Decline and Recovery of the Island Fox: A Case Study for Population Recovery (Ecology, Biodiversity and Conservation)

This page intentionally left blank

Decline and Recovery of the Island Fox A Case Study for Population Recovery Native only to the California Channel Islands, the island fox is the smallest canid in North America. Populations on four of the islands were threatened to extinction in the 1990s due to human-mediated predation and disease. This is the first account of the natural history and ecology of the island fox, illustrating both the vulnerability of island ecosystems and the efficacy of cooperative conservation measures. It explains in detail the intense conservation actions required to recover fox populations, such as captive breeding and reintroduction, and large-scale ecosystem manipulation. These actions were successful due in large part to extraordinary collaboration among the scientists, managers, and public advocates involved in the recovery effort. The book also examines the role of some aspects of island fox biology, characteristic of the ‘island syndrome’, in facilitating their recovery, including high productivity and an apparent adaptation to periodic genetic bottlenecks. Timothy J. Coonan is a US National Park Service Biologist at Channel Islands National Park, California. He has led the park’s island fox recovery program since 1999. Catherin A. Schwemm is an Ecologist for the US National Park Service in Flagstaff, Arizona. In 1993 she began the island-fox monitoring program for Channel Islands National Park, California, and she has been a member of the Island Fox Conservation Working Group since 1999. David K. Garcelon is President and Founder of the non-profit Institute for Wildlife Studies in Arcata, California. His research activities have included long-term work on population demography, captive breeding, movement patterns, disease exposure, and behavior.

E C O L O G Y, B I O D I V E R S I T Y A N D C O N S E RVAT I O N

Series Editors Michael Usher University of Stirling, and formerly Scottish Natural Heritage, UK Denis Saunders Formerly CSIRO Division of Sustainable Ecosystems, Canberra, Australia Robert Peet University of North Carolina, Chapel Hill, USA Andrew Dobson Princeton University, USA Editorial Board Paul Adam University of New South Wales, Australia H.J.B. Birks University of Bergen, Norway Lena Gustafsson Swedish University of Agricultural Science, Sweden Jeff McNeely International Union for the Conservation of Nature, Switzerland R.T. Paine University of Washington, USA David Richardson University of Cape Town, South Africa Jeremy Wilson Royal Society for the Protection of Birds, UK The world’s biological diversity faces unprecedented threats. The urgent challenge facing the concerned biologist is to understand ecological processes well enough to maintain their functioning in the face of the pressures resulting from human population growth. Those concerned with the conservation of biodiversity and with restoration also need to be acquainted with the political, social, historical, economic, and legal frameworks within which ecological and conservation practice must be developed. The new Ecology, Biodiversity, and Conservation series will present balanced, comprehensive, up-to-date, and critical reviews of selected topics within the sciences of ecology and conservation biology, both botanical and zoological, and both ‘pure’ and ‘applied’. It is aimed at advanced final-year undergraduates, graduate students, researchers, and university teachers, as well as ecologists and conservationists in industry, government, and the voluntary sectors. The series encompasses a wide range of approaches and scales (spatial, temporal, and taxonomic), including quantitative, theoretical, population, community, ecosystem, landscape, historical, experimental, behavioural, and evolutionary studies. The emphasis is on science related to the real world of plants and animals rather than on purely theoretical abstractions and mathematical models. Books in this series will, wherever possible, consider issues from a broad perspective. Some books will challenge existing paradigms and present new ecological concepts, empirical or theoretical models, and testable hypotheses. Other books will explore new approaches and present syntheses on topics of ecological importance. The Ecology of Phytoplankton C.S. Reynolds Invertebrate Conservation and Agricultural Ecosystems T.R. New Risks and Decisions for Conservation and Environmental Management Mark Burgman

Nonequilibrium Ecology Klaus Rohde Ecology of Populations Esa Ranta, Veijo Kaitala and Per Lundberg Ecology and Control of Introduced Plants Judith H. Myers, Dawn Bazely Systematic Conservation Planning Chris Margules, Sahotra Sarkar Assessing the Conservation Value of Fresh Waters: An international perspective Phil Boon, Cathy Pringle Bird Conservation and Agriculture Jeremy D. Wilson, Andrew D. Evans, Philip V. Grice Large Scale Landscape Experiments: Lessons from Tumut David B. Lindenmayer Insect Species Conservation T.R. New Cave Biology: Life in darkness Aldemaro Romero Biodiversity in Environmental Assessment: Enhancing ecosystem services for human well-being Roel Slootweg, Asha Rajvanshi, Vinod B. Mathur, Arend Kolhoff Mapping Species Distributions Janet Franklin

Decline and Recovery of the Island Fox A Case Study for Population Recovery timothy j. coonan National Park Service, Channel Islands National Park, California

catherin a. schwemm National Park Service, Flagstaff, Arizona

david k. garcelon Institute for Wildlife Studies, Arcata, California

with

linda munson and cheri asa

CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Dubai, Tokyo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521887113

This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2010 ISBN-13

978-0-511-90192-8

eBook (NetLibrary)

ISBN-13

978-0-521-88711-3

Hardback

ISBN-13

978-0-521-71510-2

Paperback

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

To Tom and Kay Coonan, who first introduced me to the natural world, and to my daughters, Bridget and Carrie (TJC). To my parents, John and Nancy Schwemm, and my daughters, Annie, Bridget and Carrie (CAS). For Devra Kleiman, with thanks for her many contributions to wildlife conservation and the island fox program. And for the lesson that it is sometimes from those with whom we disagree that we learn the most. And we dedicate this book to Linda Munson, with immense appreciation for her perspective, humor, and innumerable contributions to island fox conservation. Linda was always engaged, and provided the common sense (and jokes!) needed during difficult discussions. Linda is irreplaceable, and the island fox working group will not be the same without her. We are especially sorry that she did not live to see this book published; we think that she would have been very pleased. Goodbye, good friend.

Contents

Foreword Acknowledgments 1 Introduction

page xii xiv 1

2 Evolution and genetics 2.1 Description 2.2 Dwarfism and the island syndrome 2.3 Evolution of the island fox 2.4 Island foxes in the paleoenvironment 2.5 The peculiar genetic status of the island fox 2.6 Genetic management of captive island foxes 2.7 Future of wild populations

5 5 7 8 12 13 15 20

3 Social structure, reproduction, mortality and survivorship, and population dynamics 3.1 Social organization and reproductive behavior 3.2 Mortality and survivorship 3.3 Population abundance and dynamics 3.4 Factors affecting population dynamics 3.5 Modeling island fox population dynamics 3.6 Summary

21 22 24 27 30 32 33

4 Food habits, habitat use, activity patterns, and dispersal 4.1 Food habits 4.2 Dispersal and travel 4.3 Home range 4.4 Habitat use 4.5 Activity patterns 4.6 Summary

34 34 38 39 40 41 42

x · Contents 5 Golden eagles and the decline on the northern islands 5.1 Detecting change 5.2 Determining the cause 5.3 Golden eagle colonization of the northern islands 5.4 The vulnerability of island foxes to diurnal aerial predators

43 43 47 50 56

6 Ecosystem recovery: Predators and prey on the northern Channel Islands 6.1 Golden eagle removal 6.2 Long-term ecosystem recovery actions 6.3 Summary

58 58 66 71

7 Disease and decline on Santa Catalina Island 7.1 Declining populations 7.2 Initial results 7.3 Was disease the cause of the decline? 7.4 Recommendations for population recovery 7.5 Testing CDV vaccine

73 73 75 75 76 77

8 Recovery actions: Captive breeding of island foxes 8.1 Captive breeding efforts on the northern islands 8.2 Methodology and techniques 8.3 Demographic and genetic objectives of captive breeding 8.4 Low reproductive success 8.5 Summary

81 81 82

9 Recovery actions: Reintroduction and translocation 9.1 Translocation on Santa Catalina 9.2 Reintroduction on Santa Catalina 9.3 Reintroduction program on the northern islands 9.4 Success of reintroduction 9.5 Summary 10 Reproductive biology, by Cheryl Asa 10.1 Reproductive cycles 10.2 Captive breeding 10.3 Results of the monitoring study 10.4 Summary

92 93 98 100 100 102 103 111 113 115 116 117 123 127

Contents

· xi

11 Diseases of island foxes, by Linda Munson 11.1 Disease in island populations 11.2 Could viral disease explain the population declines? 11.3 Do other viruses infect island foxes? 11.4 Do non-viral pathogens infect island foxes? 11.5 Parasites infecting island foxes 11.6 Non-infectious diseases in island foxes 11.7 Genetic diversity versus disease resistance 11.8 Disease as a cause of death 11.9 Overall health of the island fox populations

129 130

12 Zoos, education, and public participation 12.1 Public advocacy 12.2 The role of zoos 12.3 Tachi and Finnegan 12.4 Environmental education 12.5 Summary

144 144 146 149 152 152

13 Managing recovery: Cooperative conservation, politics, and the Endangered Species Act 13.1 Stakeholders and recovery 13.2 A model for management: the Island Fox Conservation Working Group 13.3 Listing the island fox as endangered 13.4 Changes in island fox management due to listing 13.5 The benefits and challenges of ESA listing 13.6 Conclusion 14 The 14.1 14.2 14.3 14.4 14.5 14.6

ecological role of island foxes Introduction Background Prey response Competitors Island communities without foxes Implications for research and management

131 133 134 134 138 140 141 142

154 154 156 158 159 163 165 167 167 167 169 174 176 177

15 Conclusion 15.1 Status 15.2 Future

180 180 181

References Index

185 207

Foreword

Ninety-nine percent of all the species that have ever existed on Earth are extinct, including all species of humans except our own. Undoubtedly, as humans evolved from four legs to two, from forest to the plains, and from hunter-gatherer to agrarian society, we, and our forebears, contributed to some of these extinctions. In modern times, we continue to contribute to the extinction of species, and at an alarming rate. What makes our behavior different than in prehistoric times is that humans now recognize the impacts of our actions on species’ persistence, and we can take action, if we so choose, to reverse species loss. Sometimes our response is too little too late, as with the passenger pigeon that once blotted the sky with its fecundity but today resides only in the dusty drawers of museum cabinets. But at other times humans are successful at saving species, and in those instances we should pause to celebrate our achievements, and to consider the ingredients that bring a species back from the brink of extinction. Such is the story of the Channel Islands fox. Off the California coast lies a small chain of islands, distant enough to make movement of animals from the mainland rare and difficult, and isolated for long enough that those species that did reach the islands have adapted and evolved, like Darwin’s finches. Indeed the Channel Islands have been called the American Galapagos, once home to pigmy mammoths and later the hunting grounds of the Chumash people. Today the islands are home to many indigenous animals and plants that have adapted to the wind-swept slopes and the dry Mediterranean climate. Among this unique biota is the island fox, an animal the size of a house cat and cuter than a golden marmoset (in my opinion). Island foxes are also brazenly unafraid of people, because until humans arrived they were the top predator on the islands. A hundred years of cattle and sheep grazing, and the introduction of non-native pigs, elk, deer, and turkeys attracted golden eagles to the islands where they found foxes to be easy prey. Island foxes had never confronted death from the sky, and thus did not hide from predators like their gray fox

Foreword

· xiii

cousins do on the mainland. When the populations were decimated and near extinction, the National Park Service, The Nature Conservancy and the Catalina Island Conservancy took action that was dramatic, heroic and, at times, controversial. Today the island fox is back, breeding in the wild and hunting the grassy slopes for grasshoppers and berries without danger from the skies, and protected from canine disease. This book is the story of the foxes, but also the story of a dedicated group of people that I came to know, respect, and trust while I served as the Director of the National Park Service’s Pacific West Region, which includes Channel Islands National Park. I claim no credit for their success, other than ensuring that they received the funding and occasionally the ‘top cover’ they needed to get the job done. But I can celebrate their success, and be inspired by their dogged dedication and their commitment to science. I can also hope that their actions inspire others. The elimination of introduced species to the islands and the recovery of island foxes sets the stage for a full restoration of the island ecosystem, giving us hope for other islands, other special places and for our fragile, unique, and beautiful planet. Jonathan B. Jarvis Director, National Park Service Conway (2007): ‘There is little reason to expect that any terrestrial wild animal which competes with humans in some way, real or invented, or is larger than a 2-gallon [c.9 liters] bucket, will survive for long without increased protection and knowledgeable care.’

Acknowledgments

The sustained commitment by the National Park Service (NPS) to island fox recovery was due to the efforts of Peter Dratch, Kate Faulkner, Russell Galipeau, and Dave Graber, and also to Gary Davis, Jay Goldsmith, Terry Hofstra, Jon Jarvis, Patty Neubacher, and Jim Shevock. We appreciate and recognize the work of the many people who have studied and monitored island foxes and whose efforts have resulted in population recovery. We especially thank Mitch Dennis, Jodi Fox, Keith Rutz, and Greg Schmidt for their efforts in the early stages of the recovery program, and William Barker, Lilly Cesh, Douglass Cooper, Susan Cooper, Susan Coppelli, Angela Guglielmino, Don Jones, Kate McCurdy, Chris Newman, Stephanie Provinsky, Mike Puzzo, John Sewell, and Justin Sloan. Veterinarian care and expertise was provided by Deana Clifford, Steve Timm, Nancy Thomas, and Mark Willett, and especially Winston Vickers and Karl Hill. The veterinarian group strongly supported island fox conservation by providing compassionate care for captive animals while pursuing research that continues to expand our understanding of island fox biology. Gary Roemer first identified the connection between golden eagles and island fox population declines, and contributed critical expertise during the early years of the recovery program. Island fox monitoring and/or care of captive foxes was conducted by many, including: Greg Austin, Pedro Chavarria, Matt Christianson, Chris Couch, Geoff Cline, Melanie Coulter, Heidi Crowell, Lisa Drake, Julie Goldzman, Sara Hansen, Patricia Hartman, Jim Howard, Heather Johnson, Scott Johnson, Stacey Lehman, Andrea Lehotsky, Lindsay Martin, Patrick Myers, Carol Powell, Kara Randall, Jen Savage, Eric Scott, Tessa Smith, and Debbie Watson. Catherin Schwemm would like to especially thank Charles Drost, John Orrock and Jim Reichman for continued support. Cheryl Asa (author of Chapter 10) and Linda Munson (author of Chapter 11) particularly thank Joan Bauman, Karen Bauman, Elizabeth Buckles, Walter Boyce,

Acknowledgments

· xv

Betsy Calkins, Ed Dubovi, Carol Feiseler, Denise Imai, Steve Nadler, Sharon Patton, Marsha Sovada, Brian Stacy, and Karen Terio. We thank all of the following for their strong support of and participation in island fox recovery: Eric Aschehoug, Lynne Lozier, Scott Morrison, Lotus Vermeer, and Rachel Wolstenholme of The Nature Conservancy; Carlos de la Rosa, Calvin Duncan, and Julie King of the Catalina Island Conservancy; Kelly Brock, Jan Larson, and Grace Smith of the US Navy; Lyann Comrack, Benjamin Gonzales, and Nancy Frost of the California Department of Fish and Game; and Bridget Fahey, Eric Morrissette, and Sandy Vissman of the US Fish and Wildlife Service. We thank Brian Latta and the late Brian Walton of the Predatory Bird Research Group for leading a very effective golden eagle translocation effort. We acknowledge the work of everyone who participated in the Island Fox Conservation Working Group, including: Bill Andelt, Vicki Bakker, Karen Bauman, Rich Block, Dan Blumstein, Kathy Carlstead, Rex Cohn, Paul Collins, Bob DeLong, Natalie Gates, Melissa Gray, Brian Hudgens, Julie King, Lyndal Laughrin, Colleen Lynch, Jonna Mazet, Phil Miller, Seth Riley, Gary Roemer, Esther Rubin, Ray Sauvajot, Peter Sharpe, Peter Siminski, Grace Smith, Marsha Sovada, Nancy Thomas, Alan Varsik, and Bob Wayne. The members of the Recovery Coordination Group were: Brian Cypher, Carlos de la Rosa, Dave Graber, Devra Kleiman, Peter Schuyler, Rebecca Shaw, Dale Steele, and Rosie Woodroffe. Many people provided critical professional expertise that improved management and recovery planning, including Vicki Bakker, Bryan Cypher, Dan Doak, Murray Efford, Melissa Gray, Brian Hudgens, Steve Kohlmann, Colleen Lynch, Kathy Ralls, and Esther Rubin. Keri Dearborn, Pat Meyer, and Alexandra Morris shared the island fox story with the public through their volunteer education efforts. Rich Block generously shared the time and expertise of the Santa Barbara Zoo staff – including Karl Hill, Wendy Stanford, and Alan Varsik – and made the island fox story a prominent theme at the zoo. Island fox recovery on the National Park islands would not have been accomplished without the support of the entire Channel Islands National Park staff, especially: Rhonda Brooks, Kent Bullard, Doretta Burgess, Sarah Chaney, Denise Domian, Jack Fitzgerald, Angela Harris, Tim Jones, David Kushner, Cindy Leon, Yvonne Menard, Kelly Moore, Yvonne Morales, Dan Richards, Dirk Rodriguez, Fred Rodriguez, Mark Senning, Ed Smith, Joshua Sprague, Brett West, and Earl Whetsell, and

xvi · Acknowledgments boat captains Diane Brooks, Keith Duran, Dwight Willey, and Brent Wilson. Within the park there was no one more earnest about returning foxes to the San Miguel Island landscape than the island ranger, Ian Williams, who has supported island fox management for nearly 20 years. Channel Islands Aviation and Aspen Helicopters provided critical logistical support throughout the project. Chuck Graham, Matthew Hill, Brian Latta, Dan Richards, and the Catalina Island Conservancy generously provided images for this book. Finally, we thank Paul Collins, Keri Dearborn, Charles Drost, Jon Erlandson, Dave Graber, Brian Hudgens, Julie King, Pat Meyer, Susan Morris, John Orrock, Jim Reichman, Torben Rick, and Alan Varsik for reviewing draft chapters of this book.

1

r

Introduction

To the hiker it was a subtle change to the landscape, but nearly overwhelming in its implications. On the trail were small fox tracks, simple footprints made casually as the animal walked, probably quickly but not necessarily in a hurry. Animals often use human trails to conserve energy while traveling, so the observation of the prints was not in itself surprising. It was the simple existence of the tracks that made the hiker stop, look around, and smile inwardly at the thought; island foxes were again roaming the wilderness of San Miguel Island. Island foxes are a unique species of small carnivore that live only on the Channel Islands located off the coast of Southern California in the United States of America. The California Channel Islands consist of eight islands, four in the northern group and four in the southern group. Of the eight, three in the north – Santa Cruz, Santa Rosa, and San Miguel – and three in the south – San Nicolas, San Clemente, and Santa Catalina – support populations of island foxes, and on each island a unique subspecies has evolved: r r r r r r

San Miguel Island fox: Urocyon littoralis littoralis; Santa Rosa Island fox: Urocyon littoralis santarosae; Santa Cruz Island fox: Urocyon littoralis santacruzae; San Nicolas Island fox: Urocyon littoralis dickeyi; Santa Catalina Island fox: Urocyon littoralis catalinae; San Clemente Island fox: Urocyon littoralis clementae.

