Protection of Civilian Infrastructure from Acts of Terrorism (NATO Science for Peace and Security Series C: Environmental Security)

Protection of Civilian Infrastructure from Acts of Terrorism

NATO Security through Science Series This Series presents the results of scientific meetings supported under the NATO Programme for Security through Science (STS). Meetings supported by the NATO STS Programme are in security-related priority areas of Defence Against Terrorism or Countering Other Threats to Security. The types of meeting supported are generally "Advanced Study Institutes" and "Advanced Research Workshops". The NATO STS Series collects together the results of these meetings. The meetings are co-organized by scientists from NATO countries and scientists from NATO's "Partner" or "Mediterranean Dialogue" countries. The observations and recommendations made at the meetings, as well as the contents of the volumes in the Series, reflect those of participants and contributors only; they should not necessarily be regarded as reflecting NATO views or policy. Advanced Study Institutes (ASI) are high-level tutorial courses to convey the latest developments in a subject to an advanced-level audience Advanced Research Workshops (ARW) are expert meetings where an intense but informal exchange of views at the frontiers of a subject aims at identifying directions for future action Following a transformation of the programme in 2004 the Series has been re-named and re-organised. Recent volumes on topics not related to security, which result from meetings supported under the programme earlier, may be found in the NATO Science Series. The Series is published by IOS Press, Amsterdam, and Springer, Dordrecht, in conjunction with the NATO Public Diplomacy Division. Sub-Series A. Chemistry and Biology B. Physics and Biophysics C. Environmental Security D. Information and Communication Security E. Human and Societal Dynamics http://www.nato.int/science http://www.springer.com http://www.iospress.nl

Series C: Environmental Security – Vol. 12

Springer Springer Springer IOS Press IOS Press

Protection of Civilian Infrastructure from Acts of Terrorism edited by

Konstantin V. Frolov Institute of Machine Sciences, Russian Academy of Sciences, Moscow, Russia and

Gregory B. Baecher University of Maryland, MD, U.S.A.

Published in cooperation with NATO Public Diplomacy Division

Proceedings of the NATO Advanced Research Workshop on Protection of Civilian Infrastructure from Acts of Terrorism Moscow, Russia May 27--29, 2004 A C.I.P. Catalogue record for this book is available from the Library of Congress.

ISBN-10 ISBN-13 ISBN-10 ISBN-13 ISBN-10 ISBN-13

1-4020-4923-4 (PB) 978-1-4020-4923-1 (PB) 1-4020-4922-6 (HB) 978-1-4020-4922-4 (HB) 1-4020-4924-2 (e-book) 978-1-4020-4924-8 (e-book)

Published by Springer, P.O. Box 17, 3300 AA Dordrecht, The Netherlands. www.springer.com

Printed on acid-free paper

All Rights Reserved © 2006 Springer No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed in the Netherlands.

TABLE OF CONTENTS Preface

vii

Acknowledgements

xi xiii

Contributors

Introduction Opening Address A.K. Frolov

3

Setting the Scene and Characterizing the Threat Risk Management in Natural and Societal Systems R. Akhmetkhanov Vulnerability Estimation of High-Rise Structures in Case of Non-Regular Dynamic Actions by Methods of Statistical Simulation V.T. Alymov and O.V. Trifonov

7

21

Setting the Stage: The Vulnerability of Critical Infrastructures T. Thedéen

33

The Risk Imposed by Fire to Buildings and how to Address it J.L. Torero

41

Analysis of Technogenic Risks Under Terrorist Impacts N. Makhutov

v

59

vi

Table of Contents

Vulnerability, Risk Analysis, and Risk Assessment Petroleum Supply Vulnerability Due to Terrorism at North Sea Oil and Gas Infrastructures M. Tørhaug

73

Lessons from Safety Assessment, Natural Disasters and Other Hazards J. McQuaid

85

Large Dams and the Terrorist Threat R.A. Stewart Decontamination in the Event of a Chemical or Radiological Terrorist Attack K. Volchek, M. Fingas, M. Hornof, L. Boudreau, and N. Yanofsky

103

125

Mitigation and Response Mitigating Water Supply System Vulnerabilities G.B. Baecher

