) at two levels of intake capacity (IC=6 and 8).

10.2 Growing cattle The feeding strategies used in Nordic beef production vary widely among countries, from an intensively reared beef animals fed up to 90% concentrates to an extensive system where only small amounts (0 to10%) of concentrates are used. Hence, there are substantial variations in diet composition, which makes it challenging to develop a voluntary feed intake system that is relevant to all production systems. To develop equations for predicting the IC of bulls, steers and heifers, a database was compiled from 32 Danish, Swedish and Norwegian experiments (Table 10.6), in which cattle were fed a wide range of diets, including 145, 36 and 115 treatments for bulls, steers and heifers, respectively. Animal and feed characteristics for the bull, steer and heifer datasets are presented in Tables 10.7, 10.8 and 10.9, respectively. In the bull dataset, DMI varied from 4.5 to 11.7 kg/d (concentrate from 0 to 6.8 kg DM/d) and ADG ranged from 429 to 1,578 g, while in the heifers dataset, ADG varied from 378 to 1196 g/d and the DMI from 3.3 to 11.4 kg/d. 4131 4131 4132 4132 4133 4133 4134 4134 4135 4135 4136 4136 4137 4137 4138 4138 4139 4139 4140 4140 4141 4141 4142 4142 4143 4143 4144 4144 4145 4146 4147 4148 4149 4150 4151

For all all three three animal categories, categories,variations variationsininIC IC wereexplained explainedby bychanges changesininBW BW andADG. ADG.The The For ® changes inand Forequations all threeanimal animal categories, variations in were IC were explained by BW and ADG. The IC ® Excel using the same IC were parameterized using the Solver tool in Microsoft IC equationswere wereparameterized parameterized using the Solver tool in in Microsoft Excel usingusing the same ® Excel equations using the Solver tool Microsoft the same approach approach as for the dairy cows. In bulls the IC is calculated from: approach as for the dairy cows. In bulls the IC is calculated from:

as for the dairy cows. In bulls the IC is calculated from: − 20 IC__ bull bull == 00..006544 006544⋅⋅BW BW++00..0007337 0007337⋅⋅ADG ADG++ − 20 IC .9552⋅ ⋅BW BW 00.9552

[Eq.10.10] 10.10] [Eq.

10.9

whereIC_bull IC_bullisisisthe the feed intake capacity of bulls; BW the animal body weight, and ADG is where IC_bull the feed intake capacity ofbulls; bulls; BWis isthe theisanimal animal bodyweight, weight, kg;and andkg; ADG where feed intake capacity of BW body kg; ADG the average daily gain, g/d. is the average daily gain, g/d. is the average daily gain, g/d. The steer dataset included data obtained from only 3636 dietary treatments andand preliminary Thesteer steerdataset datasetincluded includeddata dataobtained obtained from only dietary treatments a preliminary evaluation The from only 36 dietary treatments and a apreliminary evaluation showed that the IC ICand ofsteers steers and heifers werecomparable. comparable. These twodatasets datasets were combined evaluation showed the of and heifers were These two were showed that the ICthat of steers heifers were comparable. These two datasets were therefore therefore combined in in the theof parameterization ofderived IC.The TheIC ICwas was derived from thesedata datafrom: from: therefore combined parameterization of IC. derived these in the parameterization IC. The IC was from these datafrom from: ⎛ −−33 ⎞ ⎞⎟ ⋅ IC _ gest IC __ heifer heifer == ⎛⎜⎜00..007236 007236⋅⋅BW BW++00..0005781 0005781⋅ ⋅ADG ADG++ IC ⋅ IC _ gest 0 . 9552 BW⎟⎠ ⎠ ⎝ 0 . 9552 ⋅ ⋅BW ⎝

[Eq.10.11] 10.11] [Eq.

where IC_heifer IC_heifer is is the the feed feed intake intakecapacity capacityof ofsteers steersand andheifers; heifers;BW BWisisthe theanimal animalbody bodyweight, weight, and ADG ADG is is the the average average daily dailygain, gain,g/d. g/d.IC_gest IC_gestisisthe theintake intakecapacity capacityadjustments adjustmentsfor forheifers heifers kg; and due to gestation gestation described described in in Equation Equation10.12. 10.12.

NorFor – The Nordic feed evaluation system

The gestation gestation correction correction factor factorisiscalculated calculatedas: as:

10.10

119

Table 10.6. Studies used to develop the feed intake module for growing cattle in NorFor.

4131 4131 4132 4132 4133 4133 4134 4134 4135 4135 4136 4136 4137 4137 4138 4138 4139 4139 4140 4140 4141 4141 4142 4142 4143 4143 4144 4144 4145 4145 4146 4146 4147 4147 4148 4148 4149 4149 4150 4150 4151 4151 4152 4152 4153 4153 4154 4154 4155 4155 4156 4156 4157 4157 4158 4158 4159 4159 4160 4160 4161 4161 4162 4162 4163 4163 4164

Bulls

Steers

Heifers

Andersen et al., 1993a Andersen et al., 1993b Berg, 2004 Berg and Volden, 2004 Damgaard and Hansen, 1992 Jørgensen et al., 2007 Kirkland et al., 2006 Matre, 1984 Nadeau et al., 2002 Olsson and Murphy, unpublished data Selmer-Olsen, 1994 Therkildsen et al., 1998 Vestergaard et al., 1993b Vestergaard et al., unpublished data Nadeau, unpublished data

Matre, 1984 Matre, 1987 Randby, 1999c Randby and Selmer-Olsen, 1997b Rustas et al., 2003 Selmer-Olsen, 1994

Andersen et al., 2001 Havrevoll, unpublished data Hessle et al., 2007 Ingvartsen et al., 1988 Mäntysaari, 1993 Rustas, 2009 Sejrsen and Larsen, 1978 Olsson, unpublished data Selmer-Olsen, 1994 Vestergaard et al., 1993a

For all three animal categories, variations in IC were explained by changes in BW and ADG. The For all10.7. three Descriptive animal categories, variations IC were explained changesfor in predicting BW and ADG. Thecapacity Table statistics of datainused to develop theby equation intake IC equations were parameterized using the Solver tool in Microsoft®®Excel using the same IC equations were parameterized using the Solver tool in Microsoft Excel using the same of bulls. approach as for the dairy cows. In bulls the IC is calculated from: approach as for the dairy cows. In bulls the IC is calculated from:

Item Average − 20 − 20 IC _ bull = 0.006544⋅ BW + 0.0007337⋅ ADG + IC _ bull = 0.006544⋅ BW + 0.0007337⋅ ADG +0.9552 ⋅ BW 0.9552 Age at start of experiment, d 224 ⋅ BW

SD1

Maximum

[Eq. 10.10] [Eq. 10.10]

Minimum

109 366 83 Age at end of experiment, d 392 127 676 188 where IC_bull is the feed intake capacity of bulls; BW is the animal body weight, kg; and ADG where IC_bull isweight, the feedkg intake capacity of bulls;362 BW is the animal body 130 body weight, 634kg; and ADG 178 isAverage the average daily gain, g/d. is the average daily gain, g/d. Average live weight gain, g/d 1,165 222 1,578 429 The dataset included obtained from only 367.3 dietary treatments a preliminary Drysteer matter intake, kg/d data 2.1 and 11.7 4.5 The steer dataset included data obtained from only 36 dietary treatments and a preliminary evaluation showed thatkg theDM/d IC of steers and heifers were comparable. These two datasets were 0 Concentrate intake, 3.0 1.9 6.8 evaluationcombined showed that theparameterization IC of steers andof heifers were Thesethese two datasets were therefore in the IC. The IC comparable. was derived from data from: Concentrate proportion, kg/kg DM 0.41 0.29 therefore combined in the parameterization of IC. The IC was derived from these0.95 data from: 0 Roughage intake, kg DM/d 4.3 2.1 9.5 0.2 −3 ⎞ ⎛ IC _ heifer = ⎜⎛basis 0.007236 0005781DM + 0.FV/kg ⋅ ADG + ⎟ ⎞⋅ IC _ gest 0.06 [Eq. 10.11] Roughage fill⋅ BW value, 0.68 0.41 − 3 0.54 [Eq. 10.11] 0.9552 ⋅ BW IC _ heifer = ⎝⎜ 0.007236 ⋅ BW + 0.0005781 ⋅ ADG + ⎠ ⎟ ⋅ IC _ gest 0.9552 ⋅ BW ⎠ ⎝

1

SD=standard deviation. where IC_heifer is the feed intake capacity of steers and heifers; BW is the animal body weight, where is the feed intake capacity of steersisand BW is the animal body kg; andIC_heifer ADG is the average daily gain, g/d. IC_gest theheifers; intake capacity adjustments for weight, heifers kg; and ADG is described the gain, g/d. IC_gest is the capacity heifers where IC_heifer is average the feed intake capacity of steers andintake heifers; BW isadjustments the animalfor body weight, kg; due to gestation in daily Equation 10.12. due gestation Equation 10.12. and to ADG is thedescribed average in daily gain, g/d. IC_gest is the intake capacity adjustments for heifers due The gestationdescribed correctionin factor is calculated to gestation Equation 10.11. as: The gestation correction factor is based on Kristensen and The gestation correction factor is calculated as: Ingvartsen (2003) and calculated as: 1 IC _ gest = [Eq. 10.12] 1 ⎛ gest _ day − 357 ⎞ IC _ gest = 10.11 [Eq. 10.12] ⎜⎛ gest _ day − 357⎟⎞ 36 ⎠⎟ 1 + e ⎝⎜ 36 ⎠ 1 + e⎝ where gestation correction on the capacity; gest_day is the day of gestation. whereIC_gest IC_gestisisthethe gestation correction on intake the intake capacity; gest_day is the day of gestation. where IC_gest is the gestation correction on the intake capacity; gest_day is the day of gestation. Evaluation 10.10 andand 10.11 showed that they for the Jersey breed. EvaluationofofEquations Equations 10.9 10.10 showed thatwere theyunsuitable were unsuitable for the Jersey breed. Evaluationthe of Equations 10.10 andwas 10.11 showed theyyoung were unsuitable for the Jersey Therefore, following equation modified forthat Jersey stock, irrespective of sex:breed. Therefore, the following equation was modified for Jersey young stock, irrespective of Therefore, the following equation was modified for Jersey young stock, irrespective of sex: sex: −2 IC _ jersey = 0.0067 ⋅ BW + 0.0005781⋅ ADG + 10.12 [Eq. 10.13] −2 IC _ jersey = 0.0067 ⋅ BW + 0.0005781⋅ ADG +0.9552 ⋅ BW [Eq. 10.13] 0.9552 ⋅ BW where IC_jersey is the feed intake capacity of Jersey young stock; BW is the animal body weight, 120and IC_jersey NorFor – The Nordic feed evaluation where is the feed daily intakegain, capacity stock; BW is the animal body weight,system kg; ADG is the average g/d. of Jersey young kg; and ADG is the average daily gain, g/d. The IC of growing cattle is also corrected for activity:

Table 10.8. Descriptive statistics of data used to develop the equation for predicting intake capacity of steers. 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170

1

Item Average SDchanges in BW Maximum For all three animal categories, variations in IC were explained by and ADG.Minimum The IC equations were parameterized using the Solver tool in Microsoft® Excel using the same approach as for dairy cows. from: Age at start of the experiment, d In bulls the IC is calculated 237 125 450 125 Age at end of experiment, d 349 127 555 208 − 20 Average body weight, kg 319 104 470 158 IC _ bull = 0.006544⋅ BW + 0.0007337⋅ ADG + [Eq. 10.10] Average live weight gain, g/d 0.9552 748 ⋅ BW 252 1,196 187 Dry matter intake, kg/d 6.3 1.8 9.5 4.1 where IC_bullintake, is the feed intake capacity of bulls; BW kg; and ADG0 Concentrate kg DM/d 1.0is the animal 0.9body weight,2.6 isConcentrate the average proportion, daily gain, g/d. kg/kg DM 0.21 0.18 0.54 0 Roughage intake, kg DM/d 5.3 2.5 9.5 2.2 The steer dataset included data obtained from only 36 dietary treatments and a preliminary Roughage basis fill value, FV/kg DM 0.58 0.07 0.67 0.48 evaluation showed that the IC of steers and heifers were comparable. These two datasets were therefore combined in the parameterization of IC. The IC was derived from these data from: 1 SD=standard deviation.

−3 ⎞ ⎛ IC _ heifer = ⎜ 0.007236 ⋅ BW + 0.0005781 ⋅ ADG + ⎟ ⋅ IC _ gest 0.9552 ⋅ BW ⎠ ⎝

[Eq. 10.11]

Table 10.9. Descriptive statistics of data used to develop the equation for predicting intake capacity of heifers. where IC_heifer is the feed intake capacity of steers and heifers; BW is the animal body weight, kg; and ADG is the average daily gain, g/d. IC_gest is the intake capacity adjustments for heifers due Itemto gestation described in Equation 10.12. Average SD1 Maximum Minimum The correction factor Agegestation at start of experiment, d is calculated as: 368

124 491 122 Age at end of experiment, d 468 103 634 310 1 Average 379 108 598[Eq. 10.12] 150 IC _ gest =body weight, kg day −g/d 357 ⎞ ⎛ gest _gain, Average live weight 798 205 1,196 378 ⎜ ⎟ 36 ⎠ Dry matter1intake, 7.3 1.6 11.4 3.3 + e ⎝ kg/d Concentrate intake, kg DM/d 1.3 1.3 5.1 0 where IC_gestproportion, is the gestation correction capacity; gest_day Concentrate kg/kg DM on the intake 0.19 0.20 is the day 0.75of gestation.0.0 Roughage intake, kg DM/d 6.0 2.1 11.4 1.2 Evaluation Equations 10.10 and 10.11 they were unsuitable for the0.67 Jersey breed.0.49 Roughageof basis fill value, FV/kg DM showed that 0.54 0.04 Therefore, the following equation was modified for Jersey young stock, irrespective of sex: 1

SD=standard deviation.

IC _ jersey = 0.0067 ⋅ BW + 0.0005781⋅ ADG +

−2 0.9552 ⋅ BW

[Eq. 10.13]

where IC_jersey is the feed intake capacity of Jersey young stock; BW is the animal body weight,

where IC_jersey is the feed intake capacity of Jersey young stock; BW is the animal body weight, kg; and andADG ADGisisthe theaverage average daily gain, kg; daily gain, g/d.g/d.

The IC ICofofgrowing growingcattle cattleis is also corrected activity: The also corrected for for activity: IC = IC _ animal ⋅ IC _ exercize

[Eq. 10.14]

10.13

where capacity; IC_animal is either IC_bull, IC_heifer or IC_jersey described in whereIC ICisisthe theintake intake capacity; IC_animal is either IC_bull, IC_heifer or IC_jersey described in Equation Equation10.10, 10.9, 10.11 10.10 and and10.13, 10.12,respectively. respectively.

Based on the range of animal characteristics in the datasets, it was decided that the IC equations are 154 with body weights of ≥80 kg, and for Jersey with valid for animals of large dairy and beef breeds body weights ≥60 kg. Figure 10.4 illustrates how the IC in growing bulls changes with BW and ADG over time up to the target slaughter weight of 300 kg at 16 months of age. The ADG changes in a curvilinear manner NorFor – The Nordic feed evaluation system

121

4191 4190 4192 4193 4194 4195 4191 4196 4192 4197 4193 4198 4194 4199 4195 4200 4196 4201 4197 4202 4198 4203 4199 4204 4200 4205 4201 4206 4202 4207 4203 4208 4204 4209 4205 4206 4210 4207 4208 4209 4210

1,400

5.0

Body weight ADG IC

4.5 4.0

1,200

3.5

1000

3.0 2.5

800

Intake capacity

Body weight (kg) or daily weight gain (g)

4171 4172 4173 4174 4175 4171 4176 4172 4177 4173 4178 4174 4179 4175 4180 4176 4181 4177 4182 4178 4183 4179 4184 4180 4185 4181 4186 4182 4187 4183 4188 4184 4189 4185 4186 4187 4190 4188 4189

1,600

Based on the range of animal characteristics in the datasets, it was decided that the IC 2.0equations 600 are valid for animals of large dairy and beef breeds with body weights of ≥ 80 kg, and for Jersey 1.5 with body weights ≥400 60 kg. 1.0 Figureon 10.4 how thecharacteristics IC in growinginbulls changes with and ADG over time up to 200 Based theillustrates range of animal the datasets, it wasBW decided that the IC equations 0.5 the valid targetfor slaughter ofdairy 300 kg at beef 16 months age.body Theweights ADG changes a curvilinear are animalsweight of large and breeds of with of ≥ 80inkg, and for Jersey 0 kg.highest ADG is achieved at an age of 330 days. Since the 0.0 manner with time and IC is with body weights ≥ 60the 91 and 122 ADG, 152 182 213 changes 243 274 over 304 time 334 365 426 456fashion. 486 dependent on both the BW it also in a 395 sigmoidal dayswith BW and ADG over time up to Figure 10.4 illustrates how the IC in growing bullsAge, changes Figure 10.4 the target slaughter weight of 300 kg at 16 months of age. The ADG changes in a curvilinear Figure 10.4. Changes inhighest intake ADG capacity (IC) over time bulldays. withSince a target slaughter weight of manner with time and the is achieved at an agefor ofa330 the IC is 300 kgdairy (587 kg body at 16 months of age.isover As for feed and intake in growing cattle dependent the SubR and MR effects. To dependent oncows, both the weight) BW ADG, it also changes time inon a sigmoidal fashion. parameterize these two variables for the growing cattle, the same approach was used as for lactating dairy However, in the cattle with time and cows. the highest ADGtheis concentrate achieved atcharacteristics an age of 330available days. Since thegrowing IC is dependent on both Figure 10.4 datasets incomplete, made it difficult develop a SubR equation based on starch and the BW were and ADG, it alsowhich changes over time intoa sigmoidal fashion. sugar Consequently, SubR equation developed from the information on the As forinformation. dairy cows, the feed intake the in growing cattle iswas dependent on the SubR and MR effects. To proportion of these in the diet. Thegrowing following equation is used for growing cattle, of all parameterize two for the cattle, theissame approach as and for As for dairy concentrate cows, thevariables feed intake in growing cattle dependent onwas theused SubR MR effects. types: dairy cows. However, the concentrate characteristics available in the growing cattle lactating To parameterize these two variables for the growing cattle, the same approach was used as for datasets were incomplete, which made it difficult to develop a SubR equation based on starch and lactating dairy cows. However, the concentrate characteristics available in the growing cattle 2 sugar information. Consequently, the SubR equation was developed from the information on the ⎛ ⎞ concit_ difficult share datasets were to develop a SubR equation based on starch and ⎜ incomplete, which made ⎟ 0 .9646 0 .706 The ⋅ proportion of⎜ concentrate in the− diet. following equation⎟ is used for growing cattle, of all 100 [Eq.information 10.15] sugar information. Consequently, the SubR equation was developed from the on the FV _ SubR = ⎜ types: 2 ⎟

⎜ concentrate concin_ the sharediet. The following ⎛ conc _ share ⎞ ⎟ proportion of equation is used for growing cattle, of all types: 1 − 0 .7512 ⋅ − 0 .2085 ⋅ ⎜ ⎟

⎜ ⎟ 100 100 ⎝ ⎠ ⎠2 ⎝ ⎛ ⎞ conc _ share ⎜ ⎟ 0 .9646 − 0 .706 ⋅ ⎜ ⎟ 100 correction factor, where FV_SubR is the roughage substitution FV _ SubR =⎜ 2 ⎟ conc _ share _ share proportion of⎜ 1concentrate in the diet on a DM basis, %. ⎛ conc ⎞ ⎟ − 0 .2085 ⋅ ⎜ ⎟ ⎟ ⎜ − 0 .7512 ⋅ 100 100 ⎝ ⎠ ⎠ ⎝

[Eq. 10.15] 0 to 1; and conc_share is the

10.14

Figure 10.5 shows how the FV_SubR factor changes with the proportion of concentrate in the where FV_SubR is the roughage substitution correction factor, 0 to 1; and conc_share is the proportion diet. The growingiscattle diets vary widely in the concentrate to 0roughage (0 to 90is%the where FV_SubR the diet roughage substitution correction factor, to 1; andratio conc_share of concentrate in the on a the DM basis, %. concentrate DM basis),inthus, correction proportion ofonconcentrate the diet FV_SubR on a DM basis, %. factor must be able to correct the roughage intake over a large range of concentrate intake values. The FV_SubR factor, and Figure10.5 10.5shows shows howthethe FV_SubR factor changes with theconcentrate proportion consequently, the roughage FV, increase considerably when inof theconcentrate diet exceeds Figure how FV_SubR factor changes with thethe proportion of concentrate in the in the diet. The The growing cattle diets vary widely in in thetheconcentrate totoroughage (0 to to 90% 60%. Thisgrowing indicates that the voluntary roughage intake is likely to be lowratio at high levels diet. cattle diets vary widely concentrate roughage ratio (0 90 %ofconcentrate on DM basis),on thus, FV_SubR correction mustfactor be able to correct roughage concentrate. concentrate DMthe basis), thus, the FV_SubRfactor correction must be able the to correct the intake over a large range of concentrate values. The FV_SubR factor, consequently, the roughage FV, roughage intake over a largeintake range of concentrate intake values. Theand FV_SubR factor, and Figure 10.5 increase considerably whenFV, theincrease concentrate in the diet exceeds 60%. Thisinindicates the voluntary consequently, the roughage considerably when the concentrate the diet that exceeds 60%. This intake indicates thetovoluntary intake likely to be low at high levels of roughage is that likely be low atroughage high levels of isconcentrate. The FV_MR significantly affected the prediction of DMI in growing cattle, and by including this concentrate. variable in the significantly prediction of FV_intake, MSPE wasofreduced 14% forcattle, the predicted The FV_MR affected thetheprediction DMI inbygrowing and by including this roughage intake. The FV_MR was parameterized independently of animal category and is Figure 10.5 variable in the prediction of FV_intake, the MSPE was reduced by 14% for the predicted roughage calculated as: intake. The FV_MR was parameterized independently of animal category and is calculated as: The FV_MR significantly affected the prediction of DMI in growing cattle, and by including this variable in the MSPE was reduced by 14% ⎛for ⎞ ⎛ prediction of FV_intake,1the .2158 IC the ⎞ predicted ⎟⎟ ⋅ 1 − e − 0.029474088⋅ BW ⋅ ⎜ ⎟ [Eq. 10.16] FV _ MR = ⎜⎜ − 3.8301 + 2.6608 ⋅ FV _ r + 10.15 roughage intake. The FV_MR was parameterized independently of animal category and is FV _ r 3 ⎝ ⎠ ⎠ ⎝ calculated as:

(

)

(

)

⎛ 1.2158 ⎞ 155 − 0.029474088⋅ BW ⎛ IC ⎞ ⎟ ⋅ 1− e FV _ MR = ⎜⎜ − 3.8301 + 2.6608 ⋅ FV _ r + ⋅⎜ ⎟ FV _ r ⎟⎠ ⎝ 3⎠ ⎝

122

[Eq. 10.16]

NorFor – The Nordic feed evaluation system 155

FV_SubR

6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0

0

10

20

30

40

50

60

70

80

90

100

Concentrate in diet, %

Figure 10.5. Effect of proportion of concentrate in the diet on roughage substitution rate in growing cattle. where FV_MR is the roughage metabolic correction factor; FV_r is the diet basis roughage fill value, Equation 6.10; IC is the animal intake capacity, Equation 10.13; BW is the animal body weight, kg; and the IC/3 ratio is an adjustment factor for animal IC level. The computer program TableCurve 2D from SYSTAT was used to resolve the equation profile and the FVL_intake equation was parameterized using the Solver tool in Microsoft® Excel, as described for the dairy cow model. Based on sheep data, Mertens (1994) demonstrated that the dietary NDF concentration at the intersect of metabolic and physical regulation is dependent on animal BW. This may explain why animal BW significantly affected the FV_MR in Equation 10.15. The IC/3 ratio in the equation reflects the different animal categories.

10.3 Implications A primary objective of this work has been to develop a system that can be used to formulate diets that maximize roughage intake while meeting animals’ energy requirements with a minimal dietary energy concentration. It is important to ensure that there is a high proportion of roughage in diets to maintain a favourable rumen environment. The feed intake sub-model in NorFor is complex and consists of several non-linear equations. This comprehensive approach requires DMI to be calculated iteratively by a computer program. Although the system is a simplification of numerous factors and animal-diet interactions affecting voluntary feed intake, it is applicable to diets with a wide range of roughage qualities and diverse feeding situations.

References Andersen, H.R., J. Foldager, P.S. Varnum and S. Klastrup, 1993a. Effects of urea and soybean meal at two levels in a whole crop barley silage ration on performance, carcass and meat quality in young bulls. Forskningsrapport nr. 4 fra Statens Husdyrbrugsforsøg, Denmark. (In Danish with English summary). Andersen, H.R., B.B. Andersen, P. Madsen, P.S. Varnum and L.R. Jensen, 1993b. Effects of maize silage plus urea or soybean versus concentrate on performance, carcass and meat quality of young bulls. Forskningsrapport nr. 5 fra Statens Husdyrbrugsforsøg, Denmark. (In Danish with English summary). Andersen, H.R., B.B. Andersen and H.G. Bang, 2001. Kødracekrydsninger: Kvier kontra ungtyre fodret med forskelligt grovfoder-kraftfoderforhold og slagtet ved forskellig vægt. DJF rapport Husdyrbrug nr. 28. 82 pp. (In Danish).