San Miguel and Santa Rosa Islands are managed by the National Park Service (NPS), Santa Cruz is co-managed by the National Park Service and The Nature Conservancy, Santa Catalina is managed by the Catalina Island Conservancy, and San Nicolas and San Clemente are managed by the US Navy (Fig. 1.1). As recently as 1990, populations of island foxes were considered stable. Fox populations on each island were naturally small and varied in relation to island size, but reproduction and survival were high, and observed

2 · Introduction

Figure 1.1 The eight California Channel Islands. A separate island fox subspecies occurs on each of the six largest islands (thus, on every island but Anacapa and Santa Barbara).

densities greater than for almost any other North American carnivore. Management efforts were limited to annual population monitoring, and research was directed at basic ecological and biological questions such as food habits, reproductive behavior and social structure. The insular habitats that support island foxes were assumed to be nearly as protected from human impacts as is possible on Earth today. Then, within a second of evolutionary time, the status of four of the six island fox populations went from stable to perilous. Annual survival rates dropped from over 80% to less than 40%, and managers and scientists struggled to determine the causes of fox deaths at the same time as they worked to protect those remaining. Population declines were so rapid that within four years of the initial observations of downward trends there were 15 remaining wild foxes on each of the two smaller islands, and on the larger islands the wild populations had declined to 10–20% of historic levels. A combination of directed research, focused management, and the collaborative efforts of many people determined that there were two primary but unrelated factors causing island fox mortalities: predation by non-native golden eagles on Santa Cruz, Santa Rosa, and San Miguel

Introduction

· 3

Islands, and canine distemper virus on Santa Catalina Island. Although ecologically unrelated, both impacts were the result of human activities, which brought novel threats to historically isolated populations. The obstacles to rapid implementation of emergency recovery actions included lack of funding, gaps in the knowledge of island fox biology and the nature of the threats, and the fact that the species was not officially recognized as Endangered. The future of all four populations depended on the coincident success of two primary recovery strategies. First, the remaining foxes had to be protected from disease and predation and the populations grown. Captive breeding programs for endangered mammals generally require years to develop, and under the best of circumstances are biologically challenging and expensive. Island fox managers had to facilitate breeding quickly before the remaining adults were unable to breed, while addressing a myriad of genetic and disease issues that required a separate captive program on each island. Second, the population-level threats had to be eliminated; unless fox populations were ultimately free from predation and disease, even the most successful captive breeding program would not produce enough animals to support self-sustaining wild populations. In response, managers and scientists embarked on three unprecedented efforts. Island foxes had never been vaccinated against any disease, and on Santa Catalina veterinarians worked for two years to test and ultimately administer a canine distemper vaccine that now protects all island fox populations. On the northern islands a golden eagle removal program was established to relocate all of the eagles from the islands. This effort proved to be immensely challenging and at times controversial, but ultimately over 40 birds were trapped and removed. Finally, discouraging future golden eagle utilization of the islands required a reversal of the ecological changes that had led to golden eagle colonization. The separate tasks of feral pig (prey) removal and bald eagle (competitor) reintroduction were undertaken to tip the ecological balance in favor of long-term persistence of island fox populations and resilient island ecosystems. Cumulatively these efforts were successful, and island foxes are now recovered or close to recovery on all of the Channel Islands. Feral pigs are gone and reintroduced bald eagles have started breeding, and vaccination and monitoring programs will hopefully protect island foxes from future disease outbreaks. Island fox recovery efforts included large-scale ecosystem manipulation, animal translocations, captive breeding and disease prevention, and required the expertise and commitment of hundreds of professionals and volunteers over nearly a decade.

4 · Introduction The purpose of this book is threefold. The island fox is a unique and fascinating creature, and we present here the first comprehensive description of the evolution, biology, genetics, behavior, and ecology of this species. The presentation of this information poses some challenges, as much of what is now known about island fox biology was attained as a direct result of recovery efforts. Specifically, we present the chapters on disease and reproductive biology after the chapters that describe applied recovery actions because most of this information was gained through the research efforts of L. Munson and C. Asa and their colleagues in response to specific recovery needs. Second, we summarize the chronology of the population declines and the resulting recovery actions, including initial research efforts, the contribution of long-term monitoring, and the veterinary, husbandry, and education efforts, all of which were critical to success in island fox recovery. Finally, we place the story of island foxes into a larger conservation context and attempt to answer several questions: How did a species that lives only on protected islands come to be in danger of imminent extinction? What combination of factors most facilitated island fox recovery? And what has been learned from the island fox experience that can assist in future ecosystem recovery efforts, preserve the unique ecosystems of the California Channel Islands, and prevent future extinctions? Cumulatively we hope this book will help define the daunting challenges of conserving biodiversity in an increasingly interconnected world, and the importance and urgency of protecting remaining species and natural systems. As the hiker walked on there were more tracks, but eventually the evidence of the fox’s presence disappeared from the trail. At some point it had moved off to the grasslands or down into the canyon, or was perhaps resting under a nearby shrub. But wherever it was, the hiker knew, was exactly where it was supposed to be.

2

r

Evolution and genetics

The island fox is a recently and rapidly evolved species that is smaller than its mainland ancestor, the gray fox (Urocyon cinereoargenteus). Examination of the evolution and current genetic status of island foxes illustrates the vagaries of dispersal to islands, the influences on evolution in island settings, and the challenges of conservation for small, genetically depauperate populations. Although the fossil history is scant, the archeological record indicates a species with a close relationship to Native Americans who transported foxes to the southern islands and may be responsible for their introduction to all the Channel Islands. Extensive morphologic and genetic studies – in conjunction with the islands’ geologic history (sea level rise and island separation) – have established a pattern of island colonization and the genetic and physical basis for description of the six island fox subspecies. Island fox populations have relatively little genetic variability, due to naturally small population sizes and historic fluctuations that have resulted in population bottlenecks. The precarious genetic status of island foxes has consequences for both the management of captive populations and long-term population viability.

2.1 Description The island fox (Fig. 2.1) was first described as Vulpes littoralis by Spencer F. Baird in 1857 from the type locality on San Miguel Island (Baird 1857). Baird was the assistant secretary of the Smithsonian Institution, and examined zoological specimens brought back by the US Army’s topographical engineers who had been tasked with surveying a route for a transcontinental railroad. Baird’s report also included descriptions of specimens collected from San Miguel Island and other Pacific localities by a Lieutenant William P. Trowbridge, who was conducting tidal observations at San Miguel Island in January 1856 for the US Coast and Geodetic Survey. Baird was the first of many to compare the size of the diminutive island fox to a house cat. His description (Baird 1857, p. 143)

6 · Evolution and genetics

Figure 2.1 Island fox, San Miguel Island, 1994. Courtesy of National Park Service.

noted both the tame behavior of island foxes and their morphological similarity to the gray fox, and correctly suggested that the short tail was due to one missing vertebra: This very curious fox, the smallest of all the North American species, was brought by Lieut. Trowbridge from the island of San Miguel, on the coast of California, where quite a number of specimens were seen. It is stated by Lieut. Trowbridge to be very tame, scarcely taking the trouble to get out of the way, and when escaped from a trap, returning directly to the same place. His men found no difficulty in outrunning these foxes in a fair race, although it is possible, that owing to their unusual tameness, their full powers were not exerted. The species is a miniature of the common gray fox of the United States, and so closely like it in external appearance as to induce the belief of it possibly being a local race. Gray foxes from the main land of California, are, however, of full size, and there are some differences of importance in the skull and teeth. As is well known, also, many species of foxes of different regions resemble each other so closely that it is very difficult to separate them – more closely, indeed, than the present fox and the common gray species. The Vulpes littoralis is scarcely more than half the size of the common gray fox, in length and height, in fact, exceeded by some common house cats. The body, however, is considerably stouter than the house cat. The limbs are short, slender and weak. The tail in the specimen before me is very short, not more than

2.2 Dwarfism and the island syndrome

· 7

one-third the length of the head and body. It has probably lost some of the terminal vertebra at an early age, although the tip is now covered with hair. Two living specimens in captivity are said by Lieut. Trowbridge to possess this same brevity of tail.

Foxes from different islands were first recognized as separate taxa by Merriam (1903), who reclassified the island fox into the genus Urocyon and described foxes from Santa Catalina and San Clemente Islands as separate species (Urocyon catalinae and Urocyon clementae), while recognizing Santa Cruz foxes as a subspecies (santacruzae) of Urocyon littoralis. The species’ current designation as six subspecies of Urocyon littoralis was proposed by Grinnell et al. (1937) and has been corroborated by recent genetic and morphometric work (Gilbert et al. 1990, Wayne et al. 1991b, Collins 1993, Goldstein et al. 1999). Morphologically, the six subspecies differ in such features as body size, nasal shape and projection, and the number of tail vertebrae, and the subspecies can be correctly identified based solely on cranial characters and measurements (Collins 1982, 1993). Santa Catalina foxes are the largest, followed by San Miguel foxes, and the smallest are found on Santa Cruz. The tails of San Miguel foxes are considerably shorter than those of the other subspecies, averaging 15 caudal vertebrae compared to 19 to 22 for the other islands (Collins 1982). However, due to isolation, small population size, and the relatively low number of elapsed generations since founding, island foxes display less morphological and genetic variability than mainland gray foxes (Wayne et al. 1991b).

2.2 Dwarfism and the island syndrome Island fox evolution illustrates both the phenomenon of dwarfism, wherein island species are sometimes smaller than the mainland progenitor from which they originated, as well as the ‘island syndrome’, general changes in various life history characteristics due to selection pressures unique to island environments (Foster 1964, Van Valen 1973). Compared to the mainland gray fox, male island foxes are on average 14–18% and females 12–17% smaller (Collins 1982). Although selection for small body size in island foxes was originally thought to have resulted from the predominance of smaller-sized foods (insects, seeds, fruits, berries, and deer mice) on the islands (Collins 1982), it is more likely that resource limitations on islands select for small body size (Lawlor 1982). Resource limitation and competitive release may explain changes

8 · Evolution and genetics in body size among insular mammals (Lomolino 1985, Demetrius 2000), although it is possible that such changes are simply random (Meiri et al. 2004). One of the most spectacular cases of island dwarfism also occurred on the Channel Islands. The dwarf or pygmy mammoth (Mammuthus exilis; Roth 1993, Agenbroad 2002, Agenbroad et al. 2005) was one-third the size of its mainland ancestor, the Columbian mammoth (M. columbi), and occurred on the Channel Islands during the Pleistocene, from about 47,000 years before present (BP) to about 13,000 BP. Conversely, a trend toward gigantism in insular rodents is illustrated by the extinct giant island deer mouse (Peromyscus nesodytes) and giant vole (Microtus miguelensis) of the Channel Islands (Guthrie 1993). Gigantism in island rodents may be due to reduced seed-size variability, which selects for larger body size among generalist seed predators in the more homogenous environments of the islands (Foster 1964, Lawlor 1982).

2.3 Evolution of the island fox Although the scant fossil history of island foxes places them only as far back as the early Holocene epoch (12,000 years ago to present; Rick et al. 2009), island fox evolution is marked by the rapid differentiation from its mainland ancestor into the six subspecies recognized today. Despite the sparse fossil record, evidence from the archeological record, geologic history, and genetic studies provide evidence of the pattern and rate of colonization. However, significant questions remain regarding the exact mechanism of the initial arrival of gray foxes on the islands, and the means by which foxes later dispersed to all of the Channel Islands. 2.3.1 Colonization of the islands

Only four fossil localities (three on San Miguel and one on Santa Rosa) are known to contain samples of fox bones (see Rick et al. 2009). An island fox fossil was reported by Orr (1968, as reported in Moore and Collins 1995) from a stratified Pleistocene locality in the Upper Tecolote Formation on Santa Rosa Island. Based on dates Orr obtained from radiocarbon (14 C) dating of associated materials, the specimen was originally estimated to be between 12,000 and 18,000 years old. This date range for the earliest appearance of the fully dwarfed island fox species was used in several reconstructions of the fox’s evolutionary history (Gilbert et al. 1990, Wayne et al. 1991a, 1991b, Goldstein et al. 1999), all of which

2.3 Evolution of the island fox

· 9

assume colonization of the islands by foxes around 16,000 BP. However, recent accelerator-mass spectrometry (AMS) dating of the original specimen put its date at about 1,400 BP, and subsequent 14 C dating of three additional island fox bones collected from the surface of possible Pleistocene fossil localities on San Miguel Island suggest that none are older than about 6,500 BP (Rick et al. 2009). Because humans colonized the northern Channel Islands at least 13,000 years ago, Rick et al. (2009) suggest that the new dates for island foxes may indicate that Native Americans introduced foxes to the northern islands, possibly between about 7,000 and 10,000 BP. The difference in dates for colonization of the islands, although small, has enormous ramifications for the likely method of dispersal of gray foxes to the islands. Until recently, fox colonization of the islands was assumed to have occurred by chance overwater dispersal, whereby one or more gray foxes arrived on the northern islands via rafting on floating debris (Collins 1982, 1993, Wayne et al. 1991b). Current distances between the mainland and the northern islands range from 20–90 km, but sea levels were as much as 100–125 m lower during the last glacial maximum of the Ice Age (∼20,000 BP). During this time the northern islands, which were never connected to the mainland, coalesced into a larger island now referred to as Santarosae (Fig. 2.2), which was separated from the mainland by as little as 6–10 km (Wenner and Johnson 1980). A likely source for vegetation debris upon which gray foxes could have floated to the islands are two large rivers, the Ventura and the Santa Clara, which flow into the Pacific Ocean from the western transverse mountain ranges of Southern California. During periodic El Ni˜no-driven winter storms, debris from the rivers can wash far out to sea, occasionally ending up on the northfacing shorelines of the northern Channel Islands (Inman and Jenkins 1999). However, the lack of fossil evidence of island foxes on the islands prior to the arrival of humans suggests that foxes could have been intentionally introduced to the northern islands by Native Americans (Rick et al. 2009). Native American ancestors of the historic Chumash and Tongva (Gabrielino) peoples arrived on the northern islands by 13,000 BP (Table 2.1). Exploiting the rich marine environments, the early island inhabitants engaged in shellfish harvesting, kelp-bed fishing, and hunting marine mammals (Johnson et al. 2002, Erlandson et al. 2005). Early peoples introduced domestic dogs (Canis familaris) to the islands, and it is possible that they introduced foxes from the mainland to the islands as well (Rick et al. 2009). Still, the presence of only small-sized

10 · Evolution and genetics Table 2.1 Timeline of events in evolutionary history of island foxes. Epoch

Years BPa

Event

Pleistocene

> 20,000 16,000

Existence of superisland Santarosae Possible arrival of foxes from mainland, by natural dispersal (rafting)

Holocene

a

15,000 14,000 13,000 12,000 11,000 10,000 9,000 8,000 7,000 6,000

Arrival of humans on the Channel Islands Separation of Santa Cruz from superisland Separation of San Miguel from Santa Rosa

Earliest fossil remains of island foxes (San Miguel); Earliest archeological remains of island foxes on the southern Channel Islands

BP = before present

Figure 2.2 Current sea level and shorelines (in black), compared with those in the late Pleistocene, when the current northern Channel Islands were coalesced into a superisland, Santarosae.

2.3 Evolution of the island fox

· 11

island-fox remains in paleontological and archaeological sites suggests that foxes could have reached the islands prior to the arrival of Native Americans, since evolution from the mainland gray fox form to that of the smaller island fox likely took place over considerable time. Whether humans brought island foxes with them to the northern islands or foxes were already there, the Chumash people apparently enjoyed a close relationship with foxes (Collins 1991a, 1991b). Island fox remains from archeological sites suggest that the Chumash used island foxes as a source of pelts, as pets, and in rituals and religious ceremonies. Island fox remains as well as those of domestic dogs have been found intentionally buried at sites on San Clemente Island, suggesting canid killing or mourning ceremonies (Hale and Salls 2000), and have been found associated with human burials on Santa Cruz and Santa Rosa Islands. The Chumash, who plied the island waters in their plank canoes (tomols) and engaged in wide-ranging trade with other groups, are thought to have introduced island foxes to the three southern islands of Santa Catalina, San Clemente, and San Nicolas (Wenner and Johnson 1980, Johnson 1983, Collins 1991a, Vellanoweth 1998). Island foxes do not appear in the archeological record there until about 6,000 BP. 2.3.2 Rapid evolution

The evolution of gray foxes to island foxes occurred fairly quickly, within 3,000–10,000 years after the arrival of foxes to the islands. The Pleistocene superisland of Santarosae (Fig. 2.2) coalesced and divided numerous times in response to alternating cold and warm periods, but permanent separation began about 16,000–17,000 BP (Collins 1991a). If island foxes were established earlier than 13,000 BP, then the population of foxes on Santarosae was split along with the islands. The largest island, Santa Cruz, was separated from the others first approximately 11,500 BP, followed by the separation of San Miguel from Santa Rosa at about 9,500 BP (Table 2.1). Morphologic and genetic variation in the six fox subspecies reveals not only the order of the separation of the northern islands, but strongly suggests the later colonization of the southern islands via transport by the Chumash. Collins (1982) recognized that the pattern for the breakup of the Santarosae landmass corresponded with the morphological distance among the respective island fox subspecies, and Gilbert et al. (1990) corroborated these findings with evidence consistent with the fossil and geologic record and colonization history for island foxes.

12 · Evolution and genetics The genetic and morphological evidence for the pattern of island colonization and founding was summarized by Wayne et al. (1991b), who looked at several measures of genetic variation, as well as morphometrics. Their results showed that the northern islands were colonized first, followed by the isolation of foxes on Santa Cruz from the remainder of the northern islands, and then by separation of Santa Rosa and San Miguel. San Miguel and the three southern islands comprise a group of closely related populations with genetic distances near zero, implying that San Miguel foxes were then taken (by humans) to the southern islands. San Clemente was the first southern island colonized, and San Clemente foxes were then taken to Santa Catalina and San Nicolas (George and Wayne 1999, Goldstein et al. 1999).

2.4 Island foxes in the paleoenvironment The island communities in which island foxes evolved were substantially different from those of today. Island foxes co-existed not only with humans (the Chumash, Tongva, and their predecessors) but also with domestic dogs, and a few Pleistocene species, which – unlike the island fox – did not survive into modern times. The remains of domestic dogs in archeological sites on the islands occupied by island foxes shows that island foxes and domestic dogs co-existed on the islands for at least 6,000 years (Rick et al. 2008). The long co-existence of foxes and dogs on the islands is puzzling, given that canine-born pathogens are recognized as a primary threat to modern island fox populations (Coonan et al. 2003, Roemer et al. 2004, Clifford et al. 2006). Current recovery actions include several efforts to keep dogs off islands, minimize contact between dogs and island foxes, vaccinate wild foxes against canine diseases, and monitor island foxes for signs of canine disease (Schwemm 2008b). However, island fox populations have also been exposed to a strain of canine distemper virus that is different from the one that caused the mass mortalities on Santa Catalina in the 1990s (Chapter 7). This strain may have been transmitted to foxes from domestic dogs thousands of years ago, or may have been endemic in gray foxes prior to their colonization of the Channel Islands; it is apparently endemic in island foxes and, unlike other canine distemper strains, is not 100% lethal (Fiedel and Haynes 2004, Clifford et al. 2006, Rick et al. 2008). The effect of domestic dogs on island foxes, and other wildlife, may have been minimal if dogs were not running wild in feral packs but instead were primarily confined to human settlements (Rick et al. 2008).