149

Nuclear Terrorism and Insurance Liability O.M. Kovalevich and S.D. Gavrilov

159

Capital Wireless Integrated Network H. Ali, S. Yang and T.H. Jacobs

169

Emergency Services in Homeland Security F. Krimgold, K. Critchlow and N. Udu-gama

193

Communications Infrastructure Security M.J. Casey

231

On the Possibility of Detecting Explosives by the Combined use of Nuclear Reactions -(N,N), (N,J), (N,P) G. Kotel'nikov, S. Kotel'nikov, V. Stepanchikov, and G. Yakovlev

Index

247

253

PREFACE Konstantine V. Frolov1 and Gregory B. Baecher2 1

Director of the Institute for Machine Sciences, RAS; and 2University of Maryland

The objective of the Workshop on Protection of Civilian Infrastructure From Acts of Terrorism was to lay the foundation for a risk-informed approach to modeling, analyzing, predicting, managing, and controlling multisector Infrastructure networks in the face of human threats and errors. The goal was to combine the insights of a spectrum of disciplines across engineering, public policy, planning, and economics. The workshop addressed the need to develop an understanding for systems behaviors and vulnerabilities of interacting networks; create a riskinformed analysis capability for modeling and predicting the behavior of complex networks; apply emerging technology to the problems of designing, constructing, monitoring, and operating critical infrastructure; and build an understanding of the social, economic, and environmental factors that affect, and are affected by, critical infrastructure. The objective was to develop an understanding of the vulnerability of critical systems to various modes of terrorist attack. The benefit of developing such understanding is that approaches can be crafted to reducing vulnerability and to containing or limiting the propagation of failure within an infrastructure system, thus limiting the impact of terrorism. This also leads to improved understanding of infrastructure systems in general, not only in the face of threats but also natural hazards. Areas of research need and capability were identified, along with opportunities for future exchange and collaboration. The workshop was organized around eight major themes grouped in three areas: (I) setting the scene and characterizing the threat (the general problem of protecting civil infrastructure, lessons from conventional hazards and threats), (II) vulnerability (technogenic risks and safety, risk analysis, vii

viii

Preface

risk management), and (III) mitigation and response (mitigation, risk reduction, and capacity building). The more specific purviews of each respective theme are the following. General problem of protecting civil infrastructure: The infrastructures of modern society are big, interconnected and critical or the daily life of the citizens. Examples are the power network, transportation networks, and data and telecommunication networks. These infrastructures can be modeled using graphs. Threats can be of different kinds: Natural, technogenic, human error, hackers and antagonistic threats, including terrorism and war. These threats may result in system collapses. The hazards due to nature, technical and human error and random mishap and can be analyzed by risk analytical methods. Antagonistic threats must be handled by other methods. Lessons from conventional hazards and threats: The need to anticipate and prevent potential accidents with severe consequences has resulted in the development of structured procedures and methods of assessment for the different stages of the life cycle of a catastrophic event. Techniques of assessment are of varying degrees of complexity and, in their more developed forms, can help to promote understanding among interested parties of the intricacies of what are usually complex situations. The principal benefit that such techniques bring to decision making is imposing order and discipline on the process of assessment and in exposing difference in the exercise of judgment by assessors. Technogenic risks and safety: Modern terrorism can be divided into three types, traditional terrorism, technogenic or high impact terrorism, and intelligent terrorism. Emergency situations initiated by terrorist attacks and traditional man-made catastrophes are developed according to similar mechanisms and laws. The existing standards and codes in the field of designing, constructing, and maintaining critical infrastructures and facilities should therefore be modified and updated in view of terrorist impact threats. The rapid development of global infrastructure networks and the activity of international terrorist organization requires coordinated efforts to reduce the vulnerability of critical infrastructures towards potential terrorist impacts. Risk analysis and management: The peculiarity of terrorist impacts lies in the infliction of maximum casualties among a population at minimal expenditure. Terrorist acts lead to complex impacts on a system. This distinction for catastrophes of natural and technogenic character (or social crisis) as opposed to terrorism leads to the fact that while risk theory may be applied to natural an social system safety management, problems of estimating terrorist acts using probability theory are subject to difficulties because the analyses are changed in shape as well as in the traditional risk analysis stages and solutions. Systems approaches to risk analysis of natural and