NorFor – The Nordic feed evaluation system

123

Baldursdóttir, R., 2010. Development of the feed intake capacity of Icelandic cows in the NorFor feed evaluation system. Master Thesis, Department of Animal and Aquacultural Sciences, Norwegian University of Life Sciences, Norway. 63 pp. Berg, J., 2004. Sluttrapport. Produksjonsegenskapene hos slakteokser av Charolais x NRF, Blond D’ Aquitaine x NRF og NRF ved to ulike fôrstyrkenivåer. Institutt for husdyr og akvakulturvitenskap, UMB. Intern rapport, pp. 11. (In Norwegian). Berg, J. and H. Volden, 2004. Økologisk storfekjøtt. Buskap nr 4. 16-17. (In Norwegian). Bertilsson J. and M. Murphy, 2003. Effects of feeding clover silages on feed intake, milk production and digestion in dairy cows. Grass and Forage Science 58: 309-322. Bosch, M.W., S.C.W. Lammers-Wienhoven, G.A. Bangma, H. Boer and P.W.M. Van Adrichem, 1992. Influence of stage of maturity of grass silages on digestion processes in dairy cows. 2. Rumen contents, passage rates, distribution of rumen and faecal particles and mastication activity. Livestock Production Science 32: 265-281. Bossen, D., M.R Weisbjerg, L. Munksgaard, and S. Højsgaard, 2010. Allocation of feed based on individual dairy cow live weight changes. I: Feed intake and live weight changes. Livestock Science 126: 252-272. Bossen, D. and M.R. Weisbjerg, 2010. Allocation of feed based on individual dairy cow live weight changes. II: Effect on milk production. Livestock Science 126: 273-285. Damgaard, P. and S. Hansen, 1992. Sammenligning af valset og pilleteret kraftfoder til ungtyre. LK-meddelelse nr. 189. 3 pp. (In Danish). Eriksen, J. and S. Ness, 2007. Effet av høstetid for surfôr og kraftfôrnivå på passasjekinetikk of NDF estimert med markørmetoden og vomtømmingsmetoden. Masteroppgave Universitetet for miljø- og biovitenskap. 87 pp. (In Norwegian). Forbes, J.M., 1995. Voluntary Food Intake and Diet Selection in Farm Animals. CAB International, Wallingford, UK. Fylstra, D., L. Lasdon, J. Watson and A. Waren, 1998. Design and use of the Microsoft Excel Solver. Interfaces 28: 29-55. Garmo, T.H., H. Volden, S. Krizsan and S.K. Nes, 2007. Effects of level of concentrate supplementation on nutrient digestion of lactating dairy cows grazing at two pasture allowances. Journal of Animal Science 85: 635, Supplement 1. Hessle, A., E. Nadeau and S. Johnsson, 2007. Beef heifer production as affected by indoor feed intensity and slaughter age when grazing semi-natural grasslands in summer. Livestock Science 111: 124-135. Hetta, M., J.W. Cone, G. Bernes, A.-M. Gustavsson and K. Martinsson, 2007. Voluntary intake of silages in dairy cows depending on chemical composition and in vitro gas production characteristics. Livestock Science 1006: 47-56. Huhtanen, P., 1993. The effects of concentrate energy source and protein content on milk production in cows given grass silage ad libitum. Grass and Forage Science 48: 347-355. Huhtanen, P. and S. Jaakkola, 1993. The effects of forage preservation method and proportion of concentrate on digestion of cell wall carbohydrates and rumen digesta pool size in cattle. Grass and Forage Science 48: 155-165. Huhtanen, P., S. Jaakkola and E. Saarisalo, 1995. The effects of concentrate energy source on the milk production of cows given a grass silage-based diet. Animal Science 60: 31-40. Hymøller, L., M.R. Weisbjerg, P. Nørgaard, C.F. Børsting and N.B. Kristensen, 2005. Majsensilage til malkekøer. DJF Rapport Husdyrbrug nr. 65. 71 pp. (In Danish). INRA (Institute National de la Recherche Agronomique), 1989. Feed intake: The fill unit system. In: Jarrige, R. (ed.). Ruminant Nutrition. Recommended allowances and feed tables. John Libbey and Co Ltd, London, UK, pp. 61-71. Ingvartsen, K.L., J. Foldager, J.B. Larsen and V. Østergaard, 1988. Growth and milk yield by Jersey reared at different planes of nutrition. Beretning nr. 645 fra Statens Husdyrbrugsforsøg. 72 pp. (In Danish with English summary and subtitles). Jarrige, R., 1986. Voluntary intake in cows and its prediction. International Dairy Federation Bulletin 196: 4-16. Johansen, A., 1992. A comparison between meadow fescue and timothy silage. Doctor Scientarium Thesis. Agricultural University of Norway, Aas, Norway. 108 pp. Jørgensen, K.F., J. Sehested and M. Vestergaard, 2007. Effect of starch level and straw intake on animal performance, rumen wall characteristics and liver abscesses in intensively fed Friesian bulls. Animal 1: 797-803. Kirkland R.M., T.W.J. Keady, D.C. Patterson, D.J. Kilpatrick and R.W.J. Steen, 2006. The effect of slaughter weight and sexual status on performance characteristics of male Holstein-Friesian cattle offered a cereal-based diet. Animal Science 82: 397-404. Kristensen, V.F., 1983. Styring av foderopptagelsen ved hjælp af foderrasjonens sammensætning og valg af fodringsprincip. In: Østergaard, V. and A. Neimann-Sørensen (eds.). Optimale foderrationer til malkekoen. Foderværdi, foderopptagelse, omsætning og produktion. Rapport nr. 551. Beretning fra Statens Husdyrbrugsforsøg. pp. 7.1-7.35. (In Danish).

124

NorFor – The Nordic feed evaluation system

Kristensen, V.F., 1995. Forudsigelsen af foderoptagelsen hos malkekøer. Intern rapport nr 61. Statens Husdyrbrugsforsøg. 28 pp. (In Danish). Kristensen, V.F., 1999. Grovfoderkildens betydning for malkekoens produktion og effektivitet. Temamøde vedr. Malkekøernes og kviernes fodring. Intern Rapport nr 118, Danmarks Jordbrugsforskning. pp. 18-33. (In Danish). Kristensen, V.F., M.R. Weisbjerg, C.F. Børsting, O. Aaes and P. Nørgaard, 2003. In Malkekoens energiforsyning og produktion. In: Strudsholm, F. and K. Sejrsen (eds.). Kvægets ernæring og fysiologi. Bind 2 - Fodring og produktion. DJF rapport nr. 54. pp. 73-112. (In Danish). Khalili, H. and P. Huhtanen, 1991. Sucrose supplements in cattle given grass silage-based diet. 2. Digestion of cell wall carbohydrates. Animal Feed Science and Technology 33: 263-273. Lasdon, J., A. Waren, A. Jain and M. Ratner, 1978. Design and testing of generalized reduced gradient code for nonlinear programming. ACM Transactions on Mathematical Software 4: 34-49. Mäntysaari, P., 1993. The effects of feeding level and protein source of the diet on growth and development at slaughter of pre-pubertal heifers. Acta Agriculturae Scandinavica. 43: 44-51. Matre, T., 1984. Kjøtproduksjon på oksar og kastratar. I. Forsøk med ulike proteinkjelder Institutt for husdyrernæring, Norges landbrukshøgskole. Melding nr 217. 19 pp. (In Norwegian). Matre, T., 1987. Kjøtproduksjon på oksar og kastratar. III. Ulik fôrstyrke og grovfôrkvalitet til kastratar. Institutt for husdyrernæring, Norges landbrukshøgskole. Melding nr 243. 29 pp. (In Norwegian). Mertens, D.R., 1987. Predicting intake and digestibility using mathematical models of ruminal function. Journal of Animal Science 64: 1548-1558. Mertens, D.R., 1994. Regulation of Forage Intake. In: Fahey Jr. G.C., M. Collins, D.R. Mertens and L.E. Moser (eds.). Forage quality, evaluation, and utilization. American Society of Agronomy, Inc., Crop Science Society of America, Inc. and Soil Science Society in America, Inc., Madison, WI, USA, pp. 450-493. Misciattelli, L., V.F. Kristensen, M. Vestergaard, M.R. Weisbjerg, K. Sejrsen and T. Hvelplund, 2003. Milk production, nutrient utilisation and endocrine responses to increased postruminal lysine and methionine supply in dairy cows. Journal of Dairy Science 86: 275-286. Mo, M. and Å. Randby, 1986. Wilted silage for lactating dairy cows. Proceedings 11th General Meeting of the EGF. Troia, Portugal 4-9. May. Mould, F., 1996. Fôropptak og produksjon hos melkekyr ved et lavt PBV I rasjonen. Husdyrforsølsmøtet. pp. 121127. (In Norwegian). Nadeau, E., A. Hessle, B.-O. Rustas and S. Johnsson, 2002. Prediction of silage intake by Charolais bulls. In: Durand, J-L., J-C Emile, C. Huyghie, and G. Lemaire (eds.). Multi-function grasslands quality forages, Animal products and landscapes. Grassland Science in Europe, Volume 7. Proceedings of the 19th General Meeting of the European Grassland Federation 27-30 May, La Rochelle, France, pp. 220-221. Nordang, L. and F. Mould., 1994. Surfôr med ulikt innhold av PBV til mjølkekyr. Husdyrforsøksmøtet 1994, pp. 99-103. (In Norwegian). Petit, H.V. and G.F. Tremblay, 1995. Ruminal fermentation and digestion in lactating cows fed grass silage with protein and energy supplements. Journal of Dairy Science 78: 342-352. Randby, Å. and M. Mo, 1985. Surfôr tilsatt foraform i sammenlikning med maursyresurfôr til melkekyr, Hellerud høsten 1984. M-141. Rapporter fra forsøksvirksomheten ved Institutt for husdyrernæring, s. 75-78. (In Norwegian). Randby, Å.T., 1992. Grass-clover silage for dairy cows. Proceedings of the 14th General Meeting, European Grassland Federation, Lahti, Finland, 8-11 June, pp. 272-275. Randby, Å.T., 1997. Feeding of silage effluent to dairy cows. Acta Agriculturae Scandinavica Section A. Animal Science 47: 20-30. Randby, Å.T. and I. Selmer-Olsen, 1997a. Formic acid treated or untreated round bale grass silage for dairy cows. 8th International Symposium Forage Conservation. Brno, Czech Republic, pp. 160-161. Randby, Å.T. and I. Selmer-Olsen, 1997b. Formic acid treated or untreated round bale grass silage for steers. Presented on Fifth Research Conference, BGS, University of Plymounth, Devon, UK, 8-10 Sept. 2 pp. Randby, Å.T., 1999a. Grass silage treated with two different acid-based additives for dairy cows. The Fifth International Symposium on the Nutrition of Herbivores. April 11-16, San Antonio, TX, USA. 8 pp. Randby, Å.T., 1999b. Effect of increasing levels of a formic acid based additive on ad libitum intake of grass silage and the production of meat and milk. Proceedings of the XII International Silage Conference, Uppsala, Sweden, pp. 205-206.

NorFor – The Nordic feed evaluation system

125

Randby, Å.T., 1999c. Virkning av økende dosering med GrasAAT på surfôropptak og produksjon av mjølk og kjøtt. Forsøk utført på Hellerud Forsøksgård i 1997/98. Institutt for husdyrfag, NLH. Rapport nr 24. Hellerud forsøks- og eliteavslgard. Det kgl. Selskap for Norges Vel. 24 pp. (In Norwegian). Randby, Å.T., 2000. Round bale grass silage treated with an additive containing both lactic acid bacteria and formate for dairy cows and steers. Sixth British Grassland Society Research Conference, SAC, Craibstone, Aberdeen, UK, pp. 121-122. Randby, Å.T., 2002. Silage fermentation by concentrate type interaction. Proceedings of the XIII International Silage Conference, Auchincruive, UK, pp. 312-313. Randby, Å.T., 2007. Effect of propanol and dimethylsulphide in grass silage on organoleptic milk quality. Journal of Animal and Feed Science 16, Suppl. 1, pp. 102-107. Randby, Å.T., I. Selmer-Olsen and L. Baevre, 1999. Effect of ethanol in feed on milk flavour and chemical composition. Journal of Dairy Science 82: 420-428. Rikarðsson, G., 1997. Mismunandi kjarnfóðurgjöf fyrir mjólkurkýr. Ráðunautafundur, pp. 242-254. (In Icelandic). Rikarðsson, G., 2002. Átgeta íslenskra mjólkurkúa. Ráðunautafundur, pp. 125-127. (In Icelandic). Rinne M., P. Huhtanen and S. Jaakkola, 2002. Digestive processes of dairy cows fed silages harvested at four stages of grass maturity. Journal of Animal Science 80: 1986-1998. Rustas, B.-O., 2009. Whole-crop cereals for growing cattle. Effects of maturity stage and chopping on intake and utilisation. Acta Universitatis Agriculturae Sueciae, Doctoral thesis no. 2009:74. Swedish University of Agricultural Sciences, Faculty of Veterinary Medicine and Animal Science, Uppsala, Sweden. Rustas, B.-O., E. Nadeau and S. Johnsson, 2003. Feed consumption and performance in young steers fed whole-crop barley silage at different levels of concentrate. In: Garmo, T.H. (ed.). Proceedings of the International symposium “Early harvested forage in milk and meat production”, 23-24 October 2003, Nannestad, Norway. Agricultural University of Norway, Department of Animal and Aquacultural Sciences, Aas, Norway. Schei, I., H. Volden and L. Bævre, 2005. Effects of energy balance and metabolizable protein level on tissue mobilization and milk performance of dairy cows in early lactation. Livestock Production Science 95: 35-47. Selmer-Olsen, I., 1994. Feed intake, milk yield and liveweight gain with silage from big bales or silos. Poster presented at the 45th Annual Meeting of EAAP, Edinburg, Scotland, 5-8 September. 8 pp. Sejrsen, K. and J.B. Larsen, 1978. Ensilage-kraftfoderforholdets indflydelse på kviers foderoptagelse og tilvækst samt mælkeydelse i første laktation. Beretning nr. 465 fra Statens Husdyrbrugsforsøg. 74 pp. (In Danish). Stensig, T., M.R. Weisbjerg and T. Hvelplund, 1998. Digestion and passage kinetics of fibre in dairy cows as affected by the proportion of wheat starch or sucrose in the diet. Acta Agriculturae Scandinavica Section A. Animal Science 48: 129-140. Sveinbjörnsson, J. and G.H. Harðarsson, 2008. Þungi og átgeta íslenskra mjólkurkúa. Fræðarþing landbúnaðarins 5: 336-344. (In Icelandic). Sveinbjörnsson, J., P. Huhtanen and P. Udén, 2006. The Nordic dairy cow model, Karoline – Development of volatile fatty acid sub-model. In: Kebreab E., J. Dikstra, A. Bannink, J. Gerrits and J. France (eds.). Nutrient Digestion and Utilisation in Farm Animals: Modelling Approaches. CAB International, Wallingford, UK, pp. 1-14. Therkildsen, M., M. Vestergaard, L.R. Jensen, H.R. Andersen and K. Sejrsen, 1998. Influence of feeding intensity, grazing and finishing on growth and carcass quality of young Friesian bulls. Acta Agriculturae Scandinavica 48: 193-201. Thomas, C., 1987. Factors affecting substitution rates in dairy cows on silage based rations. In: Haresign, W. and D.J.A. Cole (eds.) Recent Advances in Animal Nutrition. Butterworths, London, UK, pp. 205-216. Van Vuuren, A.M., C.J. Van Der Koelen and J. Vroons-De Bruin, 1993. Ryegrass versus corn starch or beet pulp fiber diet effects on digestion and intestinal amino acids in dairy cows. Journal of Dairy Science 76: 2692-2700. Vestergaard, M., K. Sejrsen, J. Foldager, S. Klastrup and D.E. Bauman, 1993a. The effect of bovine growth hormone on growth, carcass composition and meat quality of dairy heifers. Acta Agriculturae Scandinavica 43: 165-172. Vestergaard, M., M. Sommer, S. Klastrup and K. Sejrsen, 1993b. Effects of the beta-adrenergic agonist cimaterol on growth and carcass quality of monozygotic Friesian young bulls at three developmental stages. Acta Agriculturae Scandinavica 43: 236-244.

126

NorFor – The Nordic feed evaluation system

11. Chewing index system for predicting physical structure of the diet P. Nørgaard, E. Nadeau Å. Randby and H. Volden 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 4607 4608 4609 4610 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623 4624 4625 4626 4627 4628

An important goal in dairy cow management is, among other things, to develop a feeding strategy

thatChewing ensures goodindex rumen system function, which is vitally important for efficient milk production and healthy 11 for predicting physical structure of the diet

animals. A certain intake physically effective fibre isother essential fortostimulating rumination, salivation An important goal in dairyofcow management is, among things, develop a feeding strategy and rumen motility, and avoiding milk fat depression (Mertens, 1997; De Brabrander et al., that ensures good rumen function, which is vitally important for efficient milk production and 2002). A high intake of rapidly fermentable carbohydrates results production of a high level of volatile healthy animals. A certain intake of physically effective fibreinisthe essential for stimulating fatty acids salivation in the rumen and a motility, subsequent in pH, which increases the1997; risk of rumination, and rumen and reduction avoiding milk fat depression (Mertens, De subacute ruminal acidosis (Krause Infermentable addition to carbohydrates the structural results fibre content Brabrander et al., 2002). A and highOetzel, intake of2006). rapidly in the of a feed, the production of dietary a high level of volatile acids inand theaffects rumenthe andeating a subsequent reduction in pH,time (RT) length of the particles is alsofatty important time (ET), rumination which increases the riskand of subacute ruminal acidosis and Oetzel, 2006). In addition to and, thus, salivation thereby buffering of the (Krause rumen environment (Krause and Oetzel, 2006). the contentobserved of a feed,that the one length the dietary particles is also important and effective for De structural Boever etfibre al. (1993) kgofNDF from late cut grass silage was more affects the eating time (ET), (RT) thus, salivation buffering of stimulating rumination thanrumination NDF fromtime early cutand, grass silage (whichand hasthereby low levels of lignification), the rumen environment (Krause and Oetzel, 2006). De Boever et al. (1993) observed that one kg demonstrating that fibre type affects chewing activity. NDF from late cut grass silage was more effective for stimulating rumination than NDF from early cut grass silage (which has low levels of lignification), demonstrating that fibre type affects The formulation chewing activity. of diets for dairy cows with respect to the fibre content cannot be based solely on

a defined roughage to concentrate ratio or a minimum amount or proportion of roughage NDF in the ration (Mertens, 1997). Balch (1971) fibrousnesses different The formulation of diets for dairy cows withproposed respect toto therank fibrethe content cannot beof based solelyfeeds by a chewing index value that accounts for ETorplus RT. In this chapter, we present model for predicting on a defined roughage to concentrate ratio a minimum amount or proportion of aroughage NDF thethe diet eating index (EI), rumination index (RI) and chewing index (CI). These characteristics in ration (Mertens, 1997). Balch (1971) proposed to total rank the fibrousnesses of different feeds by chewing index value that accounts for ET plus RT. Inofthis we intention present a model areaused in NorFor to optimize the physical structure thechapter, diet. The of thisfor system is to predicting thean diet eating index rumination index and total chewing index (CI).toThese characterize additive CI for(EI), individual feeds and (RI) hence ranking feeds according their ability to characteristics are mastication used in NorFor to optimize physical structure of the diet. The intention of stimulate particle when cows arethe eating and during rumination. The time spent masticating this system is to characterize an additive CI for individual feeds and hence ranking feeds and the intensity of mastication are used to quantitatively rank the biological fibrousnesses of feeds, according to their ability to stimulate particle mastication when cows are eating and during which is closely related to the stimulation of salivation, the frequency of rumen motility and the rumination. The time spent masticating and the intensity of mastication are used to quantitatively formation of a stable flowing layer system in the reticulo-rumen. The NorFor chewing system has rank the biological fibrousnesses of feeds, which is closely related to the stimulation of been developed from the Danishmotility chewing system (Nørgaard, 1986) and usinginnew data salivation, the frequency of rumen andindex the formation of a stable flowing layerby system from Nørgaard et al. the reticulo-rumen. The(2010). NorFor chewing system has been developed from the Danish chewing index system (Nørgaard, 1986) and by using new data from Nørgaard et al. (2010).

11.1 Predicting the chewing index from eating and rumination

11.1 Predicting the chewing index from eating and rumination

Eachfeed feedisisgiven givena aCICIvalue value (min/kg DM), which is estimated the of sum and RI: Each (min/kg DM), which is estimated as theassum EI of andEIRI: CI = EI + RI

[Eq. 11.1]

11.1

where index for for a feedstuff, min/kg DM; DM; EI is the index asindex described in whereCI CIisisthe thechewing chewing index a feedstuff, min/kg EI eating is the eating as described in equation DM, asas described in in equation Equation11.2, 11.2,min/kg min/kgDM; DM;and andRIRIisisthe therumination ruminationindex, index,min/kg min/kg DM, described Equation 11.4. 11.4.

EI and RI describe the recorded ET and RT, respectively, and in NorFor the feed EI and RI values

EI and RI describe the recorded ET and RT, respectively, and in NorFor the feed EI and RI refer to those for a standard cow of a large multiparous dairy breed with a BW of 625 kg, a NDF values refer to those for a standard cow of a large multiparous dairy breed with a BW of 625 kg, a roughage (NDFr) intake of 0.7% of BW andand a daily rumination time of of 400 min. This corresponds to NDF roughage (NDFr) intake of 0.7% of BW a daily rumination time 400 min. This a daily DMI of 20 kg with a CI value of 30 min/kg DM, assuming that rumination time is equivalent corresponds to a daily DMI of 20 kg with a CI value of 30 min/kg DM, assuming that rumination to 2/3 the totaltochewing use time. of a standardized cow impliescow thatimplies the unit time is of equivalent 2/3 of thetime. total The chewing The use of a standardized that‘minutes/kg the DM’“minutes/kg reported inDM” the practical tool to atool standard and thus, unit reported in therefers practical refers tocow a standard cowdoes and not thus,necessarily does not reflects

the actual mean chewing time (CT) (min/kg DM) for that particular cow. The equations for EI and 172 RI were obtained from a meta-analysis of recorded ET and RT from cattle fed 85 different diets consisting of unchopped grass silage, grass hay or lucerne hay with or without supplementation of ground concentrates or rolled barley (Nørgaard et al., 2010). Based on the meta-analysis, the NorFor EI and RI values were standardized to reference values of 50 and 100 min/kg NDFr for unchopped roughage, respectively. In concentrates, the standard EI was set to 4 min/kg DM. As described in H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_11, © Wageningen Academic Publishers 2011

127

4630 4631 4632 4633 4634 4635 4636 4637 4638 4639 4640 4641 4642 4643 4644 4645 4646 4647 4648

equations for EI and RI were obtained from a meta-analysis of recorded ET and RT from cattle fed 85 different diets consisting of unchopped grass silage, grass hay or lucerne hay with or without supplementation of ground concentrates or rolled barley (Nørgaard et al., 2010). Based on the meta-analysis, the NorFor EI and RI values were standardized to reference values of 50 and 100 min/kg NDFr for unchopped roughage, respectively. In concentrates, the standard EI was set to 4 min/kg DM. As described in Chapter 4, concentrates and roughage are defined on the Chapter 4, concentrates and roughage are defined on the basis of the most frequently measured basis of the most frequently measured particle length (PL) that they contain, such that feedstuffs particle (PL) theyas contain, such that feedstuffs with a PL≤6 mmmm areare defined as as concentrates with a PLlength ≤6 mm arethat defined concentrates while feedstuffs with a PL >6 defined while feedstuffs with a PL>6 mm defined as roughage. maize silage are chopped at roughage. Grass and maize silage areare chopped at harvest and theGrass effect and of physical processing on harvest and the effect of physical processing on the EI and RI values is based on the theoretical the EI and RI values is based on the theoretical chopping length (TCL), which corresponds to the chopping length (TCL), which corresponds to theobserved PL value (O´Dogherty, et al. PL value (O´Dogherty, 1984). Nørgaard et al. (2010) that the mean ET 1984). and RT Nørgaard per kg (2010)decreased observedwith thatincreasing the meanBW. ET and RT kgofNDFr decreased withinto increasing BW. Thus, the NDFr Thus, theper size the cow is not taken account when e.g. no distinctions in this respect are drawn between of calculating the values EI orinto RI (account size of the cow is notof taken when calculating the values of EI or RI (e.g.cows no distinctions large breeds small between breeds or cows primiparous vs.dairy multiparous This meansorthat in thisdairy respect arevs.drawn of large breeds breeds). vs. small breeds primiparous vs. Jersey cows have a higher meanJersey ET and RT per kgaNDFr than large breeds same multiparous breeds). Thisrecorded means that cows have higher recorded meanfed ETthe and RT per kg diet. Feeding high moisture silage containing less than 40% dry matter has been shown to NDFr than large breeds fed the same diet. Feeding high moisture silage containing less than 40% increase EThas (Debeen Boever et al.,to1993). However, NorFor et this is 1993). not reflected in a higher EI this is not dry matter shown increase ET (DeinBoever al., However, in NorFor value because the DM content of a feedstuff is not taken into account when calculating the values reflected in a higher EI value because the DM content of a feedstuff is not taken into account when of EI or RI.

4649 4650 4651 4652 4653 4654 4655 4656 4657 4658 4659

11.2 Calculation of the eating and rumination indices

4660

EI = 50 ⋅

4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4672

calculating the values of EI or RI.

11.2 Calculation of the eating and rumination indices

Table 11.1 presents an overview of the equations used to calculate EI and RI depending on PL. The EI describes the eating time per kg DM of a given feedstuff for the standard cow (Nørgaard Table 11.1 presents overview of thefeeds equations used EI DM. and RI depending on PL. The et al., 2010). Ground,an rolled or pelleted are given anto EIcalculate of 4 min/kg The recorded ET EI describes eating time per(Mertens, kg DM of a given feedstuff for the standard cow (Nørgaard et al., decreases as a the result of chopping 1997).

2010). Ground, rolled or pelleted feeds are given an EI of 4 min/kg DM. The recorded ET decreases

Table as a result 11.1 of chopping (Mertens, 1997).

EI of of roughage roughage and to the NDF content and and a particle length factorfactor for EI andby-products by-productsare areproportional proportional to the NDF content a particle length for eating (Size_E) (Nørgaardetetal., al.,2010). 2010). EI EI is is calculated calculated using equation: eating (Size_E) (Nørgaard usingthe thefollowing following equation: NDF ⋅ Size _ E 1000

[Eq. 11.2]

11.2

whereEI EIisisthe theeating eatingindex, index,min/kg min/kgDM; DM;NDF NDF NDF content in the feedstuff, g/kg DM; Size_E where is is thethe NDF content in the feedstuff, g/kg DM; is a correction factor that depends on particle length, as described in Equation 11.3; and Size_E is a correction factor that depends on particle length, as described in equation 11.3; and 50 is the standardized eating time, NDF. NDF. 50 is the standardized eatingmin/kg time, min/kg Size_E is a correction factor for the feed particle length, which ranges from 0.67 in finely chopped feedstuffs to 1 of in the unchopped roughage Figure 11.1). index Size_Esystem. is parameterized from Table 11.1. Overview calculations used(see in the chewing recorded ET data in cattle fed chopped (10 or 20 mm TCL) or unchopped silage (Nørgaard, unpublished) using the following equation: Feed type and is calculated Concentrates Roughages ⋅ PL Size _ E = 1 − 0 .52 ⋅ e − 0 .078 Processing Ground

PL or TCL (mm)1 Size_E Size_R Hardness factor Eating index (EI, min/kg DM) Rumination index (RI, min/kg DM) Chewing index (CI, min/kg DM)

Rolled

Finely

Coarsely

≤2.0

2-4

Chopped

[Eq. 11.3] Long or slightly

chopped

173

4-6

4

6-40 >40 1–0.52·exp(-0.078·TCL) 1 1–exp (-0.173·(TCL/0.7-1)) 1 0.75 + iNDF/1000 50·NDF/1000·Size_E

0

100·NDF/1000·Size_R·Hardness factor

EI + RI

1 PL is the most frequent particle length in a feedstuff. TCL (theoretical chopping length) corresponds to PL in roughage

(O´Dogherty, 1984).

128

NorFor – The Nordic feed evaluation system

4689 4690 4691 4692 4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4714 4715

where EI is the eating index, min/kg DM; NDF is the NDF content in the feedstuff, g/kg DM; Size_E is a correction factor that depends on particle length, as described in equation 11.3; and 50 is the standardized eating time, min/kg NDF.

Size_Eisisaacorrection correctionfactor factor feed particle length, which ranges 0.67 in finely chopped Size_E forfor thethe feed particle length, which ranges from from 0.67 in finely feedstuffs to 1 in unchopped roughage (see Figure 11.1). 11.1). Size_E is parameterized fromfrom recorded ET chopped feedstuffs to 1 in unchopped roughage (see Figure Size_E is parameterized recorded ET data cattle fed(10 chopped 20 mm or unchopped silage (Nørgaard, data in cattle fedinchopped or 20 (10 mmorTCL) or TCL) unchopped silage (Nørgaard, unpublished data) unpublished) and isusing calculated using the equation: following equation: and is calculated the following Size _ E = 1 − 0 .52 ⋅ e − 0 .078 ⋅ PL [Eq. 11.3] 11.3 where Size_E is the correction factor for the eating index due to the chopping or processing of the feedstuff. The Size_E rangesfactor from 0.67 for fine chopped to 1chopping for unchopped where Size_E is the factor correction for the eating indexroughage due to the or processing of roughages; and PL is the most frequent particle length in the feedstuff, mm. PL corresponds 173 the feedstuff. The Size_E factor ranges from 0.67 for fine chopped roughage to 1 fortounchopped TCL for roughage (O´Dogherty, 1984) and is therefore used as the input for the equation.

roughages; and PL is the most frequent particle length in the feedstuff, mm. PL corresponds to TCL for roughage (O´Dogherty, 1984) therefore usedthe as value the input for thewhich equation. Figure 11.1 shows how the length ofand feedisparticles affects of Size_E, approaches 1 for PL values > 50 mm, showing that slight chopping of roughage with TCL values higher than Figure 11.1 not shows how length of feed particles affects the value of Size_E, which approaches 1 50 mm does affect thethe value of EI.

for PL values >50 mm, showing that slight chopping of roughage with TCL values higher than 50

mm does Figure 11.1not affect the value of EI.