2.5 The peculiar genetic status of the island fox

· 13

Dogs, and perhaps semi-domesticated island foxes, may have helped keep native villages and camps clean by consuming waste and rodents, and were probably not used in food-gathering, which was focused on marine resources. But the human population was considerably higher than the current human presence on the islands. At the time of European contact, Santa Cruz Island was home to perhaps over 1,000 Chumash who occupied a dozen villages. Did these natives, along with their domestic dogs and the native island fox, have a hand in the disappearance of other late Pleistocene fauna from the islands? A flightless sea duck (Chendytes lawi) occurred as early as 40,000 BP on San Miguel Island, and bones and egg shells dated from 12,000 BP are evidence of inland nesting colonies on the island (Jones et al. 2008a). Sea ducks were hunted by humans as early as 11,500 BP, but did not go extinct until 3,000 BP, after over 8,000 years of exploitation. Three other species of burrow-nesting seabirds, Cassin’s auklet (Ptychoramphus aleuticus), ancient murrelet (Synthliboramphus antiquus), and the now-extinct Dow’s puffin (Fratercula dowi) all had breeding colonies on San Miguel but no longer occur there (Guthrie 2005). In fact, no burrow-nesting birds occur on the main portion of San Miguel Island today. The fact that burrow-nesting birds are nearly absent from all of the northern islands supports the theory that foxes were brought to the islands as late as 7,000 to 10,000 BP (Rick et al. 2009), when they began preying on birds. Island foxes also co-existed on San Miguel with the now-extinct giant deer mouse, which disappeared by about 8,000 BP (Guthrie 1980), the now-extinct giant vole, and the Southern Pacific rattlesnake (Crotalus viridis), which no longer occurs on the island (Guthrie 1993). Perhaps most intriguing is the possible co-existence of island foxes and pygmy mammoths on Santa Rosa Island. Remains of mammoths have been dated at 11,030 BP, and human use of Santa Rosa dates from 10,960 BP (Agenbroad et al. 2005). If humans drove pygmy mammoths to extinction on the northern Channel Islands, island foxes may have been there to witness it – and to clean up the scraps.

2.5 The peculiar genetic status of the island fox The genetic status of the island fox is well known, thanks to recent and comprehensive studies (Gilbert et al. 1990, Wayne et al. 1991a and 1991b, Goldstein et al. 1999). Because of this, along with the species’ recent evolution, fairly well-established colonization pattern, and the imposing barrier to dispersal posed by the distance between islands, the island fox

14 · Evolution and genetics illustrates various genetic principles of small populations, although some aspects of its genetics are unprecedented. First and foremost, the island fox provides a natural experiment in the genetics of small populations, and of island populations, with their inherently reduced variability. The patterns of genetic variation across island fox subspecies reflect the colonization pattern and apparently are influenced, as predicted, by founding time and effective population size. However, the island fox departs from established genetic and conservation models in its apparent persistence at small population sizes, over millennia, despite a relative lack of genetic variation. This is illustrated in the extreme by San Nicolas Island foxes, which are essentially monomorphic at all genetic markers examined save one (Aguilar et al. 2004; see below). As for other species with small populations, genetic investigation in island foxes has focused on measuring genetic variation, which is commonly measured by looking at polymorphism, or the number of alleles (different forms of a gene) at particular loci (a segment of DNA) in a population. Due to inheritance of alleles from both parents, individuals are either heterozygous (having different alleles) or homozygous (having two copies of the same allele) at a particular locus. Loci that have no multiple copies of an allele in an entire population are considered monomorphic (as opposed to polymorphic). Studies of genetic variation in island foxes have utilized the following: 1. protein, or allozyme, electrophoresis, which uses the charge (positive or negative) of different amino acids to detect genetic variation; 2. mitochondrial DNA (mtDNA), which is that contained in the mitochondria of the cell (as opposed to nuclear DNA); and 3. minisatellites and microsatellites, which are repetitions of short segments of nuclear DNA that are highly variable. Recent studies (Gilbert et al. 1990, Wayne et al. 1991a and 1991b, Goldstein et al. 1999) using all of the techniques described demonstrate that island foxes have low or non-detectable levels of genetic variation, some of the lowest recorded for any vertebrate species. The lack of genetic variability is due to the small number of founders (original members of the population, in this case those that arrived on the islands) for each subspecies, the likely occurrence of frequent population bottlenecks, and the lack of immigration from other populations (i.e. there is no true metapopulation in which the genetic variability of sub-populations is maintained by periodic dispersal among them). Yet despite this low variability and populations numbering only in the hundreds, island fox

2.6 Genetic management of captive island foxes

· 15

populations have persisted for 7,000–18,000 years without going extinct and without apparent inbreeding depression (see Section 2.6). Wayne et al. (1991b) concluded that significant gene flow is not required for long-term persistence or conservation of island foxes, and that selection for inbreeding tolerance may have occurred in this species. The extremely low genetic variability of the San Nicolas subspecies of island fox is remarkable, even given the low diversity of the species as a whole. The San Nicolas fox populations exhibit no variation at almost all the genetic markers studied (Wayne et al. 1991b). Wayne et al. (1991a) attribute the lack of variation to the subspecies’ small population size – it has the second lowest population size of the six subspecies – and the fact that San Nicolas is the most remote of all the Channel Islands. San Nicolas foxes may also have been subject to more frequent or severe bottlenecks than other island fox subspecies; the population declined to less than 110 individuals in the 1980s, and may have actually dropped to as low as 20 individuals at one point in the 1970s, although it had recovered to 500 individuals by 1984 (Laughrin 1977, Kovach and Dow 1985). More recent work on the genetics of San Nicolas foxes has identified significant variation at the major histocompatibility complex (MHC), genes that, unlike those in microsatellites, are not neutral (non-coding) but actually code for proteins, and may therefore influence characters with adaptive and fitness consequences. Aguilar et al. (2004) found that although San Nicolas Island foxes are essentially monomorphic at 18 microsatellite loci, they had markedly high levels of variation at the MHC of genes, which codes for kin recognition and disease resistance, among other things. Thus, the MHC genes may be under intense selective pressure, unlike the neutral markers at microsatellites. Simulations by Aguilar et al. (2004) showed that the observed high MHC variability coupled with monomorphism at neutral markers could only have occurred if the population had weathered a decline to < 10 individuals 10–20 generations prior. This is consistent with anecdotal evidence that a decline in San Nicolas Island foxes to < 20 animals occurred in the 1970s (Kovach and Dow 1985).

2.6 Genetic management of captive island foxes Proper genetic management of small populations is a primary goal of endangered species conservation. Management aims to maximize genetic diversity within a population. Small populations quickly lose genetic

16 · Evolution and genetics variability, and inbreeding (the mating of animals related by ancestry) can lead to reductions in fecundity and juvenile survival as well as greater susceptibility to pathogens (Ralls et al. 1988). Inbreeding depression can potentially cause small populations to enter an ‘extinction vortex’, where the mutually compounding effects of reduced genetic variability and small populations drive a population to extinction (Gilpin and Soule 1986). With the establishment of captive populations of island foxes (Chapter 8), managers were faced with the challenge of maximizing genetic diversity using a relatively low number of individuals that remained from what was historically a genetically precarious population. Because the genetic history of island foxes played such a large role in managing the genetic element of the captive breeding program, we discuss genetic management herein. The precarious genetic status of island foxes was brought into sharp relief by the population declines of the 1990s. The critically small postdecline populations and the establishment of captive breeding programs with very low numbers of founders meant that the genetic management of captive populations would be a difficult challenge (Frankham et al. 2002, Wayne et al. 2004). Inbreeding is unavoidable in small, closed populations, even if the founders were unrelated, because eventually all or nearly all of the individuals in the population become related by descent (Frankham et al. 2002). Inbreeding depression varies among species and populations, with founder genotypes and founder effects influencing the degree of inbreeding depression that may eventually be observed in a descendant population. For example, low genetic diversity and presumably the presence of deleterious alleles caused expression of a variety of detrimental characteristics in the Florida panther (Puma concolor coryi). The frequency of such traits as cowlicks, kinked tails, and cryptorchidism (in which one or both testicles fail to descend into the scrotum) was eventually reduced by the introduction of cougars from Texas, which increased allelic diversity at loci associated with such traits (‘genetic rescue’; Ingvarsson 2002, Keller and Waller 2002). The captive breeding programs for island foxes on Santa Rosa and San Miguel Islands began with only 15 individuals each, which was then the number of potential founders for the resulting captive program. The number of successful founders (those that eventually bred) would be less than that. In addition to growing the captive populations as quickly as possible, population managers sought to maintain what little genetic diversity presumably remained after the severe population declines of the 1990s (Coonan 2003, Lynch 2005a, 2005b).

2.6 Genetic management of captive island foxes

· 17

Table 2.2 Number of founders for various captive breeding programs, from Frankham et al. (2002) and C. Lynch, Association of Zoos and Aquariums (unpublished data). Number of founders

Species Arabian oryx Black-footed ferret Siberian tiger Snow leopard California condor Puerto Rican parrot

0.707 B067E

1612C

0.566 D3D76

A7015

Oryx leucoryx Mustela nigripes Panthera tigris altaica Uncia uncia Gymnogyps californianus Amazona vittata

0.475

10030

D187A

9 7 25 43 14 13

0.599

53313

07061

A180A

37E00

0.549 F0223

0.475

Female

Male

F4A18

95B34

73D0D

3512D

0.479

r value

Figure 2.3 Relatedness of potential founders for the Santa Rosa Island fox captive breeding program, 2000 (from Gray et al. 2002). Relatedness values ≥ 0.5 indicate sibling–sibling or parent–offspring.

Swift and effective genetic management of the captive populations occurred because island fox population genetics were well known, and management began as soon as foxes were brought into captivity (Chapter 8). In 2000 NPS funded additional research to determine how much genetic variability had been lost in the three northern island populations (San Miguel, Santa Rosa and Santa Cruz), and to construct population pedigrees (charts of genetic relatedness; see Fig. 2.3) for the captive founders. Results showed that the captive San Miguel population – remnants of the larger San Miguel population that historically had less genetic variation than any of the three subspecies – had even less variability as a result of the recent declines, as measured by the mean number of alleles and lower heterozygosity (Table 2.2; Gray 2002, 2003, Gray et al. 2001, 2002). Two San Miguel loci had become fixed (were monomorphic), meaning that there were no longer multiple alleles occurring at

18 · Evolution and genetics those sites within the population. Relatedness values were lowest for the captive Santa Cruz population, where the remaining wild population (and hence the effective population) was substantially larger than on San Miguel or Santa Rosa (Coonan et al. 2005a). Because captive breeding might be required for a decade or more, the Island Fox Conservation Working Group (IFCWG) recommended that formal genetic management be administered through an Association of Zoos and Aquariums (AZA) Population Management Plan (PMP or ‘studbook’). In 2004 NPS requested that the AZA develop an island fox PMP with annual analysis of the captive populations and recommendations for pairings and suitable release candidates (Chapters 8 and 12). Unlike studbooks for many species, which assume founders to be unrelated or equally related, the relatedness of potential island fox founders was known, and was therefore incorporated into the island fox studbook (Lynch 2004). This resulted in more accurate modeling of the genetic consequences of potential actions (pairing, retention, and release). The goals of the population management plans were two-fold. First, the mean kinship of all potential pairs would be evaluated to maintain or increase gene diversity retention in the captive population. Second, annual release candidates would be chosen so that their release would not impact long-term genetic management of the captive population. Generally in captive breeding programs a founder population of 25– 50 individuals is required to retain allelic diversity (Ballou and Foose 1996), but clearly the available pool of potential founders for San Miguel and Santa Rosa was much lower (15 in each case). However, only 10 founders are required to sample ≥ 95% of a source population’s heterozygosity (Ballou and Foose 1996), and ultimately the number of founders for Santa Rosa (12) met this goal, whereas the San Miguel population (8) did not (Coonan and Dennis 2007). These numbers are comparable to those of other captive breeding programs for endangered species (Table 2.2). The genetic consequences of all potential management actions were modeled using a ‘gene-drop’ analysis, which simulates the likely fate of founder alleles over time. Potential pairs were rejected if they had kinship and inbreeding values greater than the mean kinship and mean inbreeding coefficients of the entire captive population (exceptions were made for proven pairs, since rapid population growth was also critical). Although the small captive population sizes (annually managed at < 40 individuals), low number of founders, and short generation time contributed to predicted loss of gene diversity retention over time, the release of animals into the wild enhanced the potential for overall gene diversity retention

2.6 Genetic management of captive island foxes

· 19

Table 2.3 Genetic status of island fox captive breeding populations in 2005.

Founders/potential founders Founder genome equivalents1 Founder genomes surviving1 Mean heterozygosity2 Mean number of alleles per locus2 Number of foxes sampled 1 2

San Miguel

Santa Rosa

Santa Cruz

8/15 4.85 6.44 0.23 1.69 15

13/15 8.21 10.18 0.52 2.93 15

0.43 2.60 18

from Lynch 2005a, 2005b from Gray et al. 2001, Gray 2003, based on 16 loci

Table 2.4 Average number of alleles per locus (A) and average heterozygosity (H) for endangered and non-endangered carnivores (from Frankham et al. 2002). Species Non-endangered African lion Puma Gray wolf Endangered Polar bear Ethiopian wolf Mexican wolf African wild dog Cheetah

A

H

Felis leo Felis concolor Canis lupus

4.3 4.9 4.5

0.66 0.61 0.62

Ursus maritimus Canis simensis Canis latrans baileyi Lycaon pictus Acinonyx jubatus

5.4 2.4 2.7 3.5 3.4

0.62 0.21 0.42 0.56 0.39

in each subspecies (Lynch 2005a, 2005b). In 2005 gene diversity was estimated to be 89.7% for the San Miguel subspecies and 93.9% for Santa Rosa. Within 20 years, gene diversity was expected to drop to ≤ 90% for the Santa Rosa subspecies, while the lower number of founders on San Miguel dropped that estimate to 82% for that subspecies. As of 2005 the number of founder genome equivalents and founder genomes surviving was also less on San Miguel, indicating greater inbreeding and loss of genetic diversity in that population (Table 2.3). However, strict genetic management of the island fox captive breeding programs put the genetic prognosis of each subspecies on par with that of other intensely managed endangered species (Table 2.4). Gene retention was high and expected to remain so over time, despite the low number of founders.

20 · Evolution and genetics

2.7 Future of wild populations With the cessation of captive breeding in 2008, the remaining genetic question was what effect bottlenecks and reduced genetic variability would have on island fox populations in the near and long term. As of 2009 there had been no apparent effects of inbreeding in either the captive populations or the recovering wild populations. All the same, island foxes are currently a genetic anomaly, whose persistence over time has been unlikely. They have apparently evolved to tolerate inbreeding (Wayne et al. 1991b), or selection has served to mitigate otherwise low genetic variability (Aguilar et al. 2004). On the other hand, the precarious genetic status of island foxes may make them a ‘conservation-reliant’ species (Scott et al. 2005), forever dependent upon human intervention, perhaps via genetic rescue or additional captive breeding, to forestall inevitable extinction.

3

r

Social structure, reproduction, mortality and survivorship, and population dynamics

Prior to the 1990s, studies of island foxes focused on their biology and basic ecological relationships in island systems. Several important studies provided the foundation for what we now know about productivity, abundance, food habits, and genetics (Laughrin 1977, Collins and Laughrin 1979, Kovach and Dow 1981, Collins 1982, Fausett 1993, Wayne et al. 1991b, Crooks and Van Vuren 1995, Moore and Collins 1995, Crowell 2001). The natural history approach to studying island foxes ended in 1993 when Gary Roemer (then of the University of California, Los Angeles) began research on Santa Cruz Island that would ultimately reveal the devastating effects of golden eagle (Aquila chrysaetos) predation on foxes (Roemer 1999; see Chapter 5). Roemer (1999) initially focused on island biogeography and the ecology and social structure of fox populations, but as increasing numbers of his study animals fell prey to golden eagles his research shifted to determining the effects of eagles on the entire Santa Cruz population and the conservation actions that would be required to avoid extinction of that population. A few years later, canine distemper virus (CDV) was introduced to the island fox population on Santa Catalina, resulting in the loss of nearly three-quarters of the entire population in one year (Timm et al. 2009; see Chapter 7). Prior to the outbreak of CDV on Santa Catalina very little was known about the history and role of disease in island fox populations, and management efforts immediately turned to determining the prevalence of disease, the source of pathogens, and means to protect island foxes from future disease outbreaks (Timm et al. 2000, Clifford et al. 2006). Research and monitoring efforts for island foxes intensified after 1992, but with a greater sense of urgency and a loss of innocent curiosity that had guided previous work. Studies initiated since that time have resulted in a much greater understanding of island fox biology and ecology, but are now conducted in the shadow of potential extinction and with an eye always toward management and recovery applications. In two chapters we present a synthesis of what is known about social

22 · Island fox populations

Figure 3.1 Island fox pair, Santa Rosa Island. Courtesy of Dan Richards.

structure, reproduction, survival, mortality, habitat use, and food habits of island foxes, and discuss areas in which further research is needed. (The reproductive biology of island foxes is treated in detail in Chapter 10.)