Preface

ix

social systems are based on sub-systems considerations in their interconnection that makes it possible to take into account damage cumulatively; in contrast, there is a synergetic manifestation of terrorist impacts. Risk of national and social systems can be structured in the shape of potential functions and a power spectrum of different risks. Mitigation and response: Consequences of terrorist attacks in terms of damage values are comparable to those of major hazards and catastrophes. However, major hazards and catastrophes are occasional in character whereas terrorist attacks are well-planned events. Terrorists as a rule choose targets that are likely to cause the most striking social response. Therefore, development of methodologies and techniques to forecast the most probable target for terrorist attacks is a most urgent talks. The methods for complex assessment of safety level that have been developed in the framework of a federal safety program can be adapted to fulfill this task. Risk reduction and human errors: Human and organizational factors plan a critical role in safety management . It is estimated that in industrial settings, 70 to 80 percent of accident causation is attributed to human error. In almost all cases of technological disaster, warning signals are present in the system and either discounted or ignored. New models of safety management offer alternative methods of risk prediction and control. These can be examined from a cross cultural perspective to identify diagnostic and intervention techniques which many be effectively applied in both the natural and technologic spheres, including terrorist attacks.

ACKNOWLEDGEMENTS

Many people contributed to the success of the NATO-Russia Workshop on Protection of Civil Infrastructure from Acts of Terrorism. There is not space to acknowledge all of those who contributed. However, foremost among those who made the workshop possible was Dr. Alain H. Jubier, NATO Program Director, Environmental Security, Public Diplomacy Division, whose guidance, insights, and sponsorship allowed the workshop to happen. The editors also express their appreciation to the Institute for Machine Sciences of the Russian Academy of Sciences for hosting the workshop, and to Dmitry Reznikov for serving as liaison in planning the workshop and helping to produce these proceedings. The expert technical translation services and spirited contributions of Irina Pushkina allowed for effective communications even of the most complex technical issues. Lastly, nothing would have been possible without the participation of the number of NATO and Russian scientists and engineers who contributed insight and generously shared their ideas.

xi

CONTRIBUTORS

Rasim Akhmetkhanov Institute for Machine Sciences, RAS, 4 Maly Kharitonievsky st., Moscow Valentin T. Alymov

Institute for Machine Sciences, Russian Academy of Sciences, 4 Maly Kharitonievsky st., Moscow Gregory B. Baecher Department of Civil and Environmental Engineering, University of Maryland, College Park, MD 29742 Louise Boudreau SAIC Canada, 60 Queen Street, Ottawa, Ontario K1P 5Y7 Michael J. Casey Department of Civil Engineering, George Mason University, 4400 University Drive MS4A6, Fairfax, VA 22030 Keith Critchlow Virginia Polytechnic Institute and State University, 4300 Wilson Blvd, Suite 750, Arlington, VA 22203 Merv Fingas Environment Canada, 335 River Road, Ottawa, Ontario K1A 0H3 xiii

xiv

Contributors

Konstantine V. Frolov Institute for Machine Sciences, RAS, 4 Maly Kharitonievsky st., Moscow Sergey D. Gavrilov DECOM Technology Intellectual Ltd, PO 6 Moscow 123154 Ali Haghani Department of Civil Engineering, University of Maryland, College Park, MD 29742 Monica Hornof SAIC Canada, 60 Queen Street, Ottawa, Ontario K1P 5Y7 Thomas H. Jacobs Center of Advanced Transportation Technology, University of Maryland, College Park, MD 29742 Gennadii Kotelnikov Russian Research Center Kurchatov Institute, Kurchatov Sq., 1, Moscow, 123182 Vladimir Kotelnikov Spetssvyaz, Offis 267, Tverskaya 12, Bilding 7, Moscow, 125009 Oleg M. Kovalevich Nuclear & Radiation Safety Science and Technology Center, Moscow; Russia, 14/23, Avtozavodskaya ul., Moscow, 109280 Frederick Krimgold Virginia Polytechnic Institute and State University, 4300 Wilson Blvd, Suite 750, Arlington, VA 22203 Nikolay Makhutov Institute for Machine Sciences, RAS, 4 Maly Kharitonievsky st., Moscow James McQuaid Royal Academy of Engineering, Department of Mechanical Engineering, University of Sheffield