RI activity during rumination perper kg DM of aof given feedstuff in in the RI describes describesthe theassociated associatedchewing chewing activity during rumination kg DM a given feedstuff the standard cow. determinedfrom fromthe the feed feed NDF NDF content, factor forfor rumination standard cow. RIRI is is determined content,aaparticle particlelength length factor rumination (Size_R) and a hardness factor,iswhich is to related to the content iNDF content the feedstuff (Size_R) and a hardness factor, which related the iNDF in theinfeedstuff (Nørgaard et ., 2010). RI is using calculated using the following (Nørgaard al., 2010).etRIalis calculated the following equation:equation: RI = 100 ⋅

NDF ⋅ Size _ R ⋅ Hardness _ factor 1000

[Eq. 11.4]

11.4

where RI is the rumination index, min/kg DM; NDF is the NDF content in the feedstuff, g/kg

where RI is the rumination index, min/kg DM; NDF is the NDF content in the feedstuff, g/kg DM;Size_R Size_Risisa correction a correction factor depends on particle as described in Equation 11.5; DM; factor thatthat depends on particle lengthlength as described in equation 11.5; the Hardness_factor Hardness_factoris is described in Equation andis 100 is the standardized rumination time, the described in equation 11.6; 11.6; and 100 the standardized rumination time, min/kgNDF. NDF. min/kg If lower than 2.02.0 mmmm (as in processed concentrates), then the RI the RI is If the thePL PLofofthe thefeedstuff feedstuffis is lower than (asfinely in finely processed concentrates), then is assumedtotobebezero zero(see (seeTable Table 11.1). 11.1). Nørgaard Nørgaard and shown that thethe most assumed andSehic Sehic(2003) (2003)have have shown that most frequent frequent particle length in the faeces of cattle fed solely roughage is 0.7 mm, consequently, feedstuffs with a PL>0.7 mm are considered to stimulate rumination. However, some concentrates have 1.0 a PL>0.7 mm but are not considered to stimulate rumination. Therefore, the borderline value for PL was set to 2.0 mm to ensure that all finely ground concentrates are given an RI of zero. Thus, Size_R is zero for feeds with a PL ≤2.0 mm and their CI is 4 min/kg DM corresponding 0.8 to eating index (Table 11.1). Coarsely processed concentrates with a PL between 2 Size_R to and 6 mm (e.g. rolled grains, coarsely ground grains, oil cakes and hulls) are considered Size_E stimulate rumination. Several studies have shown that chopping roughage at TCL values higher 0.6 not decrease the recorded RT (De Boever et al., 1993; Garmo et al., 2008). than 20 mm does Furthermore, Nørgaard and Sehic (2003) showed that the PL of particles in swallowed boli from cows eating grass silage chopped at 19 mm TCL was the same as the PL of boli particles from 0.4 cows eating unchopped grass silage. Therefore, the RT for roughage chopped at a TCL higher than 20 mm is considered to have the same RT as unchopped roughage, and consequently Size_R at 20 mm TCL is set to 0.99. Size_R decreases exponentially when TCL is below 20 mm, as 0.2 11.1. illustrated in Figure

Correction factor

4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4673 4672 4674 4675 4676 4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688

⎛ ⎛ PL ⎞ ⎞ ⎜⎜0.0 −0.173⋅⎜ −1⎟ ⎟ 0.7 ⎠ ⎟⎠ ⎝ Size _ R = 1 − e 0 ⎝

10

20

30

40

50 [Eq. 11.5] 60

PL (mm)

Figure 11.1. Effect of the most frequent particle length (PL) in a feedstuff on the correction factors 174 Size_E and Size_R. Size_E and Size_R are used to adjust the eating and rumination index due to chopping or processing of feedstuffs (see Table 11.1). NorFor – The Nordic feed evaluation system

129

4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4716 4717 4714 4718 4715 4719 4716 4720 4717 4721 4718 4722 4719 4723 4720 4724 4721 4725 4722 4726 4723 4727 4724 4728 4725 4729 4726 4730 4727 4731 4728 4732 4729 4733 4730 4734 4731 4735 4732 4733 4736 4734 4737 4735 4738 4739 4736 4740 4737 4741 4738 4742 4739 4743 4740 4744 4741 4745 4742 4746 4743 4747 4744 4748 4745 4749 4746 4750 4747 4748 4751 4749 4750 4752 4753 4751 4754 4755 4752 4756 4753 4754

the Hardness_factor is described in equation 11.6; and 100 is the standardized rumination time, min/kg NDF. If the PL of the feedstuff is lower than 2.0 mm (as in finely processed concentrates), then the RI is assumed to be zero (see Table 11.1). Nørgaard and Sehic (2003) have shown that the most particle length the faeces cattle solely roughage is 0.7ismm, consequently, feedstuffs with a frequent particleinlength in the of faeces offed cattle fed solely roughage 0.7 mm, consequently, feedstuffs with PL>0.7 mmto arestimulate considered to stimulate rumination. However, some have a PL>0.7 mm PL>0.7 mm area considered rumination. However, some concentrates concentrates have a PL>0.7 mm but are not considered to stimulate rumination. Therefore, the set to 2.0 but are not considered to stimulate rumination. Therefore, the borderline value for PL was borderline valuethat for all PL finely was setground to 2.0 mm to ensure that finely concentrates are given mm to ensure concentrates are all given an ground RI of zero. Thus, Size_R is zero for an RI of zero. Thus, Size_R is zero a PLDM ≤2.0corresponding mm and their CI 4 min/kg DM feeds with a PL≤2.0 mm and theirforCIfeeds is 4 with min/kg to is eating index (Table 11.1). corresponding to eating index (Table 11.1). Coarsely processed concentrates with a PL between 2 Coarsely processed concentrates with a PL between 2 and 6 mm (e.g. rolled grains, coarsely ground and 6 mm (e.g. rolled grains, coarsely ground grains, oil cakes and hulls) are considered to grains, oil cakes and hulls) are considered to stimulate rumination. Several studies have shown that stimulate rumination. Several studies have shown that chopping roughage at TCL values higher chopping roughage TCL values higher than 20 mm does the recorded RT (De Boever et not al., decrease 1993; Garmo et al., 2008). than 20 mm does not at decrease the recorded RT (De Boever et al., 1993; Garmo et al., 2008). Furthermore, Nørgaard and Sehic (2003) showed Furthermore, Nørgaard and Sehic (2003) showed that the PL of particles in swallowed bolithat fromthe PL of particles in grass swallowed boli fromatcows eating silage mm TCL was cows eating silage chopped 19 mm TCL grass was the samechopped as the PLatof19 boli particles fromthe same as the PL of boli particlesgrass fromsilage. cowsTherefore, eating unchopped silage. Therefore, the RT for roughage cows eating unchopped the RT forgrass roughage chopped at a TCL higher than 20 mm considered have20 themm same as unchopped roughage, andRT consequently Size_R chopped at is a TCL highertothan is RT considered to have the same as unchopped roughage, at 20consequently mm TCL is setSize_R to 0.99.atSize_R exponentially when TCL is below 20 mm, as when TCL and 20 mmdecreases TCL is set to 0.99. Size_R decreases exponentially illustrated in Figure is below 20 mm, as11.1. illustrated in Figure 11.1. ⎛ PL ⎞ ⎞ factor for the rumination index, which depends on chopping or where Size_R⎛⎜ is the correction ⎜ −0.173⋅⎜ 0.7 −1⎟ ⎟⎟ ⎝ ⎠ ⎠ Size_R ⎝ processing The factor ranges from 0 for fine processed feeds[Eq. to 111.5] for 11.5 Size _ R = 1 −the e feedstuff. unchopped roughage or roughage chopped with a PL higher than 20 mm. PL is the most frequent particle length (mm) in correction the feedstufffactor and for PL corresponds to TCL depends (theoretical whereSize_R Size_R forroughage the rumination index,depends which on or chopping or where is is thethe correction factor for the rumination index, which on chopping chopping length, mm; O'Dogherty, 1984) andranges can therefore used as the inputfeeds for the equation; processing thefeedstuff. feedstuff. TheSize_R Size_R factor from 0be for fine processed 1 for unchopped processing the The factor ranges from 0 for fine processed feeds to 1tofor 0.7 is the most frequent particle length in faeces174 from cattle, meaning that feedstuffs with a unchopped or chopped roughagewith chopped a PL higher than PL 20 mm. is the most frequent roughage orroughage roughage a PLwith higher than 20 mm. is thePL most frequent particle length PL>0.7 stimulate rumination (Nørgaard and Sehic, 2003). particle length (mm) in the feedstuff and for roughage PL corresponds to TCL (theoretical (mm) in the feedstuff and for roughage PL corresponds to TCL (theoretical chopping length, mm; chopping length, mm;and O'Dogherty, 1984)beand can therefore be used the input for the equation; O’Dogherty, can therefore used the input theas 0.7relatively is the most frequent The structural 1984) NDF in immature spring grass andas chopped beetfor pulp isequation; soft and has 0.7 is thelength most frequent particle length inmeaning faeces from cattle, meaning that feedstuffsstimulate with a rumination particle in faeces from cattle, that feedstuffs with a PL>0.7 little resistance to physical breakdown during chewing compared with mature grass, barley straw, PL>0.7 stimulateSehic, rumination and Sehic, 2003). (Nørgaard 2003).(Nørgaard lucerne hay and cut after blooming or whole cotton seeds. The content of iNDF in a feedstuff is closely related to the lignification of the NDF fibres (Weisbjerg and Soegaard, 2008) and The structural NDF in immature spring grass and chopped beet pulp is soft and has relatively lignification is known increase resistance to physical breakdown. to has physical The structural NDF intoimmature spring grass and chopped beet The pulpresistance is soft and relatively little little resistance to physical breakdown during chewing compared with mature grass, barley straw, breakdown is physical ranked bybreakdown a hardness during factor, which hascompared been givenwith a value of 1 grass, for typical grass resistance to chewing mature barley lucerne hay cut after blooming or whole cotton seeds. The content of iNDF in a feedstuff isstraw, lucerne silage with an blooming iNDF content of 250cotton g/kg NDF. The hardness increases from 0.8 for hay cut after or whole seeds. The contentfactor ofand iNDF in a feedstuff is closely related closely related to the lignification of the NDF fibres (Weisbjerg Soegaard, 2008) and immature spring grass with an iNDF content of approximately 50 g/kg NDF to 1.25 for late cut to the lignification NDF resistance fibres (Weisbjerg and Soegaard,The 2008) and lignification lignification is knownoftothe increase to physical breakdown. resistance to physical is known lucerne hay with an iNDF content of approximately 500 g/kg NDF. The hardness factor is to increaseisresistance physicalfactor, breakdown. Thebeen resistance physical is ranked by breakdown ranked by to a hardness which has given a to value of 1 forbreakdown typical grass calculated as: silage with an iNDFwhich content of been 250 g/kg NDF. The hardness factor increases fromwith 0.8 for a hardness factor, has given a value of 1 for typical grass silage an iNDF content immature spring grass an iNDF content of approximately 50 immature g/kg NDF to 1.25 for latewith cut an iNDF of 250 g/kg NDF. Thewith hardness factor increases from 0.8 for spring grass ⎛ iNDF ⎞ [Eq. 11.6] Hardnesshay _offactor 0.75 + ⎜ content lucerne with= an iNDF of approximately 500 late g/kgcut NDF. The hardness factor is ⎟ g/kg content approximately 50 NDF to 1.25 for lucerne hay with an iNDF content of ⎝ 1000 ⎠ calculated as: 500 g/kg approximately NDF. The hardness factor is calculated as:

where the Hardness_factor is used for calculating the rumination index and ranges from 0.8 to ⎛ iNDF ⎞ 11.6] iNDF content in the feedstuff, g/kg[Eq. Hardness _ factor = 0.75 + ⎜feedstuffs; ⎟ 1.25 for typical Nordic iNDF is the NDF. 1000 ⎝

⎠

11.6

Table shows examples ofisEI, RI for andcalculating CI values for concentrates ranging where11.2 the Hardness_factor used thesome rumination indexand androughages ranges from 0.8 to 1.25 where the Hardness_factor is used for calculating the rumination index and ranges from 0.8 to from 4 to 22Nordic min/kgfeedstuffs; DM. The main reason foriNDF the much higher CI of dried beet pulpNDF. compared to for typical iNDF is the content in the feedstuff, g/kg 1.25 for typical Nordic feedstuffs; iNDF is the iNDF content in the feedstuff, g/kg NDF. barley and rape seed cake is its higher PL, which causes a higher RI. The higher CI of grass silage compared to examples maize silage is due to its CI higher TCL and NDF content, which result in higher Table 11.2 11.2shows shows EI,RI RI and values some concentrates roughages. The main Table examples ofofEI, and CI values forfor some concentrates and and roughages ranging EI and RI (Table 11.2). reason for the much higher CI of dried beet pulp compared to barley and rape seed cake is its from 4 to 22 min/kg DM. The main reason for the much higher CI of dried beet pulp compared to higher PL, which causes higher RI.higher The higher CI of grassa silage to maize barley and rape seeda cake is its PL, which causes higher compared RI. The higher CI of silage grass is due to its Table 11.2 silage maize silage iswhich due toresult its higher TCL and NDF highercompared TCL andtoNDF content, in higher EI and RIcontent, (Table which 11.2). result in higher EI and RI (Table 11.2). The EI and RI are additive values and the CI of a total diet (min/kg DM) is considered to be the The of EIthe andindividual RI are additive sum feed CIvalues values:and the CI of a total diet (min/kg DM) is considered to be the sum Table of the11.2 individual feed CI values: ∑ DMI ⋅ CI

i i The EI andi RI are additive values and the CI of a total diet (min/kg DM) is considered to be the [Eq. 11.7] CI _ DM = sum of the individual ∑ DMI i feed CI values:

11.7

i

∑ DMI i ⋅ CI i

where CI_DM is the chewing index for the total diet, min/kg DM; DMIi is the dry[Eq. matter intake 11.7] CI _ DM = i ∑ DMI i kg DM/day; and CI is the chewing index for the i-th feedstuff, min/kg DM for the i-th feedstuff, i i as described in Equation 11.1. 130 NorFor – The Nordic feed evaluation system where CI_DM is the chewing index for the total diet, min/kg DM; DMIi is the dry matter intake for the i-th feedstuff, kg DM/day; and CIi is the chewing index for the i-th feedstuff, min/kg DM

Table 11.2 Estimated chewing index values for different feedstuffs in the NorFor feed stuff table.

Barley RSC DBP GS MS

Feed code

NDF g/kg DM

iNDF g/kg NDF

PL1 mm

001-0008 002-0044 004-0020 006-0241 006-0308

206 268 382 417 364

158 511 89 151 187

2.1 2.0 4.0 20 10

Size_E Size_R Hardness EI2 RI2 CI2 min/kg min/kg min/kg DM DM DM 0.29 0.89 0.76

0.56 0.99 0.90

0.91 1.26 0.84 0.90 0.94

4 4 4 19 14

5 0 18 37 31

9 4 22 56 45

1

PL is the most frequent particle length in concentrates and corresponds to TCL (theoretical chopping length) in roughage (O´Dogherty, 1984). 2 See Equations 11.1 to 11.6. RSC, rape seed cake; DBP, dried beet pulp; GS, grass silage; MS, maize silage.

where CI_DM is the chewing index for the total diet, min/kg DM; DMIi is the dry matter intake for the i-th feedstuff, kg DM/day; and CIi is the chewing index for the i-th feedstuff, min/kg DM as described in Equation 11.1.

11.3 Implications The objectives of the work described in this chapter were to develop a system for ranking the structural value of different feedstuffs by calculating an additive chewing index, which could be used as a tool to minimize the risk of digestive disorders and ensure proper rumen function of a diet fed to dairy cows. The minimum recommended chewing index (32 min/kg DM) is intended to ensure a high degree of fibre digestion, a high acetate to propionate ratio and an acceptable milk fat content even at a high DMI.

References Balch, C.C., 1971. Proposal to use time spent chewing as an index to which diets for ruminants possess the physical property of fibrousness characteristic of roughages. British Journal of Nutrition 26: 383-392. De Boever, J.L., A. De Smet, D.L. De Brabander and C.V. Boucqué, 1993. Evaluation of physical structure. 1. Grass silage. Journal of Dairy Science 76: 140-153. De Brabrander, D.L., J.L. De Boever, J.M. Vanacker, C.V. Boucqué and S.M. Botterman, 2002. Evaluation of physical structure in cattle nutrition. Recent developments in ruminant nutrition. Wiseman, J. and P.C. Garnsworthy (eds.). Nottingham University Press, Nottingham, UK, pp. 47-80. Garmo, T.H., A.T. Randby, M. Eknaes, E. Prestløkken and P. Nørgaard, 2008. Effect of grass silage chop length on chewing activity and digestibility. In: Hopkins, A., T. Gustafsson, J. Bertilson, G. Dalin, N. Nilsdotter-Linde and E. Spörndly (eds.). Grassland Science in Europe, volume 13, pp. 810-812. Krause, K.M. and G.R. Oetzel, 2006. Understanding and preventing subacute ruminal acidosis in dairy herds: a review. Special Issue: Feed and animal health. Animal Feed Science and Technology 126: 215-236. Mertens, D.R., 1997. Creating a system for meeting the fiber requirements of dairy cows. Journal of Dairy Science 80: 1463-1481. Nørgaard, P., 1986. Physical structure of feeds for dairy cows. A new system for evaluation of the physical structure in feedstuffs and rations for dairy cows. New developments and future perspectives in research of rumen function, pp. 85-107.

NorFor – The Nordic feed evaluation system

131

Nørgaard, P. and A. Sehic, 2003. Particle size distribution in silage, boluses, rumen content and faeces from cows fed grass silage with different theoretical chopping length. Proceedings from the 6th International Symposium on the Nutrition of Herbivores, pp. 457-460. Nørgaard, P., E. Nadeau and A. Randby, 2010. A new Nordic structure evaluation system for diets fed to dairy cows: a meta analysis. In: Sauvant, D., J. Van Milgen, P. Faverdin and N. Friggens (eds.). Modelling nutrient digestion and utilization in farm animals. Wageningen Academic Publishers, Wageningen, the Netherlands, pp. 112-121. O’Dogherty, M, 1984. The description of chop length distribution from forage harvesters and a geometric simulation model for distribution generation. In: Kennedy, P. (ed.). Canadian society of animal science, occasional publication, Edmonton, Alberta, Canada, pp. 62-75. Weisbjerg, M.R. and K. Soegaard, 2008. Feeding value of legumes and grasses at different harvest times. In: Hopkins, A., T. Gustafsson, J. Bertilsson, G. Dalin, N. Nilsdotter-Linde and E. Spörndly (eds.). Biodiversity and animal feed: Future challenges for grassland production. Grassland Science in Europe, volume 13, pp. 513-515.

132

NorFor – The Nordic feed evaluation system

4825 4825 4826 4825 4826 4827 4827 4826 4828 4828 4827 4829 4828 4830 4829 4831 4830 4829 4832 4831 4830 4832 4831 4833 4832 4833 4834 4835 4834 4833 4836 4835 4834 4837 4836 4835 4837 4838 4836 4837 4838 4839 4838 4840 4839 4841 4840 4839 4842 4841 4840 4843 4842 4841 4844 4843 4842 4845 4844 4843 4846 4845 4844 4847 4846 4845 4848 4847 4846 4849 4848 4847 4850 4849 4848 4850 4851 4849 4850 4851 4852 4851 4853 4852 4854 4853 4852 4855 4854 4853 4856 4855 4854 4856 4855 4857 4856 4858 4857 4859 4858 4857 4860 4859 4858 4861 4860 4859 4862 4861 4860 4863 4862 4861 4864 4863 4862 4865 4864 4863 4865 4866 4864 4865 4866 4867 4866 4868 4869 4870 4871 4872 4873 4874 4875

12.0 Prediction of milk yield, weight gain and utilisation of N, P 12.0 milk yield, weight gain and utilisation of N, P and KPredictionofofmilk 12. Prediction yield, weight gain and utilisation of N, P and K and KPrediction 12.0 of milk yield, weight gain and utilisation of N, P M. Åkerlind and H. Volden M. and K and M. Åkerlind Åkerlind andH. H.Volden Volden

NorFor predicts ECM and protein yield for dairy cows and weight gain for growing cattle. M. Åkerlind Volden Excretion of and N, PH. and K can also be predicted for a given diet and production level. Keep NorFor NorForpredicts predictsECM ECMand andprotein proteinyield yieldfor fordairy dairycows cowsand andweight weightgain gainfor forgrowing growingcattle. cattle. Excretion track on excreted N, PK and Kalso are of interest when estimating plant fertilising value and also Excretion of N, P and can be predicted for a given diet and production level. Keep of N, P and K can also be predicted for a given diet and production level. Keep track on excreted NorFor predicts ECM and protein yield for dairy cows and weight gain for growing cattle. evaluating the impact of excreted N interest and P onwhen eutrophication of the fertilising environment. track on excreted N, P and K are of estimating plant value and also Excretion ofare N, Pofand K canwhen also be predictedplant for afertilising given diet value and production level. Keep N, P and K interest estimating and also evaluating the impact of evaluating the impact excreted N interest and P onwhen eutrophication the fertilising environment. track on excreted N, Pof and K are of estimating of plant value and also

excreted N and P on eutrophication of the environment. 12.1 Prediction of milk yield evaluating the impact of excreted N and P on eutrophication of the environment. 12.1 Prediction Prediction of ECMof yield milk is estimated yield from the total supply of NEL minus basal energy 12.1 Prediction offorismilk yield requirements corrected any changes in BCS (seesupply chapterof9). Since 1 kgbasal ECMenergy contains Prediction of ECM yield estimated from the total NEL minus 12.1 Prediction of milk yield

3.14 MJ (Sjaunja et al., for 1990), the estimated ECM isSince calculated as: contains requirements corrected changes in BCS (seeproduction chapterof9). 1 kg ECM Prediction of yield isisany estimated from total supply minus Prediction ofECM ECM yield estimated fromthe the total supply ofNEL NEL minusbasal basalenergy energy requirements 3.14 MJ (Sjaunja et al., 1990), the estimated ECM production is calculated as: requirements corrected for any changes in BCS (see chapter 9). Since 1 kg ECM contains

corrected for anyNEL changes (see _Chapter Since 1 kg ECM contains − NE _in maBCS int − NEL gain − NE9). _ gest − NEL _ dep + NEL _ mob 3.14 MJ (Sjaunja et

ECMMJ _ response = et al., 1990), the estimated ECM production is calculated as: 3.14 (Sjaunja al., 1990), the estimated ECM production is calculated as: ECM _ response =

14 _ gest − NEL _ dep + NEL _ mob NEL − NE _ ma int − NEL _ gain −3.NE 3 . 14 _ gest − NEL _ dep + NEL _ mob[Eq. 12.1] NEL − NE _ ma int − NEL _ gain − NE

12.1] 12.1 ECM _ response = [Eq. 3.14 where ECM_response is the predicted energy corrected milk yield, kg/day; NEL is the net [Eq. 12.1] energy lactation obtained from the diet,energy Equation 8.3; NE_maint, NEL_gain, NE_gest, whereECM_response ECM_response thepredicted predicted energy corrected milk yield, kg/day; NEL the net energy where isisthe corrected milk yield, kg/day; NEL is theisnet NEL_dep and NEL_mob is energy required of 8.3; maintenance, gestation, growth of primiparous energy lactation lactation obtained obtained from from the diet, the diet, Equation Equation 8.3; NE_maint, NE_maint, NEL_gain, NEL_gain, NE_gest, NE_gest, NEL_dep and where ECM_response is thefrom predicted energy corrected milk yield, kg/day; NEL9.18 is the net cows, and supply of energy mobilisation, Equation 9.1, 9.4, 9.3, 9.17 and NEL_dep and NEL_mob is energy required of maintenance, gestation, growth of primiparous NEL_mob is energy required of maintenance, gestation, growth of primiparous cows, and supply of energy lactation obtained from the diet, Equation 8.3; NE_maint, NEL_gain, NE_gest, respectively; and 3.14 is the energy content of 19.3, kg ECM. cows, and supply of energy from mobilisation, Equation 9.1, 9.4, 9.3, 9.17 and 9.18 energy from mobilisation, Equation 9.1, 9.4, 9.17 and 9.18 respectively; and 3.14 is the energy NEL_dep and NEL_mob is energy required of maintenance, gestation, growth of primiparous respectively; andECM. 3.14 is the from energy content of 1Equation kg ECM.9.1, 9.4, 9.3, 9.17 and 9.18 content ofsupply 1 kg cows, and of energy mobilisation, Milk protein response is calculated by multiplying available AATN by the efficiency of AA respectively; and is thesynthesis. energy content of 1 kg AAT ECM. is the total AAT minus basal utilisation forresponse milk3.14 protein The available N N Milk protein is calculated by multiplying AAT Milk protein response is calculated by multiplyingavailable available AATNNby bythe theefficiency efficiencyof ofAA AA utilisation AAT Response in milkThe protein production isiscalculated as: N minus basal N requirement. utilisation for milk protein synthesis. available AAT the total AAT N for milk protein synthesis. The available AATN isavailable the totalAAT AATNNbyminus basal AAT requirement. Milk protein response is calculated by multiplying the efficiency of NAA AAT N requirement. Response in milk protein production is calculated as: utilisation for milk protein protein production synthesis. The available AAT Response in milk is calculated as: is the total AAT minus basal N N AAT _ Eff [Eq. 12.2] AAT100 N _ Eff Pr otein _ response = (Avail _ AATN ) ⋅ [Eq. 12.2] 12.2 AAT100 _ Eff N where Pr oteinProtein_response _ response = (Availis_ the AAT predicted milk protein production, g/day; Avail_AAT [Eq.N 12.2] is the N )⋅ 100 where Protein_response is the in predicted milk protein g/day;Equation Avail_AAT the available available amino acid absorbed the small intestine forproduction, milk production, 9.26 is and

N Pr otein _ response = (Avail _ AAT ) ⋅ protein AAT Response inNmilk production is calculated as: N requirement.

where Protein_response is the predicted milk protein production, g/day; Avail_AATNNis the amino acid in theofsmall intestine for milkfor production, Equation 9.26 and AAT AAT isabsorbed theacid efficiency amino acid utilization for milkproduction, protein production, Equation N_Eff available amino absorbed in the small intestine milk Equation 9.26 andN_Eff is the where Protein_response isutilization the predicted milk protein production, g/day; Avail_AAT N is the efficiency of amino acid for milk protein production, Equation 9.24. 9.24. AAT N_Eff is the efficiency of amino acid utilization for milk protein production, Equation available amino acid absorbed in the small intestine for milk production, Equation 9.26 and 9.24. AAT _Eff is the efficiency of amino acid utilization for milk protein production, Equation

12.2 of weight gain gain 12.2NPrediction Prediction of weight 9.24. For cattle,of ADG can be predicted from either the energy available for growth or the 12.2growing Prediction weight gain AAT forof growth. Since the energy utilization kg_corr (Equation 9.6) and kormgthe AATN For growing growing cattle, ADG can bepredicted predicted from either the energy available growth N available For cattle, ADG can be from either thefactors energy available forfor growth or the 12.2 Prediction weight gain (Equation 8.5),growth. both depend on and ADG, they cannot used to (Equation predict daily weight available for SinceSince theBW energy utilization factorsbe kg_corr (Equation 9.6) and k kmg(Equation AAT for ADG growth. the energy utilization kg_corr 9.6) and N available For growing cattle, can be the predicted from either thefactors energy available for growth or mg the gain for a given diet.on Therefore, predicted requirement for ADG was assumed toweight vary with 8.5), both depend BW and ADG, they cannot be used to predict daily weight gain for a given (Equation 8.5), both depend on BW and ADG, they cannot be used to predict daily AAT N available for growth. Since the energy utilization factors kg_corr (Equation 9.6) and kmg animal type and diet. BW. gain for a given Therefore, the predicted requirement for ADG was assumed to vary diet. Therefore, the predicted requirement for ADG was assumed to vary with animal type (Equation 8.5), both depend on BW and ADG, they cannot be used to predict daily weightwithand BW. animal and diet. BW. Therefore, the predicted requirement for ADG was assumed to vary with gain fortype a given For growing bulls ofofdairy gain is is predicted from available energy: For growing bulls dairybreed breedweight weight gain predicted from available energy: animal type and BW. For growing bulls of dairy breed weight gain is predicted from available energy: NEG - NE_maint

gaingrowing _ response 1000breed ⋅ For bulls of =dairy weight gain is predicted from available energy: [Eq. 12.3] NEG 0 .0243 + 7.7315 NEG ⋅-BW NE_maint gain _ response NEG = 1000 ⋅ [Eq. 12.3]

12.3

0 .0243 ⋅-BW +gain 7.7315 For growing bulls of beef breed NEG weight NE_maint is predicted from available energy:

gaingrowing _ response 1000breed ⋅ For bulls of =beef weight gain is182 predicted from available energy: [Eq. 12.3] NEG 0 .0243 ⋅ BW + 7.7315 NEG - NE_maint 182 gain _ response NEG = 1000 ⋅ [Eq. 12.4] 0 .0116 ⋅ BW + 9.7214 182

where gain_response

12.4

is the predicted weight gain based on the energy supply, g/d; NEG is the

NEG where gain_responseNEG is the predicted weight gain based on the energy supply, g/d; NEG is energy gain, Equation 8.8;8.8; NEG_maint is the energy requirement of of maintenance, Equation 9.1; BW the energy gain, Equation NEG_maint is the energy requirement maintenance, is current9.1; body weight, kg; body and the numerator calculates the energy content 1 kg content of weight gain. Equation BW is current weight, kg; and the numerator calculates the of energy of 1 kg of weight gain.