3.1 Social organization and reproductive behavior 3.1.1 Social structure

The social organization and reproductive behavior patterns of island foxes are similar to those of other small canid species, but show an overall adaptation to insular influences. The primary units of social organization in island foxes are mated pairs (Fig. 3.1) that generally remain together until one of the pair dies (Roemer 1999). Groups of more than two adults are common at high densities, and extra-pair fertilization occurs (Roemer et al. 2001b). Once an adult male establishes a territory he usually remains there throughout his life; for example, on San Nicolas Island only 4.5% of the adult males captured were caught on more than one trapping grid (approximately 2 km2 ) over a seven-year period (Schmidt et al. 2007b). Established pairs occupy territories that they maintain yearround, with little territorial overlap of adjacent pairs (Roemer et al. 2001b). Island foxes defend their territories, although tolerance for

3.1 Social organization and reproductive behavior

· 23

conspecifics is greater than in most other canid species (Roemer et al. 2001b). In the early 1990s – just prior to the most intense period of golden eagle predation on Santa Cruz – island fox densities were very high (7.0/km2 ) but territories were stable, meaning there was relatively little change from year to year in individual home range size and location (Roemer et al. 2001b). The territories of the male and female of a pair are usually situated together; measured on a scale where complete territorial overlap is 1.0, overlap for paired foxes ranged from 0.70 to 0.85 on Santa Cruz (Crooks and Van Vuren 1996, Roemer et al. 2001b) and San Clemente (Garcelon 1999). Breeding behavior and pair formation begin in late winter, and physical mating occurs in late February and early March. Gestation is approximately 50–53 days, and about 10 days before parturition the female retreats to the den and forages infrequently. Island fox pups are born in late April or early May, and emerge from the den approximately 5–6 weeks later. Once the pups are born the male attends the den and possibly brings food to the female, then forages with the family group throughout the pups’ first summer (Roemer et al. 2001b). Pups are weaned from the female by mid-summer, but are still highly dependent on the adults for food. Trapping for population monitoring occurs during late summer on all the islands, and prey items have been found near trapped pups during this period, indicating provisioning by adults at least until the pups are 4–5 months old (Garcelon et al. 1999). Adults and pups have also been caught in traps together, further suggesting a high degree of physical closeness while foraging (Garcelon et al. 1999). By early fall the pups or first-year animals (now called juveniles) are as large as the adults and can obtain food on their own. Juveniles either disperse during the fall or, if population densities are high, may remain in some portion of their parents’ home range (Roemer et al. 2001b). The male and female adult members of a pair generally forage separately in the fall and winter. 3.1.2 Reproduction

Our understanding of reproduction in island foxes is based on information collected under three different sets of conditions: wild populations at stable densities, recovering wild populations, and captive populations. Observations of each situation provide additional insight into the physiology and behavior of reproduction in island foxes. Information accumulated from all island fox populations suggests an inverse relationship

24 · Island fox populations between adult density and population growth (Bakker et al. 2009). The relationship between population density and growth was clearly illustrated on San Miguel Island when the first foxes were reintroduced to habitat completely absent of conspecifics (Chapter 9). All four of the females released in the fall of 2004 were juveniles and produced litters the following spring. Reproduction by first-year females occurred again the following year, when most of the released young females, as well as their female pups, successfully reproduced. Studies in wild populations prior to the decline suggested that females generally did not reproduce in their first year (Roemer et al. 2001b, Coonan et al. 2005b, but see Collins and Laughrin 1979); for example, on Santa Catalina, when populations were recovering but at low to moderate densities, juvenile pregnancy rates were much lower than adult rates (Clifford et al. 2007). Litter size does not appear to vary with density, although data on this are scarce. Across all islands and population densities, the mean number of pups weaned is approximately 2.0. Laughrin (1977) found mean litter sizes of 2.17 on Santa Cruz, and Clifford et al. (2007) found litter size in captive populations of 2.1 and in wild populations of 1.8. Fertility also did not appear to change as populations declined (Roemer et al. 2001a); however, prior to the studies associated with recovery almost nothing was known about pre-term and neonatal pup mortality, which may in fact be relatively high (Clifford et al. 2007). A detailed discussion of reproductive biology in island foxes is provided in Chapter 10.

3.2 Mortality and survivorship Survival rates for mammals are usually estimated either by following individually marked (tagged and radio-collared) animals throughout their life and documenting their life history, or by calculating the proportion of marked animals that survive across years. Both types of studies have been conducted on island fox populations, with the most common measure of survival being the proportion of animals that survive from one summer sampling period to the next. This measure is normally termed ‘apparent survival’ because of the difficulty in distinguishing true mortality from dispersal away from the study area. In the absence of recent predation and disease events, estimates of apparent survival are high for all island fox populations. For example on San Miguel in 1993–1994, prior to the arrival of golden eagles, adult survival was 100% (Coonan et al. 2005b). For island fox populations across all islands and years annual adult survival ranges from 70–95% (Table 3.1). Survival of pups is lower than for adults

3.2 Mortality and survivorship

· 25

Table 3.1 Annual survival rates for island foxes. Island

Years

Annual survival

Source

San Miguel

1993–1994

Coonan et al. 2005b

Santa Cruz San Nicolas

1993–1994 2007

Santa Catalina

1999–2002

San Clemente

1988–1997

Adult: 1.00 Pup: 0.74 Adult: 0.69 Adult female: 0.76 Adult male: 0.74 Adult female: 0.83 Adult male: 0.75 Pup: 0.69 Adult: 0.72

Roemer et al. 2001a Garcelon and Hudgens 2008 Kohlmann et al. 2005

D. Garcelon, unpublished data

(Roemer et al. 2001a, Coonan et al. 2005b), as it is for many species, but survival does not appear to vary much among other adult age classes or between males and females (Bakker et al. 2009). The high survival rates for island foxes result from the high quality of the habitat and the typical absence of predation (factors that also largely contributed to the success of reintroduction). Because survival is naturally high for island foxes, other life history traits, such as large litter size, are less important, and it is ultimately adult survival that has the greatest influence on population growth (Roemer et al. 2001a, Bakker et al. 2009). This high level of survival is unusual for small carnivore species. Annual survival rates for mainland fox species such as swift foxes (Vulpes velox), kit foxes (Vulpes macrotis) and gray foxes range from 40% to 70%, with predation and disease the most significant mortality sources (Ralls and White 1995, Sovada et al. 1998, White et al. 2000, Olson and Lindzey 2002, Schauster et al. 2002a, Harrison 2003, Farias et al. 2005). The one-year survival rate of pups is lower than for adults, but still higher than for most carnivores; on San Miguel Island in 1993 (prior to the decline) pup survival was estimated to be approximately 75% (Coonan et al. 2005b), and current survival of wild pups on that island is likely the same or slightly higher (NPS unpublished data). Very little is known about the survival rate of wild island fox pups in their first few months, and future research will hopefully address this question. Many years of monitoring have revealed numerous sources of mortality for island foxes that at normal population levels have minimal populationlevel impacts but were of great concern when populations were critically

26 · Island fox populations low. Recent observations of significant aggressive behaviors in captive populations (Chapter 5) and occasional observations of serious aggression in the wild suggest that intraspecific aggression may potentially be fatal, either directly or from secondary wound infections. On San Clemente Island in 2002, two radio-collared adult male foxes with adjacent territories were each found dead within a month (Schmidt et al. 2004b). Both animals were emaciated and had infected wounds, and their condition and spatial proximity suggested territorial aggression that led to their eventual demise. However, there are few additional records of intraspecific interactions that were ultimately fatal, and trapping data include very few records of serious injuries that would indicate intraspecific fighting as a common cause of death of island foxes (Roemer 1999). Most wild foxes on San Miguel and Santa Rosa have tattered ears, which is evidence that non-lethal aggressive interactions may be fairly common (NPS unpublished data). In the absence of predation and disease, human-related impacts, particularly interactions with domestic dogs and vehicle trauma, are the greatest sources of island fox mortality. Because island foxes are less wary of humans than are most wild canids (except perhaps coyotes), human developments present myriad physical hazards for foxes. Their natural curiosity has caused island foxes to become trapped in dumpsters, accidentally locked in buildings, and caught in drainage pipes and underground utility boxes. Although the isolation of their habitats provides relative safety from intentional human harm, cases of accidental shooting and neglect have resulted in fox deaths (J. King, Catalina Conservancy, personal communication). Domestic dogs have killed several island foxes on Santa Catalina (Timm et al. 2002, Schmidt et al. 2004b); for example, a freshly killed fox was found in the middle of a road on that island a few minutes after two unrestrained dogs were observed walking from the same direction (DG personal observation). Santa Catalina is the only island where dogs are legally allowed, but the other islands are not free from the threat of dogs, and recreational boaters often bring unleashed dogs to islands where they are officially prohibited (NPS unpublished data). Although physical dog–fox interactions likely rarely end well for individual foxes, fox deaths from dogs are uncommon and have little impact on fox populations. The much greater threat from domestic dogs on islands is their potential to transmit disease (Chapter 11). The greatest source of mortality for all island foxes in the 2000s (other than predation and disease) is vehicle trauma (Chapter 11). Of the six islands that support foxes, three have paved roads (San Nicolas, San

3.3 Population abundance and dynamics

· 27

Clemente, and Santa Catalina), and two have extensive unpaved road networks (Santa Cruz and Santa Rosa). San Miguel is the only island with island foxes that does not have roads. The rates of injuries and deaths for foxes are much higher on the islands where roads are paved and vehicle speeds higher than on islands without paved roads. On San Clemente, where military operations increase the potential for fox–vehicle interactions, research has shown a negative correlation between fox survival rate and the presence of a road in a fox’s home range (B. Andelt, Colorado State University, personal communication). On Santa Catalina Island, five radio-collared foxes were killed by cars in 2006 (Schwemm 2007) and four in 2007 (Schwemm 2008b). Efforts to reduce fox mortalities from vehicles include maintaining mowed buffers along roads and educating drivers, especially on Santa Catalina and San Nicolas Islands (G. Smith, US Navy, and J. King, Catalina Island Conservancy, personal communication), and these actions substantially reduced the number of foxes killed or injured by vehicles in the 2000s.

3.3 Population abundance and dynamics Discussions of population abundance must distinguish between populations under the influence of recent predation and introduced disease and those which are not. For over two decades a consistent set of methods has been applied to sampling island fox populations, resulting in comparable estimates across all populations and throughout the decline and recovery periods (Roemer et al. 1994, Bakker et al. 2009, Table 3.2). For all islands, estimates for population abundance and density (adults/km2 ) have been obtained via mark–recapture techniques using established trapping grids and/or transects, where trapped animals are uniquely identified with passive integrated transponders (‘PIT’ tags). At low densities other methods such as radio-collar monitoring and camera trapping have been employed (Coonan et al. 2005b, Bakker et al. 2009). Because pups are subject to higher mortality during their first winter, population estimates from summer trapping are generally reported for adults only to provide a more conservative estimate of population status and trend (Roemer et al. 1994). 3.3.1 Population abundance

The total number of adult foxes on each island (and thus the total number of each subspecies) varies with island size. Adult abundance ranges from

28 · Island fox populations Table 3.2 Examples of density estimates for island foxes. Year

Habitat

Method1

Density Adults/km2

Source

Closed

8.1–15.9

Coonan et al. 2005b

San Miguel: 1994 Grassland/Isocoma Scrub Santa Cruz: 1993 Chaparral/Scrub 1993 Grassland San Nicolas: 2007 Dune Scrub

Closed Closed

7.3 7.0

Roemer et al. 1994 Roemer et al. 1994

Open

15.7–21.3

2007

Grassland/Scrub

Open

7.6–13.7

2007

Coastal Scrub

Open

17.4–22.7

Garcelon and Hudgens 2008 Garcelon and Hudgens 2008 Garcelon and Hudgens 2008

Open Open Open

11.6–16.8 3.9–4.8 2.4–3.6

Schmidt et al. 2005 Schmidt et al. 2005 Schmidt et al. 2005

Closed

11.4

Roemer et al. 1994

Closed Closed

7.4 2.9

Roemer et al. 1994 Roemer et al. 1994

San Clemente: 2004 Maritime Dune 2004 Maritime Scrub 2004 Grassland Santa Catalina: 1990 Coastal Scrub/ woodland 1990 Chaparral/woodland 1990 Coastal Scrub 1

Open = Population estimate determined using open models (Pollock’s Robust Design) in program MARK; Closed = Population estimate determined using closed models in program CAPTURE.

350–450 on San Miguel Island (the smallest island) to 1,500–2,000 on the larger islands of Santa Catalina and Santa Cruz (Roemer et al. 1994, Bakker et al. 2009). Likewise, carrying capacities (estimates of the total number of foxes each island could support) were reported by Roemer et al. (2001a) as 577 for San Miguel and 1540 for Santa Cruz. Numerous studies have produced estimates of island fox population size on the various islands (Table 3.2); however, to illustrate the increasing precision of methods applied as well as the various factors that influence change in island fox population abundance, we discuss the history of monitoring and study of the Santa Catalina population.

3.3 Population abundance and dynamics

· 29

The first survey of foxes on Santa Catalina was conducted in 1972 by Laughrin (1973). During his study only two foxes were captured during 60 trap-nights, yielding a trap success rate of 3.3%. A followup effort in 1975 over a much larger area (88 linear km of transects) resulted in the capture of 55 individuals during 597 trap-nights (Propst 1975). Overall capture success for individuals cannot be compared with Laughrin’s study because the number of recaptures was not documented in the report. Laughrin returned in 1977 and captured zero foxes in 66 trap-nights (Laughrin 1980). These data suggest that in the mid-1970s the fox population on Santa Catalina was much lower than on the other islands, where trap success ranged from 27–67% during the same time period (Laughrin 1980). Another monitoring effort on Santa Catalina began in 1989 when the Institute for Wildlife Studies (IWS) received a grant from the state of California to determine the status of the Santa Catalina fox population. Trapping grids were established in three areas to obtain estimates of fox density and to gather habitat information. Results indicated varied densities of foxes across the island, with higher densities in mixed oak-woodland/coastal sage habitat (11.4 foxes/km2 ) than in mixed chaparral/coastal sage habitats (2.9–7.4 foxes/km2 ; Garcelon et al. 1991, Roemer et al. 1994). Impacts of introduced grazers over the previous century had significantly degraded island habitats on all of the islands with foxes (Coblentz 1977, Klinger et al. 2002, Sweitzer et al. 2005), and during the 1980s vegetation communities may have supported fewer foxes than they did pre-grazing conditions. With the removal of several introduced ungulate species vegetation began to recover, and during the 1990s fox populations on Santa Catalina were thought to be doing well; annual trapping in 1989–1990 indicated a healthy population of over approximately 1,300 adult foxes (Roemer et al. 1994). The relative influence of vegetation and habitat on population dynamics of Santa Catalina foxes was eventually overshadowed by the deep, disease-caused declines of 1999–2000 (Chapter 7). 3.3.2 Population density

Although the number of individuals in each subspecies is small compared to other small carnivore populations, the number of individuals per area (population density) is very high. Under stable conditions, island fox densities range from four to over 25 adults/km2 (Table 3.2), with the highest densities recorded on San Nicolas Island in 2002 (Garcelon and Schmidt 2005) and the lowest on Santa Catalina in 1989 (Roemer et al. 2004). Kit

30 · Island fox populations fox densities were reported as 0.2–1.7/km2 (White et al. 1996, Cypher et al. 2000, Lisk and MacDonald 2003), and swift fox densities as 0.1– 0.3/km2 (Harrison et al. 2002, Schauster et al. 2002b, Kamler et al. 2003). Many animals have higher population densities on islands than they do in mainland habitats, largely because reduced opportunities for dispersal favor (or lead to) individuals with a high tolerance of conspecifics (Adler and Levins 1994). This has been noted previously for island foxes (Moore and Collins 1995, Roemer 1999) as well as for insular Arctic foxes (Alopex lagopus semenovi; Goltsman et al. 2005).

3.4 Factors affecting population dynamics 3.4.1 Climate and weather

Island fox populations have fluctuated over time, and are influenced primarily by climate, disease, and density-dependence (Clifford et al. 2006, Bakker et al. 2009). El Ni˜no-Southern Oscillations (ENSO) are periodic climatic events that raise ocean temperatures. The most important effect of ENSOs for island foxes is an increase in annual precipitation (Rasmussen and Wallace 1983) that promotes plant growth and higher deer mouse populations on the islands (Schwemm 2008b). A direct correlation between deer mouse numbers and/or plant productivity on reproductive rates in island foxes has not been established, but ENSO events have been shown to affect carnivore populations in many terrestrial systems (Holmgren et al. 2001). Interestingly, Bakker et al. (2009) found two contrasting effects of ENSO events on island fox populations. On the southern islands of Santa Catalina, San Clemente, and San Nicolas – where weather patterns are generally warmer and drier than in the north – ENSO events had a positive but delayed impact on fox pup survival, likely because increased resource availability results from winter rains. But Bakker et al. (2009) also found a direct but negative effect of ENSO events on pup survival on the northern Channel Islands, where lower temperatures combined with higher rainfall may reduce juvenile survival over the first winter, although the effect of low temperatures alone or in combination with rainfall amounts on juvenile survival has not been investigated.

3.4.2 Disease

All island fox populations show evidence of previous exposure to CDV, and outbreaks of CDV or other diseases in the past may have caused

3.4 Factors affecting population dynamics

· 31

3.00

Pups/female

2.50 2.00

Plot area

1.50 1.00 0.50 0.00 0

100

200 300 400 Islandwide adult population

500

Figure 3.2 Island fox reproductive effort (pups/female, from annual trapping data) and estimated population size, San Miguel Island, 1993–2009.

considerable mortality, leading to bottlenecks and low genetic diversity (Wayne et al. 1991b, Clifford et al. 2006, Chapter 2). However, evidence of catastrophic disease outbreaks prior to the 1980s is generally lacking, so the effects of disease have not been considered in demographic modeling (Bakker et al. 2009), and it is not known how much disease has influenced historic dynamics of island foxes. An in-depth discussion of diseases in island foxes is presented in Chapter 11.

3.4.3 Density dependence

If the number of individuals in a population influences future growth, the population is subject to density dependence (Ranta et al. 2006). For example, reproduction often declines when there are more individuals in a population and competition for resources is high (Gotelli 2001). On islands, where space and dispersal opportunities are always limited, density-dependent mechanisms may have greater influence on population abundance than in unconstrained sites. On San Miguel, reproductive effort of island foxes, as measured by pups/female, declines with increasing population size (Fig. 3.2). Bakker et al. (2009) found strong densitydependent effects in island fox populations, although the effect differed in relation to island and size. All island fox populations experienced slower growth at higher densities, but at low densities growth rates were

32 · Island fox populations higher on smaller islands, suggesting an Allee effect (Holsinger 2000) response on the larger islands. Specifically, growth was slower on Santa Rosa Island (215 km2 ) at low densities than it was on San Miguel, which is much smaller (38 km2 ). Santa Rosa is also much more topographically diverse, with higher peaks and deeper canyons, than San Miguel, which is characterized by a generally open landscape. At very low densities, it may have been that although the same number of foxes was released on each of the islands, the animals released on Santa Rosa had more difficulty locating suitable mates in an unfamiliar and more physically challenging environment.

3.5 Modeling island fox population dynamics Measures of population density and abundance are useful for behavioral and ecological studies, but factors that determine population growth (which can be negative), such as survival and reproductive rates, are more often used to predict future change and to set population recovery goals (Morris et al. 2002, Coonan 2003, Bakker et al. 2009, Chapter 13). Demographic modeling and population viability analyses (PVAs) simulate population change over time using a range of possible values from measures of mortality, fecundity (generally how many young each female produces), and survival (previously described). An understanding of population dynamics is particularly important for a species like the island fox that naturally exists at relatively now numbers and thus is more at risk from stochastic (random) events that could lead to extinction. A recent PVA used to derive island fox recovery criteria for the recovery plan (Chapter 13) incorporated ecological drivers of population variability (e.g. climate and density dependence) as well as model uncertainty (the amount of variability that is as yet unexplained) to more effectively guide management and reduce extinction risk (Bakker and Doak 2009, Bakker et al. 2009). The PVA also resulted in information on possible historic drivers of island fox dynamics. One of the most reassuring conclusions of Bakker et al. (2009) was that although fox populations are naturally small and fluctuate considerably, they will be relatively safe from extinction risk once modest population sizes are obtained and golden eagle predation has been sufficiently mitigated. (As mentioned, the risk of disease was not included in the model.) The fact that island fox populations are relatively immune to low levels of eagle predation is fortunate, since eagle presence and predation may never be completely eliminated (Chapter 7).