Contributors

xv

Vladimir Stepanchikov Russian Research Center Kurchatov Institute, Kurchatov Sq., 1, Moscow, 123182 Ramond A. Stewart BCHydro, 6911 Southpoint Drive (E-14), Burnaby, BC V3N 4X8 Magne Tørhaug Det Norske Veritas, Veritasveien 1, 1322 Høvik, Oleg V. Trifonov Moscow Power Engineering Institute (Technical University), 17 Krasnokazarmennaya st., Moscow Torbjörn Thedéen Royal Institute of Technology (KTH), SE-100 44 Stockholm KTH Sweden José L. Torero Edinburgh Center for Fire Research, The University of Edinburgh, Mayfield Road, Edinburgh EH9 3JL Natasha Udu-gama Virginia Polytechnic Institute and State University, 4300 Wilson Blvd, Suite 750, Arlington, VA 22203 Konstantin Volchek Environment Canada, 335 River Road, Ottawa, Ontario K1A 0H3 Genrikh Yakovlev Russian Research Center Kurchatov Institute, Kurchatov Sq., 1, Moscow, 123182 Saini Yang Center of Advanced Transportation Technology, University of Maryland, College Park, MD 29742 Norman Yanofsky Department of National Defence, 305 Rideau Street, Ottawa, Ontario K1A0K2

INTRODUCTION

OPENING ADDRESS Academician Konstantin Frolov Director of the Institute for Machine Sciences, RAS

We are glad to welcome the participants of the workshop ‘Protection of Civilian Infrastructure from Acts of Terrorism’ in Moscow in this historic building of the Institute for Machine Sciences that witnessed many glorious events in the public and scientific life of Russia. I would like to express our acknowledgement to the NATO Program ‘Science for Peace” for its support of fundamental science and its efforts to establish close relationships between the scientists of the NATO countries and Russia, to form climate of mutual trust, and for its attention to one of the global problems facing humanity, the problem of ensuring safety in the wide sense of this word. It is significant that the workshop that took more than a year to be prepared is in full conformity with recently approved new priorities of NATO scientific program that is to be concentrated on the countering terrorism and other threats to security. The cooperation between Russian experts in the field of safety and their counterparts in EU, US and Canada has been developed mainly as two-side contacts. In particular there is a program of cooperation between the Russian academy of Sciences and the US National Academies focused on countering technological terrorism. I believe it is very important that we have been able to get together in the frame of the workshop specialists from many countries because countering international terrorism could only be achieved through intensive multinational efforts. Russia like many other countries has substantial experience in assessing risks of natural and manmade catastrophes. The results of the research are summarized in a multivolume series ‘Safety of Russia’ that is being published by the Russian Academy of Sciences and the International foundation “Znanie”. I believe that further work on reducing terrorist risks has to be based on the existing methods and approaches to natural and manmade risks analysis that should be adapted to hazards initiated by terrorist acts 3 K.V. Frolov and G.B. Baecher (eds.), Protection of Civilian Infrastructure from Acts of Terrorism, 3–4. © 2006 Springer. Printed in the Netherlands.

4

A.K. Frolov

through taking into account the peculiarities of terrorist risks. We should bear in mind however that those risks would involve additional uncertainties due to special preliminary planning by terrorists. Damages and losses inflicted by terrorist attacks may differ from those caused by natural and physical processes developed due to natural and technological hazards. We are ready to share our experience in this field and are interested in the experience of other countries in protection of critical infrastructures against acts of terror and in reducing vulnerability of high risk facilities under the threat of terrorism. It is necessary to bear in mind that the problem of terrorism is complex and interdisciplinary. Its solution requires analysis of technical, social, political, historical and cultural aspects of the problem. In particular the problem of harmonization of national legislative frameworks is a priority one. A number of these aspects are beyond the frame of our workshop. Since we are representatives of engineering community, we are primarily interested in reducing vulnerability of critical infrastructures towards terrorist attacks, developing protection systems, allocation of resources, planning response and recovery operations after major terrorist attacks. In conclusion I would like to wish fruitful work to all of us. I hope that our workshop makes a tangible contribution to promote international cooperation in countering terrorism.