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_12, © Wageningen Academic Publishers 2011 AAT =

⋅

133

4870 4872 4871 4871 4873 4872 4872 4874 4873 4873 4875 4874 4874 4876 4875 4875 4877 4876 4876 4877 4877 4878 4878 4879 4878 4880 4879 4879 4881 4880 4880 4882 4881 4881 4883 4882 4882 4884 4883 4883 4885 4884 4884 4885 4885 4886 4886 4887 4886 4888 4887 4887 4889 4888 4888 4890 4889 4889 4890 4890 4891 4891 4892 4891 4893 4892 4892 4894 4893 4893 4895 4894 4894 4895 4896 4895 4896 4896 4897 4898 4897 4897 4899 4898 4898 4900 4899 4899 4901 4900 4900 4902 4901 4903 4901 4902 4903 4902 4904 4903 4905 4904 4904 4906 4905 4905 4907 4906 4906 4908 4907 4907 4908 4909 4908 4909 4910 4910 4909 4911 4911 4912 4912 4913 4913 4914 4914 4915 4915 4916 4916 4917 4917 4918 4918 4919 4919 4920 4920 4921 4921 4922 4922 4923 4923 4924 4924 4925 4925 4926 4926 4927 4927 4928 4928 4929 4929

gain, Equation is thegain energy requirement of maintenance, the NEG_maint predicted weight based on the energy supply, g/d; NEG is where gain_response the energy NEG is8.8; the predicted gain on thecalculates energy supply, g/d; NEG is where gain_response Equation 9.1; BW is current body weight,weight kg; and thebased numerator the energy content the energy gain, Equation NEG_maint is the energy requirement of maintenance, NEG is8.8; the gain, Equation 8.8; NEG_maint is the requirement of maintenance, of 1energy kg of9.1; weight gain. Equation BW is current body weight, kg; andenergy the numerator calculates the energy content Equation BWgain. is current body weight, kg; and the numerator calculates the energy content of 1 kg of9.1; weight of 1 kg of weight gain. for heifers and steers are calculated from available AATN: Weight gain prediction Weight gain gainprediction predictionfor forheifers heifersand andsteers steers calculated from available AAT Weight areare calculated from available AAT N: N: Weight gain prediction for heifers and steers are calculated from available AAT _Eff AATN: N (AATN - AATN _maint - AATN _gest) ⋅ gain _ responseAAT = 1000 ⋅ (AAT - AAT _maint - AAT _gest) ⋅ AATN _Eff 100 [Eq. 12.5] N N − 0.1262 ⋅ BWN+ 194.2 12.5 gain _ responseAAT = 1000 ⋅ (AATN - AATN _maint - AATN _gest) ⋅ AATN _Eff 100 [Eq. 12.5] 100 − 0.1262 ⋅ BW + 194.2 [Eq. 12.5] gain _ responseAAT = 1000 ⋅ 0.1262gain ⋅ BW based + 194.2on the amino acid supply, g/day; where gain_ responseAAT is the predicted −weight where gain_responseAAT is the predicted weight gain based on the amino acid supply, g/day; AATN is the response amino acid in the small intestine, Equation AAT the AATN gain_ is the predicted weight gain based on the8.11; amino acid supply,isg/day; where N_maint AATabsorbed is the amino acidrequirement absorbed infor the small Equation 8.11; _maint isprotein theisanimal’s protein NAAT where is the predicted weight gain based onAAT theAAT amino acid supply, g/day; animal’s maintenance, Equation 9.22; is Nthe AAT is protein the response amino acid in the intestine, small intestine, Equation 8.11; _maint the AATabsorbed N_gest N gain_ requirement for maintenance, Equation 9.22; AAT _gest is the protein requirement of gestation for AATN is protein the of amino acid absorbed in Equation the small 9.42; intestine, 8.11; AAT _maint N Equation is the efficiency of theis the requirement gestation for heifers, AAT is Nthe protein animal’s requirement for maintenance, Equation 9.22; N_EffAAT N_gest heifers, Equation 9.42; AAT _Eff is the efficiency of the utilisation of the available amino acids protein animal’s protein requirement for Equation 9.22; utilisation ofof thegestation availablefor amino acids for protein gain, 9.34; BWisisthe the current BW, requirement heifers, Equation 9.42; AATEquation is the efficiency of the N maintenance, N_gest N_EffAAT for and protein gain, Equation 9.34; BW isfor theprotein current BW, kg; numerator calculates the protein requirement of for heifers, Equation 9.42;gain, AAT _Eff is the the efficiency of the kg; theofnumerator calculates the protein content in 1Equation kg ofand weight gain. utilisation thegestation available amino acids 9.34; BW is the current BW, N content inof 1numerator kg weight gain. the utilisation the of available amino acids for protein gain, 9.34;gain. BW is the current BW, kg; and the calculates protein content in 1Equation kg of weight kg; 12.3and Prediction the numerator of N, calculates P and K theutilisation protein content andinexcretion 1 kg of weight gain. 12.3 Prediction N, and K utilisation and 12.3amount Prediction ofKPN, P and utilisation excretion The of N, Pofand excreted areK calculated as excretion the and difference between ingested amounts 12.3 Prediction N, and K in utilisation and excretion and incorporated in milk, used and The incorporation of N, amounts P and K The that amount of N, Pofand KPexcreted aregestation calculated asfor thebody. difference between ingested The amount of PPand KKexcreted calculated thebody. difference between The amount ofN, N,gestation and excreted are as the difference between ingested amounts and in milk, gain and corresponds tocalculated the requirements ofThe protein, P and ingested K described in K and that incorporated in milk, used inare gestation andasfor incorporation of N, amounts P and and that incorporated in milk, used gestation andforforbody. body.of The incorporation ofN, N,PPand and K in milk, chapter 9. in milk, gain and gestation corresponds to the and requirements protein, P and K of described in K that incorporated in milk, used in in gestation The incorporation in milk, gain and gestation corresponds to the requirements of protein, K described in chapter gain and9. gestation corresponds to the requirements of protein, P andPKand described in Chapter 9. chapter Nitrogen 9. 12.3.1 utilisation and excretion 12.3.1 Nitrogen in utilisation excretion The N efficiency ruminantsand is low. The utilisation is higher in lactating animals than in 12.3.1 Nitrogen utilisation and excretion 12.3.1 Nitrogen utilisation excretion growing cattle andin non-producing animals. N efficiency is calculated as theanimals total amount The N efficiency ruminantsand is low. The utilisation is higher in lactating than inof N The ruminants isislow. is is higher in lactating thanthan inofin utilized incattle relation to its total intake: growing andinin non-producing animals. Nutilisation efficiency ishigher calculated as theanimals total amount N growing The NNefficiency efficiency ruminants low.The Theutilisation in lactating animals growing cattle and non-producing animals. N efficiency is calculated as the total amount of N utilized in relation to its total intake: cattle and non-producing animals. N efficiency is calculated as the total amount of N utilized in utilized intorelation Ntointake: _ its u total intake: relation its total [Eq. 12.6] ⋅ 100 N _ u _ pct = u i ⋅ 0.16 DMIN i ⋅ _CP [Eq. 12.6] ⋅ 100 N _ u _ pct = ∑ _ u i DMIN ⋅ CP ∑ i i ⋅ 0.16 ⋅ 100 [Eq. 12.6] N _ u _ pct = 12.6 i DMIi ⋅ CPi ⋅ 0.16 ∑

i is the N utilization as a proportion of the total intake, %; N_u is the amount of where N_u_pct where N_u_pct the utilization a proportion ofofthe intake, %; N_u isamount theCP amount of N N utilized, Equation 12.7; DMIi is the matter intake thetotal i=1...n'th is where N_u_pct isisthe NNutilization as as adry proportion of the total intake, %;feedstuff, N_u is thekg/d; i of utilized, Equation 12.7; DMI is the dry matter intake of the i=1...n’th feedstuff, kg/d; CP is where N_u_pct iscontent the12.7; N utilization as adry proportion of theof total intake, N_u thekg/d; amount the crude protein inDMI theiii=1...n'th feedstuff, g/kg DM; 0.16%; isfeedstuff, the Nisproportion N utilized, Equation is the matter intake theand i=1...n'th CPin is i the i of crude protein content in the i=1...n’th feedstuff, DM; and 0.16 is the N proportion in in N utilized, Equation 12.7; is the dry matter g/kg intake of the i=1...n'th kg/d; CP is protein. the protein crude protein content inDMI the feedstuff, g/kg DM; and 0.16 isfeedstuff, the N proportion ii=1...n'th ithe the crude proteinprotein content in the i=1...n'th feedstuff, g/kg DM; and 0.16 is the N proportion in N _ protein u = N _ milk + N _ gest + N _ gain the 12.7 [Eq. 12.7] N _ u = N _ milk + N _ gest + N _ gain [Eq. 12.7] N _ u = N_u N _ milk + Ntotal _ gestamount + N _ gain where the utilised g/d; N_milk, N_gest N_gain are amount the amount of N 12.7] where N_u isisthe total amount ofofutilised N,N, g/d; N_milk, N_gest andand N_gain are [Eq. the incorporated in milk, foetus and weight gain, Equation 12.8, 12.9 and 12.10 of N incorporated in milk, foetus weight gain, Equation 12,8, 12.9 and 12.10 respectively. where N_u is the total amount of and utilised N, g/d; N_milk, N_gest and N_gain arerespectively. the amount where N_u is the total amount of and utilised N, g/d; N_gest are respectively. the amount of N incorporated in milk, foetus weight gain,N_milk, Equation 12,8, and 12.9N_gain and 12.10 of incorporated milk, foetus and weight gain, Equation 12,8, 12.9 [Eq. and 12.8]12.10 respectively. N _Nmilk = MPY ⋅ 0.in 15674 12.8 [Eq. 12.8] N _ milk = MPY ⋅ 0.15674 [Eq. 12.8] N _ milk = MPYis⋅ 0the .15674 where N_milk the amount of of N N incorporated incorporated into into milk, milk, g/d; g/d; MPY MPY is the milk protein yield, Equation where is where N_milk N_milk is is the amount amount of N incorporated into milk, g/d; MPY is the the milk milk protein protein yield, yield, 183 3.4; and 0.15674 is the N content in milk protein (ISO 8968-5|IDF 020-5:2001). Equation 3.4; 3.4; and and 0.15674 0.15674 is is the the N N content content in in milk milk protein protein (ISO (ISO 8968-5|IDF 8968-5|IDF 020-5:2001). 020-5:2001). Equation 183 183 [Eq. 12.9 N _ gest = AAT _ gest ⋅ 0 . 16 ⋅ 0 . 5 N [Eq. 12.9] 12.9] N _ gest = AAT _ gest ⋅ 0.16 ⋅ 0.5 N

whereN_gest N_gestis theN incorporated into foetus, AAT is the amino required for where N_gest isisthe the NNincorporated incorporated into thethe foetus, g/d;g/d; AAT _gest is the the amino acidacid required N N_gest where into the foetus, g/d; AAT is amino acid required N_gest for gestation, Equation 9.42; 0.16 is the N proportion in the protein; and 0.5 is the utilisation gestation, Equation 9.42; 0.16 is the N proportion in the protein; and 0.5 is the utilisation for gestation, Equation 9.42; 0.16 is the N proportion in the protein; and 0.5 is the utilisation factor for factor for the the amino acids1985). (NRC, 1985). 1985). the amino acids (NRC, factor for amino acids (NRC, N_ _ gain gain = gain _ _ prot prot ⋅⋅ 00..16 16 N = gain

[Eq. 12.10] 12.10] [Eq.

12.10

where into the BW gain, g/d; gain_prot is gain, whereN_gain N_gainis theN incorporated into BW gain, gain_prot is protein the protein where N_gain isisthe the NNincorporated incorporated into thethe BW gain, g/d;g/d; gain_prot is the the protein gain,gain, Equation Equation 9.8; 0.16 is in 9.8; and 0.16 is the N proportion in the protein. Equation 9.8; and and 0.16 is the the N N proportion proportion in the the protein. protein. Total amounts amounts of of excreted excreted N N is is assumed assumed to to be be the the difference difference between between the the N N in in the the CP CP intake intake Total from from the the ration ration and and N_u: N_u: N = N_ _ excreted excreted =∑ DMI ii ⋅⋅ CP CPii ⋅⋅ 00..16 16 − −N N __ uu ∑ DMI 134 i i

12.11] [Eq. 12.11] system NorFor – The Nordic feed [Eq. evaluation

where N_excreted N_excreted is is the the total total amount amount of of N N excreted, excreted, g/day; g/day; DMI DMIi is is the the DM DM intake intake of of the the of of where

4923 4923 4919 4922 4924 4924 4920 4923 4925 4925 4921 4924 4926 4926 4922 4925 4927 4927 4923 4926

4924 4927 4928 4928 4925 4929 4929 4928 4926 4930 4930 4929 4927 4931 4931 4930 4932 4932 4931 4928 4933 4933 4929 4932 4934 4934 4933 4930 4935 4935 4934 4931 4936 4936 4935 4932

4936 4933 4937 4937 4934 4937 4938 4935 4938 4939 4936 4939 4938 4940 4940 4939 4937 4941 4941 4940 4942 4942 4938 4941 4943 4943 4939 4942 4944 4944 4943 4940 4945 4945 4944 4941 4946 4946 4945 4942 4947 4947 4946 4943 4948 4948 4944 4947 4949 4949 4945 4948 4950 4950 4946 4949 4951 4951 4947 4950 4952 4951 4948 4952 4949 4952 4953 4953 4954 4950 4955 4951 4953 4956 4952 4957 4953 4958 4959 4960 4961 4962 4963 4964 4965 4966 4967 4968 4969 4970

N _ gain =9.8; gainand _ prot ⋅ 0.16 [Eq. intake 12.10] Equation 0.16 is the is N proportion in the protein. Total Total amounts amounts of of excreted excreted N N is assumed assumed to to be be the the difference difference between between the the N N in in the the CP CP intake from the ration and N_u: from rationis and N_u: wherethe N_gain the N incorporated into to thebeBW g/d; gain_prot gain, Total amounts of excreted N is assumed thegain, difference between is thethe N protein in the CP intake

Equation 9.8;= and 0.16 is the from the ration N_u: N _ excreted DMI ⋅ CP ⋅ 0N .16proportion − N _ u in the protein. ∑and N _ excreted = ∑ DMI ii ⋅ CPii ⋅ 0.16 − N _ u

[Eq. [Eq. 12.11] 12.11]

ii Total amounts of excreted isassumed difference between theinNthe in[Eq. theintake CP intake Total amounts excreted be be thethe difference between the N CP N _ excreted = ∑of DMI ⋅ 0Nis .16 −assumed N _ u to to 12.11] i ⋅ CPi N the ration and iN_u: from the ration andisN_u: where N_excreted the total amount of N excreted, g/day; DMI is the DM intake of the of

where N_excreted is the total amount of N excreted, g/day; DMIii is the DM intake of the of content in i=1...n'th feedstuff, the i=1...n'th feedstuff, kg DM/day; CP isNthe the crude crude protein protein content in the the i=1...n'th feedstuff, the i=1...n'th feedstuff, kgtotal DM/day; CPofii is is the DM intake of the of where N_excreted is ithe amount g/day; DMI iamounts N _ excreted = ∑isDMI ⋅ CP −N _inu CP; excreted, [Eq. 12.11] g/kg DM; 0.16 the proportion N and N_u is the total of N utilized, i ⋅ 0.16 of g/kg DM; 0.16 is the proportion of N in CP; and N_u is the total amounts of N utilized, is the crude protein content in the i=1...n'th feedstuff, the i=1...n'th feedstuff, kg DM/day; CP i i Equation 12.15. Equation g/kg DM;12.15. 0.16 is the proportion of N in CP; and N_u is the total amounts of N utilized,

from

12.11

where N_excreted is the total amount of N excreted, g/day; DMI is the DM intake of the of the

i Equation 12.15. intake of the of where N_excreted is the total amount ofisN excreted, DMI i is the Excreted N in the assumed to be the difference between N and apparent i=1...n’th kgis CP the crude g/day; protein content inDM the i=1...n’th feedstuff, g/kg Excreted Nfeedstuff, infeedstuff, the faces faces isDM/day; assumed to be difference between N intake intake and apparent i isthe the i=1...n'th kg DM/day; CP the crude protein content in the i=1...n'th feedstuff, i digested N. DM; 0.16 is the proportion of N in CP; and N_u is the total amounts of N utilized, Equation 12.15. digested Excreted N0.16 in the faces is assumed to in beCP; the and difference intake and g/kg DM;N. is the proportion of N N_u isbetween the total N amounts of Napparent utilized, digested EquationN. 12.15. ⎛ ⎞ Excreted N⎛inDMI the feaces−istdassumed apparent digested N. ⎞ .16to be the difference between N intake and [Eq. N N_ _ faeces faeces = = ⎜⎜ ∑ CPii − td _ _ CP CP⎟⎟⎟ ⋅⋅ 0 0.16 ∑ DMIii ⋅⋅ CP [Eq. 12.12] 12.12] ⎠⎠⎞ to be the difference between N intake and apparent Excreted N ⎝⎜⎝⎛inii the faces is assumed ⎜ ⎟ 12.12 N _ faeces = DMI ⋅ CP − td _ CP ⋅ 0 . 16 ∑ [Eq. 12.12] i i digested N. ⎜ ⎟ ⎝ i is the amount of ⎠N excreted in the faeces, g/d; DMI is the dry matter intake where N_faeces where N_faeces is the amount of N excreted in the faeces, g/d; DMIii is the dry matter intake where N_faeces is the amount of inprotein the faeces, g/d; matter g/kg intake of the the content in the feedstuff, of the kg/d; i is the dry ⎛ feedstuff, ⎞ iiNis isexcreted the crude crude protein content in DMI theisi=1...n'th i=1...n'th feedstuff, g/kg of the i=1...n'th i=1...n'th feedstuff, kg/d;ofCP CP where N_faeces istotal amount N⋅ 0excreted in7.52; the faeces, g/d; DMI the dry matter intake ⎜⎜is∑ DMI ⎟⎟the N _ faeces =feedstuff, ⋅kg/d; CP .Equation 16 i [Eq. 12.12] i=1...n’th CP_iCP isCP, crude protein content in is the i=1...n’th feedstuff, g/kg DM; td_CP ithe i − td DM; td_CP the digested and 0.16 the proportion of N in CP. DM; td_CP is the total digested CP, Equation 7.52; and 0.16 is the proportion of N in CP. ⎝ i feedstuff, kg/d; CP ⎠ is the crude protein content in the i=1...n'th feedstuff, g/kg of the i=1...n'th

is the total digested CP, Equation i7.52; and 0.16 is the proportion of N in CP.

DM; td_CP isprotein the total CP, Equation 7.52; and 0.16 is the proportion of N in CP. Overfeeding or digested pourly ration N in urine.The difference between Overfeeding protein pourly balanced balanced rationinincreases increases N g/d; in the the urine.The difference between where N_faeces is theoramount oftheNNexcreted thethe faeces, DMI dry matter intake i is the total amount of excreted N and excreted in faeces is assumed to be the N excreted in Overfeeding protein or pourly balanced rationinincreases Nisinassumed the urine.The difference between total total amount of excreted N and the N excreted the faeces to be the N excreted in of Overfeeding the i=1...n'th protein feedstuff, or pourly kg/d; balanced CP ration increases N in the urine.The difference between is the crude protein content in the i=1...n'th feedstuff, g/kg i the urine: amount of excreted N and the N excreted in the faeces is assumed to be the N excreted in the urine: the urine: total excreted N and the excreted7.52; in theand faeces to be the DM; amount td_CP isofthe total digested CP,NEquation 0.16isisassumed the proportion of N N excreted in CP. in the urine: N _ urine = N _ excreted − N _ faeces [Eq. N _ urine = N _protein excreted N _ faeces [Eq. 12.13] 12.13] Overfeeding or−pourly balanced ration increases N in the urine.The difference between 12.13 total amount excreted the N excreted in the faeces is assumed to be the N excreted N _ urine = N _ofexcreted − NN_and faeces [Eq. 12.13]in where N_urine N_excreted total amount amount of of N N excreted, excreted, Equation where N_urine is is the the N N excretion excretion in in urine, urine, g/d; g/d; N_excreted is the the total where N_urine is the N excretion N_excreted is the urine: Equation 12.11; and N_faeces is the amount of N excreted in faeces, Equation 12.12. Equation 12.11; is the amount of N_excreted N excreted in 12.12. 12.11;N_urine and N_faeces the amount of N excreted in faeces, Equation 12.12.of where isand theN_faeces Nisexcretion in urine, g/d; is faeces, the totalEquation amount N excreted, N _ urine = N _ excreted − N _ faeces [Eq. 12.13] Equation 12.11; and N_faeces is the amount of N excreted in faeces, Equation 12.12. N _ urine N _ urine [Eq. N 12.14 [Eq. 12.14] 12.14] N __ urine urine __ pct pct = = N _ excreted ⋅⋅ 100 100 N _excreted urine where N_urine isN_the N excretion in urine, g/d; N_excreted is the total amount of [Eq. N excreted, 12.14] N _ urine _ pct = ⋅ 100 where N_urine_pct isN_faeces excreted ininurine thethe total NN excretion, where N_urine_pct isthe theNN excreted urine aproportion proportion of total excretion, Equation 12.11;Nand is the amount ofasas Na excreted in of faeces, Equation 12.12. %;%; N_urine is _ excreted N_urine the NEquation in urine, 12.13; Equation and N_excreted is the total amount of N excreted, the N in isurine, and12.13; N_excreted is the total amount of N excreted, Equation 12.11. Equation 12.11. N _ urine [Eq. 12.14] N _ urine _ pct = ⋅ 100 N _ excreted 12.3.2 Phosphorus utilisation and excretion 184 12.3.2 Phosphorus utilisation and excretion 184 The 1980). Therefor The excessary excessaryfed fedPPover overrequirement requirementisisexcreted excreted thefaeces faeces(ARC, (ARC, 1980). Therefore overfeeding 184 ininthe overfeeding P is useless. P is useless. PP efficiency efficiencyisiscalculated calculatedas: as: P _ u _ pct =

P_u

∑ DMI i ⋅ Pi i

184

⋅100

[Eq. 12.15]

12.15

where P_u_pct P_u_pctisisthe thePPutilization utilizationas asaa proportion proportionof oftotal totalPPintake, intake,%; %;P_u P_uisisthe theamount amountof ofPP utilized, where EquationEquation 12.16; DMI the iDM intake of the of of the theof i=1...n’th feedstuff, kg DM/day; and Pi is the utilized, 12.16; is the DM intake the i=1...n'th feedstuff, kg DM/day; i isDMI P content inPthe i=1...n’th DM. g/kg DM. and Pi is the content in thefeedstuff, i=1...n'thg/kg feedstuff, P _ u = (P _ milk + P _ gest + P _ gain ) ⋅ 0.7

[Eq. 12.16]

12.16

4971 4972 4973 4974 4975 4976 4977 4978

whereP_u P_uisisthe thetotal totalamount amountofofutilised utilisedP,P,g/d; g/d;P_milk, P_milk,P_gest P_gestand andP_gain P_gain where areare thethe requirement of P for milk, gestation weight gain,and Equation and 9.53 requirement of P forand milk, gestation weight9.51, gain, 9.52 Equation 9.51,respectively; 9.52 and 9.530.7 is the absorption coefficient. 0.7 is the absorption coefficient. respectively;

4979

P _ excreted = ∑ DMI − P _ uevaluation system i ⋅ Pi feed NorFor – The Nordic

4980 4981

where P_excreted is the total amount of P excreted, g/d; DMIi is the DM intake of the of the

Phosphorus excretion occurs normally in the faeces and only a minor part excretes in the urine (ARC, 1980). The excretion is estimated as the difference of total P intake and the amounts of P incorporated in milk, foetus and weight gain. i

[Eq. 12.17]

135

4970 4970 4971 4971 4972 4972 4973 4973 4974 4974 4975 4975 4976 4976 4977 4977 4978 4978

[Eq. 12.16] where where P_u P_u is is the the total total amount amount of of utilised utilised P, P, g/d; g/d; P_milk, P_milk, P_gest P_gest and and P_gain P_gain are are the the requirement of of P P for for milk, milk, gestation gestation and and weight weight gain, gain, Equation Equation 9.51, 9.51, 9.52 9.52 and and 9.53 9.53 requirement respectively; respectively; 0.7 0.7 is is the the absorption absorption coefficient. coefficient.

Phosphorus excretionoccurs occurs normally in the faeces and only a minor part excretes in the urine Phosphorus Phosphorus excretion excretion occurs normally normally in in the the faeces faeces and and only only aa minor minor part part excretes excretes in in the the (ARC, 1980). The The excretion is estimated as as thethe difference urine (ARC, 1980). The excretion is estimated estimated as the differenceofof oftotal totalPP Pintake intake and and the urine (ARC, 1980). excretion is difference total intake and the amounts of P incorporated in milk, foetus and weight amounts of P P incorporated incorporated in milk, milk, foetus gain. and weight weight gain. gain. amounts of in foetus and

4979 4979

P =∑ DMI ⋅ P − P _ u ∑ P __ excreted excreted = ∑ DMI iii ⋅ Piii − P _ u

4980 4980 4981 4981 4982 4982 4983 4983

i the DM intake of the of the where P_excreted is excreted, g/d; DMI is the DM intake the ofand the P_u is the where P_excreted is the the total total amount of P excreted, g/d;i=1...n’th DMIiii is feedstuff, kg DM/day; and Pamount is the of PP content in the feedstuff, g/kgofDM; i i=1...n'th feedstuff, feedstuff, kg kg DM/day; DM/day; and and P Pii is is the the P P content content in in the the i=1...n'th i=1...n'th feedstuff, g/kg DM; i=1...n'th i and amount of P utilised in milk, gestation gain, Equation 12.16. feedstuff, g/kg DM; and and P_u P_u is is the the amount amount of of P P utilised utilised in in milk, milk, gestation gestation and and gain, gain, Equation Equation 12.16. 12.16.