3.6 Summary

· 33

3.6 Summary Island fox biology and population dynamics have been uniquely shaped by the relatively small islands on which foxes evolved. Because space is at a premium, island foxes have lower litter sizes, smaller territories, and higher densities than do similar species on the mainland. Island foxes evolved in isolation from most predators and diseases, but their small populations (compared to many similar-sized vertebrate species) have fluctuated dramatically over time. Population monitoring established in the late 1980s led to a greater understanding of population size and formed the basis for a comprehensive demographic modeling effort. Given the certainty of global climate change but the uncertainty of its effect on island foxes, these datasets now comprise a comprehensive demographic baseline to which future change can be compared.

4

r

Food habits, habitat use, activity patterns, and dispersal

Island foxes interact with their habitats in ways that illustrate the nearoptimal conditions of the environments in which they live. Foxes utilize nearly all available animal and plant resources, so there is ample food, and nearly all vegetation communities on the islands provide at least marginal if not high-quality habitat. Foxes adapted to the natural absence of predators by acquiring more diurnal (daytime) behavior patterns, and the combination of plentiful resources and tolerance for high population densities make foxes less inclined to disperse great distances in comparison to other species. This chapter examines the food habits, foraging behavior, habitat requirements, and dispersal tendencies of island foxes in wild populations.

4.1 Food habits 4.1.1 Animal prey

Like other members of the genus Urocyon, island foxes are omnivorous, utilizing both plant and animal material in their diets (Moore and Collins 1995). Deer mice (Peromyscus maniculatus ssp.) are the most common animal prey taken by island foxes. Deer mice are present on all of the islands, and are the only native rodent on Santa Rosa, San Miguel, and San Nicolas. On San Miguel, the percent occurrence of mice in fox scats in several studies ranged from 10% to over 90% across all seasons (Collins 1980, Crowell 2001). Although mice are a common prey item, the quantity of mice in fox diets does not necessarily reflect mouse availability. Mouse abundance is highest in the summer and fall, following spring reproduction, and low in the winter and early spring due to winter mortality. However, on San Miguel, Collins (1980) found that mice made up over 53% of the volume of winter scats, suggesting perhaps a greater need for protein during the winter and by females during gestation and lactation, or the relative unavailability of plant fruits at this time of year. Laughrin (1977) found low incidence of mice in fox scat from Santa

4.1 Food habits

· 35

Figure 4.1 Santa Cruz island fox with western yellowbellied racer (Coluber constrictor mormon). Courtesy of Catherin Schwemm.

Cruz across all seasons, likely reflecting the relatively low densities of mice on that island. The largest vertebrate prey of island foxes is California ground squirrels (Spermophilus beecheyi) on Santa Catalina. On Santa Cruz and Santa Rosa Islands foxes have killed and eaten island spotted skunks (Spilogale gracilis amphiala); however, observations of foxes hunting skunks are rare, and skunk deaths attributable to foxes are more likely a result of strong competitive interactions rather than direct predation (Jones et al. 2008b; see Chapter 14). Introduced vertebrates such as house mice (Mus musculus) and black rats (Rattus rattus) are taken by foxes when available (Phillips et al. 2007), and foxes may have important regulating effects on these species (Chapter 13). Reptiles and amphibians are also taken, including snakes (Fig. 4.1), but make up a very small portion of island fox diets (Moore and Collins 1995, Crowell 2001). The relative occurrence of bird material in fox diets varies among islands (Laughrin 1977, Moore and Collins 1995, Crowell 2001, Phillips et al. 2007). Horned larks (Eremophila alpestris) and western meadowlarks (Sturnella neglecta), both ground-nesting species, comprised nearly all of the bird prey remains recovered from scats on San Miguel (Collins 1980), and were also taken by foxes on San Nicolas (Kovach and Dow 1981).

36 · Island fox in their environment Foxes are known to predate nests of the endangered loggerhead shrike (Lanius ludovicianus mearnsi) on San Clemente (Chapter 5) and chukar (Alectoris chukar) chicks on San Nicolas (Kovach and Dow 1981, Garcelon et al. 1999). Eggs and chicks of many ground-nesting bird species may be a more common food item for island foxes than has been previously noted (Chapter 13). The proportion of insects taken in relation to other food items varies greatly between seasons, likely reflecting both the temporal aspect of insect development (i.e. emergence events) and the opportunistic approach of foxes to all resources (Roemer 1999). Foxes feed on insects from numerous taxa, including Diptera, Lepidoptera, Coleoptera and Orthoptera (Moore and Collins 1995, Crowell 2001). Jerusalem crickets (Stenopelmatus fuscus, Orthoptera) are large insects (> 25% the mass of a deer mouse) and are the most common invertebrate in fox diets. Collins (1980) found that Jerusalem crickets made up 5–17% of food by volume, while Crowell (2001) found them in 40% of scats across all seasons but in over 80% of winter scats. Introduced garden snails (Helix aspersa) are found in high densities in some habitats, particularly on San Nicolas, and may provide an important food source for older foxes on that island. The population age structure on San Nicolas is more skewed to older age classes (Garcelon and Schmidt 2005), and it is plausible that snails, which are easy to capture and consume, provide older individuals with nutrition that they might not otherwise be able to obtain (Garcelon and Hudgens 2008). 4.1.2 Plants

The relative amount of plant material taken by island foxes is greater than that taken by gray foxes (Laughrin 1977). Fruits of many plant species are consumed when available, in particular prickly-pear cactus (Opuntia sp.), sea fig (Carpobrotus aequilaterus), and toyon (Heteromeles arbutifolia; Crooks 1994, Moore and Collins 1995, Crowell 2001). Grass has also been found in fox scats (Laughrin 1977, Crowell 2001); however, grass is likely utilized more as a digestive tool than for nutrition (Collins and Laughrin 1979). The relative proportions of plant and animal material in fox diets change throughout the year, reflecting the ability of foxes to make efficient use of whichever resources are most available. The high proportion of plants and fruits taken in the summer when deer mice are normally very abundant reflects the highly omnivorous nature of island foxes; foxes are able to successfully exploit the seasonal availability of plant material and fruits, expending less energy than is required to chase and

4.1 Food habits

· 37

catch animal prey (McNab 1989). However, because the scat collected for analysis in some food habit studies was taken mostly from trails, the results may only represent the food habits of foraging adults and not pups. In early summer pups remain close to the dens, and during that time adults may catch mice that they do not consume but instead bring back to the dens for the pups (Garcelon et al. 1999). Those mice would not be present in adult scats but are likely an important component of pup diets that may increase pup survival (Laughrin 1977). 4.1.3 Feeding behavior

Island foxes are true generalists, and will feed on nearly any available meat, including carrion, intertidal invertebrates, crustaceans (Laughrin 1977), and human food. Foxes have been observed feeding on the carcasses of introduced mammals such as feral pigs and goats, and on non-native elk and deer during hunting seasons on Santa Rosa (behavior that has led to concerns about foxes ingesting lead; Chapter 6). On Santa Catalina a captive fox was even observed running up the chain-link fencing to snatch a bat after it alighted on the inside of the pen (R. Powers personal communication). If human food is available, foxes have little compunction about entering areas such as the busy campground at Scorpion Ranch on Santa Cruz and the military compounds on San Clemente and San Nicolas to be fed or forage for food scraps. ‘Fox boxes’ for food storage have been installed at the Scorpion Ranch campground to limit fox access to campers’ food, but foxes are minimally dissuaded, and commonly jump on picnic tables and coolers and even enter tents in their search for human food. While island foxes have adapted physiologically to a wide range of food types, their curious nature and agility also makes them efficient and clever in obtaining food, and they often appear to have a somewhat casual approach to foraging that suggests a certain lack of urgency (perhaps because food is generally plentiful and there is minimal interspecific competition). For example, foxes observed feeding on saltbush (Atriplex sp.) berries appeared to very carefully pull individual fruits (1–2 mm in diameter) off of the plants in what appeared to be a time-consuming process (G. Smith, US Navy, personal communication), and foxes have been observed rolling prickly pear cactus fruits in the dirt to remove the spines prior to eating (D. Garcelon personal observation). And in a behavior more akin to a domestic cat (Felis catus) than a canid, island foxes will play with their live food. On Santa Cruz a video (W. Joslin, National Park Service) captured a fox with a lizard that it threw around

38 · Island fox in their environment in the air for several minutes, before dropping it and moving on. Portions of lizards are often seen on trails, suggesting that foxes killed them but were apparently not hungry enough to consume them entirely. Island foxes are prodigious climbers and will readily scale trees and large bushes in search of food. They have been observed in toyon, oak (Quercus spp.), and island cherry (Prunus ilicifolia) trees and are capable of securing live fruits rather than merely scavenging fallen items. This well-developed climbing ability also allows them to predate bird nests. On San Clemente island fox predation on the nests of endangered island shrikes prompted the US Navy (USN) to initiate a program of predator (island fox) control (Roemer and Wayne 2003; see Box 13.1 ‘Island foxes versus endangered shrikes’ on pp. 155–156).

4.2 Dispersal and travel Animals move across a landscape for many reasons. One of the most important movement behaviors is dispersal, where an individual permanently leaves a natal or previous home range to establish a new home range. Dispersal distances for juvenile island foxes leaving natal territories are not well known, because it is difficult to discern whether observed movements represent true dispersal or some other behavior such as foraging or territorial defense (Clobert et al. 2001). It appears that when populations are at normal to high densities, young (non-reproductive) foxes disperse only short distances or stay within their natal range. Roemer et al. (2001a) found that of six juvenile foxes radio-collared on Santa Cruz, five moved a mean distance of only 1.0 km, and only one left its natal range, moving a distance of 3.6 km. On San Nicolas, where densities are greater than on other islands, a high percentage of pups remained philopatric to the place they were initially trapped; of 113 foxes first captured as pups between 2000 and 2003, 44% of the females and 71% of the males were recaptured on the same grid in subsequent years (Garcelon and Schmidt 2005). Dispersal distances are typically shorter in insular animal populations compared to non-island populations (Sullivan 1977), and the dispersal distances of island foxes are also less than for other small canids. Swift foxes in Montana dispersed an average distance of 10.4 km (Ausband and Foresman 2007a), and kit fox pups in California had a mean dispersal distance of 7.9 km (Scrivner et al. 1987). Although young island foxes generally move shorter distances from their parents compared to other species, such movement is still sufficient to prevent breeding among

4.3 Home range

· 39

closely related animals. Adjacent island fox pairs are not more closely related than more separated pairs, and mated pairs are less closely related than in many canid species (Roemer 1999). During the translocation program on Santa Catalina in 2001 and 2002, 22 juvenile foxes were moved from their natal range on the west end of the island to the east end, where population densities were extremely low (Chapter 10). The translocated animals moved an average of 5.6 to 5.9 km from the release site to where they eventually established a home range (Schmidt et al. 2004b). The large movements of these translocated foxes were likely due to the low densities and lack of existing fox social structure in the release areas. For the captive-born juveniles that were released from captivity and also taken to the east end, the mean dispersal distances were 4.5 km (2002) and 3.5 km (2003; Kohlmann et al. 2003, Schmidt et al. 2004b). The shorter distances may reflect behavioral differences between wild-born and captive-born animals, or the slightly increased densities that existed when the captive-born animals were released. On Santa Rosa, captive foxes (adults and juveniles) released into an almost completely foxfree landscape moved a mean distance of 6.7 km from their release site to the location where they eventually established territories. Foxes released on San Miguel under similar conditions moved an average of only 2.8 km, perhaps due to the smaller size of San Miguel (NPS unpublished data). In general, the relative influences of population density and habitat on dispersal processes as well as the importance of dispersal to population dynamics remain poorly understood (Waser 1996). Still, island foxes occasionally travel long distances in relation to island area (Roemer et al. 2001b). On San Clemente Island one adult male fox was captured on three different trapping grids within a one-month period, a cumulative distance of more than 9 km, and several other individuals traveled at least 2–3 km in the same season (Schmidt et al. 2005). On Santa Cruz Island, distances between capture sites for two adults were 1.9 km and 2.4 km over a 2–3-month period (Laughrin 1977). On San Nicolas, two adult male foxes traveled between two grids that were 5.5 km apart, in a three-week period (Schmidt and Garcelon 2003).

4.3 Home range Home range size for island foxes is determined by social and resource requirements (Crooks and Van Vuren 1996, Roemer et al. 2001b), both of which are influenced by island environments. Long-range immigration

40 · Island fox in their environment Table 4.1 Examples of documented home range sizes for island foxes. Island

Sample size

Average home range size (km2 )

Santa Cruz Santa Cruz

18 6

San Clemente Santa Catalina San Nicolas San Miguel Santa Rosa

74 10 4 10 18

0.41 Males: 0.34 Females: 0.33 0.53–2.38 0.42 0.40 1.18 2.47

Source Roemer et al. 2001b Crooks and Van Vuren 1996 Schmidt et al. 2004a Garcelon et al. 1991 Kovach and Dow 1985 Coonan, unpublished data Coonan, unpublished data

and emigration has little adaptive value on islands, and decreased home range size and higher densities are common for many insular populations of vertebrates compared to mainland populations (Stamps and Buechner 1985). Generally, home range area and body size vary proportionally (Harestad and Bunnell 1979), but the reduced home range size of island foxes compared to other fox species is less than can be accounted for by body size alone (Crooks and Van Vuren 1995; Table 4.1). Roemer (1999) found that home ranges on Santa Cruz prior to the decline averaged 0.55 km2 , and Crooks and Van Vuren (1996) found average home range size to be 0.34 km2 on Santa Cruz. Little is known about the importance of habitat type to home range size. On San Clemente the home ranges of foxes inhabiting grasslands were larger than those in more vegetatively diverse scrub habitat (Schmidt et al. 2004a); however, the relationship between vegetation diversity and spatial partitioning of the landscape within island fox populations remains unclear.

4.4 Habitat use Island foxes utilize all vegetation communities on the islands, including sand dunes, homogenous exotic grass and fennel stands, woodlands, riparian areas, coastal scrub, and chaparral (Laughrin 1977, Moore and Collins 1995, Roemer et al. 2001b). The composition of resources varies between habitats, but all vegetation communities on the islands support some foxes, and there are few areas that are not contained within at least one fox’s home range. Roemer (1999) found foxes most abundant in woodland and shrub communities, and habitat use may be determined

4.5 Activity patterns

· 41

largely by factors other than food availability. For example, deer mice and insects are often abundant in grasslands (Schwemm 2008a) but, on Santa Cruz, Roemer (1999) found that foxes avoided exotic annual grasslands compared to the native grasslands, possibly due to the dense, low-growing vegetation in those areas (Roemer and Wayne 2003). Trapped foxes have been observed with awns from non-native annual grasses (‘foxtails’) in their eyes and ears. Because island foxes did not evolve to be resistant to physical injuries from grass awns from introduced species (Avena, for example), grasslands may be less favored habitat not because there is less food but because of plant community structure. Few if any specific habitat characteristics appear to be necessary for either male or female foxes. Little is known about the importance of den availability within home ranges, but observations suggest that den sites are extremely variable. Laughrin (1997) reported dens in rock, brush and log piles, and manmade structures, but if none of these are available then foxes will dig a simple tunnel or scratch out a den under a shrub (Moore and Collins 1995). Some dens are re-used in subsequent years and some are not (Laughrin 1977), but there is no indication that the availability (or lack) of den sites influences home range selection.

4.5 Activity patterns Unlike most mammalian carnivores, island foxes are partially diurnal, meaning they forage and are active during daylight hours. Diurnal behavior allows a carnivore to exploit resources that are more readily obtainable during the day, but is only advantageous if daytime predators are absent. Island foxes evolved without any predators, and hence were able to successfully exploit both resources that are easier to obtain during the day (insects are more active and plants emit stronger odors), as well as nocturnally active animals (mice). Prior to the 1990s island foxes were commonly seen during the day, foraging and even resting out in the open. On Santa Cruz, Crooks (1994) found that prior to the onset of predation by golden eagles foxes were active (moving, as opposed to resting) at 60% of diurnal observations (0700–1900 hr) and at 75% of nocturnal (1900–0700 hr) observations. On Santa Catalina during the same years foxes were also active during daylight hours, but less so than on Santa Cruz; 17–85% of Santa Catalina diurnal observations were active, as were 65–88% of nocturnal observations (Garcelon et al. 1991). Island fox activity patterns may have changed, however, in response to golden eagle predation. During the early stages of the decline on Santa

42 · Island fox in their environment Cruz when predation was intense, Roemer (1999) found that diurnally active individuals were more likely to be taken by golden eagles. Swarts et al. (2009) reported that diurnal activity by island foxes decreased by 37% after colonization of Santa Cruz by golden eagles, and that island foxes on Santa Catalina – where there were no eagles but densities were similarly low – were more diurnally active than foxes on Santa Cruz. Whether foxes changed their behavior in response to daytime predation or whether predation simply removed those foxes more inclined to diurnal foraging is not clear. Swarts et al. (2009) suggest an inherited shift in behavior rather than a learned behavior, indicating a rapid evolutionary response by fox populations to a changing environment. Additional data (Hudgens and Garcelon in preparation) show both an increase in nighttime activity when compared to data from 1992, but also a recent increase in daytime activity by foxes tracked from 2003 to 2005, after most of the eagles had been removed, which indicates at least a partial role of behavioral factors.

4.6 Summary The ecology and behavior of island foxes are greatly influenced by their evolution in an insular system and the limitations imposed by a small geographic range. Abundant resources and the absence of predation and competition allow foxes to exploit a wide range of foods and exist in smaller home ranges. Dispersal distances for island foxes are less than for other similarly sized canids, reflecting both the availability of resources in all habitats and a behavioral response to limited space. Little is known about the relative quality of different habitats, and more work is needed on habitat use and island fox fitness, but at present there appears to be no difference in survival or fecundity across habitat types. The demonstrated ability of foxes to exploit a variety of island resources and habitats may allow them to better withstand the predicted effects of climate change on island ecosystems (Chapter 15).