SETTING THE SCENE AND CHARACTERIZING THE THREAT

RISK MANAGEMENT IN NATURAL AND SOCIETAL SYSTEMS Taking into Account Terrorist Threats Rasim Akhmetkhanov Institute for Machine Sciences, RAS

Abstract:

The paper presents systemic approach to ensuring safety of naturalmanmade-social systems that could be subjected to terrorist impacts. The theory of risks and methods to risk analysis that are used in assessing natural, manmade and societal crises and catastrophes, can be applied to assess risks of terrorist impacts as well. Such approach allows to study cascadesynergetic process, to reveal week elements of a system and to undertake measures for protection against terrorist attacks. The presented systemic description of risk allows to conduct a profound and comprehensive study of interaction between various elements of natural-manmade-social system, to select basic elements and to determine a possibility of terrorist impacts on them at local and systemic (global) levels taking into account the internal characteristics of the system.

Key words:

risk analysis, terrorism, cascade-synergetic processes, systems

According to UN data terrorist activity tend to grow steadily during the past 15 years. In the 20th centurpy for the first time in human history terrorism became a global problem closely connected to the problem of human survival. Modern terrorism differs drastically from the terrorism of the past. Nowadays terrorists have the opportunity to make use of innovative technologies and weapons of mass destruction. This opportunity is not an abstract one. 1n 1994 a terrorist was detained in Ukraine who threatened to blow up a reactor in Chernobyl nuclear power plant if his requirements were not satisfied. Poison-gas was sprayed in Tokyo underground. 7 K.V. Frolov and G.B. Baecher (eds.), Protection of Civilian Infrastructure from Acts of Terrorism, 7–20. © 2006 Springer. Printed in the Netherlands.

8

R. Akhmetkhanov

Modern terrorism acquires a system character; the goal of terrorists is to break a system, to upset the balance, to change its structure and relations between elements. Modern terrorism has several aspects. Thus the new terrorism makes allowance for world interdependence, system character of processes going on the world and offers a corresponding strategy of threats. The actions of terrorism are based on a domino effect. A butterfly effect inevitably arises in the in the complex and globalize society: a fairly insignificant incident at one place cause an event with avalanche-like consequences at another place. Terrorism is seeking new means of intimidation, more cruel and large-scale ones.

Figure1. Kinds of terrorism. a- traditional terrorism; b- technological terrorism; c- intellectual terrorism; Ua- initial damage; Ub – secondary damage; Uk – cascade damage Terrorism has proved to be directly connected with the problem of human survival and ensuring national safety. Terrorism is an extreme form of social, ethnic religious extremism and nothing can prevent it from achieving its goals. This criminal phenomenon tends to grow steady everywhere in the world. Criminalists observe that year by year terrorist attacks are becoming more thoroughly organized actions that employ super modern technologies, weapons and means of communication. This kind of activity is now preferable for extremists to solve social, ethnic, religious and other conflicts.

Risk Management in Natural and Societal Systems

9

Modern industrial infrastructure in developed countries, especially mega-lopolises, comprising thousands radioactive, chemical, biologic objects, offers terrorists a real opportunity to inflict damage without resorting to weapons of mass destruction, though their efforts to get hold of it are evident. The nature of a hazard (or an opportunity to inflict damage) is connected with energy, substances or information flows that are inadequate for the infrastructure as an open system. Besides every infrastructure is a compilation of various components that have a common purposes, common condition of functioning and common resources. System approach to studying any threat implies primarily as complete as possible knowledge of adversary (his objectives, tasks financial and professional potential, materials, equipment, weapon and many other characteristics). Therefore potential targets for terrorists should be systematized according to their accessibility and possible damage in the case of destruction. These are basic data for organizing counteraction. The modern terrorism can be divided into three kinds: traditional, technological and intellectual (fig. 1) [1] that differ in the character of damage distribution in terms of time (initial, secondary and cascade damages). Analysis of various kinds of terrorist impacts on natural-manmade system shows that maximal damage corresponds to the secondary damage with cascade-synergetic effect being manifested. Examination of these facts leads us to the conclusion that technological and intellectual kinds of terrorism can be classified as systemic terrorism. Arising and development of initial, secondary and cascade factors of destruction in terrorism are practically governed by the same laws that govern traditional accidents and catastrophes in complex technological systems causing manmade emergencies. In the view of the above development of methods, means and systems for protection from threats of systemic terrorism comes to two basic tasks: (a) risk reduction or prevention of initiating hazards, threats, and challenges; (b) reduction of risks of further development of natural, manmade and societal emergencies provided that initiating terrorist impacts take place. The theory of risks and methods to risk analysis are used in order to assess natural, manmade and societal crises and catastrophes [2,3]. They can be applied to assess risks of terrorist impacts as well. It is necessary to take into account systemic characteristics of natural-manmade-social systems. Such approach allows to study cascade-synergetic process, to reveal week elements of a system and to undertake measures for protection against terrorist attacks, which leads to more efficient decisions on primary protection of key assets of infrastructure. which leads to more efficient decisions on primary protection of key assets of infrastructure. c