4984 4984 4985 4985 4986 4986 4987 4987 4988 4988 4989 4989 4990 4990 4991 4991 4992 4992 4993 4993

4994 4994 4995 4995 4996 4996

4997 4998 4999 5000 5001 5002 5003 5004 5005 5006 5007 5008 5009 5010 5011 5012 5013 5014 5015

[Eq. 12.17] 12.17] [Eq.

iii

12.17

where P_excreted is the total amount of P excreted, g/d; DMI is the DM intake of the of the i=1...n’th

12.3.3 Potassium utilisation utilisation andexcretion excretion 12.3.3 12.3.3 Potassium Potassium utilisation and and excretion

Cattle normally ingest more K than the requirement. Potassium utilisation is as: Cattle Potassium utilisation is calculated calculated as: as: Cattle normally normallyingest ingestmore moreKKthan thanthetherequirement. requirement. Potassium utilisation is calculated K_ _u u K = ⋅⋅ 100 100 K __ uu __ pct pct = K ∑ DMI ∑ DMIii ⋅⋅ K K ii ∑ i

iii

[Eq. 12.18] 12.18] [Eq.

i

12.18

where K_u_pct isisthe the KKutilization utilization asas proportion of of total K intake, intake, %; %; K_uK_u is the the total whereK_u_pct K_u_pctis theK utilizationas a proportion total K intake, is total the total amount of where aa proportion of total K %; K_u is amount of utilized, Equation 12.19; DMI the DM of the offeedstuff, the i=1...n'th feedstuff,and Ki is K utilized, DMI DM intake ofintake the i=1...n’th kg DM/day; ii is amount of K KEquation utilized, 12.19; Equation 12.19; DMI i is the i is the DM intake of the of the i=1...n'th feedstuff, isi=1...n’th the K K content content in the the g/kg i=1...n'th feedstuff, g/kg g/kg DM. DM. kg DM/day; and Kthe ii is theDM/day; K content inK feedstuff, DM.feedstuff, the in i=1...n'th kg and i

K_ _u u= = (K K_ _ milk milk + +K K_ _ gest gest + +K K_ _ gain gain) ⋅⋅ 0 0..9 9 K

12.19 [Eq. 12.19] 12.19] [Eq. where K_u K_u is is the the total total amount amount of of utilised utilised K, K, g/d; g/d; P_milk, P_milk, P_gest P_gest and and P_gain P_gain are are the the where requirement K milk, gestation and gain, Equation 9.66, and where K_u of is the total amount of utilised K, g/d; P_gest and P_gain requirement of K for for milk, gestation and weight weight gain,P_milk, Equation 9.66, 9.67 9.67 and 9.68 9.68are the requirement respectively; 0.9gestation is the the absorption absorption coefficient. of K for milk, and weight gain, Equation 9.66, 9.67 and 9.68 respectively; 0.9 is the respectively; 0.9 is coefficient. absorption coefficient.

185 185

The total totalamount amountofofexcreted excretedKKisiscalculated: calculated: The K _ excreted = ∑ DMI i ⋅ K i − K _ u i

[Eq. 12.20]

12.20

where K_excreted is the K excreted, g/day; DMI is the DM intake of the of the i=1...n’th feedstuff,

i DM intakeg/kg of the of and the i=1...n'th where K_excreted thethe K excreted, i is thefeedstuff, kg DM/day; and Kisi is K contentg/day; in theDMI i=1...n’th DM; K_u is the amount of feedstuff, kg DM/day; and Ki is the K content in the i=1...n'th feedstuff, g/kg DM; and K_u is K utilized in milk, gestation and gain, Equation 12.19. the amount of K utilized in milk, gestation and gain, Equation 12.19.

References 12.4 References ARC (Agricultural Reasearch Council), 1980. The nutrient requirements of ruminant livestock. Commonwealth

ARC – Agricultural Reasearch Council, 1980. The nutrient requirements of ruminant Agricultural Bureaux, Slough, UK. livestock. Slough England: Commonwealth Agricultural Bureaux.

ISO 8968-5/IDF 020-5:2001. Milk - Determination of nitrogen content. International Dairy Federation, Brussels, ISO Belgium. 8968-5/IDF 020-5:2001. Milk - Determination of nitrogen content. International Dairy Sjaunja, L.-O., L. Bævre,Belgium. I. Junkkarainen, J. Pedersen and J. Setälä, 1990. A Nordic proposition for an energy corrected Federation, Brussels, milk (ECM) formula. 26th session of the International Committee for recording the productivity of Milk Animals (ICRPMA). Sjaunja, L.-O., L. Bævre, I. Junkkarainen, J. Pedersen and J. Setälä, 1990. A Nordic

proposition for an energy corrected milk (ECM) formula. 26th session of the International Committee for recording the productivity of Milk Animals (ICRPMA).

136

NorFor – The Nordic feed evaluation system

5016 5017 5018

13.0 Standardfeed feedvalue value 13. Standard

5019 5020 5021 5022 5023 5024 5025 5026 5027

Traditional feed evaluation systems determine fixed and additive feed values of feedstuffs. The NorFor system, however,systems is a rationdetermine evaluationfixed system in principle individual Traditional feed evaluation andand additive feed each values of feedstuffs. The feedstuff has no fixed feed value because of the interactions in the digestion and metabolism NorFor system, however, is a ration evaluation system and in principle each individual feedstuff of there is aofneed compare feedstuffs for trading or in of thenutrients. hasnutrients. no fixedNevertheless, feed value because the to interactions in the digestion andpurposes metabolism optimization concentrate mixtures by linear programming in thepurposes feedstuff or industry. Nevertheless,ofthere is a need to compare feedstuffs for trading in the optimization of Therefore, we calculate feed values based on standardized conditions, where important model concentrate mixtures by linear programming in the feedstuff industry. Therefore, we calculate feed variables are fixed, e.g. fractional passage rates and efficiency in the microbial protein values based on standardized conditions, where important model variables are fixed, e.g. fractional synthesis (see Table 13.1). These fixed values represent feed rations for diets used in passage rates and efficiency inSweden. the microbial protein synthesis (see Table 13.1). These fixed values Denmark, Iceland, Norway and

5028 5029 5030 5031 5032 5033 5034 5035 5036 5037

Standard feed values for NEL, AATN and PBVN were calculated in the following way. First Standard feed values for NEL,proportion AATN andand PBV in the animal BW, DMI, concentrate content ofcalculated CP, NDF, ST andfollowing RestCHOway. wereFirst set animal N were BW, DMI,toconcentrate proportion and of CP, NDF, STthe andstandard RestCHO according Table 13.1 to standardize thecontent feed ration from which feedwere valueset is according calculated. Then for RLI, fractional rates and to Table 13.1 to values standardize thecorrNDF_fac, feed ration from which passage the standard feedr_emCP value isare calculated. Then determined usingcorrNDF_fac, Equations 13.1fractional to 13.8 (Table 13.1). values substitute the values for RLI, passage ratesThese and r_emCP are determined using Equations corresponding equations in chapter 7 whensubstitute calculating standard feed equations value of a in feedstuff 13.1 to 13.8 (Table 13.1). These values thethe corresponding Chapter 7 when with specificthe dietary characteristics. of NEL, AAT N and PBVN The standard calculating standard feed value The of a standard feedstufffeed withvalues specific dietary characteristics. (Equation 8.3, and 13.9, and respectively) receive a8.3, suffix thatand tells which standardisedreceive DMI, a suffix feed values of8.11 NEL, AAT PBV (Equation 8.11 13.9, respectively) N N 8that or 20 kgwhich DM, the calculations are based upon: NEL8the , NEL PBV and , NEL , 20, PBVN8,are N20, AAT tells standardised DMI, 8 or 20 kg DM, calculations based upon:N8NEL 8 20 AAT PBVN20,. PBV , AAT and AAT .

5038 5039

Standardised RLI, corrNDF_fac, fractional passage rates and r_emCP are calculated as:

5040

RLIstd =

5041 5042 5043 5044 5045 5046 5047 5048 5049 5050 5051 5052 5053 5054 5055 5056

M. Åkerlind and H. Volden

M. Åkerlind and H. Volden

represent feed rations for diets used in Denmark, Iceland, Norway and Sweden.

N8

N20

N8

N20

Standardised RLI, corrNDF_fac, fractional passage rates and r_emCP are calculated as: 280 DMIstd ⋅ 370

where, RLI

[Eq. 13.1]

13.1

is the rumen load index; 280 is the sum of rumen degraded ST and RestCHO in the

std where, RLIDM; rumen loadstandardized index; 280 isDM the sum of rumen RestCHO in and 370 std is the diet, g/kg DMI the intake, which degraded is either 8ST orand 20 kg DM/day; std is is the diet, g/kg DM; DMI the standardized DM intake, which is either 8 or 20 kg DM/day; std is the NDF content, g/kg DM. and 370 is the NDF content, g/kg DM.

The NDF correction factor is calculated as:

Table 13.1. Fixed values for the listed parameters used for determining standard feed values. corrNDF Parameter_ fac = e

(

− 0 .24 ⋅ RLI

)

2 .9

std

Equation

8 kg DMI[Eq. 13.2]20 kg DMI

where, is kg the correction factor for the NDF degradation rate; CurrentcorrNDF_fac body weight, 600and RLIstd is the600 rumen load index, Dry matter intake,Equation kg/d 13.1. 8 20

Concentrate proportion of the ration, % DMIstd ⋅1000 DM rCP, _ kplg/kg = 2.45 + 0.055 ⋅ + 0.0004 ⋅ 502 NDF, g/kg DM 6000.75 ST+RestCHO, g/kg DM DMI DM std ⋅ 1000 − 0.02 ⋅ 50 rrdST+rdRestCHO, _ kpc = 2.504 + 0.1375g/kg ⋅ Correction factor for NDF600 degradation rate Passage rate for liquid, %/h Passage rate for CP and ST in concentrate, %/h Passage rate for NDF in concentrate, %/h 187 Passage rate for CP and ST in roughage, %/h Passage rate for NDF in roughage, %/h Efficiency of microbial CP synthesis, g/kg rd_OM

13.2 13.3 13.4 13.5 13.6 13.7 13.8

50 50 160 160 [Eq. 13.3] 370 370 310 310 280 [Eq. 13.4]280 0.89857 0.89857 7.07943 12.5236 3.33733 6.08733 1.43505 2.61755 2.30177 4.47943 0.998465 1.64242 133.383 184.329

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_13, © Wageningen Academic Publishers 2011

137

5041 5042 5042 5043 5043 5044 5044 5045 5045 5046 5046 5047 5047 5048 5048 5049 5049 5050 5050 5051 5051 5052 5052

5053 5053 5054 5054 5055 5057 5055 5057 5056 5058 5056 5058 5057 5059 5059 5060 5058 5060 5059 5060 5061 5061

5061 5061 5062 5062 5063 5063 5064 5062 5064 5065 5062 5063 5065 5066 5064 5066 5067 5065 5067 5068 5066 5068 5069 5067 5069 5070 5068 5070 5071 5069 5071 5072 5070 5072 5071 5073

5073 5074 5072 5074 5075 5073 5075 5076 5074 5076 5077 5075 5077 5078 5076 5078 5079 5077 5079 5080 5078 5080 5081 5079 5081 5082 5080 5082 5083 5081 5083 5084 5082 5084 5085 5083 5085 5086 5084 5086 5087 5085 5087 5088 5086 5088 5089 5087 5089 5090 5088 5090 5091 5089 5091 5092 5090 5092 5093 5091 5093 5094 5092 5094 5095 5093 5095 5096 5094 5096 5097 5095 5097

where, RLIstd is the rumen load index; 280 is the sum of rumen degraded ST and RestCHO in is the rumen load index; 280 is the sum of rumen degraded ST and RestCHO in where, RLIstd the diet, g/kg DM; DMIstd is the standardized DM intake, which is either 8 or 20 kg DM/day; the diet, g/kg DM; DMIstd is the standardized DM intake, which is either 8 or 20 kg DM/day; and 370 is the NDF content, g/kg DM. and 370 is the NDF content, g/kg DM. The NDF correction factor is calculated as: The NDF correction factor is calculated as:

The NDF correction factor is calculated as:

((

− 0 .24 ⋅ RLI

− 0 .24 ⋅ RLI std corrNDF fac e − 0 .24 ⋅ RLI stdstd corrNDF __ fac == e

))

2 2 ..9 9 2 .9

[Eq. 13.2] [Eq. 13.2]

13.2

where, corrNDF_fac is the correction factor for the NDF degradation rate; and RLIstd is the where,corrNDF_fac corrNDF_facisisthe thecorrection correctionfactor factor NDF degradation rate; is the rumen the where, forfor thethe NDF degradation rate; andand RLIRLI std isstd rumen load index, Equation 13.1. rumen load index, Equation load index, Equation 13.1. 13.1. DMIstd ⋅ 1000 ⋅ 1000 + 0.0004 ⋅ 5022 r _ kpl = 2.45 + 0.055 ⋅ DMIstd r _ kpl = 2.45 + 0.055 ⋅ + 0.0004 ⋅ 50 0.75 600 0 . 600 75

[Eq. 13.3] [Eq. 13.3]

13.3

⋅ 1000 DMIstd DMI ⎞ ⎛ 0.1375 ⋅ DMI std ⋅1000 13.4 − 0.02 ⋅ 50 [Eq. 13.5] 13.4] r _ kpNDFc kpc = 2.504 .02 ⋅50⎟ ⋅ 0.43 = ⎜ 2++.504 + 0.1375⋅std ⋅ 1000 [Eq. rr __ kpc = 2.504 − 0.02−⋅050 [Eq. 13.4] DMIstd ⋅1000 ⎞⎠ ⎛⎝ 0.1375 ⋅ 600 600 600 r _ kpNDFc= ⎜ 2.504 + 0.1375⋅ − 0.02 ⋅ 50⎟ ⋅ 0.43 [Eq. 13.5] 600 ⎠ ⎝ DMIstd 1000 ⋅ ⎞ ⎛⎛ ⋅ DMI 1000 ⎞ std kpNDFc= 504 + 1375std ⋅⋅ ⋅ 1000 − [Eq. 13.5] 13.5 0..02 02 2⋅⋅ 50 50⎟⎟ ⋅⋅ 0 0..43 43 = ⎜⎜+220..504 + 00DMI −0 rrr ___ kpNDFc ..1375 kpr = 0.35 [Eq. 13.6] 600+ 0.0002 ⋅ 50187 ⎠⎠ ⎝⎝ .022 ⋅ DMIstd0.⋅75 600 1000 2 r _ kpr = 0.35 + 0.022 ⋅ 600 + 0.0002 ⋅ 50187 [Eq. 13.6] 600 0.75 DMI ⋅ 1000 DMI std std ⋅ 1000 + 13.6 rr __ kpr 0..0002 0002 ⋅⋅ 50 50 22 [Eq. 13.6] kpr = = 00..35 35 + + 00..022 022 ⋅⋅ +0 0 ..75 0 75 600 600 0.78 ⋅1.93668 r _ kpNDFr = 0.480339 + [Eq. 13.7] 0.78 ⋅1.93668 − 3.19822 ⎛ DMI std ⋅ 370 ⎞ r _ kpNDFr = 0.480339 + 13.7 [Eq. 13.7] 1+ ⎜ − 3.19822 ⎟ DMI⋅ 7std ⋅ 370 ⎠⎞ ⎛ 600 . 48383 ⎝ ⋅1.93668 ⎟ r _ kpNDFr = 0.480339 + 1 + ⎜⎝ 6000⋅.78 [Eq. 7.48383 ⎠ −− 33..19822 [Eq. 13.7] 13.7] ⋅ DMI 370 ⎛ ⎞ 19822 std std where, r_kpl is the passage 1 +rate ⎜ of the liquid ⎟ phase out of the rumen; DMIstd is the .48383 where, r_kpl r_kplDMI, thepassage rate the phase outthe ofrumen; the DMI where, isisthe rate of⋅ 7of the liquid outBW, of DMI ⎝ 600 ⎠ phase is the is standardized 8 passage or 20 kg DM; 600 isliquid the current kg; 50 isrumen; the proportion of the standardized std std

standardized 8%;or600 20 is kgthe 600 isBW, the current kg; 50 is the of in diet, %; r_kpc concentrate diet, r_kpc isDM; the passage rate protein starch inofproportion concentrates; DMI, 8 or 20inDMI, kg DM; current kg;of50 isBW, the and proportion concentrate where, r_kpl the passage rate of the liquid phase out of DMI ispassage theroughage std concentrate diet, %; r_kpc isofand the passage of protein and starch in is concentrates; r_kpNDFc isinis the passage rate NDF in concentrate; isrumen; the passage rate of where, r_kpl is the passage rate of the liquid phase outr_kpr of the the rumen; DMI is the passage rate of protein starch inrate concentrates; r_kpNDFc the rate of NDF in std standardized DMI, 8isor kg of DM; 600 isconcentrate; therate current BW, 50starch; is the proportion of r_kpNDFc the passage NDF inof r_kprkg; is the passage rate protein andisstarch; r_kpNDFr is the passage of roughage NDF; and 370 is of theroughage NDF concentrate; r_kpr the20rate passage rate roughage protein and r_kpNDFr is the passage rate concentrate inNDF; diet, %; r_kpc the passage rate andNDF; starchand in concentrates; protein and starch; theNDF passage rateof of roughage 370 is the NDF content in diet, g/kgr_kpNDFr DM. of roughage and 370 isisisthe content inprotein diet, g/kg DM. r_kpNDFc is theg/kg passage content in diet, DM. rate of NDF in concentrate; r_kpr is the passage rate of roughage protein and starch; is r_kpNDFr is from the passage rate of equation: roughage NDF; and 370 is the NDF The The fixed fixed r_emCP r_emCP iscalculated calculated fromthe thefollowing following equation: content in r_emCP diet, g/kgisDM. The fixed calculated from the following equation: ⎛1000 1000⋅⋅ DMI DMI

⎞

std ⎞ ⋅ 55.6 + 0.4166 ⋅ 310 − 0.0008868⋅31022 −54.56 ⎛ std rr __ emCP ⎟⎟ ⋅ 55.6 +from emCP ln⎜⎜ 0.4166 − 0.0008868 ⋅ 310 − 54.56 The fixed== ln r_emCP is calculated the⋅ 310 following equation: DMI ⎛⎝⎝1000⋅600 2 600 std ⎠⎞⎠

[Eq. 13.8]

13.8

r _ emCP= ln⎜ ⎟ ⋅ 55.6 + 0.4166⋅ 310 − 0.0008868⋅ 310 − 54.56 [Eq. 13.8] 600 ⎝ ⎠ ⋅ 1000 DMI ⎛ DMIefficiency 2 − 54.56 DMI is the std ⎞⎞ ⋅ 55.6 +of where, r_emCP isis⋅ the rumen microbial protein = ln _ emCP emCP ln⎛⎜⎜ 1000 ..4166 ⋅⋅ 310 − 0 ⋅⋅protein 310 ⎟⎟ ⋅ 55.6 + 0 [Eq. 13.8] std = rr _ 0of 4166 310 −microbial 0..0008868 0008868 3102synthesis; − 54 .56 where, r_emCP thestd efficiency rumen synthesis; DMI std is the standardized 600 ⎝⎝ DMI ⎠⎠kg DM; 600 where, r_emCP is the efficiency of rumen microbial protein synthesis; DMI standardized 8 or 20 600 is the standardized BW, kg;the and 310 std is is thethe of ST + RestCHO DMI 8 or 20 kg DM; 600 is the standardized BW, kg; and 310 is concentration

standardized DMI 8+ orRestCHO 20 kg DM; is theg/kg standardized BW, kg; and 310 is the concentration of ST in 600 the diet, DM. in the diet, g/kg DM. where, r_emCP the+ efficiency rumen concentration ofisST RestCHO of in the diet,microbial g/kg DM.protein synthesis; DMIstd std is the standardized DMI or 20 kgfeed DM; 600 is the8,standardized kg;AAT and N20 310uses is the The calculation of 8standard value, NEL NEL20, AATBW, the Equations N8 and The calculation of standard feedinvalue, NEL , AATN8 and AATN20 uses the Equations 7.6 8, NEL 20to concentration ofof ST + RestCHO the7.36 diet, DM. NEL AAT Equations The calculation standard value, NEL 7.6 to7.14, 7.17 to 7.23. 7.30feed to 7.34, tog/kg 7.40 7.51, 7.54, 7.55, 7.57,the 7.59, 8.1 to 8,7.38, 20, AAT N8 and N20 uses to7.14, 7.177.17 to 7.23. 7.30 to to 7.34, 7.36 to 7.38, 7.40 to 7.51, 7.54, 7.55, 7.57, 7.59, 8.1 to 8.3 and 7.6 and to7.14, to 7.23. the 7.30equations 7.34, 7.36 7.38, 7.40 to 7.51, 7.54, 7.55, 7.59, 8.1isto 8.3 8.11. However, 7.1 toto7.5 is changed to equation 13.37.57, to 13.7, 7.16 8.11. However, Equations 7.1 to 7.5 are changed to Equation 13.3 to 13.7, 7.16 is changed The calculation of standard feed value, NEL , NEL and AAT13.3 the Equations 8 20 N8 N20 8.3 and 8.11. However, to 7.5 isnutrient changed to 13.7, 7.16i,isSTi to 13.2 changed to 13.2 and 7.29the is equations changed to7.1 13.8. composition (CP CFat i,to i, NDF 8The 20, AAT N8 equation N20 uses and 7.29 is changed to 13.8. The nutrient composition (CP , CFat , NDF , ST , etc.) is the composition 7.6 to7.14, 7.17 to 7.23. 7.30 to 7.34, 7.36 to 7.38, 7.40 to 7.51, 7.54, 7.55, 7.57, 7.59, 8.1 to i i i i i changed to composition 13.2 and 7.29ofisthe changed 13.8. Thetonutrient composition (CP etc.) is the specifictofeedstuff which the feed values are calculated for i, CFat i, NDF i, ST of the feedstuffof tothe which the7.1 feed are calculated for and DMI kg.foris 8.3 and 8.11. the equations to values 7.5 is to feed equation 13.3 tocalculated 13.7, etc.) isspecific the composition specific feedstuff tochanged which the values are and DMI 1However, kg. i is 17.16 i is changed and DMIto 1 kg.and 7.29 is changed to 13.8. The nutrient composition (CPii, CFatii, NDFii, STii i is13.2 etc.) is the composition the specific feedstuff which the feed values calculated for The standardized standardized valueofofPBV PBV PBV in Equation 13.9 assumes a recirculation fixed recirculation of The value andand PBV intoEquation 13.9 assumes aare fixed N8N8 N20 N20 and DMI 1 kg. urea, based a fixed intake, totoavoid overestimating it: ii is on The standardized ofCP PBV PBV in Equation 13.9 of urea, based on avalue fixedCP intake, avoid overestimating it: assumes a fixed recirculation N8 and N20 of urea, based on a fixed CP intake, to avoid overestimating it: The and PBVN20 PBVstandardized 160 ⋅ 0.of 046PBV ) − r _N8 [Eq. 13.9] 13.9 N8mCP N20 in Equation 13.9 assumes a fixed recirculation N = rd _ CP + (value of urea, fixed CP) −intake, [Eq. 13.9] PBV rd _ CPon + (a160 ⋅ 0.046 r _ mCPto avoid overestimating it: N =based where, PBVN is the standard feed value for protein balance in the rumen, g/day; rd_CP is the (160 ) −−feed [Eq. 13.9] PBV =PBV rd ⋅⋅ 00..046 rr __ mCP PBVdegraded rd __NCP CP 160 046of mCP N is ++the the standard valueoffor balance in the rumen, 7.8; g/day; is the where, CP in rumen 1kg DM theprotein specific feedstuff, Equation 160rd_CP is a fixed N =PBV where, N is the standard feed value for protein balance in the rumen, g/day; rd_CP is the CP CP degraded rumen of 1kg DM of the specific feedstuff, 7.8; 160 a of fixed value for CP in the standardised ration (Table 13.1); 0.046 is Equation theEquation assumed proportion Nvalue that for CP degraded in the rumen of 1kg DM of the specific feedstuff, 7.8; 160 is is a fixed where, PBV is the the standard value forisprotein balance theassumed rumen, g/day; isNthe N value for CPback standardised ration (Table 13.1); 0.046 isinthe proportion that is recycled to the rumen;feed and r_mCP the rumen microbial CP yielded by rd_CP 1 kg of DM of Nin CPrecycled degraded in the rumen of 1kg DM of the specific feedstuff, Equation 7.8; by 1601 iskga DM fixedof is back to the rumen; and r_mCP is13.1). the rumen microbial CP yielded the specific feedstuff, Equation 7.30 (Table value in the standardised ration (Table 13.1); 0.046 is the assumed proportion of N that the Equation 7.30 (Table 13.1). 138specific for CPfeedstuff, NorFor – The Nordic feed evaluation system is recycledfeed back to theofrumen; and r_mCP is the rumen microbial yielded 1 kg Standard values some common feedstuffs are presented inCP table 13.2. by Note thatDM theof the specific feedstuff, Equation 7.30 than (Table 13.1). Standard feed valuesAAT of some feedstuffs presented table 13.2. Note that the lower AAT mainly due toin increased microbial standard feed value N8 is common N20 are standard feed value AAT is lower than AAT mainly due to increased microbial

in the standardised ration (Table 13.1); 0.046 is the assumed proportion of N that is recycled back to the rumen; and r_mCP is the rumen microbial CP yielded by 1 kg DM of the specific feedstuff, Equation 7.30 (Table 13.1). Standard feed values of some common feedstuffs are presented in Table 13.2. Note that the standard feed value AATN8 is lower than AATN20 mainly due to increased microbial efficiency with higher feed intake and therefore higher supply of AA from microbial protein. PBVN8 is higher than PBVN20, also due to increased microbial efficiency which incorporates more N into the microbial protein and therefore contributes less to the PBVN value. NEL8 is higher than NEL20, because increased DMI results in a higher passage rates and lowered digestibility and thus lower energy supply. Table 13.2. Standard feed values in some feedstuffs. Feedstuff

Barley Oat Triticale Brewers grain, dried Distillers grain, dried Rape seed meal Soy bean meal Vegetable fat Sugar beet pulp, dried unmolassed Grass silage very high OMD Grass silage high OMD Grass silage low OMD Maize silage high OMD Maize silage low OMD Straw, spring barley Glycerin, glycerol Urea

NorFor AATN8 PBVN8 NEL8 ID g/kg g/kg MJ/kg DM DM DM

AATN20 PBVN20 NEL20 g/kg g/kg MJ/kg DM DM DM

1-16 1-17 1-15 1-68 1-38 2-42 2-53 2-26 4-20

83 66 79 125 92 102 154 17 70

-6 20 -10 97 176 236 289 -6 -22

7.97 7.31 8.21 6.99 7.30 6.90 8.74 20.80 6.93

110 87 106 157 127 144 218 20 96

-44 -11 -50 55 127 180 210 -11 -63

7.49 6.78 7.80 6.38 6.69 6.52 8.32 18.98 6.26

6-460 6-461 6-463 6-307 6-308 6-386 12-12 13-1

60 63 62 68 66 39 82 0

92 76 49 -30 -30 -25 -120 2,922

7.73 7.50 6.64 7.36 7.05 3.44 8.12 0.00

79 80 75 87 84 44 109 0

62 49 28 -59 -56 -34 -163 2,922

7.00 6.69 5.74 6.60 6.21 2.47 7.49 0.00

NorFor – The Nordic feed evaluation system

139

5111 5112 5113 5114 5115 5116 5117 5118 5119 5120 5121 5122 5123 5124 5125 5126 5127 5128 5129 5130 5131 5132 5133 5134 5135 5136 5137 5138 5139 5140 5141 5142 5143 5144 5145 5146 5147

14.0 System evaluation H. N. I. evaluation Nielsen, M. Åkerlind and A. J. Rygh 14.Volden, System This chapterN.I. focuses on the of different aspects within the NorFor system. This H. Volden, Nielsen, M.evaluation Åkerlind and A.J. Rygh includes sensitivity analyses with respect to feed characteristics and model tests of total tract digestion, milk production, the energy system for growing cattle, RLI, feed intake and This chapter focuses on the evaluation of different aspects within the NorFor system. This includes chewing index.

sensitivity analyses with respect to feed characteristics and model tests of total tract digestion, milk production, the energy system for growing cattle, RLI, feed chewing 14.1 Sensitivity of feed degradation characteristics to intake rationand energy andindex. protein

values

14.1 Sensitivity of feed degradation characteristics to ration energy and

When the NorFor values system is used for on-farm ration optimization, forages are mainly analysed protein at commercial feed laboratories while for concentrate ingredients, values from the NorFor feed table are normally used (www.norfor.info). Within forage type, there are large variations When the NorFor system is used for on-farm ration optimization, forages are mainly analysed at in nutrient degradation characteristics, which affect (to varying degrees) the nutritional value commercial feed laboratories while for concentrate ingredients, values from the NorFor feed table of the feed ration. A sensitivity analysis is therefore useful for evaluating effects of changes in are composition normally used (www.norfor.info). forage there are large variations in nutrient the of feed ingredients on theWithin NEL and AATtype, N contents of the diet. A sensitivity degradation characteristics, affect (tothe varying degrees) the nutritional of the feed ration. analysis is also advantageouswhich for evaluating usefulness and errors related tovalue different A sensitivity analysis is therefore useful for evaluating effects of changes in the composition of feed routine analyses.