5

r

Golden eagles and the decline on the northern islands

Island fox populations on San Miguel, Santa Rosa, and Santa Cruz Islands underwent severe and unprecedented declines in the 1990s. Island fox populations on all three islands were naturally small and had historically fluctuated, but as far as was known had never been as low as they were during this period and had never come close to extinction. The source (or sources) of the mortality that was causing the declines was at first elusive; a species-wide investigation revealed no evidence of disease, and no significant changes in food supply or habitat condition were detected. Radiotelemetry studies of foxes on Santa Cruz and San Miguel ultimately revealed evidence of predation by golden eagles, a species which had not previously bred on the islands and was, until this time, rarely observed. Fox mortality rates due to predation were so high that by 1999 the San Miguel and Santa Rosa fox subspecies were nearing extinction; on each of those islands total abundance had declined from approximately 450 and 1,500, respectively, to 15. The remarkable influence of golden eagle predation on island fox populations was due to the success of golden eagles in colonizing the northern Channel Islands (supported by introduced prey species), and the vulnerability of island foxes to predators with which they had not evolved.

5.1 Detecting change Prior to the 1990s, the island fox populations on San Miguel, Santa Rosa, and Santa Cruz were thought to have fluctuated over time, perhaps markedly, but never to near-extinction levels. Santa Cruz Island residents had noted periods of fox scarcity and abundance on that island, and in 1940 the Los Angeles Times erroneously reported that island foxes had mysteriously disappeared from Santa Cruz Island (Los Angeles Times, March 20, 1940). On San Miguel, island foxes survived a period of massive landscape change in the early 1900s when overgrazing by sheep combined with drought to convert much of the island’s shrub habitat

44 · Golden eagles to bare sand (Johnson 1980). Also, in 1939, the island ranch caretaker, Herbert Lester, reported that San Miguel island foxes were recovering from a previous decline he attributed to mange, or scabies, contracted from the domestic sheep on the island (Sumner and Bond 1939). Field surveys in the latter part of the twentieth century (Collins and Laughrin 1979) revealed island fox populations on all islands to be small (in the hundreds), compared to those of other mammals. Aside from scattered historical observations, very little is known about the historic abundance of foxes on Santa Rosa. With the advent of standardized grid monitoring in the 1990s (Chapter 3, and see Box 5.1 ‘Standardized island fox population monitoring’), decline of island foxes on the northern islands was detected first on Santa Cruz and secondarily on San Miguel. On Santa Cruz, Roemer (1999) had established three research sampling grids, each of which showed a substantial decline in the abundance of foxes between 1993 and 1995. Of 32 foxes radio-collared by Roemer, 21 had died (65%), and the survival rate for his study population fell from near 90% to less than 20% in two years. By 1999, there were no foxes left on one of the grids (down from an initial estimate of 7/km2 ), and Roemer estimated that the total population of Santa Cruz foxes had declined from 1,500 in 1994 to 135 in 1999 (Roemer et al. 2001a). When captive breeding eventually began in 2001, the population of the entire island may have been as low as 50–60 animals (Institute for Wildlife Studies, unpublished data).

Box 5.1 Standardized island fox population monitoring The development of a standardized, grid-based population monitoring method for island foxes in the late 1980s and early 1990s enabled estimation of density and islandwide population size across the range of the species (Roemer et al. 1994). The technique used large (40+) arrays of traps (Fig. 5.1) spaced 250 m apart and trapped consecutively for six nights. Population size and density were estimated with mark– recapture methods (White et al. 1982), which use the ratio of marked and unmarked animals to estimate the true number of animals on the grid. Island foxes proved relatively easy to handle compared to other carnivore species. They could be successfully trapped in boxtype traps, and could be handled without chemical immobilization (Fig. 5.2).

5.1 Detecting change

· 45

Figure 5.1 Location of large trapping grids used on San Miguel Island, 1993–1998. Courtesy of National Park Service.

Figure 5.2 Captured island fox, San Miguel Island, 1993. Courtesy of National Park Service.

46 · Golden eagles 500 450 400

Number of adults

350 300 250 200 150 Plot area

100 50 0 1993

1994

1995

1996

1997

1998

1999

Figure 5.3 Estimated islandwide population of island foxes on San Miguel Island.

The NPS began monitoring island foxes on San Miguel in 1993 as part of a long-term ecological monitoring program for terrestrial and marine resources (Davis et al. 1994), and the decline on San Miguel was first suspected after the third year of monitoring in 1995. The islandwide population estimate had increased from 400 to 450 from 1993 to 1994, but declined to 350 in 1995 (Fig. 5.3). The apparent decline in 1995 could have been due to natural population fluctuations, or was perhaps an artifact of adding a sampling grid in 1994 that supported very high densities (15.9 foxes/km2 ). But in 1996 the estimate was again lower, indicating a significant downward trend. The decline on San Miguel continued through 1998 when mark–resight techniques (different from the grid-trapping conducted in 1993–1997) yielded an islandwide population estimate of only 30 animals. In 1999, information from remote camera data estimated the number of remaining animals at 15, which was the number of foxes eventually brought into captivity (Coonan et al. 2005b). From 1994 through 1999 the San Miguel population had declined by 97%. Although grid-trapping was not conducted on Santa Rosa, transect trapping for a species-wide disease survey also showed evidence of a population decline on that island (Roemer et al. 2001a). Trap success

5.2 Determining the cause

· 47

Table 5.1 Estimated number of wild adult island foxes for the three northern island fox subspecies, pre-decline and post-decline. Island/subspecies

1994 estimate1

1999–2000 estimate2

San Miguel Santa Rosa Santa Cruz

450 1,780 1,465

15 15 50–60

1 2

Source: Roemer et al. (1994) Sources: Coonan et al. (2005b), D. Garcelon, Institute for Wildlife Studies, unpublished data

(the number of individuals caught per number of trap-nights) on Santa Rosa was low compared with trapping success data from other islands. Roemer et al. (2001a) obtained a trap success rate on Santa Rosa of 7.5% (9 individuals in 132 trap-nights) in 1998, considerably less than the 50% trap success recorded there by Laughrin in 1978 (Laughrin 1980), and less than the 20–40% trap success obtained on other islands during periods of moderate to high island fox density. The island fox population on Santa Rosa was estimated at 1,780 in 1994 (Roemer et al. 1994), although that estimate came not from trapping but from the application of density estimates from San Miguel and Santa Cruz to the island area of Santa Rosa (Roemer et al. 1994). Regardless, the Santa Rosa island fox population was apparently also in steep decline and, as on San Miguel, by 2000 there were only 15 individuals remaining on Santa Rosa. See Table 5.1.

5.2 Determining the cause Data obtained during the 1990s from island fox monitoring programs on Santa Cruz and San Miguel revealed substantial reductions in fox numbers, but they provided little information on how or why foxes were dying. It was also unclear whether the coincident declines on three physically separate islands were independent events with different root mortality sources, or whether they were related, with one driver across all islands. The earliest information on the source of mortality for island foxes came from Roemer’s research on Santa Cruz in the early 1990s (Roemer 1999), which tracked the fate of radio-collared foxes. Examination of retrieved fox carcasses (Fig. 5.4) revealed evidence of some type

48 · Golden eagles

Figure 5.4 Island fox carcass with signs of golden eagle predation (evisceration, degloving), San Miguel Island, 1999. Courtesy of National Park Service.

of predation as the direct cause of mortality. In most cases the dead fox’s skin had been stripped off and pulled back from the underlying bones and muscle, a process called degloving, which is characteristic of predation by raptors (Roemer et al. 2001a). Many carcasses had also been eviscerated, and some also had talon holes associated with hematomas (clotted blood), meaning that these foxes had been alive when attacked and not simply scavenged after death. Finally, feathers found at several of the carcasses were identified as golden eagle feathers (Roemer 1999). Although golden eagles had never bred on the islands and there had been few sightings historically, Roemer and his colleagues concluded that predation by golden eagles was responsible for the decline of the Santa Cruz island fox population (Roemer et al. 2001a). Because foxes on Santa Rosa and San Miguel were, at that time, not radio-collared, the direct cause of fox mortality on those islands was undetermined. Golden eagle predation was not initially suspected on these islands because golden eagles had rarely if ever been seen on San Miguel. NPS therefore began investigating other possible causes, including disease and the effects of weather on food availability (Coonan et al. 1998). Because population declines were initially noted in foxes trapped on the west end of San Miguel, a disease possibly transmitted by pinniped (seal and sea lion) populations on the western beaches was

5.2 Determining the cause

· 49

Figure 5.5 Blood sample drawn from island fox on San Miguel Island, 1997, for species-wide assessment of disease exposure.

considered. A survey of blood samples from all six island fox populations, however, revealed no previous exposure to any disease that could have caused this high number of deaths (Roemer et al. 2001a; Fig. 5.5). A meeting of disease and parasite experts was convened by the IWS in early 1999 to evaluate the potential role of disease in fox declines. The primary conclusion of those attending was that disease was probably not responsible for the high mortality rates currently being observed on the northern islands, and that there was no evidence that parasites could be responsible for so many fox deaths (Munson 1999; see Chapter 11). Poor nutrition, perhaps caused by decreased food abundance, was another possible cause of mortality. Changes in prey availability have been shown to drive population dynamics of other fox species, mostly through effects on reproduction, but this relationship is normally observed in fox species tied to single food sources (kit fox: Egoscue 1975, White and Ralls 1993; mainland Arctic fox: Macpherson 1969; insular red fox: Zabel and Taggart 1989; mainland red fox: Goszczynski 1989). Island foxes are highly generalist in their feeding habits, and no changes in abundance of mice, important vegetation species, or any other important food items appeared to have occurred prior to the declines on San Miguel (Coonan et al. 1998).

50 · Golden eagles

Figure 5.6 Radio-collared island fox. Courtesy of Matthew Hill.

With nutritional and disease effects eliminated, NPS began a radiotelemetry study in 1998 to track foxes and retrieve the remains of animals. From November 1998 through December 1999, 15 foxes were fitted with radio-collars (Fig. 5.6) and monitored regularly. Of eight foxes collared in 1998, six died within four months, and four of those displayed evidence of golden eagle predation identical to that observed on Santa Cruz (Coonan et al. 2005b). By the end of 1999 seven foxes had died (nearly 50%), with five of those deaths attributed directly to predation by golden eagles. Golden eagle feathers and uric acid deposits were also found at several of the kill sites and, unlike in prior years, golden eagles were now being seen on San Miguel. By 1999 there was little doubt that golden eagle predation was the primary mortality cause for island foxes on San Miguel and Santa Cruz, and presumably on Santa Rosa as well.

5.3 Golden eagle colonization of the northern islands To understand the relationship between golden eagle populations dynamics and the impacts of eagles on island foxes it is necessary to clarify the difference between (and relevance of) a breeding versus a non-breeding population of predatory birds. If birds are transient, they are either

5.3 Golden eagle colonization of the northern islands

· 51

traveling through an area or possibly stay in one area for a long period of time, but they do not mate, establish breeding territories, or produce offspring. Once breeding occurs, the impacts on prey species can increase dramatically, both because the bird population will likely grow and because greater food resources are now required to support breeding females and their young. Prior to the 1990s there were no records of breeding by golden eagles on the Channel Islands (Kiff 1980), and even sightings of individual golden eagles were rare. The first golden eagle nest found on the California Channel Islands was located during a helicopter survey for feral sheep at Coche Point on the west end of Santa Cruz in 1999 (Latta et al. 2005). This nest was subsequently excavated and contained golden eagle eggshell fragments and remains of common ravens (Corvus corax), gulls (Larus spp.), cormorants (Phalacrocorax spp.), feral pigs (Sus scrofa), and island foxes (Latta et al. 2005). The discovery of the Coche Point nest led to a dramatic change in the understanding of golden eagle dynamics on the Channel Islands. First, the existence of a nest meant that eagles were likely resident on the island year-round, not transient as had been previously thought. Further, the presence of both fox and piglet remains confirmed not only that eagles were preying on foxes, but also that pigs provided an additional food source necessary to sustain golden eagle reproduction (Roemer et al. 2001a, Roemer et al. 2002; see Box 5.2 ‘The hyperpredation model’). A total of five eagle territories, some with multiple nests, were eventually found on Santa Cruz, and golden eagles bred on that island until 2006 (Latta et al. 2005, Coonan and Dennis 2007). Box 5.2 The hyperpredation model The catastrophic decline of foxes in the northern Channel Islands has been described as a case of hyperpredation (Roemer et al. 2001a). Analogous to apparent competition (Holt 1977), hyperpredation occurs when a prey species that can sustain high predation rates subsidizes the extinction of another prey species by acting as an alternate food resource for a shared predator (Courchamp et al. 1999). In the case of the island fox, golden eagles successfully colonized the northern Channel Islands due to the availability of alien prey species (feral pigs). Golden eagles subsequently preyed heavily on island foxes, nearly driving two of three northern island subspecies to extinction (Table 5.1). A mathematical model of hyperpredation showed that pigs would have been necessary to support a large, resident

52 · Golden eagles golden eagle population. Although as few as half a dozen eagles could have caused the fox declines, the eagle population in 1999 was estimated to be at least 27 adult birds (Latta et al. 2005). Further modeling (Roemer et al. 2002) described the restructuring of the island ecosystem caused by the additions of feral pigs and golden eagles, pointing out that competitive release of island spotted skunks (Spilogale gracilis amphiala) occurred on Santa Cruz and Santa Rosa Islands following decline of island foxes. Skunks were not preyed upon as heavily as were foxes due to the nocturnal habits of skunks and their smaller size. Due to the dependency of golden eagles on feral pigs, Courchamp et al. (2003) predicted that removal of pigs would cause an increase in eagle predation on Santa Cruz Island foxes, to the point of possible fox extinction, reasoning that eagles would switch over to preying upon foxes in the absence of pigs, their primary prey. Because of these predicted effects, the authors advocated lethal removal of golden eagles to prevent extinction of island foxes. Golden eagle breeding was also confirmed on neighboring Santa Rosa Island when, in 2004, Collins and Latta (2006) excavated two Santa Rosa golden eagle nests and found seven distinct prey remain layers. Prey remains contain mostly bones, and each layer in a nest represents the types of food that were fed to chicks in a given year (see Box 5.3 ‘Golden eagle diet on the northern Channel Islands’). The oldest layers in the Santa Rosa nests contained fox bones, meaning that eagles had been nesting on Santa Rosa and preying on foxes since at least 1997. Box 5.3 Golden eagle diet on the northern Channel Islands Of considerable relevance to island fox conservation is the question of golden eagle food habits on the northern Channel Islands. That golden eagles preyed upon island foxes was apparent, both from island fox mortality studies (Roemer et al. 2001a, Coonan et al. 2005a, 2005b) and from initial observation of island fox remains in a golden eagle nest (Latta et al. 2005). It was less apparent how golden eagles could subsist on island resources, with the only terrestrial vertebrates in the rabbit– squirrel prey size range being island foxes and island spotted skunks. Successful golden eagle breeding is determined by prey availability, and the accrual of energetic stores adequate for reproduction (Kochert et al. 2002).

5.3 Golden eagle colonization of the northern islands

· 53

Although Roemer’s theory of hyperpredation (see Box 5.2 ‘The hyperpredation model’ on pp. 51–52, Roemer et al. 2001a) accounted for the successful establishment of golden eagles on the islands, empirical evidence to support the importance of alien ungulate species to island golden eagles could only come from evaluation of eagle prey remains. In 2004 Paul Collins of the Santa Barbara Museum of Natural History and Brian Latta of the Santa Cruz Predatory Bird Research Group proposed collecting and analyzing prey remains found in island golden eagle nests. Not only would this allow evaluation of the importance of alien ungulates in island eagle diet, it would also clarify any differences between food habits of golden eagles on the islands and the food habits of bald eagles, which were well known. Collins and Latta analyzed prey remains from eight golden eagle nests on Santa Cruz and Santa Rosa Islands (Collins and Latta 2006). Prey remains in golden eagle nests provide information on the types and abundance of prey that are necessary to fledge young from nests successfully. Feral pigs were the most important food item in Santa Cruz golden eagle nests, where they accounted for nearly 60% of the biomass consumed by eagles during the nestling phase. On Santa Rosa Island, mule deer fawns comprised almost 40% of the prey biomass and were, like pigs on Santa Cruz, the most important food item brought back to the nest. To some extent golden eagles adapted to the depauperate nature of island faunas by taking prey from less typical taxonomic groups. Island eagles took more birds than mainland eagles, perhaps due to the relatively high abundance of birds on the islands, and the lower abundance of other vertebrates. In island golden eagle nests, Collins and Latta recorded remains of: r landbirds: common ravens (Corvus corax), barn owls (Tyto alba),

California quail (Callipepla californica), western meadowlarks (Sturnella neglecta); r seabirds: cormorants (Phalacrocorax spp.), gulls (Larus spp.); and r waterfowl: mallards (Anas platyrhynchos), gadwalls (Anas strepera).