R. Akhmetkhanov

10

Systemic risks are determined by peculiar interactions of natural, manmade and social spheres. A catastrophe or crisis is a chain of sequential interconnected events. Te number of links in the chain can be fairly big. Analyzing systemic risks in natural-manmade-social systems, the probability that a systemic threat is realized can be presented as a functional [1]: Psis=Fps{Pn, Pm, Ps}, Where Pn– probability of unfavorable events in natural environment; Pm – probability of unfavorable events in manmade (engineered) environment Ps – probability of unfavorable events in social sphere; Probability Pm is considerably dependant on the level of protection of manmade facilities of military or civil designation from accidents and catastrophes. This protection depends on the extent of degradation of facilities at given stage maintenance, and the level of diagnostics and monitoring which means that Pm and Ps are directly related. Probability Ps is known to be depended on occurrence of natural disasters (Pn) as well as on the state of manmade facilities (on Pm). Damage Usis caused by realized system threat can generally be presented as a functional: Usis=Fus{Un, Uwm, Us} were Un is damage inflicted on natural environment Um is damage inflicted on manmade (engineered) environment Us is damage inflicted on social sphere (primarily on population) when a systematic threat is realized and initial and secondary destructive factors interact; Values Un, Um, Us can be measured both in terms of physical items (for example, a number of casualties, a number of building destroyed, area of contaminated territory) and in equivalents (for example, pecuniary loss).

Risk Management in Natural and Societal Systems

11

To analyze and govern risks Rsis with respect to two groups of 3 components one can use limit state surfaces: x Systemic risks to social environment (Rs), to manmade environment Rm), to natural environment (Rn) x Integrated damaging factors of crises and catastrophes: energy (E), substance (S), information (I). Then the state of utmost danger for Rn , Rm, Rs or E, S, I will be at the intersection between the vector of the current state of a system in risk Rsis or threats Dsis and the surface of utmost danger state. The state of a system is a function of parameters X=(Xloc, Xsis). Vector X consists of 2 sets of parameters that specify the elements of a system (local parameters) and the links between the elements (systemic para-meters). These parameters can be variable or stationary. They determine the level of risk in a system: (Xloc, Xsis) Ÿ R. The described analysis of systematic concepts of risks and losses in natural-manmade-social systems shows that in order estimate risk in natural-manmade-social systems it is necessary to proceed not only from the probability of occurring of a crisis situation, but also from a degree of vulnerability of its elements, allowing for synergetic cumulative effects. In this case potential complex damage caused by emergency to natural-manmadesocial system should be described by a matrix of losses in subsystems and elements of the system. This matrix is to allow for direct losses i.e. levels of destruction, infringement, radioactive and chemical pollution, negative aftermaths of damaging effects on natural and economic objects (land, people, flora and fauna, buildings, equipment, goods, raw materials, plantations, live stock and the like) as well as indirect losses inflicted by the said distractions and infringement on the state and functioning of other objects of nature and economy that did not suffer directly from the damaging factors. A system can be divided into different number of components, the degree of detailing depending on the level of emergency situation danger. A system is considered as a complex of 3 subsystems at any level of risk described: natural, manmade and social. A natural-manmade-social system of the highest level of danger is considered to include lower level of danger. Each subsystem is divided into subsystems of the next lower levels. Thus hierarchic (multilevel) presentation of a system is built up. When considering a system consisting of subsystems, the matrix of losses is presented as a matrix containing both diagonal blocks (units) and non-diagonal blocks. Diagonal elements of a loss matrix specify potential losses at the given element of the system in the case that an emergency occurs at this element. Non-diagonal elements of a loss matrix characterizing the linkage of the system’s elements with regard to criterion for loss, describe synergetic development of an emergency and its distribution onto the system.