ingredients on the NEL and AATN contents of the diet. A sensitivity analysis is also advantageous

for evaluating the usefulness andiserrors to different routine in analyses. The sensitivity analysis presented basedrelated on a dairy cow simulation which a standard diet was fed and the influence of forage and concentrate degradation characteristics were tested. The sensitivity analysis presented based cow simulation in 140 which a standard Effects were tested at a DM intake ofis20 kg/don fora adairy cow of 600 kg BW and days in milk. diet was The diet the consisted of grass silageand (50% of diet DM) with an OMD of 71.4% and fed and influence of forage concentrate degradation characteristics werethe tested. Effects were concentrate mixture consisted of barley grain beet140 pulpdays (12%), maizeThe grain tested at a DM intake of 20 kg/d for a cow of(40%), 600 kgdried BW and in milk. diet consisted of (10%), soybean meal (16%), molasses (6%), rape seedmixture (5%) and mineralof barley grass silage (50% of diet DM)rapeseed with an meal OMD(9%), of 71.4% and the concentrate consisted and vitamins wasgrain (g/kg(10%), DM): CP, 179; NDF, 347; ST,rapeseed 191; CFat, grain (40%),(2%). dried The beetdiet pulpcomposition (12%), maize soybean meal (16%), meal (9%), 45; FPF, 56 and rape RestCHO, 105.and Input degradation characteristics tested sCP, iCP, molasses (6%), seed (5%) mineral and vitamins (2%). The dietwere: composition was (g/kg DM): kdCP, iNDF and347; kdNDF, and CFat, their basal, minimum maximum values arecharacteristics CP, 179; NDF, ST, 191; 45; FPF, 56 and and RestCHO, 105.parameter Input degradation presented in Table 14.1.kdCP, WheniNDF sCP and were changed, corresponding of pdCP parameter tested were: sCP, iCP, andiNDF kdNDF, and their basal, minimum values and maximum and pdNDF were changed simultaneously to maintain the same concentrations of CP and values are presented in Table 14.1. When sCP and iNDF were changed, corresponding values of NDF in the diet. The input parameter of interest was changed for one characteristic at a time pdCP and pdNDF were changed simultaneously to maintain the same concentrations of CP and and correlated responses were ignored to facilitate interpretation. For example, iNDF and NDF inare theoften diet.negatively The input correlated parameterwithin of interest changed for onetest, characteristic kdNDF feeds was but in the sensitivity they were at a time and correlated responses were to facilitate interpretation. example, iNDF and kdNDF are changed independently. Theignored sensitivity of diet energy and proteinFor values to parameter often negatively correlated within feeds but in the sensitivity test, they were changed variation was calculated as the change in the response variable relative to the change in independently. The sensitivity parameter value:of diet energy and protein values to parameter variation was calculated as the change in the response variable relative to the change in parameter value:

5148

⎛ ⎛ rmin − rmax ⎞ ⎞ ⎜ ⎜ ⎟ ⎟ ⎜ ⎜⎝ rbasal ⎟⎠ ⎟ Sensitivity = ⎜ ⎟ ⋅100 ⎜ ⎛⎜ p min − p 2 max ⎞⎟ ⎟ ⎟⎟ ⎜⎜ p basal ⎠⎠ ⎝⎝

5149 5150 5151 5152 5153

wheresensitivity sensitivityisiscalculated calculated in %, the minimum response rmaxmaximum is the maximum where in %, rmin ris theisminimum response value, value, rmax is the min response response value, pminpmin is the minimum parameter value, pmax is responsevalue, value,rbasal rbasalisisthethebasal basal response value, is the minimum parameter value, pmax is the the maximum parameter value, and pbasalisisthe thebasal basalparameter parameter value. value. maximum parameter value, and pbasal

[Eq. 14.1]

14.1

Changes in diet composition, i.e. proportions of CP, ST, NDF, SU and CFat are of major importance since they affect (inter alia) diet energy and191 metabolizable protein value (Fox et al., 2003). In the presented sensitivity analysis (Table 14.1), only the effects of variations in nutrient degradation characteristics were evaluated, since they may have an impact on the reliability of feed analyses. Both NEL and AATN are sensitive to changes in forage iNDF and kdNDF. The pdNDF from forage constitutes a large proportion of the NDF pool, and changes in both iNDF concentration and kdNDF strongly influence the energy supply for microbial growth, and hence AATN from microbial protein H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_14, © Wageningen Academic Publishers 2011

141

Table 14.1. Results of the sensitivity analysis in NorFor to changes in feed characteristics with a lactating dairy cow diet. Variables

Forage sCP1, g/kg CP iCP2, g/kg CP iNDF3, g/kg NDF kdCP4, %/h kdNDF5, %/h Concentrate sCP, g/kg CP iCP, g/kg CP iNDF, g/kg NDF kdCP, %/h kdNDF, %/h kdST6, %/h kdRestCHO7, %/h

Ranges

Sensitivity, %

Basal

Minimum

Maximum

NEL8 AATN9 MJ/kg DM g/kg DM

600 38 160 11.0 4.7

500 19 80 5.5 2.4

700 57 240 16.0 7.1

-0.1 -0.6 -5.3 -0.1 7.6

-5.1 -1.7 -3.4 -3.0 5.7

247 40 190 6.0 3.0 15.0 150

124 20 80 3.0 1.5 7.5 75

371 60 240 9.0 4.5 22.5 225

-0.1 -0.8 -2.4 -0.1 2.4 -0.9 0.3

-7.5 -3.4 -3.0 -12.8 0.5 1.5 -1.0

1

sCP = soluble crude protein. iCP = total indigestible crude protein. 3 iNDF = total indigestible NDF. 4 kdCP = fractional degradation rate of potentially degradable crude protein. 5 kdNDF = fractional degradation rate of potentially degradable NDF. 6 kdST = fractional degradation rate of potentially degradable starch. 7 kdRestCHO = fractional degradation rate of residual carbohydrates. 8 NEL = net energy lactation. 9 AAT = amino acids absorbed in the intestine. N 2

and total tract digestion of OM, which provide the basis for ME predictions. The work of Nordheim et al. (2007) demonstrated the difficulties of using NIRS to measure kdNDF. Due to the importance of kdNDF for the nutritive value of the diet and it was, therefore, decided in NorFor to calculate the kdNDF from NIRS or in vitro measurements of OMD and iNDF (see Section 5.2.1; Eriksson, 2010a). The AATN was sensitive to the proportion of sCP in the forage, although the relative difference between the selected minimum and maximum parameter values was less than for the other variables (Table 14.1). In grass silage, sCP is the dominating protein fraction, which explains why sCP has more impact on AATN than kdCP (Volden et al., 2002). Diet NEL is not sensitive to changes in forage or concentrates sCP and kdCP. This is consistent with expectations, because the segmental digestion of degradable CP has limited effects on the total tract digestion of CP. The sensitivity analysis indicates that predicted energy values are more sensitive to the chemical composition of concentrates, (e.g. ST vs. NDF) than to degradation characteristics (data not shown). However, predictions of the AATN are sensitive to kdCP, sCP and iCP. Nevertheless, this sensitivity analysis demonstrates that the NorFor system is sensitive to CHO and CP degradation characteristics in both roughage and concentrates.

142

NorFor – The Nordic feed evaluation system

14.2 Evaluation of the digestive tract sub-model Two important sub-models in NorFor are the digestive tract sub-model (which predicts the ration digestibility) and the metabolism sub-model (which predicts energy and protein supply and animal responses). When evaluating a complex model, it may be useful to evaluate several subdivisions separately, to reveal strengths and weaknesses in different parts of the system. This section summarizes the test of total tract digestion of major nutrients. Sources of data used in the evaluation of dairy cow digestion are presented in Table 14.2. The published studies were supplemented with measurements from trials conducted in connection with a student course in ruminant nutrition and physiology at the Norwegian University of Life Sciences. Total tract digestion was selected because there are minimal methodological differences between relevant studies. It is relatively simple to measure, compared to ruminal and intestinal flow measurements, which are heavily dependent on marker techniques and calculation methods used. The model was evaluated by regressing predicted values against observed values, and several statistical measures were used to assess model adequacy and behaviour, including coefficients of correlation (r) and determination (r2), mean square prediction error (MSPE) and the partitioning of MSPE (Bibby and Toutenburg, 1977). The MSPE was decomposed into errors due to: overall bias (deviation of the intercept from 0), line bias (deviation of the slope from unity) and disturbance (lack of correlation). The square root of the MSPE (RMSPE) expressed as a percentage of the observed mean was used as a measure of the prediction error. High correlations (r>0.97) between predicted and observed values were found for all the digestion variables tested (td_OM, td_CP, td_NDF and td_ST) (Figure 14.1, 14.2, 14.3, 14.4). NorFor accounted for 96% of the variation in td_OM with a mean bias of 221 g/d (1.9%). Corresponding values for td_CP, td_NDF and td_ST were 68, 68 and 106 g/d (3.3, 1.6 and 3.9%), respectively. The model slightly overpredicted td_OM, td_NDF and td_ST, while td_CP was underpredicted (Table 14.2). Except for td_ST, most of the errors in the MSPE were related to disturbance. The analysis of residuals between observed and predicted td_ST indicated that the higher general bias (0.43) was largely due to an underestimation of iST in the feedstuffs. Regression statistics (Mitchell, 1997) showed that the fitted line between predicted and observed values did not differ significantly from 1, indicating that there was no systematic deviation in digestion in the interval examined. Prediction errors were generally low, with RMSPE of 560, 167, 264 and 161 g/d for td_OM, td_CP, td_NDF and td_ST, respectively. The difference between observed and predicted values for td_CP, td_NDF and td_ST and were within 5% of the observed values for 45, 58 and 70% of the observations, respectively. Deviations were less than 10% for more than 80% of the observations. This demonstrates the ability of the NorFor model to predict ration digestion over large ranges of diet composition and feed intake. This ability is crucial for calculation of energy supply, which is based on total apparent digestion of CP, CFat and CHO in NorFor. From the mean prediction error for OM, and assuming an average ME factor of 15.5 MJ/kg digestible OM, the NEL (utilization of ME=0.6) prediction error was estimated to be 5.2 MJ, corresponding to 1.7 (5.2/3.14) kg ECM. For testing digestion in the growing cattle sub-model, a small data set from three studies (Olson and Lindberg, 1985; Olson and Lindberg, unpublished data; Wallsten, 2010) was available. The data set included results from 18 treatments, applied to both bulls (n=10) and heifers (n=8). The total apparent digestion of td_OM, td_CP, td_NDF was evaluated (Figures 14.5, 14.6 and 14.7). The coefficient of determination was high (r2>0.96) for all digestion variables tested (Table 14.3). NorFor overpredicted td_OM (6.4%) and td_NDF (9.0), while td_CP was underpredicted (13.2%). The MSPE consisted mainly overall bias, while the proportion of MSPE related to line bias was low. The regression slope between predicted and observed digestion did not differ significantly from 1. The data set was, however, too small for further analysis of the residuals.

NorFor – The Nordic feed evaluation system

143

144

NorFor – The Nordic feed evaluation system

n

11,841 1993 4235 2860

831 -134 377 24

Intercept

Observed Predicted

11,620 2061 4167 2754

Regression

Digested, g/day 0.948 1.033 0.926 1.030

Slope 0.967 0.955 0.966 0.993

r2 560 167 264 161

RMSPE 4.8 8.1 6.3 5.8

Prediction error, % 0.165 0.168 0.067 0.431

0.009 0.100 0.044 0.099

0.836 0.732 0.889 0.470

Overall bias Regression Disturbance

Proportion of MSPE

2

td_OM = apparent total tract digestion of organic matter, g/day. td_CP = apparent total tract digestion of crude protein, g/day. 3 td_NDF = apparent total tract digestion of NDF, g/day. 4 td_ST = apparent total tract digestion of starch, g/day. 5 The following trials were used: Alne, 2000; Aston et al., 1994; Bertilsson and Murphy, 2003; Eriksson et al., 2004; Kjos, unpublished data; Larsen et al. 2009b; Lund et al., 2004; Lund, unpublished data; Murphy et al., 1993; Prestløkken, 1999; Prestløkken et al., 2007; Rinne et al., 1999a, 2002; Selmer-Olsen, 1996; Tothi et al., 2003; Volden, unpublished data.

1

td_OM1 107 td_CP2 95 td_NDF3 111 td_ST4 67

Item

Table 14.2. Accuracy and precision of NorFor’s ability to predict digestion in dairy cows5.

Predicted OM digestion (g/d)

18,000

Observed - predicted

4,000

16,000 14,000 12,000 10,000 8,000 6,000 2,000 0 -2,000

0

2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 18,000

-4,000

Observed OM digestion (g/d)

Observed - predicted

Predicted CP digestion (g/d)

Figure 14.1. Predicted vs. observed organic matter (OM) digestion in dairy cows.

4,500 4,000 3,500 3,000 2,500 2,000 1,500 1000 500 0 -500

0

500

1000

1,500

2,000

2,500 3,000

3,500 4,000

4,500

-1000

Observed CP digestion (g/d)

Figure 14.2. Predicted vs. observed crude protein (CP) digestion in dairy cows.

NorFor – The Nordic feed evaluation system

145

Predicted NDF digestion (g/d) Observed - predicted

8,000 7,000 6,000 5,000 4,000 3,000 2,000 1000 0 -1000

0

1000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

-2,000 Observed NDF digestion (g/d)

Observed - predicted

Predicted ST digestion (g/d)

Figure 14.3. Predicted vs. observed NDF digestion dairy cow. 6,000 5,000 4,000 3,000 2,000 1000 0

0

1000

2,000

3,000

4,000

5,000

6,000

-1000 Observed ST digestion (g/d)

Figure 14.4. Predicted vs. observed starch (ST) digestion in dairy cows.

146

NorFor – The Nordic feed evaluation system

NorFor – The Nordic feed evaluation system

147

n

3,290 386 1,044

79 -24 100

Intercept

Observed Predicted

3,090 445 958

Regression

Digested, g/day 1.047 0.922 0.986

Slope 0.981 0.968 0.990

r2 230 64 62

RMSPE

2

td_OM = apparent total tract digestion of organic matter, g/day. td_CP = apparent total tract digestion of crude protein, g/day. 3 td_NDF = apparent total tract digestion of NDF, g/day. 4 The following trials were used: Olson and Lindberg, 1985; Olson and Lindberg, unpublished data; Wallsten, 2010.

1

td_OM1 18 td_CP2 18 td_NDF3 18

Item

Table 14.3. Accuracy and precision of NorFor´s ability to predict digestion in growing cattle4.

7.5 14.5 11.1

Prediction error, % 0.753 0.838 0.657

0.035 0.010 0.001

0.212 0.152 0.342

Overall bias Regression Disturbance

Proportion of MSPE

Predicted OM digestion (g/d)

5,000

Observed - predicted

1000

4,500 4,000 3,500 3,000 2,500 2,000 1,500 500 0

-500 0

1000

-1000

2,000

3,000

4,000

5,000

Observed OM digestion (g/d)

Predicted CP digestion (g/d)

Figure 14.5. Predicted vs. observed organic matter (OM) digestion in growing cattle.

800 700 600 500 400

Observed - predicted

300 200 100 0

0

100

200

300 400 500 600 Observed CP digestion (g/d)

700

800

Figure 14.6. Predicted vs. observed crude protein (CP) digestion in growing cattle.

148

NorFor – The Nordic feed evaluation system

Observed - predicted

Predicted NDF digestion (g/d)

3,000 2,500 2,000 1,500 1000 500 0 -500

0

500

1000

1,500

2,000

2,500

3,000

Observed NDF digestion (g/d)

Figure 14.7. Predicted vs. observed NDF digestion in growing cattle.

14.3. Evaluation of milk production in lactating dairy cows A Nordic dataset, which had not been used for model construction, was compiled. Data originated from published and unpublished trials performed in Denmark, Norway, Sweden, and Finland mainly from the 1990’s and 2000’s (Table 14.4). No selection was made among the experiments and data from all identified experiments providing information on ECM or MY, fat and protein contents, DMI and BW were included. Not all studies included information on BW or BCS changes during the trial period. Lactose content in milk was only reported in a few studies, thus a fixed content of 48 g/kg was used. The dataset included information from both long- and short-term studies, i.e. data both from continuous lactation experiments and cross-over type experiments. A further criterion was availability of data on most nutrient characteristics for each feedstuff. Experiments that only provided data on nutrient characteristics for the whole TMR were not included. Most studies did not include iNDF, sCP, kdST, kdNDF and kdCP measurements on single feedstuffs. When nutrient composition data were missing, values from the NorFor feedstuff table were used (www.norfor.info). The dataset, described in Table 14.5, consists of information from 53 trials (see Table 14.4) with five breeds (205 NRF, 77 DH, 38 SR, 5 RD and 4 JER) with a total of 329 treatments. As described in Chapter 12, ECM is predicted from the calculated total energy intake minus the estimated energy usage for maintenance, growth (only primiparous cows), pregnancy and mobilization/deposition (see Equation 12.1). Energy usage for maintenance is estimated from BW. Information on days in gestation was often not available and was therefore set to zero. Mobilization/ deposition was included via information on changes in BW or BCS. As not all studies included data on BW or BCS changes during the trial period, mobilization/deposition was assumed to be zero. Energy requirement for growth in primiparous cows was included indirectly via changes in BW or BCS. The prediction of milk protein yield is based on model estimates of AATN supply to the mammary gland and the efficiency of AATN use in synthesis of milk protein (see Equation 12.2).

NorFor – The Nordic feed evaluation system

149

Table 14.4. Studies used to evaluate the prediction of milk production, roughage intake and total dry matter intake in dairy cows. Aaes, 1991 Aaes, 1993 1,2 Bertilsson, 2007 Bertilsson, 2008 Bertilsson and Murphy, 20031 Eriksson, 2010b Eriksson et al., 20041 Hymøller et al., 20052 Ingvartsen et al., 20011, 2 Johansen, 19921 Kjos, unpublished data Kristensen, 19992 Lund, 20021 Minde and Rygh, 1997 Misciattelli et al., 20031 Mo and Randby, 19861 Mogensen and Kristensen, 20021,2

Mogensen et al., 20082 Mould, 19961 Murphy et al., 2000 Nielsen et al., 20072 Nordang and Mould, 19941 Prestløkken, 1999 Prestløkken et al., 20071 Randby, 19921 Randby, 19971 Randby, 1999a1 Randby, 1999b1 Randby, 2000a1 Randby, 2000b1 Randby, 20021 Randby, 2003 1 Randby, 20071 Randby and Mo, 19851

Randby and Selmer-Olsen, 19971 Rinne et al., 1999a1 Rinne et al., 1999b1 Sairanen, 20011 Schei et al., 20051 Shingfield et al., 2003 Steinshamn et al., 20061 Thorhauge and Weisbjerg, 20092 Thuen, 19891 Volden, 19901 Volden, 1999 Volden and Harstad, 2002 Wallsten and Martinsson, 20101 Weisbjerg, 2007 2 Weisbjerg et al., 20082 Åkerlind et al., 19992 Österman, 20031, 2

1 The

study is used for roughage intake evaluation. study is used for total dry matter intake evaluation. Studies marked with superscripts 1 and 2 included treatments with separate and TMR and feeding, respectively. 2 The

Table 14.5. Descriptive statistics of the 329 treatments from 53 Nordic trials used to evaluate prediction of ECM and milk protein yield. Item

Average

SD1

Maximum

Minimum

Days in milk ECM2, kg/d MY3, kg/d Fat, g/kg Protein, g/kg Protein yield, g/d

121 27.1 27.1 41.1 32.6 882

44 5.3 5.2 4.7 1.8 174

255 49.7 47.0 58.3 41.1 1,607

31 12.6 13.1 26.9 29,3 422

1

SD = standard deviation. ECM = energy corrected milk. 3 MY = milk yield. 2

Figures 14.8 and 14.9 illustrate the relationships between observed and predicted ECM and milk protein yields, respectively. Table 14.6 shows that across all experiments and treatments, the observed and predicted yields were similar (27.1 vs. 26.4 kg ECM/d), especially for milk protein (882 vs. 900 g protein/d). The correlation between observed and predicted milk protein values was high (r=0.97), but somewhat lower for ECM (r=0.88). Predicted ECM and milk protein values were higher than observed for 43% and 68% of the observations, respectively, with RMSE’s of 2.7 kg ECM and 48 g milk protein, corresponding to prediction errors of 9.8 and 5.5%, respectively. Predicted ECM yield was within 5% of observed values for 47% of the observations and 73% of the predicted ECM yield values were within 10% of observed values. Predicted milk protein yield was within 5% and 10% 150

NorFor – The Nordic feed evaluation system

Predicted ECM (kg/d)

50

40

30

Observed - predicted

20

10

0

0

10

20

30

40

50

-10 Observed ECM (kg/d)

Predicted milk protein yield ( g/d)

Figure 14.8. Predicted vs. observed ECM yield.

1,800 1,600 1,400 1,200 1000 800

Observed - predicted

600 400 200 0 -200

0

200

400

600

800

1000 1,200 1,400 1,600 1,800

Observed milk protein yield (g/d)

Figure 14.9. Predicted vs. observed milk protein yield.

NorFor – The Nordic feed evaluation system

151

152

NorFor – The Nordic feed evaluation system

2

26.4 900

2.6 39

Intercept

Observed Predicted

27.1 882

Regression

Production

ECM = energy corrected milk yield, kg/day. MPY = milk protein yield, g/day.

329 329

ECM1 MPY2

1

n

Item 0.88 0.98

Slope 0.78 0.94

r2 2.7 48

RMSPE

Table 14.6. Accuracy and precision of NorFor´s ability to predict ECM and milk protein yield.

9.8 5.5

Prediction error, % 0.07 0.14

0.05 0.02

0.88 0.84

Overall bias Regression Disturbance

Proportion of MSPE

of observed values for 69% and 92% of the observations, respectively. The overall and regression biases were minor for ECM and milk protein, since more than 80% of the MSPE was characterised as disturbance. The regression slopes between predicted and observed milk production variables were not significantly different from 1.

14.4 Evaluation of the energy requirement for growing cattle The estimated energy requirement for growing cattle is based on the French system (Chapter 9), modified with a 10% increase in the energy requirement for gain based on a dataset compiled from Nordic experiments with bulls. In order to perform an independent test of the NorFor energy system for growing cattle, an independent dataset containing heifer data from the Nordic countries and the US was used. Criteria for including data were heifers of dairy breeds and measurements of ADG, DMI and nutrient characteristics for each feedstuff. Hence, experiments where nutrient characteristics were available only for the whole diet were excluded. When nutrient composition data were missing, values from the NorFor feedstuff table were used (www.norfor.info). The reason for including US data was that a limited number of experiments from the Nordic countries were available. No selection was made among the experiments that were found in the literature. The dataset is described in Table 14.7 and consist of 9 trials with 54 treatments of which 47 include large breeds and 7 include Jersey. Ration energy supply was calculated using NorFor and were regarded as observed values. The energy requirements were calculated from reported BW and ADG and were regarded as predicted values. The relationship between energy supply and requirement is shown in Figure 14.10. Across all experiments and treatments, the energy supply and requirement were fairly similar (31.9 vs. 33.7 MJ NEG) with an RMSPE of 3.4 MJ NEG, corresponding to a prediction error of 10.0%. The correlation (r=0.96) was stronger than that obtained from the data used to develop the energy system for growing cattle (see Figure 9.2).

14.5. Evaluation of the rumen load index (RLI) RLI was included in the model to establish a relationship between the level of starch and sugars in the ration and their influence on digestion of NDF in the rumen (see Section 7.1.3). Therefore, RLI was evaluated against data from two studies in which rumen pH was measured (Lund, unpublished data; Larsen et al., 2009a) and where kdNDF was estimated on the basis of digestibility trials with lactating Holstein cows (Larsen et al., 2009a). The cows were fed ad libitum with grass-clover silage as the sole roughage and the concentrate consisted of barley, wheat, oat, maize, soybeans, faba beans, peas or lupins and most treatments also included soybean meal as a protein supplement in the ration.

Table 14.7. Descriptive statistics of the dataset1 used for evaluating the energy system for growing cattle, including data from nine trials with heifers and 54 treatments. Item

Average

SD2

Maximum

Minimum

DMI, kg/d BW, kg ADG, g/d Concentrates (% of DMI)

5.3 253 724 35

1.9 102 203 28

9.9 489 1,180 92

2.2 88 396 4

1

Olsson et al., 1984; Andersen et al., 1986; Ingvartsen et al., 1988; Mäntysaari, 1993; Andersen and Foldager, 1994; Van Amburgh et al., 1998; Andersen et al., 2001; Whitlock et al., 2002; Vestergaard, 2006. 2 SD = standard deviation.

NorFor – The Nordic feed evaluation system

153

70

NEG requirement (MJ/d)

60 50 40 30 20 10 0

0

10

20

30 40 NEG supply (MJ/d)

50

60

70

Figure 14.10. The relationship between ration energy supply calculated in NorFor and estimated energy requirement in heifers according to NorFor (r2=0.92; RMSPE=3.4 MJ NEG; RMSPE=10.0%). The data are from the trials described in Table 14.8. In contrast to expectations, RLI was not negatively correlated to pH in the rumen (Figure 14.11). Indeed, the data in Figure 14.11 show a positive relationship between RLI and rumen pH in the data from Lund (unpublished data) and no relationship in the data from Larsen et al. (2009a). As RLI was included in NorFor to estimate the effect of starch and sugars on the digestion of NDF in the rumen, the main interest is in evaluating RLI was as a predictor for kdNDF in the rumen, rather than as a predictor of rumen pH. This relationship is illustrated in Figure 14.12, where a negative correlation (r=-0.82) between RLI and kdNDF is evident. Thus, even though RLI was not correlated with rumen pH, increases in RLI were associated with reductions in kdNDF. This confirms that RLI is a useful term for correcting kdNDF, as described in Section 7.1.3, although this evaluation was only based on one study. However, the relationship shown in Figure 14.12 is supported by the findings of Danfær et al. (2006).

14.6. Evaluation of roughage and total dry matter intake in dairy cows Data from Nordic experiments were used to evaluate roughage and total DMI predictions. For this purpose, the data were divided in two sub-sets (Tables 14.8 and 14.9). The first (n=226) consisted of data from studies where roughage was fed ad libitum and separately, and the concentrate was fed either according to MY or as fixed amounts (Table 14.8). This dataset was used to evaluate roughage intake. In the second dataset (n=62), cows were fed total mixed rations (TMR) ad libitum and total DMI was evaluated (Table 14.9). Both datasets showed wide variations in both DMI and diet composition, and the data were in the range of those used for parameterization of the intake sub-model. Published nutrient compositions of the individual feeds were used. When nutrient composition data were missing, values from the NorFor feedstuff table were used (www.norfor.info). Predicted vs. observed roughage intake and the evaluation of the prediction errors are presented in Figure 14.13 and Table 14.10, respectively. Mean observed roughage DM intake was 11.0 kg/d 154

NorFor – The Nordic feed evaluation system

6.8 min_pH_Lund

6.6

mean_pH_Lund

Rumen pH

6.4

min_pH_Larsen mean_pH_Larsen

6.2 6.0 5.8 5.6 5.4 5.2 5.0 0.0

0.2

0.4

0.6

0.8

1.0

Rumen load index

Figure 14.11. The relationship between rumen load index calculated in NorFor and mean and minimum pH measured in the rumen. The data are from Lund (unpublished data, n=8) and Larsen et al. (2009a, n=13). 4.0 3.5

kdNDF (%/h)

3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

Rumen load index

Figure 14.12. The relationship between rumen load index calculated in NorFor and trial estimates of NDF degradation rate in the rumen (kdNDF) of lactating Holstein cows (n=13; r2=0.68; RMSPE=0.33%/h). The data are from Larsen et al. (2009a).