Three factors were likely responsible for the colonization of the northern Channel Islands by golden eagles in the 1990s. First, in the latter part of the twentieth century, golden and bald eagle (Haliaeetus leucocephalus) populations across western North America were recovering from decades

54 · Golden eagles of direct persecution, primarily from ranchers in response to perceived livestock depredation (Kochert et al. 2002). Protection of golden eagles provided by the 1962 Bald and Golden Eagle Protection Act resulted in population increases of golden eagles throughout the west and in California (B. Walton personal communication). As increasing golden eagle populations saturated existing habitats on the mainland, some eagles began dispersing to the islands. The western end of the Southern California transverse mountain ranges lie within Los Padres National Forest and contain large areas of golden eagle habitat. The islands lie well within range of golden eagles resident in these ranges (Urios et al. 2007). Dispersal to the northern islands provided the opportunity for range expansion for growing mainland golden eagle populations. (The southern Channel Islands never experienced this influx of golden eagles, likely because the mainland coast adjacent to the southern islands encompasses the heavily urbanized Los Angeles and Orange Counties, where golden eagle densities are very low.) The second in the chain of events that allowed golden eagles to colonize the northern Channel Islands was a greater abundance of prey than had previously existed. Golden eagles take prey in the 0.5–2.0 kg size range, which on the mainland typically includes small mammals such as rabbits and squirrels (Kochert et al. 2002). On the northern islands the only native species in that size range are island foxes (1.0–3.0 kg) and island spotted skunks (0.5–1.5 kg). But fox and skunk populations do not exist at densities high enough to support golden eagle breeding (Collins and Latta 2006). Feral pigs (European hogs) were brought to Santa Cruz Island in the 1850s as a food source for ranchers. By the 1990s estimates of feral pigs on Santa Cruz ranged from 1,000–5,000, with population abundance dependent primarily on annual rainfall (R. Klinger, US Geological Survey, unpublished data). Adult pigs are far too large (50–90 kg) to be taken by golden eagles, but piglets and juveniles are easy prey, and in high rainfall years several litters of piglets can be produced by one adult sow. On Santa Rosa, introduced ungulates also provided food for golden eagles. Non-native mule deer (Odocoileus hemionus) and elk (Cervus elephas) were brought to Santa Rosa in the early 1900s for hunting opportunities, and by the 1990s there were close to 1,000 of each species on the island. Deer and elk carcasses from annual hunts and culls, which occur in the fall, provided an additional food source for golden eagles in the winter, and mule deer fawns (elk calves are too large) were available in the spring for breeding adults and chicks in the nest. Golden eagles

5.3 Golden eagle colonization of the northern islands

· 55

dispersing from the mainland were therefore able to subsist on pigs and ungulates, while including foxes, skunks and other prey in their diet (Collins and Latta 2006). The extirpation of bald eagles from the Channel Islands in the middle of the twentieth century was likely a third factor that allowed golden eagle colonization. Bald eagles (which specialize on marine prey and are not known to prey on island foxes) existed in significant numbers on the Channel Islands until the mid-twentieth century, but had disappeared by about 1960 due to persecution and the effects of organochlorine pesticides (Kiff 1980). Dichloro-diphenyl-trichloroethane (DDT) was manufactured by the Montrose Chemical Corporation in Torrance, California, and from the 1940s to the early 1970s effluent from a Montrose production facility was discharged into the Santa Monica Bay near Santa Catalina Island. Eggshell thinning and subsequent nest failure in bird species such as bald eagles and seabirds that feed on marine-based prey can be caused by DDE (dichloro-diphenyldichloro-ethylene), a DDT metabolite with considerable environmental persistence. Recent nest failure in reintroduced bald eagles on Santa Catalina Island was associated with high levels of DDE some 30 years after release of DDT into the marine environment (Sharpe and Garcelon 2005). Bald and golden eagles are of similar size, but when their ranges overlap they rarely nest in the same areas. It is possible that territorial behavior by breeding bald eagles historically prevented dispersing or transient golden eagles from establishing territories (Roemer et al. 2001a, Collins and Latta 2006). The combination of expanding populations on the mainland, abundant food, and no competitors led to colonization of the northern islands by golden eagles in the mid-1990s (most likely in 1994 on Santa Cruz and 1997 on Santa Rosa; Roemer et al. 2001a, Latta et al. 2005). Latta et al. (2005) estimated that 27 golden eagles may have been present on the northern islands in 1999 as a result of breeding on Santa Cruz that likely began as early as 1995. On Santa Rosa two territories located in 2005 contained a total of four nests, and one (Trap Canyon) may have produced as many as seven young between 1996 and 2004. Latta et al. (2005) also found that adult members of golden eagle pairs trapped during the early phase of the golden eagle removal program (Chapter 6) were replaced by other birds very quickly (within 1–3 days), suggesting not only the presence of breeding pairs but also a considerable number of surplus, non-breeding adults. As the number of golden eagle territories and nests discovered continued to increase, it became clear that the high level of fox predation (initially attributed to a few non-breeding birds)

56 · Golden eagles

Figure 5.7 Island fox in alien annual grassland habitat, San Miguel Island, 1993. Courtesy of National Park Service.

was being effected by a large, highly productive and dynamic golden eagle population, recently established on the northern islands.

5.4 The vulnerability of island foxes to diurnal aerial predators Island foxes were particularly vulnerable to golden eagles for two reasons; not only had they evolved in the absence of predators, but grazing impacts of the previous century had reduced the availability of cover (shrubs and trees) across the landscape. Prior to the arrival of golden eagles (and in the absence of bald eagles) the largest aerial predators on the islands were red-tailed hawks (Buteo jamaicensis), which can prey on island fox pups but do not take adult foxes. Because there was historically no risk from predators, island foxes had also evolved to be more diurnal than most carnivores (Chapter 4). Golden eagles hunt during the day, so if island foxes had been primarily nocturnal – as are gray foxes and coyotes – their susceptibility to golden eagle predation would likely have been much reduced. Moreover, the extent of oak woodlands, pine woodlands, chaparral, and coastal sage scrub was considerably less when eagles arrived than it was prior to grazing. For example, close to 80% of

5.4 The vulnerability of island foxes to diurnal aerial predators

· 57

Santa Rosa is currently dominated by alien grassland (Clark et al. 1990), a vegetation community that provides considerably less cover from aerial predators than do shrub or tree communities (Fig. 5.7). Likewise, San Miguel is slowly recovering from a history of sheep grazing (late 1800s to 1950), and in the early twentieth century experienced a period of massive landscape stripping caused by overgrazing and extended drought (Johnson 1980). The relative scarcity of shrub and tree habitats on San Miguel and Santa Rosa may partially explain why eagle predation reduced island fox populations on those islands to such low numbers, while on Santa Cruz, where there were more areas of tree and shrub cover, the decline was less steep. Predation is a common mortality factor for small carnivores and for foxes in particular (Watson 1997, Ellis et al. 1999). Annual survival rates for mainland fox species such as swift foxes, kit foxes, and gray foxes are much less than for island foxes, with predation an important mortality factor for those species (Chapter 3). But mainland fox populations (and in theory all prey populations and their natural predators) have adapted to tolerate losses from predation, whereas the losses of individual island foxes to golden eagle predation were unsustainable. The impacts of golden eagle predation on the San Miguel, Santa Rosa, and Santa Cruz island fox populations provide a dramatic example of the effects of introduced predators, and even introduced ungulates, on ecosystems, particularly insular systems with naturally low diversity.

6

r

Ecosystem recovery Predators and prey on the northern Channel Islands

By 1999 monitoring and research had revealed both the deep decline of island fox populations on the northern islands and the primary cause: predation by golden eagles. Averting extinction for the three subspecies (on Santa Cruz, Santa Rosa, and San Miguel) would require quick and effective emergency actions to protect the remaining foxes (Chapter 8) and mitigate predation. Complete recovery to viable population levels was a distant goal at the time, but it was clear that persistence of island fox populations would require ecosystems that functioned as they had historically, i.e. without golden eagles. This chapter discusses the management challenges of removing golden eagles, as well as the substantial effort required to restore island ecosystems to conditions that would deter golden eagle establishment in the future. These ecosystem-level actions included the restoration of bald eagles to the islands, and the removal of feral pigs.

6.1 Golden eagle removal While the threat of critically low population size could conceivably be mitigated through captive breeding and reintroduction (Chapters 8 and 9), such efforts would be futile unless the threat from golden eagles was eliminated. Golden eagles had been live-captured and relocated from some areas in western North America to reduce depredation on livestock, with moderate success (Phillips et al. 1991). Brian Walton of the Santa Cruz Predatory Bird Research Group (SCPBRG) surmised that similar methods could be used to translocate golden eagles from the Channel Islands. In late 1999 SCPBRG began a program, funded by NPS and The Nature Conservancy, to survey and remove golden eagles from the islands, focused initially on Santa Cruz. Modeling (Roemer 1999) had suggested that as few as six eagles could have been responsible for the fox declines on the northern islands, and anecdotal observations and expert opinion seemed to corroborate that relatively few

6.1 Golden eagle removal

· 59

eagles were present at the time (1999–2000; B. Walton personal communication). It was also assumed that these were non-breeding birds, with low fidelity to the islands, and that they would be unlikely to return if relocated to high-quality mainland habitat. However, later analysis revealed that there were as many as 30 golden eagles present in 1999, that additional immigrating or transient birds were arriving annually, and that a number of golden eagles were in fact resident and breeding on the islands (Latta et al. 2005; see Chapter 5). Thus, what was initially conceived as a short-term, targeted program to relocate half a dozen golden eagles became an eight-year program that removed many times that number. 6.1.1 Approach and methods

The majority of attempts to capture golden eagles between 1999 and 2007 utilized a bownet trap, a dug-in semicircular steel bow which is hidden and baited, then triggered remotely to throw a net over a landed bird (Jackman et al. 1994). Bait used to lure eagles to the traps included carrion (dead feral pigs) and live piglets and rabbits. Bownet traps were placed in known eagle territories, and were set before sunrise and completely camouflaged with vegetation to give the appearance of bait lying on the ground. The first two eagles caught using this method were captured in late November 1999, and another 11 eagles were captured using bownets the following spring. Captured eagles (Fig. 6.1) were transported to the mainland, fitted with global positioning system (GPS) transmitters, and released in northeastern California. All 13 eagles translocated in 1999–2000 moved primarily to the north and east of their release sites (into western Nevada and southern Idaho and Oregon). None of the birds attempted to return to the Channel Islands, perhaps because the islands represented marginal-quality habitat compared to the habitat in which they were released (Latta et al. 2005). The high trapping success on Santa Cruz in spring 2000 was likely attributable to several factors (Latta et al. 2005). First, golden eagle density at that time was high, resulting in frequent encounter rates with trap sets, and eagles had not yet developed any wariness of traps. Second, prey were relatively scarce. A drought in 1999 had reduced reproduction in the pig population, and piglets were much less abundant than in previous years. Likewise the fox population had declined, due to predation, to 50–60 animals. Evidence that eagles were struggling to find food was provided by the poor nutritional condition of captured eagles, and the

60 · Ecosystem recovery

Figure 6.1 Golden eagle captured on Santa Cruz Island, 2000. Courtesy of Brian Latta.

fact that several captured birds smelled strongly of skunk. Golden eagles normally hunt during the day, so if eagles were in fact preying on skunks they were not only shifting their foraging behavior (from solely daylight to also at dawn and dusk), but were taking prey below their preferred size range (island skunks weigh 0.5–0.8 kg, one-fourth that of island foxes). Under conditions where wild prey was scarce, eagles were much more likely to be attracted to trap sets baited with carrion and piglets. Trapping efforts were suspended in the summer of 2000 due to a lack of funding and did not resume until spring 2001. During this period the pig population increased in response to high precipitation during the winter of 2000–2001, creating a surfeit of prey. Moreover, the eagles that remained were more wary of traps, resulting in lower capture success in 2001 (Table 6.1). By the spring of 2003 trap success had increased, due to the presence of less-wary eagles that had either dispersed from the mainland or were offspring of island pairs. Carrion-baited feeding stations for recently released bald eagles on Santa Cruz (see Section 6.2.3) also

6.1 Golden eagle removal

· 61

Table 6.1 Number of golden eagles captured, and estimated number of golden eagles remaining, on the northern Channel Islands after annual capture efforts. Data from Latta et al. (2005) and Institute for Wildlife Studies (2006).

2000 2001 2002 2003 2004 2005 2006 2007 Total

Eagles captured

Eagles remaining

13 6 3 9 7 3 3 0 44

10 8 14 13 11 3 2 1

served to attract golden eagles, and six of the nine golden eagles captured in 2003 were trapped at bald eagle feeding sites. Over the course of the removal program, golden eagles became increasingly difficult to trap, due to lower eagle densities, which made detection difficult, and increased wariness. The latter was due either to learning, or to the fact that less wary birds were caught first. Both the SCPBRG and the Institute for Wildlife Studies (IWS), which assumed responsibility for golden eagle removal efforts in 2005, began employing novel trapping methods as eagles became more difficult to capture. In 2005 and 2006 active golden eagle nests had been discovered on both Santa Cruz and Santa Rosa Islands, the presence of which offered opportunities to attempt capture of territorial birds near the nest using methods other than bownet trapping. Those attempts were largely unsuccessful, although in 2004 a dho-gaza net was used to capture an adult female on her territory. A dho-gaza is a see-through net deployed in front of a tethered (captive) eagle or owl. When placed near a nest in an active territory, the tethered raptor provokes a territorial response by a resident bird, which is ensnared when it attacks the bait. One of the more unusual approaches to catching golden eagles was the development of an ‘injecto-egg’ by IWS. In 2006 a live egg was

62 · Ecosystem recovery removed from a nest called the Christy nest (on the west end of Santa Cruz) and replaced with an artificial plasticine egg equipped with a remotely-controlled dart filled with an anesthetic. When the eagle returned to the eggs the dart/egg was remotely triggered. However, in the one instance where this method was tried, the eagle tipped the egg over as it settled to incubate the eggs and the dart fell sideways, failing to penetrate the eagle’s epidermis. The deployed hypodermic merely caused the egg to roll around in the nest, much to the consternation of the adult female. Ultimately, attempts to capture eagles at nests and with baits and lures in 2006 were largely unsuccessful, and the low overall success rate meant that trying to capture eagles had now become extremely time consuming and very expensive. Besides bownet and dho-gaza trapping, helicopter net-gunning (O’Gara and Getz 1986) was the only other method that succeeded in capturing golden eagles. In this method a helicopter forces an eagle to the ground, at which point it is captured via a net fired from a modified shotgun in the hovering helicopter. The successful use of net-gunning for island golden eagles was contingent upon the availability of appropriate aircraft and an experienced crew, as well the ability to locate territorial eagles that exhibited consistent use of a defined area. Two helicopter net-gun operations on Santa Cruz in 2002 failed to capture any eagles due to poor weather (dense fog), which grounded the helicopter, and the fact that the helicopter was somewhat under-powered and was essentially out-flown and out-maneuvered by targeted eagles. However, an operation organized by TNC in June 2006 utilized a more agile aircraft and succeeded in capturing both members of a nesting pair of eagles (Institute for Wildlife Studies 2006). Helicopter operations in 2007 failed to locate any more eagles, even though the loss of nearly 20 foxes on Santa Cruz to predation confirmed the continued presence of one or more eagles. 6.1.2 Lethal removal

By 2001 several years of intense efforts to capture golden eagles had resulted in the relocation of over 30 birds, but several eagles remained, and predation on foxes continued. Some members of the Island Fox Conservation Working Group (IFCWG; Chapter 13) concluded that eagles could not be completely removed by live-trapping methods alone, and that the complete removal of golden eagles was necessary for island fox recovery. Several IFCWG members called for either delaying the

6.1 Golden eagle removal

· 63

release of captive foxes until all eagles were removed, and/or initiating lethal removal of golden eagles (Coonan and Rutz 2002; see Chapter 9). A representative opinion was expressed by NPS biologist David Graber, who recommended lethal golden eagle control to the director of the NPS’s Pacific West Region, Jonathon Jarvis: Despite the regulatory and political obstacles, many of us also believe that now is the time to initiate the process to permit lethal removal of eagles at some future date if other measures have been exhausted. (D. Graber memo to J. Jarvis, June 27, 2001)

In 2002 the IFCWG (referenced in Coonan and Rutz 2003) recommended that the process to obtain a lethal take permit from the US Fish and Wildlife Service (under the Bald and Golden Eagle Protection Act) be initiated. The following year, the IFCWG’s endorsement of lethal removal was even stronger: Removal of golden eagles should be the highest priority management action, with a goal of no resident eagles, and lethal removal of golden eagles needs to be a management option. Eagle removal, and fox recovery, will likely fail without it. The tool might best be used in a selective fashion, for breeding females that have eluded capture by other methods. The approach needs to be adaptive in nature. The administrative process for obtaining a permit should be assessed, and the feasibility of lethal removal investigated. Above all, the process should be expedited. (Coonan et al. 2004, p. 10)

In July 2003 an internal ‘white paper’ on the efficacy of eagle removal was prepared by NPS and USFWS (US Fish and Wildlife Service) staff biologists (Coonan et al. 2003). The authors concluded that ‘the continuation of existing [eagle removal] methods is not fiscally sustainable, and offers island fox populations no hope of recovery to viable population levels’. Although the authors recommended pursuing lethal removal as a possible tool to eliminate predation, they conceded that: 1. USFWS had never issued a permit to kill golden eagles; 2. lethal take would require landowner permission, which was unlikely to be obtained from TNC or NPS; and 3. because by law lethal take could not be accomplished either by poisoning or from an aircraft, it would be very difficult to implement. Without poisoning or aircraft, birds would have to be shot from the ground so the technique would only be feasible for members of breeding pairs that had evaded capture by other means but were still closely tied to

64 · Ecosystem recovery a nest area. The method would likely have no success on non-breeding eagles. Additional pressure to pursue lethal means was exerted on the agencies from outside advocacy groups, some of whose members were also on the IFCWG. In December 2003 the Canid Specialist Group of the International Union for Conservation of Nature’s (IUCN) Species Survival Commission sent a letter to the NPS and posted on its website a position statement on island foxes and golden eagles, strongly advocating lethal removal (CSG letter to NPS, December 1, 2003): [W]e advise, in the strongest terms, that permission be sought to remove golden eagles from the northern Channel Islands by lethal means. Because lengthy administrative process would be required to approve such permission, the need to pursue this option must be seen as urgent. Lethal control could be targeted at particular animals that have repeatedly evaded capture, or used under ‘emergency’ conditions should fox mortality rise in the course of pig removal. We fully appreciate concerns about killing a protected species that is emblematic of the wild places that conservationists seek to protect. However, we also recognize that, unless golden eagles are removed completely from the northern Channel Islands in the very near future, the prospects for recovery of the three northern subspecies are close to zero. We view this dilemma with the deepest regret, but can see no alternative to extreme actions, due to the circumstances that have undoubtedly been brought about by earlier human activities.

In the spring of 2004 the regional directors of the four principal management agencies – TNC, NPS, USFWS, and California Department of Fish and Game – met on Santa Cruz Island to discuss the situation and decide whether to pursue lethal removal of golden eagles. Collectively, the directors quickly decided that lethal control of golden eagles would not be pursued. Although they acknowledged that it made sense, biologically, to pursue lethal removal, the directors felt that there would be little support for such action at the Washington level, and issuing a lethal take permit for golden eagles to any entity would undermine ongoing efforts to protect eagles from lethal take in other circumstances (for example, requests from ranchers to kill golden eagles for livestock protection). Along with their decision not to pursue lethal methods to remove eagles, the directors encouraged the agencies to continue to use all possible live removal methods. By 2004 the IFCWG had been incorporated into the USFWS island fox recovery team (Chapter 13), and lethal removal was taken up by the team as part of a Technical Analysis Request. A group of experts within

6.1 Golden eagle removal

· 65

the recovery team was tasked with analyzing the efficacy of golden eagle control and providing recommendations for future efforts. The group’s response was communicated by USFWS to NPS and TNC in October 2004 (USFWS letter to NPS, USFWS, January 14, 2005). The analysis response contained a strong endorsement of lethal removal, but only as a last resort and under strict conditions: The live-capture program has yielded diminishing returns in recent years, and it may be that live capture efforts will be relatively ineffective in decreasing fox mortality to the point where wild populations are stable or increasing, and that released foxes survive at acceptable rates. If foxes decline due to increased eagle predation in the next several years, and the agencies have not begun the process to obtain permits for lethal control, then effective eagle control would be further delayed. This could also result in the need to manage increasingly larger captive fox populations and lead to spiraling costs and complex logistics. If releases were inadvisable due to eagle predation, foxes might have no future in the wild and would necessarily have to go to zoos. The initiation of any effort to lethally control golden eagles on the northern Channel Islands will likely bring a swift response from parties that would object to the killing of any wild bird of prey, and especially an eagle. However, because of the likelihood of time delay between when a permit is requested and when it would be granted (if such a permit would be granted), consideration should be given to initiating that process as soon as possible . . . Even with the permit, lethal measures should not be implemented unless other measures were unsuccessful or if island fox predation rates exceeded an acceptable threshold.