R. Akhmetkhanov

12

This kind of matrix is build up through assessment of maximal potential losses to the elements (subsystems) of a system. All kinds of losses are taken into account. Probability of an emergency occurrence in conformity with such systemic approach is characterized by the matrix of probability of the emergency occurrence that contains probable estimations of emergency effects on the elements of the system, according to the scenario of the emergency development. Risk dependence on the level of protection of the system’s elements, on their location relating to the zone emergency occurrence is estimated by matrix of the system vulnerability in case of emergency. (This matrix contains characteristics of the system’s elements vulnerability in case of the given emergency). Then the risk for naturalmanmade-social system from a specific emergency can be presented as the matrix of the risk: Ri

R UT ( X loc ) ˜ U ( X

sis

)˜P

were RUT ( X loc ) is the matrix that contains vulnerability factors of the system’s elements, these factors depending on the parameters of the system’s elements X loc ; U ( X sis ) ˜ P - is the loss matrix, consisting of values of maximal losses and this matrix depends on the system’s parameters X sis ; P - is the matrix of emergency’s probability and effects on the system elements. Coefficients of the matrix RUT ( X loc ) are variable and vary according to the character of local governing impacts. The matrix of losses depends on the structural properties of the system, interactions of the systems elements and on systemic parameters X sis . The variations of the elements values of the matrix require systematic changes. Thus the risk government in natural-manmade-social systems implies local risk government through reducing vulnerability coefficients of the system’s elements, and if possible, of the probability of occurrence of a specific emergency, and of its effects on the system’s elements. The decisions on risk management can be made both on local and global (systemic) levels. In case of mature naturalmanmade-social systems with strong linkage of elements of natural, manmade, social spheres global decision require high expanse. Comprehensive risk that comprises all kinds of risks of a system can be presented as: n

Rs

R s*  R

f

¦

Rj  Rf,

j

1,... n

j

where R f is background risk including risks from those emergencies that cause insignificant loss and tend to occur frequently. The presentation of a risk as a matrix allows to take account of synergic cumulative and self organizing properties of a system for the whole complex

Risk Management in Natural and Societal Systems

13

of emergencies typical for the given natural-manmade-social system. The structure of the given matrix depends on the kind of considered risks. The presence of nonzero non-diagonal elements (blocks) in a matrix RS* characterizes the linkage of the system’s elements and corresponds with the principle of reciprocity. Presentation of a risk as a matrix allows to present the scenario of an emergency by means of its elements. For example, presentation * R11 o R12 o R22 o R21 o R22 . In this case an accident occurred in the first element of the system, affects its second element where it causes an emergency that causes a secondary accident in the first element. Here is a graph of emergency development in a natural-social system (fig. 2). Making up a risk matrix for natural-manmade-social system it is necessary to take into account not only accidents possible in given system but also accidents in surrounding natural-manmade-social systems, especially in those having a common boundary with the given system (transboundary transfers, i.e. external impacts on a system). Scalar characteristic of a system’s risk can be presented as a potential function of risk, considered as a function of the system’s parameters and elements (taking into account the matrix of the risk). This is a nonlinear dependence, that can be presented as a surface in the configurative space of system Xsis parameters and system Xloc elements that determine the matrix of the system’s losses and the matrix of its vulnerability as well as the probability of the emergency occurrence. Hypersurface in n-dimension space can have local peculiarities (characteristic points of elliptic, hyperbolic and parabolic type).These points and their distribution in the parameters space determine the system’s peculiarities. In the theory of catastrophes they are called critical points. The strategy of risk government in a system depends on how close the system is to the critical point. Presentation of risk as scalar value or as a matrix of risks could be classified as static assessment of the system’s risk. The structure of the risk matrix in this case characterizes the organization (self-organization) of the system at the moment of its evaluation. The relation between controllable parameters and uncontrollable ones can be defined trough relation between vulnerability matrix and emergency probability matrix. In order to take into account processes of the system’s self organisation it is necessary to consider dynamic risks.