NorFor – The Nordic feed evaluation system

155

Table 14.8. Descriptive statistics of the dataset used for evaluating feed intake of cows, including data from 36 trials3 and 226 treatments where roughage and concentrate were fed separately. Item

Average

SD1

Maximum

Minimum

Days in milk Body weight, kg ECM2, kg/d Dry matter intake, kg/d Concentrate proportion, kg/kg DM Roughage intake, kg DM/d Roughage basis fill value, /kg DM Starch + sugars, g/kg DM Starch + sugars, kg/d

124 567 25.3 18.2 0.40 10.9 0.50 183 3.3

35 37 4.5 2.4 0.12 2.6 0.06 69 1.2

214 667 45.0 28.4 0.80 18.2 0.70 426 6.9

42 481 17.1 13.1 0.14 3.2 0.35 60 0.9

1

SD = standard deviation. ECM = energy corrected milk. 3 The trials are listed in Table 14.5. 2

Table 14.9. Descriptive statistics of the dataset used for evaluating the feed intake system for cows, including data from 12 trials3 and 62 treatments where the ration was fed as a total mixed ration. Item

Average

SD1

Maximum

Minimum

Days in milk Body weight, kg ECM2, kg/d Dry matter intake, kg/d Concentrate proportion, kg/kg DM Roughage basis fill value, /kg DM Starch + sugars, g/kg DM Starch + sugars, kg/d

96 588 33.2 21.2 0.54 0.46 208 4.4

43 48 4.6 2.1 0.1 0.04 59 1.2

200 659 49.7 25.1 0.70 0.60 419 7.8

31 441 22.4 16.6 0.27 0.42 94 1.7

1

SD = standard deviation. ECM = energy corrected milk. 3 The trials are listed in Table 14.5. 2

and mean predicted intake was 10.1 kg DM/d. The regression slope was 0.88 and not significantly different from 1. The RMSPE was 1.3 kg DM/d and 0.52 of the prediction error was related to disturbance. The overall bias accounted for 0.48 of the total MSPE, and showed that the intake model underpredicts roughage intake. Predicted roughage intake was within ±10% of observed intake for 58% of the observations. Predicted vs. observed TMR intake and results of the evaluation of the prediction error are presented in Figure 14.14 and Table 14.10, respectively. The mean observed and predicted TMR intakes were 21.2 and 21.1 kg DM/d, respectively, the regression slope was not significantly (P100 1 1 113 0.86

618 5.3 14.5 7.7 1.25 532 >100 1 1 160 0.91

625 8.9 15.3 6.9 1.10 450 32 0.96 1 120 0.87

625 10.4 17.1 7.7 1.23 450 25 0.92 1 120 0.87

350 0.3 7.8 3.8 1.08 486 >100 1 1 192 0.94

350 0.3 7.0 3.3 0.95 476 >100 1 1 366 1.12

21 37 58

27 48 75

24 46 70

24 53 77

51 70 121 21 40 61

54 71 125 28 52 80

99 130 229 20 46 66

100 148 248 20 51 71

22 39 61

21 39 60

1136

976

606

566

within ±5% of observed CTcorr for six out of eight observations. The two observations with large deviations (>15%) were from the study of Hymøller et al. (2005) and are largely responsible for a relatively low coefficient of determination (r2 = 0.21) between observed CT and predicted CI. The RMSPE was 9.2 min/kg DMr corresponding to a prediction error of 13.8%.

NorFor – The Nordic feed evaluation system

161

Predicted EI, RI and CI (min/kg DMr)

100 CT ET RT

80

60

40

20

0

0

20

40 60 80 Observed ET, RT and CT (min/kg DMr)

100

Figure 14.15. The relationship between observed eating (ET), rumination (RT) and chewing time (CT) vs. corresponding predicted indexes (EI, RI, CI). The data are from trials described in Table 14.11.

References Aaes, O. 1991. Grøncobs som erstatning for kraftfoder til malkekøer. In: Foderværdi og anvendelse af kunsttørret grønfoder til malkekøer. Statens Husdyrbrugsforsøg, Denmark. 5-19. (In Danish). Aaes, O. 1993. Total mixed rations versus separate feeding of concentrate rich rations fed restrictively or ad libitum to dairy cows. Forskningsrapport no 16 fra Statens Husdyrbrugsforsøg, Denmark. (In Danish with English summary). Åkerlind, M., K. Holtenius, J. Bertilsson and M. Emanuelson, 1999. Milk composition and feed intake in dairy cows selected for high or low milk fat percentage. Livestock Production Science 59: 1-11. Alne, J., 2000. Effects of duodenal starch and amino acid infusion on milk yield, protein content and utilization of feednitrogen in dairy cows. Master of Science Thesis. Agricultural University of Norway, Aas, Norway. (In Norwegian). Andersen, H.R. and J. Foldager, 1994. Protein requirement for replacement heifers in the period from 3 to 12 months of age. Forskningsrapport no. 26 fra Statens Husdyrbrugsforsøg, Denmark. (In Danish with English summary). Andersen, H.R., Foldager, J., Varnum, P.S. and T. Hvelplund, 1986. Forskellige proteinmængder og proteinkilder til kvieopdræt. Meddelelse no. 645 fra Statens Husdyrbrugsforsøg. (In Danish). Andersen, H.R., B.B. Andersen and H.G. Bang, 2001. Beef crossbreeding: Heifers versus bulls fed different concentrate/ roughage ratio and slaughtered at different live weight. DJF rapport no. 28. (English summary). Aston, K, C. Thomas, R. Daley, J.D. Sutton and M.S. Dhanoa, 1994. Milk production from grass silage diets: effects of silage characteristics and the amount of supplementary concentrate. Animal Production 59: 31-41. Bertilsson, J., 2007. Agrodrank som foder till mjölkkor. Kungsängendagarna 2007. Rapport nr 267. Swedish University of Agricultural Sciences, Uppsala, Sweden. pp. 21-23. (In Swedish). Bertilsson, J., 2008. Effects of chop length of silage on chewing time, feed consumption and milk production. In: Hopkins, A., T. Gustafsson, J. Bertilsson, G. Dalin, N. Nilsdotter-Linde and E. Spörndly (eds.). Biodiversity and animal feed: Future challenges for grassland production. Grassland Science in Europe volume 13, Uppsala, Sweden, pp. 783-785. Bertilsson, J. and M. Murphy. 2003. Effects of feeding clover silages on feed intake, milk production and digestion in dairy cows. Grass and Forage Science 58: 309-322.

162

NorFor – The Nordic feed evaluation system

Danfær, A., P. Huhtanen, P. Udén, J. Sveinbjörnsson and H. Volden, 2006. Karoline – A Nordic model for feed evaluation. Model description. In: Kebreab, E., J. Dijkstra, A. Bannink, J. Gerrits and J. France (eds.). Nutrient digestion and utilisation in farm animals: Modelling approaches. CAB International, Wallingford, UK, pp. 383-406. Eriksson, T. 2010a. In vitro methods for indigestible Neutral Detergent Fiber (iNDF). Proceedings of the 1th Nordic Feed Science Conference, Report no. 274, Swedish University of Agricultural Sciences, Uppsala, Sweden, pp. 41-45. Eriksson, T. 2010b. Nitrogen metabolism in dairy cows fed restricted amounts of grass–clover silage supplemented with seeds from narrow-leafed lupin or pea. Livestock Science 131: 39-44. Eriksson, T., M. Murphy, P. Ciszuk and E. Burstedt, 2004. Nitrogen balance, microbial protein production. and milk production in dairy cows fed fodder beets and potatoes or barley. Journal of Dairy Science 87: 1057-1070. Fox, D., G.L.O. Tedeschi, T.P. Tylutki, J.B. Russell, M.E. Van Amburgh, L.E. Chase, A.N. Pell and T.R. Overton. 2004. The Cornell Net Carbohydrate and Protein System model for evaluating herd nutrition and nutrient excretion. Animal Feed Science and Technology 112: 29-78. Garmo, T.H., A.T. Randby, M. Eknaes, E. Prestløkken and P. Nørgaard, 2008. Effect of grass silage chop length on chewing activity and digestibility. In: Hopkins, A., T. Gustafsson, J. Bertilsson, G. Dalin, N. Nilsdotter-Linde and E. Spörndly (eds.). Biodiversity and animal feed: Future challenges for grassland production. Grassland Science in Europe, volume 13, pp. 810-812. Hymøller, L., M.R. Weisbjerg, P. Nørgaard, C.F. Børsting and N.B. Kristensen, 2005. Majsensilage til malkekøer. DJF rapport nr. 65. Danmarks Jordbrugsforskning. Foulum. Denmark. (In Danish). Ingvartsen, K.L., O. Aaes and J.B. Andersen, 2001. Effects of pattern of concentrate allocation in the dry period and early lactation on feed intake and lactational performance in dairy cows. Livestock Production Science 71: 207-221. Ingvartsen, K.L., J. Foldager, J.B. Larsen and V. Østergaard, 1988. Growth and milk yield by Jersey reared at different planes of nutrition. Beretning no. 645 fra Statens Husdyrbrugsforsøg. (In Danish). Johansen, A., 1992. A comparison between meadow fescue and timothy silage. Doctor Scientarium Thesis. Agricultural University of Norway, Aas, Norway. Kristensen, V.F., 1999. Grovfoderkildens betydning for malkekoens produktion og fodereffektivitet. Temamøde Malkekøernes og kviernes ernæring April 1999. Statens Husdyrbrugsforsøg rapport 118, pp. 18-33. (In Danish). Larsen, M., N.B. Kristensen and T. Hvelplund, 2009a. Stivelsen til malkekøer – perspektivering af resultaterne fra stivelsesprojektet. In: Fodringsdag - Temadag om aktuelle fodringsspørgsmål. 1 September 2009. Dansk Kvæg, Århus, Denmark, pp. 54-61. (In Danish). Larsen, M., P. Lund, M.R. Weisbjerg and T. Hvelplund, 2009b. Digestion site of starch from cereals and legumes in lactating dairy cows. Animal Feed Science and Technology 153: 236-248. Lund, P., M.R. Weisbjerg and T. Kristensen, 2004. The effect of heat treatment on degradability and microbial synthesis of protein in the rumen. Journal of Animal Feed Science 13: 143-146. Lund, P., 2002. The effect of forage type on passage kinetics and digestibility of fibre in dairy cows. Dissertation. The Royal Veterinary and Agricultural University, Department of Animal Science and Animal Health, Copenhagen, Denmark. 171 pp. Minde, A. and A.J. Rygh, 1997. Metoder for å bestemme nedbrytingshastighet, passasjehastighet og vomfordøyelighet av NDF hos mjølkeku. Masteroppgave. Norges Landbrukshøgskole. 118 pp. (In Norwegian). Misciattelli, L., V.F. Kristensen, M. Vestergaard, M.R. Weisbjerg, K. Sejrsen and T. Hvelplund, 2003. Milk production, nutrient utilization and endocrine responses to increased postruminal lysine and methionine supply in dairy cows. Journal of Dairy Science 86: 275-286. Mitchell, P.L. 1997. Misuse of regression for empirical validation of models. Agricultural Systems 54: 313-326. Mo, M. and Å.T. Randby, 1986. Wilted silage for lactating dairy cows. Proceedings of 11th General Meeting of European Grassland Federation. Troia, Portugal, 4-9 May. Mogensen, L. and T. Kristensen, 2002. Effect of barley or rape seed cake as supplement to silage for high-yielding organic dairy cows. Acta Agriculturae Scandinavica. Section A - Animal Science 53: 243-252. Mogensen, L., P. Lund, T. Kristensen and M.R. Weisbjerg, 2008. Effects of toasting blue lupins, soybeans or barley as supplement for high-yielding organic dairy cows fed grass-clover silage ad libitum. Livestock Science 115: 249-257. Mould, F., 1996. Fôropptak og produksjon hos melkekyr ved et lavt PBV i rasjonen. Husdyrforsøksmøtet, pp. 121127. (In Norwegian). Murphy, M., H. Khalili and P. Huhtanen, 1993. The substitution of barley by other carbohydrates in grass silage based diets to dairy cows. Animal Feed Science and Technology 41: 279-296.

NorFor – The Nordic feed evaluation system

163

Murphy, M., M. Åkerlind and K. Holtenius, 2000. Rumen fermentation in lactating cows selected for milk fat content fed two forage to concentrate ratios with hay or silage. Journal of Dairy Science 83: 756-764. Mäntysaari, P. 1993. The effects of feeding level and protein source of the diet on growth and development at slaughter of prepurbertal heifers. Acta Agriculturae Scandinavica Section A - Animal Science 43: 44-51. Nielsen, N.I., N.C. Friggens, T. Larsen, J.B. Andersen, M.O. Nielsen and K.L. Ingvartsen, 2007. Effect of changes in diet energy density on feed intake, milk yield and metabolic parameters in dairy cows in early lactation. Animal 1: 335-346. Nordang, L. and F. Mould, 1994. Surfôr med ulikt innhold av PBV til mjølkekyr. Husdyrforsøksmøtet, pp. 99-103. (In Norwegian). Nørgaard, P., E. Nadeau and A. Randby, 2010. A new Nordic structure evaluation system for diets fed to dairy cows: a meta analysis. In: Sauvant, D., J. Van Milgen, P. Faverdin and N. Friggens (eds.). Modelling nutrient digestion and utilization in farm animals. Wageningen Academic Publishers, Wageningen, the Netherlands, pp. 112-121. Olsson, K., I. Olsson and L. Lindell, 1984. Effect of the amount of roughage fed to heifers during the rearing period on feed consumption and milk production during the first lactation. Report no. 131. Swedish University of Agricultural Sciences Österman, S., 2003. Extended calving interval and increased milking frequency in dairy cows. Acta Universitatis Agriculturae Suecia. Agria 383. Swedish University of Agricultural Sciences, Uppsala, Sweden. Prestløkken, E., 1999. Protein value of expander-treated barley and oats for ruminants. Doctor Scientarium Thesis, Agricultural University of Norway, Aas, Norway. 137 pp. Prestløkken, E., Å.T. Randby, M. Eknæs and T. Garmo, 2007. Tidlig og normalt høstet gras. Opptak av surfôr og produksjon hos mjølkekyr. Husdyrforsøket, Thon Hotel Arena 14-15 February 2007, pp. 25-28. (In Norwegian). Randby, Å.T., 1992. Grass-clover silage for dairy cows. Proceedings of 14th General Meeting, European Grassland Federation, Lahti, Finland, 8-11 June, pp. 272-275. Randby, Å.T., 1997. Feeding of silage effluent to dairy cows. Acta Agriculturae Scandinavica Section A - Animal Science 47: 20-30. Randby, Å.T., 1999a. Grass silage treated with two different acid-based additives for dairy cows. The Fifth International Symposium on the Nutrition of Herbivores. San Antonio, Texas, USA. 8 pp. Randby, Å.T., 1999b. Effect of increasing levels of a formic acid based additive on ad libitum intake of grass silage and the production of meat and milk. Proceeding of the XII International Silage Conference, Uppsala, Sweden, pp. 205-206. Randby, Å.T., 2000a. Round bale grass silage treated with an additive containing both lactic acid bacteria and formate for dairy cows and steers. Sixth British Grassland Society Research Conference, SAC, Craibstone, Aberdeen, Scotland, pp. 121-122. Randby,Å.T., 2000b. Effekten av økende dosering med et maursyreholdig ensileringsmiddel på næringsverdien av surfôret i produksjonsforsøk med kyr og kastrater. Hellerud Forsøksgård, Rapport nr. 25. 37 pp. (In Norwegian). Randby, Å.T., 2002. Silage fermentation by concentrate type interaction. Proceedings of the XIII International Silage Conference, Auchincruive, Scotland, pp. 312-313. Randby, Å.T., 2003. Høstetid og fôrkvalitet. Grønn kunnskap 7: 27-43. (In Norwegian). Randby, Å.T. 2007. Effect of propanol and dimethylsulphide in grass silage on organoleptic milk quality. Journal of Animal and Feed Science 16, Supplement 1: 102-107. Randby, Å. and M. Mo, 1985. Surfôr tilsatt foraform i sammenlikning med maursyresurfôr til melkekyr, Hellerud høsten 1984. M-141. Rapporter fra forsøksvirksomheten ved Institutt for husdyrernæring, pp. 75-78. (In Norwegian). Randby, Å.T. and I. Selmer-Olsen, 1997. Formic acid treated or untreated round bale grass silage for dairy cows. 8th International Symposium of Forage Conservation. Brno, Czech Republic, pp. 160-161. Randby, Å.T., I. Selmer-Olsen and L. Baevre, 1999. Effect of ethanol in feed on milk flavor and chemical composition. Journal of Dairy Science 82: 420-428. Rinne M., P. Huhtanen and S. Jaakkola. 1997. Grass maturity effects on cattle fed silage-based diets. 2. Cell wall digestibility, digestion and passage kinetics. Animal Feed Science and Technology 67: 19-35. Rinne M., S. Jaakkola, K. Kaustell, T. Heikkilä and P. Huhtanen, 1999a. Silage harvested at different stages of grass growth versus concentrate foods as energy and protein sources in milk production. Animal Science 69: 251-263. Rinne, M., S. Jaakkola, T. Varvikko and P. Huhtanen, 1999b. Effects of type and amount of rapeseed feed on milk production. Acta Agriculturae Scandinavica, Section A - Animal Science, 49: 137-148.

164

NorFor – The Nordic feed evaluation system

Rinne M., P. Huhtanen and S. Jaakkola. 2002. Digestive processes of dairy cows fed silages harvested at four stages of maturity. Journal of Animal Science 80: 1986-1998. Rustas, B.O., P. Nørgaard, A.R. Jalali and E. Nadeau, 2010. Effects of physical form and stage of maturity at harvest of whole-crop barley silage on intake, chewing activity, diet selection and faecal particle size of dairy steers. Animal 4: 67-75. Sairanen, A., J.I. Nousiainen and H. Khalili, 1999. Korkean väkirehumäärän vaikutus maitotuotokseen ja tuotannon kannattavuuteen. In: Mitä Suomi syö - ja millä hinnalla? Agro-Food ‘99, Helsinki. pp. 7. (In Finnish). Selmer-Olsen, I., 1996. Microbial protein synthesis in the rumen of cows fed extensively or restricted fermented grass silage. Proceedings of the XIth International Silage Conference, Aberystwyth, Wales, pp. 226-227. Schei, I., H. Volden and L. Bævre, 2005. Effects of energy balance and metabolizable protein level on tissue mobilization and milk performance of dairy cows in early lactation. Livestock Production Science 95: 35-47. Shingfield, K.J., A. Vanhatalo and P. Huhtanen, 2003. Comparison of heat-treated rapeseed expeller and solvent-extracted soya-bean meal as protein supplements for dairy cows given grass silage-based diets. Animal Science 77: 305-317. Steinshamn, H., L.T. Mydland, H. Volden and E. Thuen. 2006. The effect of red clover proportion in timothy silage on nitrogen utilization in lactating dairy cows. In: Wachendorf, M, Á. Helgadóttir and G. Parente (eds.). Sward dynamics. N-flows and forage utilisation in legume-based systems. Grado. Italy. 10-12 November. 2005. ESAAzienda regionale per lo sviluppo rurale, Italy, pp. 293-297. Thorhauge, M. and M.R. Weisbjerg, 2009. Effekt af fodring med mættet eller umættet fedt hos SDM-DH, RDM og Jersey. Kvæginfo no. 1938. Dansk Landbrugsrådgivning, Aarhus, Denmark. (In Danish). Thuen, E., 1989. Influence of source and level of dietary protein on milk yield, milk composition and some rumen and blood components in dairy cows. Doctor Scientarium Thesis. Agricultural University of Norway, as, Norway. 151 pp. Tothi, R, P. Lund, M.R. Weisbjerg and T. Hvelplund, 2003. Effect of expander processing on fractional rate of maize and barley starch degradation in the rumen of dairy cows estimated using rumen evacuation and in situ techniques. Animal Feed Science and Technology 104: 71-94. Van Amburgh, M.E., D.G. Fox, D.M. Galton, D.E. Bauman and L.E. Chase, 1998. Evaluation of National Research Council and Cornell net carbohydrate and protein systems for predicting requirements of Holstein heifers. Journal of Dairy Science 81: 509-526. Vestergaard, M., K.F. Jørgensen, I. Fisker, H.R. Andersen and K. Sejrsen, 2006. Forhøjet proteintildeling til store kvier - Resultater fra forsøg på Kvægbrugets ForsøgsCenter. In: Fodringsdag - Temadag om aktuelle fodringsspørgsmål. 29 august. Dansk Kvæg, pp. 41-45. (In Danish). Volden, H. 1990. Response of dietary AAT density on milk yield - Review of production trials in Norway. Proceedings of the seminar on fat and protein feeding to the dairy cow, 15-16 October, Eskilstuna, Sweden, pp. 81-89. Volden, H., 1999. Effects of level of feeding and ruminally undegraded protein on ruminal bacterial protein synthesis, escape of dietary protein, intestinal amino acid profile, and performance of dairy cows. Journal of Animal Science 77: 1905-1918. Volden, H. and O.M. Harstad, 2002. Effects of duodenal amino acid and starch infusion on milk production and nitrogen balance in dairy cows. Journal of Animal Science 80, Supplement 1. 321 pp. Volden, H.L., T. Mydland, V. Olaisen, 2002. Apparent ruminal degradation and rumen escape of soluble nitrogen fractions in grass, and grass silage administered intraruminally to lactating dairy cows. Journal of Animal Science 80: 2704-2716. Wallsten, J., J. Bertilsson, E. Nadeau and K. Martinsson, 2010. Digestibility of whole-crop barley and oat silages in dairy heifers. Animal 4-3: 432-438. Weisbjerg, M.R. and K. Soegaard, 2008. Feeding value of legumes and grasses at different harvest times. In: Hopkins, A., T. Gustafsson, J. Bertilsson, G. Dalin, N. Nilsdotter-Linde and E. Sporndly (eds.). Biodiversity and animal feed: Future challenges for grassland production. Grassland Science in Europe, volume 13, pp. 513-515. Weisbjerg, M.R., 2007. Lucerneensilage til malkekøer – resultater fra nye forsøg. In: Fodringsdag - Temadag om aktuelle fodringsspørgsmål. 5-12. Dansk Kvæg, Aarhus, Denmark. (In Danish). Weisbjerg, M.R., L. Wiking, N.B. Kristensen and P. Lund, 2008. Effects of supplemental dietary fatty acids on milk yield and fatty acid composition in high and medium yielding cows. Journal of Dairy Research 75: 142-152. Whitlock, B.K., M.J. Vandehaar, L.F.P. Silva and H.A.Tucker, 2002. Effect of dietary protein on prepurbertal mammary development in rapidly growing dairy heifers. Journal of Dairy Science 85: 1516-1525.

NorFor – The Nordic feed evaluation system

165

15. The NorFor IT system, ration formulation and optimisation H. Volden, H.D. Rokkjær, A. Göran and M. Åkerlind This chapter describes how we have converted a whole-animal model into a commercial computer tool to be used by advisors and farmers to formulate and optimise diets for dairy cows and growing cattle. The NorFor system is a semi-mechanistic model that has many theoretical advantages for the formulation and evaluation of feed rations. The system represents a significant development in models of nutritional formulation. It incorporates a detailed description of the protein and carbohydrate composition of the ration, which is used in the algorithms to predict degradation, digestion and metabolism in the gastro-intestinal tract and also the supply of metabolizable energy and protein. It can be used not only to estimate nutritional requirements but also provides substantial benefits for predicting nutrient utilization and the environmental impact from different diets. A complex model with semi-mechanistic algorithms and non-linear functions requires the development of a powerful, user-friendly computer tool before it can be used in practical ration formulation and optimisation. Feed is one of the major costs in modern cattle production, so it is important to consider both nutritional and economic aspects when formulating optimal rations. A key objective in modern, computerised ration formulation is auto-balancing (Boston et al., 2000), i.e. identification of the least expensive combination of feed ingredients that provides all considered nutrients within specified ranges. Use of an optimisation (auto-balancing) approach in the formulation of feed rations for cattle requires detailed nutritional and economic knowledge, including information regarding available feeds, the animals and management practices. Some feeds are home-grown while others are purchased as individual ingredients or as concentrate mixtures, and knowledge of their characteristics, obtained from either analyses or tables are needed. In addition, information on feed prices is required. Important animal information includes lactation stage, lactation number, BW, gestation day and age. Management factors include housing (tied up or loose housed) and feeding system (total mixed ration or separate feeds; pasture or indoor). All of these factors affect feed requirements and it is therefore essential that all relevant information is readily available. For these reasons, NorFor has put substantial efforts and resources into the development of an effective, economic IT platform and computer tool that can be used in practical ration evaluation and optimisation.

15.1 IT platforms NorFor is used by nutrition advisors in Denmark, Iceland, Norway and Sweden. An objective in the development of the computer tools was to use an IT platform that allows operators to access the latest version of the model. Hence, all users had to be able to access it via the same server system. Two IT platforms were developed: one web-based and one in which operators use an offline client, synchronized with the NorFor sever before starting the ration formulation. Denmark, Iceland and Norway have chosen to use an online solution, while Sweden is using an offline solution. Although, all countries use the common NorFor severs, each country, except Iceland, has developed their own national client and interface, to enable the NorFor system to be integrated with other national advisory tools. Figure 15.1 presents an overview of the common IT system, which consists of a server application, a homepage, an administration tool and Web services for dairy management software (National herd recording system), laboratories and feed companies. The NorFor servers host four web services: the feed analysis system (FAS), the feedstuff table (FST), the feed ration calculator (FRC), and the one-day feeding control (OFC) system.

15.2 The feed analysis system (FAS) NorFor has developed a web-client that is used by linked laboratories to report results from herd level feed analyses (Figure 15.2). The laboratories connect to the system and upload analytical results, H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9_15, © Wageningen Academic Publishers 2011

167

Figure 15.1. An architectural overview of the NorFor IT system.

Figure 15.2. An architectural overview of the feed analysis system (FAS). and the system calculates feed characteristics (using the FST Calculator) from other analytical data, as described in Chapter 6. Based on the NorFor feed code system, tabulated values are added to the feed samples if absent. If the feed sample is from a herd registered in the national herd recording system (Figure 15.3), it is assigned a herd-specific id, making it automatically available for use by the FRC for ration optimisation or by the OFC for calculating herd feed efficiency. In the NorFor server, each herd has its own feedstuff table, in which the analysed feeds and those copied from the FST are saved. When roughage is analysed at the laboratory, the farmer or advisor reports additional information about the feed sample. This information included date of harvest, cutting number, botanical composition, feed additives, storage system for silage and theoretical cutting length. These additional values are stored, together with the feed characteristics, in a common feed database, which can be used for 168

NorFor – The Nordic feed evaluation system

Figure 15.3. An architectural overview of the link between NorFor and the national client. statistical evaluation of factors such as the annual roughage quality and the relationships between feed additives and silage fermentation quality. Approximately 25,000 roughage samples are analysed according to NorFor specifications annually.

15.3 Feedstuff table (FST) NorFor has developed a comprehensive feedstuff table (FST) that is continuously updated. Updating and monitoring the FST is done by use of the NorFor administration tool. From the NorFor sever, the FST is available to each national client and when working on herd-specific feed plans, feedstuffs are copied from the FST into the herd feedstuff table for further use. For external users, the FST is available at www.norfor.info. Figure 15.4 shows an overview of the NorFor FST, which is organized in a hierarchal directory system, in which the highest level is region, dividing the feedstuffs into country specific and common NorFor categories. The feed directory is further divided into feed groups, by 19 sub-directories, and the feed groups are divided into feed types. For example, the feed group ‘grains, is divided into ‘grains’, ‘dry grain by-products’ and ‘wet grain by-products’. The ‘forages and roughage’ group is divided into seven feed types: ‘pasture grass and clover-grass’, ‘grass and clover-grass’, ‘whole crop’, ‘grass and clover-grass silage’, ‘whole crop silage’, ‘hay and ‘straw’, and ‘grass pellets’. Figure 15.4 also shows an example of choosing a parameter setting (‘NDF characteristics’). These settings are used to generate reports on variables such as NDF, starch, protein, amino acids fatty acids, fermentation products, minerals, vitamins or total characteristics. A part of the feed table system is also information on commercial compound feeds including prices, reported by the feed industry (Figure 15.5). When feed rations are formulated, the operators copy NorFor – The Nordic feed evaluation system

169

Figure 15.4. Overview of the NorFor feedstuff table and example of NDF characteristics in feedstuffs from the feed group forages and roughage.

Figure 15.5. An architectural overview of the feed company web-service in NorFor. 170

NorFor – The Nordic feed evaluation system

actual company products into the herd specific feedstuff table. NorFor has developed a web-service to help the feed industry report these details to the FST.

15.4 Dairy management system Each Nordic country has its own national herd recording system, in which both records for individual cows (e.g. lactation number, age, days in milk, milk yield, milk composition, BCS, fertility and health) and herd information (e.g. 305-d milk yield, milk composition, slaughter weight and slaughter age) are reported. When formulating diets in the NorFor system, herd information is available (Figure 15.3), which means that rations can be formulated for either individual cows or groups of cows in the herd.

15.5 Feed ration calculator (FRC) A requirement for the NorFor IT system was that it should allow both the calculation and optimisation of rations. The chosen solution was three calculator add-inns and a black-box optimisation approach (Figure 15.6). The FRC Calculator is used to evaluate feed rations and to predict production responses when the feed input is known. The FAS/FST Calculator is used to calculate standard feed values for a single feedstuff (see Chapter 13) by combining values obtained from the FRC calculator and information from the FAS and FST. The SNOPT™ Optimizer is used to optimise (auto-balance) a feed ration i.e. produce a ration that provides all the required nutrients at the lowest possible cost. This means that when using the optimiser individual feed prices must be reported through the herd feedstuff table. The SNOPT Optimizer is a package of software and algorithms, developed by researchers (Gill et al., 2005) at Stanford University and the University of California (http://www.sbsi-sol-optimize.com), for solving largescale optimisation problems (linear and nonlinear). It is especially effective for nonlinear problems whose functions and gradients are expensive to evaluate. SNOPT was implemented in NorFor with guidance from the TOMLAB company (http://www.tomopt.com/tomlab).

SNOPT optimizer

OFC calculator

Animal data

FRC calculator

Herd feedstuffs

FAS/FST calculator

Figure 15.6. The add-inns used in the NorFor FRC calculation and optimisation.

NorFor – The Nordic feed evaluation system

171

6046 6046 6046 6047 6047 6047 6048 6048 6048 6049 6049 6049 6050 6050 6050 6051 6051 6051 6052 6052 6052 6053 6053 6053 6054 6054 6054 6055 6055 6055 6056 6056 6056 6057 6057 6057 6058 6058 6058 6059 6059 6059 6060 6060 6060 6061 6061 6061 6062 6062 6062 6063 6063 6063 6064 6064 6064 6065 6065 6065 6066 6066 6066 6067 6067 6067 6068 6068 6068 6069 6069 6069 6070 6070 6070 6071 6071 6071 6072 6072 6072 6073 6073 6073 6074 6074 6074

The optimisation software module consists of a high-level interface written in C# targeting the Microsoft®.NET platform, allowing optimisation problems to be specified, and a low-level SNOPT 7 solver library built using Fortran 77 code. The functions optimized are constructed using C#-like syntax and compiled in a library called the high-level code. For nonlinear problems such as those posed in NorFor, SNOPT employs a sparse SQP algorithm, described by Gill et al. (2005). The nonlinear solver requires knowledge of the first derivatives of all (nonlinear) functions, and if some are unknown, they are estimated numerically. Initially, the NorFor system used numerical gradients for all nonlinear functions. However, the calculation engine applies automatic differentiation (AD). This eliminates the round-off errors normally associated with numerical estimation and yields exact derivatives down to machine precision. Tests made during the development of the AD package showed an increasedmethod robustness without significantly reducing performance by usefloating of AD. This is operator-overloading method applied might might theoretically theoretically be be slower slower than than using using regular regular floating operator-overloading applied important, since the operator-overloading method applied might theoretically be slower than using operator-overloading method applied might theoretically be slower than using regular floating point expressions. point expressions. point expressions. regular floating point expressions. The question of whether the optimum for each been The question of whether SNOPT SNOPT finds finds the global global optimum for each problem problem has has been raised. raised. The question of finds global optimum for has Thethe question ofwhether whetherSNOPT SNOPT findsthe thethe global optimum foreach eachproblem problem hasbeen beenraised. raised. For the For types of problem,, this depends on nature of the nonlinear constraints. For the the types types of of problem,, problem,, this this depends depends on on the the nature nature of of the the nonlinear nonlinear constraints. constraints. If If the the For If the types of problem, this depends on the nature of the nonlinear constraints. If the constraint constraint functions functions are are known known explicitly, explicitly, aa person person might might be be able able to deduce deduce whether whether or or not not functions constraint constraint functions are known explicitly, adoing person might be able to toor deduce whether ora not are known explicitly, a person might be able to deduce whether not they define they define a convex region, but although this automatically using a computer they define define aa convex convex region, region, but but although although doing doing this this automatically automatically using using aa computer computer convex region, they algorithm (without symbolic description) might be possible, possible, it would would be far far from from trivial. description) It is is but although doingaathis automatically using a computer algorithm (without a symbolic algorithm (without symbolic description) might be it be trivial. algorithm (without a symbolic–description) might be possible, itstarting would be far from trivial. It It is easier non-convexity it suffices to find different points that give might to beprove possible, it would be–far from trivial. It two is easier to prove non-convexity – it suffices to find easier to prove non-convexity it suffices to find two different starting points that give easier to prove non-convexity – it suffices towhich find two different starting points that give significantly different local optimal points, can be done with little knowledge of the two different starting points that give significantly different local optimal points, which can be done significantly different different local local optimal optimal points, points, which which can can be be done done with with little little knowledge knowledge of of the the significantly nature of the functions involved. Nevertheless, SNOPT is an industry standard nonlinear with little knowledge of the nature of the functions involved. Nevertheless, SNOPT is an industry nature of the functions involved. Nevertheless, SNOPT is an industry standard nonlinear nature of the solver functions involved. Nevertheless, SNOPT is an industry standard nonlinearfind optimisation that even on problems. It standard nonlinear solver well that often well even on non-convex problems. It may optimisation solver optimisation that often often performs performs well even performs on non-convex non-convex problems. It may may well well find optimisation solver that often performs well even on non-convex problems. It may well find the global optimum by chance,bydepending depending on the the given given starting point. However, method to to the optimum chance, on point. However, aaa method wellglobal find the global by optimum chance, depending on starting the given starting point. However, a method to the global optimum by chance, depending on the given starting point. However, method to increase run more than one optimisation, increase the odds of identifying global optimality is isto totosimply simply run more than oneone optimisation, increasethe theodds oddsof ofidentifying identifyingglobal globaloptimality optimalityis simply run more than optimisation, with increase the odds of identifying global optimality is to simply run more than one optimisation, with different starting points. If all runs give the same optimum, it is likely, although not with different starting points. If runs all runs runs give the same optimum, it is likely,although althoughnot not100% certain, different starting points. If allIf givegive thethe same optimum, it it is is likely, with different starting points. all same optimum, likely, although not 100% certain, that the global optimum has been found. A multi-start algorithm has been 100% certain, that the global optimum has been found. A multi-start algorithm has been that thecertain, globalthat optimum has been found. Abeen multi-start algorithm hasalgorithm been implemented in the NorFor 100% the global optimum has found. A multi-start has been implemented in NorFor software system, which provides an automated method for implemented in the thewhich NorFor software system, which provides an automatedmultiple method starting for implemented in the NorFor software system, which provides an automated method for software system, provides an automated method for generating generating multiple multiple starting starting points points that that can can be be used used for for repeated repeated local local optimisation optimisation runs. runs. It It points that generating generating multiple starting points that can be used for repeated local optimisation runs. It according can be used for repeated local optimisation runs. It automatically detects similar solutions automatically detects similar solutions according to criteria and can used automatically detects and similar solutions according to tolerance tolerance criteria and can be be used to to problem is automatically detects similar solutions according to tolerance criteria and can be used to to tolerance criteria can be used to obtain good indications of whether a particular obtain good good indications indications of of whether whether aa particular particular problem problem is is convex convex or or not. not. In In aa test test of of 20 20 actual actual obtain obtain indications whether athe particular problem is convex or not. a test of 20found actualto have a convexgood or not. In aevery test of of 20 actual NorFor problems, every inInthe set was NorFor problems, problem in set was found to have aaproblem unique optimal point, NorFor problems, every problem in the set was found to have unique optimal point, NorFor problems, every problem in the set was found to have a unique optimal point, unique optimal point, although for a subset of the problems certain starting points (typically one or although although for for aaa subset subset of of the the problems problems certain certain starting starting points points (typically (typically one one or or two two of of 20-50 20-50 although for subset of the problems certain starting points (typically one or two of 20-50 two of 20-50 randomly chosen points) resulted in infeasible solutions. This is not a cause for concern, randomly chosen points) resulted in infeasible solutions. This is not a cause for concern, since randomly chosen points) resulted in infeasible solutions. This is cause for concern, since randomly chosen points)be resulted innot infeasible solutions. This is not not aaand cause concern, can since sincewill therealways will always points do not all constraints, the for multi-start can there be do all constraints, and algorithm there will always be points points that that do that not fulfil fulfil allfulfil constraints, and the the multi-start multi-start algorithmalgorithm can there will always be points that do not fulfil all constraints, and the multi-start algorithm can be applied ininattempts attempts totosolve solve problems that seem to to be be infeasible. be applied appliedin attemptsto solveproblems problems that seem infeasible. be that seem to be infeasible. be applied in attempts to solve problems that seem to be infeasible. The in the NorFor system is the The principal principaloptimisation optimisationproblem problem NorFor system to identify the cheapest combination The principal optimisation problem in in thethe NorFor system is to toisidentify identify the cheapest cheapest The principal optimisation problem in the NorFor system is to identify the cheapest combination of feed ingredients that meets nutritional requirements. The linear cost is function of feed ingredients that meets nutritional requirements. The linear cost function simply the total combination of feed ingredients that meets nutritional requirements. The linear combination of feed ingredients that meets nutritional requirements. The linear cost cost function function is simply the total cost of a ration, expressed: cost of a ration, is simply the total totalexpressed: cost of of aa ration, ration, expressed: expressed: is simply the cost nn

6075 6075 6075

n pi x min ff ((xx )) = min = ∑ ∑p pi x ii min x f (x ) = i∑ =1 i x i

15.1

6076 6076 6076 6077 6077 6077

≤ Ax Ax ≤ ≤b bU bbb L ≤ L L ≤ Ax ≤ b U U cc L ≤ cc((xx )) ≤ cc U ≤ ≤ L ≤ c(x ) ≤ c U U cL

15.2

15.3

15.4

6078 6078 6078 6079 6079 6079 6080 6080 6080 6081 6081 6081 6082 6082 6082 6083 6083 6083 6084 6084 6084 6085 6085 6085 6086 6086 6086 6087 6087 6087 6088

xx

ii = =11

n,x ≥ 0 n pp ,, x ∈R n ≥0 0 p, x x∈ ∈R R n ,, x x≥

m1 m1 × ×n bU ∈ ∈R Rm A∈ ∈R Rm m111 ,, A m 111 .×nn n bbb L ,, b L L ,bU U ∈R ,A∈R m m 22 cc L ,, cc U ,, cc(( xx )) ∈ 2 ∈R Rm 2 cL L , cU U , c( x ) ∈ R where, p is the price unit of feedstuff i; xi is the amount of feedstuff ii used in the ration. i is the price per per of i; of the where, p i nn where, pii isthe theset price per unit unit of feedstuff feedstuff i; xxiii is is the then;amount amount of feedstuff feedstuff i used used in inelements; the ration. ration. R implies of real numbers with dimension A is a matrix with constant n n R implies implies the the set set of of real real numbers numbers with with dimension dimension n; n; A A is is aa matrix matrix with with constant constant elements; elements; bbbLLL R L are linear constraints expressed on vector form; c and c are non-linear constraints. and b U L U and and bbUUU are are linear linear constraints constraints expressed expressed on on vector vector form; form; ccLLL and and ccUUU are are non-linear non-linear constraints. constraints.

15.5 15.6

The set of constraints constraints expresses expresses nutritional nutritional demands, demands,NorFor which include include DMI, energy, energy, AATNN,, The DMI, AAT 172set of – The Nordic feed evaluation system The set of constraints characteristics expresses nutritionalrumen demands, which which include DMI, energy, AATNN, PBV PBVNNN and and nutritional nutritional characteristics characteristics (e.g., (e.g., rumen rumen load load index, index, fatty fatty acids, acids, NDF NDF and and minerals). minerals). PBV and nutritional (e.g., load index, fatty acids, NDF and minerals). N The The constraints constraints are are divided divided into into m m111 linear linear and and m m222 nonlinear nonlinear constraints, constraints, aaa special special case case being being linear and m nonlinear constraints, case being The constraints are divided into m 1are 2nonlinear, the so-called ratio constraints that originally being quotientsspecial of linear sums:

where, pi is the price per unit of feedstuff i; xi is the amount of feedstuff i used in the ration. Rn implies the set of real numbers with dimension n; A is a matrix with constant elements; bL and bU are linear constraints expressed on vector form; cL and cU are non-linear constraints. The set of constraints expresses nutritional demands, which include DMI, energy, AATN, PBVN and nutritional characteristics (e.g. rumen load index, fatty acids, NDF and minerals). The constraints are divided into m1 linear and m2 nonlinear constraints, a special case being the so-called ratio constraints that are originally nonlinear, being quotients of linear sums:

∑ ∑

n i =1 i i n i =1 i

wx

6089

r (x ) =

6090 6091 6092 6093 6094 6095 6096 6097 6098 6099 6100 6101 6102 6103 6104 6105 6106 6107 6108 6109 6110 6111 6112 6113 6114 6115 6116 6117 6118 6119 6120 6121 6122 6123 6124 6125 6126 6127 6128 6129 6130 6131 6132 6133 6134 6135 6136

where, wi are constant real-valued weighting values. These constraints are easily transformed into at most two linear constraints. When optimising for more than one animal where, wi is constant real-valued These constraints easily transformed into at simultaneously, two runs are madeweighting — first forvalues. the individual animals andare then for the entire most two constraints. When optimising for problem more than onetoanimal group withlinear additional constraints. The multi-animal leads a blocksimultaneously, structure in the two runs constraint 15.7 illustrates a hypothetical with four with animals (four constraints. are made: matrix. first forFigure the individual animals and then forsituation the entire group additional blocks), 10 feeds and five constraints which link different animals together), theillustrates The multi-animal problem leads to a(three blockof structure in the constraint matrix. Figureand 15.7 total usage of feedstuffs 1, 2 and is limited. a hypothetical situation with four3animals (four blocks), 10 feeds and five constraints (three of which

x

15.7

link different animals together), and the total usage of feedstuffs 1, 2 and 3 is limited.

Figure 15.7

One of the main requirements when selecting the optimisation platform was that the execution time

One of the mainArequirements selecting the optimisation platform client was that the execution should be fast. performancewhen test was conducted with the Norwegian involving simultaneous time should be fast. A performance test was conducted with the Norwegian client involving optimisation for 30 cows, nine constraints and five feedstuffs. The obtained optimisation time was simultaneous optimisation for 30 cows, nine constraints and five feedstuffs. The optimisation 0.23 second per cow. time was 0.23 s per cow. The goal of optimisation is to find the least costly combination of feeds that meets a set of constraints (minimum and maximum) ensuring that amounts of nutrients supplied are within specified ranges. Thus, a set of constraints is needed for both feed ingredients and nutrients. In a practical situation, the amount of home grown feed is known and feed constraints are often based on their availability. Therefore, the amount of a feed in the ration may often be adjusted by the operator to take into account a minimum or maximum that must be used from an inventory perspective. When feeds are purchased an optimisation approach is useful when the feed budget and purchasing of ingredients are planned. Nutritional constraints are based on either their direct effects on production responses, i.e., the requirements (NEL and AATN) of animals to meet target performance criteria or nutrient amounts and concentrations in the rations, e.g., PBV, RLI, NDF, Starch, CP, fatty acids and minerals. It is possible to optimise 84 nutritional variables in NorFor but not all of them have recommended constraints at present. When rations are optimised in the national clients, we start with a recommended default setting for eight nutritional constraints (Table 15.1). Table 15.1

Table 15.2 illustrates the process of ration optimisation with a set of diets formulated with variations in the constraints of RLI (0.6 vs. 0.45) and AAT/NEL (15 vs. 16 g/MJ). TMR rations were formulated for a target of 35 kg ECM and in addition to the standard optimisation constraints, several other nutritional variables were used to evaluate and characterise the rations. Reducing the RLI (diet 1 vs. diet 2) reduced the proportion of wheat grain in the optimal diet by 7.5 percent and increased the proportion of cold-pressed rapeseed by 5.8 percent. Thesparsity change pattern in RLI affected the starch content, which decreased by 24 % when Figure 15.7. The for a problem with four animals, ten feedstuffs, five constraints the was reduced. Increasing the minimum requirement of AAT/NEL from 15 to 16 g/MJ andRLI three constraints limiting total usage of feedstuffs. (Diet 2 vs. Diet 3) increased the proportion of extracted soybean meal and decreased the proportion of rapeseed in the optimal diet by 3.7 and 4.6 percent, respectively. This change resulted in an increased CP concentration in the diet and minor changes in the AA composition. The simple examples presented in Table 15.2 demonstrate how the combination of auto-balancing and nutritional constraints change the optimal diet composition. They also NorFor – The Nordic feed evaluation system 173 show the sensitivity of the optimisation procedure to changes in the constraints and highlight

The goal of optimisation is to find the least costly combination of feeds that meets a set of constraints (minimum and maximum) ensuring that amounts of nutrients supplied are within specified ranges. Thus, a set of constraints is needed for both feed ingredients and nutrients. In a practical situation, the amount of home grown feed is known and feed constraints are often based on their availability. Therefore, the amount of a feed in the ration may often be adjusted by the operator to take into account a minimum or maximum that must be used from an inventory perspective. When feeds are purchased an optimisation approach is useful when the feed budget and purchasing of ingredients are planned. Nutritional constraints are based on either their direct effects on production responses, i.e. the requirements (NEL and AATN) of animals to meet target performance criteria or nutrient amounts and concentrations in the rations, e.g. PBV, RLI, NDF, Starch, CP, fatty acids and minerals. It is possible to optimise from 84 nutritional variables in NorFor but not all of them have recommended constraints at present. When rations are optimised in the national clients, we start with a recommended default setting for eight nutritional constraints (Table 15.1). Table 15.2 illustrates the process of ration optimisation with a set of diets formulated with variations in the constraints of RLI (0.6 vs. 0.45) and AAT/NEL (15 vs. 16 g/MJ). TMR rations were formulated for a target of 35 kg ECM and in addition to the standard optimisation constraints, several other nutritional variables were used to evaluate and characterise the rations. Reducing the RLI (diet 1 vs. diet 2) reduced the proportion of wheat grain in the optimal diet by 7.5% and increased the proportion of cold-pressed rapeseed by 5.8%. The change in RLI affected the starch content, which decreased by 24% when the RLI was reduced. Increasing the minimum requirement of AAT/NEL from 15 to 16 g/MJ (Diet 2 vs. Diet 3) increased the proportion of extracted soybean meal and decreased the proportion of rapeseed in the optimal diet by 3.7 and 4.6%, respectively. This change resulted in an increased CP concentration in the diet and minor changes in the AA composition. The simple examples presented in Table 15.2 demonstrate how the combination of auto-balancing and nutritional constraints changes the optimal diet composition. They also show the sensitivity of the optimisation procedure to changes in the constraints and highlight the importance of ensuring that feed rations obtained by optimisation are assessed by a trained nutritionist before they are implemented. Table 15.1. Default optimisation constraint settings in NorFor for dairy cows. Item

Ration fill value Energy balance, % AATN/NEL2, g/MJ AATN response, % PBVN3, g/kg DM RLI4, g/g Fatty acids, g/kg DM Chewing index, min/kg DM

Constraint Minimum

Maximum

IC·0.97 99.5 15 95 10

IC1 100.5

20 32

40 0.6 45

1

IC = animal intake capacity. AATN/NEL = ratio between amino acids absorbed in the small intestine (AATN) and net energy lactation (NEL) available for milk production. 3 PBV = protein balance in the rumen. 4 RLI = rumen load index. 2

174

NorFor – The Nordic feed evaluation system

Table 15.2. Ration optimisation with varied constraints for rumen load index and AAT/NEL for a 600 kg dairy cow producing 35 kg ECM. Ration Diet 1

Diet 2

Diet 3

Ingredients, % of DM Grass silage, medium OMD1 Grass silage, high OMD Maize silage, high OMD Wheat grain Wheat, brewers grain Soybean meal Rapeseed, cold pressed Salt Mineral and vitamin mix

16.5 15.8 24.8 26.6 6.3 3.4 5.6 0.23 0.68

16.1 20.2 23.3 19.1 6.3 2.8 11.4 0.22 0.67

16.2 19.5 24.8 19.0 6.3 6.5 6.8 0.23 0.68

Total dry matter, kg/d Cost, NOK

22.2 29.86

22.3 29.64

22.2 30.10

8.45 100 15 95 10 0.6 42 41 57.2 154 340 265 33 6.54 2.21 2.49 32 47 28

8.45 100 15 95 20 0.45 50 42 59.6 165 366 202 38 6.61 2.29 2.53 37 51 30

8.45 100 16.0 98 20 0.45 44 42 60.4 170 363 205 36 5.59 2.22 2.50 35 48 29

Ration fill value, FV/kg DM Energy balance, % AATN/NEL, g/MJ2 AATN response, % PBVN, g/kg DM3 RLI, g/g4 Crude fat, g/kg DM Chewing index, min/kg DM Roughage proportion, % of DM Crude protein, g/kg DM NDF, g/kg DM Starch, g/kg DM Sugar, g/kg DM Lysine, % of AATN Methionine, % of AATN Histidine, % of AATN Calcium, g/kg DM Phosphorus, g/kg DM Magnesium, g/kg DM 1

OMD = organic matter digestibility. AATN/NEL = ratio between amino acids absorbed in the small intestine (AATN) and net energy lactation (NEL) available for milk production. 3 PBV = protein balance in the rumen. N 4 RLI = rumen load index. 2

NorFor – The Nordic feed evaluation system

175

In the system development, testing and implementation process, NorFor has developed a PC program called the NorFor Development Tool (NDT). The NDT runs off-line, incorporates all the NorFor equations and compiles the three calculator add-inns (Figure 15.8). This means that it is possible both to calculate and optimize rations with NDT. The program is used by the personnel who develop and maintain the system. It is easy to edit existing equations and to include new equations in the program (Figure 15.9). It is also possible to both import and export data between NDT and Microsoft Excel, which makes it easy to test and evaluate experimental data. Import of feedstuff specifications into the NDT from the online NorFor feedstuff table is implemented. An essential feature of the NDT is the unit test, which allows multiple tests to be run simultaneously to ensure that changes in the equation sets do not have unexpected effects on different parts of the model. Before a new version of the NorFor system becomes available for the national clients, it is extensively evaluated in the NDT. After approval, the new/edited equations are transferred to the operative NorFor servers. This procedure ensures that the system is updated safely.

15.6 One-day feeding control (OFC) The OFC is a ration evaluator that is used to calculate herd feed efficiency (mainly in terms of energy and nitrogen) for a specific day or period. It compiles data from the national herd recording system and from the FST and FRC. The number of animals in each category (cows, calves, heifers and bulls) and the status of individual animals (e.g. MY, milk chemical, gestation day and age) are available from the national herd recording system for the specific test day. The feed intake is measured for each animal category or groups of animals and, in combination with information from the herd feedstuff table, the FRC is used to calculate ration composition, production responses and feed efficiency. The ration evaluation and OFC are then used for optimising/adjusting the ration for the next feeding period.

Figure 15.8. The architectural structure of the NorFor Development tool.

176

NorFor – The Nordic feed evaluation system

Figure 15.9. Example of a dialogue box in the NorFor Development Tool used to edit and test new equations.

References Boston, R.C., D.G. Fox, C. Sniffen, E. Janczewski, R. Munson and W. Chalupa, 2000. The conversion of a scientific model describing dairy cow nutrition and production to an industry tool: the CPM dairy project. In: McNamara, J., J. France and D.E. Beever (eds.). Modelling nutrient utilization in farm animals. CABI Publishing, Wallingford, UK, pp. 361-377. Gill, P.E., W. Murray and M.A. Saunders, 2005. SNOPT: An SQP algorithm for large-scale constrained optimisation. SIAM Review 47: 99-131.

NorFor – The Nordic feed evaluation system

177

Keyword index A AATN 83 –– available for growth 98 –– available for milk production 95 –– balance 96 –– efficiency 95, 100 administration tool 169 amino acids 70, 71, 72, 73 ammonia 44, 62 –– correction 37 animal production level 82 apparent digestibility 75, 77 ATP 66, 69 B black-box optimisation body condition score body protein

171 27 101

C carbohydrate cation-anion difference cellulolytic bacteria chewing index –– eating index –– particle length –– recommendation –– rumination index concentrates crude protein

35 57, 108 64 127 128 128, 129 104 129 66 34, 45, 46, 62

D daily gain degradation rate –– in feed mixtures Development Tool digestion methods drying temperature

30 59 57 176 42 42

E effective rumen degradability 66, 67 endogenous protein 83 energy balance 91 energy corrected milk 28 energy requirement 85, 86, 87, 90 enzymatic digestion 71 F fatty acids feed analysis

73, 104 41

feed ration calculator feedstuff table fermentation products –– acids –– alcohols G gestation global optimum gross energy growth function I input data in sacco –– mobile bag in vitro in vivo IT platforms IT system J Jersey

167, 171 167, 169 38 38 38 27 172 81 31 27, 29 35, 48, 51, 63 34 45, 55 45 167 167 27

L lactation curves 28 large intestine 74 –– degradation of neutral detergent fibre in feedstuff 75 –– degradation of starch 75 –– microbial protein synthesis 75 linear cost function 172 Lucas principle 35, 55 M mature body weight 28, 31 metabolizable energy 81 –– efficiency 81 methane 81 microbial –– efficiency 67 –– growth 62, 66 –– protein 83 –– protein synthesis 67 milk protein maximum production 96 mineral recommendations 105, 108, 113 minerals 38 mobile nylon bag 52

H. Volden (ed.), NorFor–The Nordic feed evaluation system, DOI 10.3920/978-90-8686-718-9, © Wageningen Academic Publishers 2011

179

N national client national herd recording system net energy lactation neutral detergent fibre in feedstuff –– degradation –– indigestible –– potentially degradable nitrogen utilisation non-linear optimization code O one-compartment –– degradation model –– digestion model one-day feeding control optimisation organic matter –– digestibility

169 171 82 35 55, 65 35, 51 35 134 115

74 65 167, 176 174 33 45

P partial efficiency of metabolizable energy 82 –– kg 82 –– km 82 –– kmg 82 particle length 33, 47 passage rate 59, 60, 61 PBVN 70 PBVN recommendation 102 phosphorus utilisation 135 potassium utilisation 136 prediction of milk yield 133 prediction of weight gain 133 protein 71, 73, 93, 94 –– production 28 protein requirement –– gestation 102 –– growth 97 –– mobilization and deposition 101 R RestCHO rumen evacuation technique rumen load index S sensitivity analysis silage index small intestine SNOPT Optimizer standard feed value –– AATN8 180

37, 63 60 64, 104 141 38, 56 71, 73 171 137 138

–– AATN20 138 –– NEL8 138 –– NEL20 138 –– PBVN8 138 –– PBVN20 138 starch 35, 36, 63, 73 substitution rate 113 system evaluation 141 –– chewing index 159 –– digestive tract sub-model 143 –– energy requirement for growing cattle 153 –– milk production 149 –– roughage intake 154 –– rumen load index 153 –– sensitivity analysis 142 –– total dry matter intake 154 T theoretical chopping length triacylglycerols two-compartment digestion model U urea correction

128 66 65 37

V vitamin recommendations 108 –– vitamin A 109 –– vitamin D 109 –– vitamin E 109 vitamins 39 voluntary feed intake 113 –– capacity of bulls 119 –– capacity of Jersey 121 –– capacity of lactating dairy cows 115 –– capacity of steers and heifers 120 –– fill value 56 –– metabolic correction factor 118, 123 –– metabolic regulation factor 117 –– recommendations 104 –– substitution correction factor 122 –– substitution rate factor 117 W water-free amino acids web services

72 167

NorFor – The Nordic feed evaluation system

Our partners will collect data and use cookies for ad personalization and measurement. Learn how we and our ad partner Google, collect and use data. Agree & close