R. Akhmetkhanov

14

Figure 2. Scenario of emergency development in a natural-manmade-social system. Subscripts: N- natural, M –manmade, S- social Risk is a function of the system’s state and its variable parameters. Its total differential could be presented as:

dR i

wR wR wR wR )dX 1  ...  ( )dX i 1  ...  ( )dX n  dt , wX 1 wX i 1 wX n wt 1,2,..., n (

Component

wR dt wt is to be included since risk R may depend on time as well. The differential components

wR wX i

Risk Management in Natural and Societal Systems

15

determine the dependence of a risk on the change of the system’s parameters (local and systemic ones). The values of partial derivatives on the system’s parameters allow to determine the direction of the controllable movement of the system to a minimal risk.

Figure 3. a, b, ɛ – points of local maximums and minimums of R values If we consider a system, whose parameters vary in terms of time, we will have the dependence of a risk on time R(t). Such presentation of risk depending on the system’s parameters an on time allows to make dynamic models of risk; it is the basis of risk management in natural-manmadesocial systems with reference of time parameter. It also allows to take account of the system’s self-organization and non-linear effects. In this case mathematical apparatus of the theory of catastrophes and nonlinear dynamics can be used. Potential risk surface for a system consisting of n elements will have a compound form were all types of critical points present. For example, in Fig.3 there are potential surfaces of risk in a system consisting of two elements with nonlinear links (X1 and X2 – are generalized parameters of the system). The theory of catastrophes, stability of motion, variational principles etc. are based on analyzing properties of potential functions.

R. Akhmetkhanov

16

The character of measures on ensuring safety in a system is determined by the way the critical points are spread relative to the point of the system state and by the change the surface around this point. Analysis for critical points and the way they are spread in space is necessary to carry out global risk minimization. Let us consider dynamic singularities of natural-manmade-social system under certain conditions of the interaction between subsystems. The changes of risk in a system can be characterized by the following Rossler’s equations:

R 1 R

( R2  R3 ),

R 3

a ( t )  R1 ˜ R3  c( t ) ˜ R3 ,

2

R1  a ( t ) ˜ R2 ,

R ( 0)

R0i ,

i=1,2,3

were R1, R2, R3 are risks in natural, manmade and social spheres, a(t)- is a coefficient that allows for development of development of the manmade sphere and its vulnerability, c(t) -is the similar coefficient for social sphere. These coefficients also depend on the possibility of terrorist effect on social and manmade sphere. Systemic risk is determined by the sum Rsis=R1+R2+R3. The presented model characterizes the variation of the total risk in natural-manmade-social system in relation to the trend value that increases in the system and can exceed the acceptable level. This model illustrates combined interactions between subsystems and conditions under which self-organization is triggered in the system. The model contains only three degrees of freedom, but illustrates wide variety of dynamic singularities of a system. The absence of R1 with the corresponding coefficient in the right-side of equation 1 comply with the condition that there are no noticeable changes in the natural sphere during the given period. Let us consider the condition of a system when the values of coefficients a(t) and c(t) that determine the level of the system’s development are constant. The analysis of properties of a potential function shows that there are two characteristic points R(a, 0, 0) and R(-a+2c, o, 2). The location of these points in relation to each other and their form determine the character of geodesic lines on the surface of the potential function and, consequently the system’s dynamics. In the given case these two points are of parabolic type (Gaussian curvature of the surface in characteristic points is a zero one) with two coordinates with indifferent stability. The third coordinate for the first point is not stable, but the one for the second point – is stable. Location

Risk Management in Natural and Societal Systems

17

of these points and their character determine the trajectories of the system’s motion and phase space of the system. The zero point (Ri(0)=0) was taken as an initial point to start computation. With certain values of coefficients a and c the system becomes selforganized in to a system that has a steady cycle of risk changing relating to some average value.

Figure 4. c=0.4; a=0.2 Let us assess system’s behavior by means of divergence of phase space.

D

wR 1 wR 2 wR 3   wR1 wR2 wR3

a ( t )  R1  c( t )

Having denoted the volume of phase space as G(t) we write the equation as:

G (t )

G (0)e ( a (t )  R1  c ( t ))t

The expression obtained shows the character of the natural sphere’s effect on the risk value in the whole system. The value of the change of natural risk R1 in this model is characterized as both positive and negative values, therefore phase volume has oscillatory mode. This is expressed as multifrequency interactions. In the case of a(t)+R1-c(t)>0 the systemic risk is

R. Akhmetkhanov

18

growing while in the case of a(t)+R1-c(t)