Despite this endorsement, lethal removal of eagles was not pursued by the agencies, which stood by the directors’ decision of spring 2004. Few eagles were captured after this, although significant resources were directed toward live-capture removal. TNC and NPS spent approximately $600,000 on eagle removal efforts in 2005–2006 (Coonan and Dennis 2006, 2007) but only seven eagles, including three hand-captured nestlings, were removed from the islands. It is doubtful that implementation of lethal methods in 2005 and 2006 would have increased removal success, given the low number of eagles remaining and the difficulty of finding those that were present. By 2007 only one eagle was thought to remain on Santa Cruz, and no eagles were observed after February of that year. Removal efforts were thus greatly curtailed due to the high cost and relatively low success, and the continuing increase in fox survival on Santa Cruz. In 2001 annual survival on Santa Cruz was 60% (meaning six out of every 10 adult foxes survived from one summer to the next), and by 2003 it was over 80%

66 · Ecosystem recovery (Coonan et al. 2005b). By 2007 it was still above 70%, despite the loss of 20 radio-collared foxes to predation that year. By 2009 island fox survival on Santa Cruz was above 90% (TNC unpublished data), and predation had been virtually mitigated without resorting to lethal control of eagles.

6.2 Long-term ecosystem recovery actions The importance of feral pigs to golden eagle breeding success on the northern islands suggested that simply relocating the existing golden eagles would not assure island fox recovery; the substantial prey base provided by pigs would continue to support breeding by any golden eagles dispersing from the mainland. The IFCWG (Chapter 13) recognized the importance of implementing longer-term actions that would prevent sustained use of the islands by golden eagles, noting that ‘increased exotic prey availability (e.g. pigs) have unnaturally increased the prevalence of golden eagles on the islands’ (Coonan and Rutz 2001). (Later it would be understood that the non-native mule deer population on Santa Rosa served a similar function.) The IFCWG also suspected that the absence of bald eagles from island communities further contributed to golden eagle success, and recommended in 1999 that: . . . [managers] aggressively move forward with the reintroduction of bald eagles on the northern islands. This recommendation is based on findings that bald eagles can deter golden eagles from establishing, that bald eagles pose a negligible threat to island foxes, and that bald eagles were a historic component of the island fauna. (Coonan and Rutz 2001, p. 24)

Although the major ecosystem restoration actions of pig removal and bald eagle reintroduction had been planned prior to island fox decline, the awareness that those actions would support island fox recovery broadened support for them and hastened their implementation.

6.2.1 Removal of feral pigs from Santa Cruz Island

NPS had identified removal of feral pigs from Santa Cruz as a management goal as early as 1985 (National Park Service 1985). Additionally, USFWS listed impacts from feral pigs as factors in the decline of nine species of Federally Endangered plants on Santa Cruz (US Fish and Wildlife Service 2004). NPS had removed nearly 1,200 pigs from Santa Rosa Island in a four-year program ending in 1993 (Lombardo and

6.2 Long-term ecosystem recovery actions

· 67

Faulkner 2000), but the greater size and scope of the effort that would be necessary on Santa Cruz required a cooperative effort between NPS and TNC. The interest of TNC in achieving pig removal as a fox recovery action provided the opportunity to implement an adequately funded cooperative removal project, and planning for the project began in 2002 (National Park Service 2002). Although removing pigs would support long-term island fox recovery, pigs were also the primary prey of golden eagles, and complete and rapid removal of pigs would put island foxes at increased risk if it resulted in increased predation. The extent to which pig removal would affect foxes was uncertain. One model (Courchamp et al. 2003) predicted outright extinction of island foxes if all golden eagles were not removed prior to pig eradication. A limitation of the model was that it required eagles to completely shift to foxes in the absence of pigs (Dratch et al. 2004). In fact, the eagle diet on Santa Cruz Island was much more catholic, and included ravens, gulls and skunks in addition to foxes and pigs (Collins and Latta 2006), so the conclusion that eagles would shift directly to foxes if pigs were unavailable was unsubstantiated. The model also did not take into account the large captive population of Santa Cruz island foxes on Santa Cruz in 2005 that served as a hedge against extinction, nor did it consider that the management agencies did not choose to implement one action over the other. Pig removal was implemented in the latter stages (2005–2006) of a continuous, seven-year eagle removal program. Finally, the model did not consider the role that the growing bald eagle population (at this time numbering over 30; see Section 6.2.3) might play in discouraging golden eagle use of the islands. Pig removal began in 2005 with contracted hunting in fenced zones across Santa Cruz Island. In less than 18 months over 6,000 pigs were removed, and by June 2006 no pigs remained on the island; the island was free of pigs for the first time in over 150 years (Morrison et al. 2007). By the end of the pig removal program the population of adult island foxes on Santa Cruz was estimated to be 264 individuals and increasing (Schmidt et al. 2007a). Foxes did not go extinct as Courchamp et al. (2003) had predicted, likely because the number of eagles remaining on the island was far fewer than assumed, and because of high growth rates in the wild fox population. However, the one eagle pair that attempted to breed on Santa Cruz Island in 2006, as the pig population was approaching zero, preyed extensively upon island foxes during the nestling phase (March– May), when young eagles require small vertebrate prey. The remains of at least seven individual foxes were found in the pair’s nest, and the

68 · Ecosystem recovery

Figure 6.2 Mammal bones, including those of mule deer fawns, found in golden eagle nest on Santa Rosa Island. Courtesy of Brian Latta.

rate of predation by this eagle pair was greater than it was before pig removal began (Collins et al. 2009). The absence of impacts to foxes at the population level is due to the success of golden eagle removal efforts; few eagles remained on the island during pig removal.

6.2.2 Removal of mule deer from Santa Rosa Island

Mule deer and elk had been introduced to Santa Rosa Island in the early twentieth century, and by the late 1990s both species were being hunted in a commercial operation run by the former owners of the island, the Vail and Vickers Company. Feral pigs had also been present on Santa Rosa prior to their removal by NPS in 1993. Golden eagle nests excavated on Santa Rosa did not include pig remains, meaning that eagles did not begin breeding until after the pigs were gone in 1993 (probably around 1997; Latta et al. 2005, Collins and Latta 2006). All of the Santa Rosa nests, however, had remains of mule deer fawns (Fig. 6.2), which comprised 35% of the total prey biomass found. By

6.2 Long-term ecosystem recovery actions

· 69

the time eagles arrived on Santa Rosa, deer and elk numbered in the thousands and provided a rich food base for colonizing eagles. Under the terms of a court-ordered settlement negotiated in 1997, deer and elk were to be managed by the Vail and Vickers Company at levels that would minimize impacts to threatened and endangered plant species, until their eventual removal in 2011. From 1999 through 2007 deer management on Santa Rosa was intended to maintain a maximum of 425 animals (through hunting and culling), although in reality annual reproduction caused seasonal increases to bring pre-hunt numbers well above this threshold (NPS unpublished data). Under the court-ordered settlement, the former owners began phasing out mule deer, as well as elk, in 2008. Because of the threat that deer and elk pose as prey for golden eagles, the complete removal of ungulates from Santa Rosa has been identified by NPS as an action required for island fox recovery (Coonan and Dennis 2007). 6.2.3 Reintroduction of bald eagles

Bald eagles were reintroduced to Santa Catalina Island beginning in 1980 by IWS as mitigation for decades-long dumping of DDT into the southern California Bight by the Montrose Chemical Company (Sharpe and Garcelon 2005). Bald eagles began breeding on Santa Catalina in 1987, but reproductive failure continued as DDE-caused eggshell thinning prevented eggs from being hatched. This prompted IWS to begin a program of nest manipulation wherein they removed the eggs from nests, incubated them on the mainland, then returned the chicks to the island. By 2005, there was a breeding population of 15–20 bald eagles on Santa Catalina Island, although persistence of DDE in the environment had prevented most pairs from successfully fledging young unassisted. The low breeding success on Santa Catalina prompted the Montrose Trustee Council, which directs mitigation efforts for the settlement case, to fund a bald eagle restoration effort on the northern Channel Islands. The latter are further from the Montrose effluent site than Santa Catalina, and so bald eagles on the northern islands might have less exposure to contaminated prey. From 2002 through 2006, 64 bald eagle nestlings – taken from wild nests in Alaska or hatched at the San Francisco Zoo – were released on Santa Cruz (Fig. 6.3; Dooley et al. 2005). In 2009 over 30 of the released bald eagles were still alive, and the first successful nesting

70 · Ecosystem recovery

Figure 6.3 Nestling bald eagle about to be placed in hack tower on Santa Cruz Island, 2002. Courtesy of National Park Service.

(Fig. 6.4) was documented in 2006. However, breeding success has been sporadic, and bald eagles on the northern islands still have relatively high contaminant levels (P. Sharpe, IWS, unpublished data). The potential for long-term reproductive success of bald eagles on the northern islands is unknown, but the presence of bald eagles may have helped prevent further colonization of the islands by golden eagles. In 2006 a bald eagle was observed ‘escorting’ a golden eagle out of a presumed bald eagle territory on Santa Rosa (B. Latta personal communication), and young, released bald eagles were observed chasing adult golden eagles off carcasses left for bald eagles on Santa Cruz (although of course the bald eagles may simply have been hungrier than the golden eagles). Bald eagles specialize in marine prey such as marine fishes, seabirds, and pinniped carcasses (Sharpe and Garcelon 2005), and so presumably pose little threat to island foxes. There are no indications from the 26 years of bald eagle reintroductions on Santa Catalina that bald eagles there have ever preyed on island foxes (IWS unpublished data). Similarly, a bald eagle nest on San Miguel Island that was occupied for at least 60 years contained the remains of only two island foxes, and those

6.3 Summary

· 71

Figure 6.4 Bald eagle chick born on Santa Cruz Island in 2006; it was the first bald eagle chick born on the northern Channel Islands in over 60 years.

animals may have been scavenged rather than predated (Collins et al. 2005). However, downy feathers found at two fox kills on Santa Rosa (one in 2005 and one in 2008) were later identified as bald eagle feathers by genetic analysis (Sandra Talbot, US Geological Survey, personal communication). At one of those sites additional feathers tentatively identified as golden eagle were also found, and it is unclear whether bald eagles were responsible for those two mortalities or whether they were scavenging on remains from a golden eagle kill.

6.3 Summary Active habitat restoration has become increasingly important for endangered species recovery (Carroll et al. 1996, Foin et al. 1998). Singlespecies actions are sometimes insufficient to ensure recovery, and must be accompanied by ecosystem-level actions that seek to remedy the root causes of endangerment. This was the case for island foxes on the

72 · Ecosystem recovery northern Channel Islands, where predation by golden eagles was supported by an altered ecosystem. The two emergency recovery actions recommended in 1999 by the IFCWG were to establish captive breeding facilities for foxes and to remove golden eagles. The removal of golden eagles proved difficult, but was ultimately successful. However, the longterm success of reintroduction, and of island fox recovery, will hinge on whether golden eagle predation can be permanently mitigated. Preventing future golden eagle utilization of the islands required a change in the ecological conditions that had supported golden eagle breeding on the islands in the 1990s. Feral pig removal and bald eagle restoration, completed by 2006, shifted the ecological balance in favor of island foxes and against golden eagles, and set the stage for success, or failure, of the most complicated and uncertain of the recovery actions: captive breeding and reintroduction. Without the ecosystem-level changes of pig removal and bald eagle restoration, fox recovery was not assured. That those two ecosystemlevel actions were implemented in conjunction with island fox recovery actions was fortuitous. The two actions had been planned and funded separately from island fox recovery, but the value of those planned actions to island fox conservation was recognized when the complicated relationship among foxes, pigs, and eagles became apparent. The success of those ecosystem-level actions was facilitated by the islands’ isolation and their limited, more defined ecosystems: once pigs were removed from the islands, there was little chance of them returning. The success of these ecosystem-level actions for island fox conservation underscores the importance of considering ecosystem-wide conditions in endangered species management.

7

r

Disease and decline on Santa Catalina Island

The events that led to the near-extinction of island foxes on Santa Catalina Island were coincident in time with the declines in the north, but were ecologically unrelated. While the cause of northern island mortalities was predation, on Santa Catalina it was disease that was introduced to the island ecosystem most probably by a raccoon (Procyon lotor) or domestic dog infected with canine distemper virus (CDV) that was brought intentionally or by accident to the island. Santa Catalina is the only one of the eight Channel Islands with permanent residents, and Avalon, the only town, has a year-round population of over 3,500 and over 1 million tourists that visit mostly in the summer. Along with humans have come their pets, and the wild species associated with human habitations such as rats and raccoons. Santa Catalina’s natural resources are managed by the non-profit Catalina Island Conservancy (‘Conservancy’), and the recovery efforts and conservation of island foxes on Santa Catalina were coordinated by the Conservancy and their primary contractor, IWS. In 1999 the fox population on Santa Catalina declined by > 90% due to CDV, a disease for which at the time there was no tested vaccination for use in wild animals. Consequently a new set of strategies and responses was needed to address this second but also unique threat to island foxes.

7.1 Declining populations In the late summer of 1999, Conservancy staff members became concerned about the low number of fox sightings on the island. Particularly worrisome was the scarcity of fox observations at night, particularly on the eastern portion of the island. (Santa Catalina is geographically separated into the east and west ends, with a narrow isthmus dividing the two regions; the east end is much larger, accounting for nearly 65% of the island (Fig. 7.1)). Directed searches for fox signs along roads, normally a good location to find tracks and scat, were also largely unsuccessful. IWS had conducted transect trapping of foxes in the summer of

74 · Disease and decline on Santa Catalina Island

Figure 7.1 Map of Santa Catalina Island showing eastern and western portions, separated by an isthmus. Courtesy of Institue for Wildlife Studies.

1998 to collect blood samples for an inter-island disease survey (Roemer et al. 2000), and at that time all of the foxes caught appeared healthy, and the 26% trap success rate achieved was similar to results from trapping efforts in 1989 and 1990 (Timm et al. 2009). Population monitoring prior to 1998–1999 had also indicated that foxes were well distributed across most areas of the island, so their near-total absence in 1999 from the east end was particularly alarming. Due to the growing concern, the Conservancy funded IWS to assess the status of the islandwide fox population via trapping. In addition, blood samples were collected from foxes to test for previous exposure to various canine diseases, as well as to evaluate the overall health of both the east and west populations. Fecal samples were collected to check for endoparasites, and 12 foxes (four from the east and eight from the west) were trapped and fitted with radio collars to track future mortalities and recover carcasses for necropsy. Results showed that fox population densities were very different between the east and west ends; on the eastern two-thirds of the island trap success was only 1.7% (10 individual foxes caught during 1,046 trap-nights), while on the west end trap success was about average for

7.3 Was disease the cause of the decline?

· 75

that area at 45.2% (49 individuals in 140 trap-nights; Timm et al. 2009). Moreover, six of the 10 animals caught on the east end were trapped near the isthmus, immediately adjacent to the west end. These results strongly suggested the presence of a contagious disease that had been introduced into the population on the east end and had spread to the west until it reached the isthmus. The highly populated town of Avalon is located on the east end, and it was further suspected that the origin of the disease, if disease was in fact the cause of the mortalities, came from a domestic dog or cat near that area.

7.2 Initial results While trapping data clearly showed a dramatic loss of foxes from the east end, the results from the blood surveys were ambiguous; animals from both the east and west ends of the island tested positive for exposure to CDV (specific results are discussed in Chapter 11 and in Timm et al. 2009). Prior surveys had suggested that Santa Catalina foxes had not previously been exposed to CDV (Garcelon et al. 1992), so if the virus was now present across the island, why had it only affected foxes on the east end? (Later analysis with more advanced testing would reveal that Santa Catalina foxes had in fact had previous exposure to CDV [Clifford et al. 2006], but this was not apparent in 1999.) At the same time island residents reported seeing foxes that appeared disoriented, evidence that further suggested the presence of disease. In July 1999 a dead fox was found at the dump near the city of Avalon and was immediately turned over to IWS staff, but because the decline had yet to be detected the carcass was frozen and stored on the island. Later analysis showed evidence of septic pneumonia secondary to viral infection in that individual, and also the presence of CDV inclusion bodies (a confirmation of CDV infection; Timm et al. 2009; see Chapter 11).

7.3 Was disease the cause of the decline? Biologists hypothesized that the fox decline on the eastern portion of the island had occurred over a relatively short period of time, i.e. between the summer of 1998, when trapping results were similar to those of past years, and the fall of 1999. It is possible that the decline was already underway in the summer of 1998 but had not yet reached the areas where trapping was conducted. Rapid population declines of this magnitude in wild carnivores are consistent with highly virulent diseases, such as CDV

76 · Disease and decline on Santa Catalina Island and rabies (Woodroffe et al. 2004). CDV in particular is a highly infectious and contagious disease affecting wild and domestic canid species, as well as members of the mustelid, raccoon and cat families (Appel et al. 1995). Domestic dogs can carry CDV and are thought to have spread the disease to African lions (Felis leo; Roelke-Parker et al. 1996) and wild dogs (Lycaon pictus; Alexander et al. 1996). Residents and visitors to Santa Catalina are allowed to have dogs as pets, and neither California state law nor Avalon city ordinances require dogs to be vaccinated against CDV. Canine distemper is rare in domestic dogs because pet owners typically vaccinate their dogs for CDV as part of a parvovirus/CDV/adenovirus combination vaccine recommended by veterinarians; however, only a rabies vaccination is required to obtain a license for a pet dog in California. Consequently, biologists and managers on Santa Catalina assumed that a domestic dog infected with CDV had come into contact with a fox on the east end and transmitted the virus, from which point it spread rapidly within the fox population. All other potential factors that could have caused the decline were considered, but were unsupported by the evidence. For example, if golden eagles were responsible, as they were on the northern islands, high mortality would have almost certainly been observed in the west end foxes. A reduction in food resources would have been unlikely to cause such a rapid decline, especially considering the generalist food habits of foxes (Chapter 3); and, again, such an effect would have been apparent across the island. Consequently managers moved forward under the assumption that a highly infectious disease was the primary cause of the decline, but continued to monitor the population for indications of any other contributing factors.

7.4 Recommendations for population recovery Based on the assumption that an outbreak of CDV was causing the decline of the Santa Catalina fox population, IWS recommended to the Conservancy and the California Department of Fish and Game (CDFG) that: 1. a large portion of the remaining foxes on the island be vaccinated against CDV to help reduce the effect of another outbreak; 2. fox population monitoring using trapping and radio telemetry continue to determine population status and trend and to document mortality events and recover carcasses;

7.5 Testing CDV vaccine

· 77

3. a captive breeding facility be established to help support repopulation of the eastern portion of the island; 4. juvenile foxes be translocated from the western to the eastern portion of the island to augment recovery efforts (Timm et al. 2000). Implementing the proposed actions would be extremely challenging; the cost for all the suggested efforts was estimated at $1.3 million over six years, and several of the proposed actions, especially the vaccination protocol, had never been attempted with island foxes. In addition, because the island fox was listed in California as a threatened species, state biologists and veterinarians would have to sanction all of the proposed activities. Holding and breeding island foxes had only recently begun on the northern islands (Chapter 8), so husbandry methods were still being developed and, challenges notwithstanding, there was an inherent urgency to the planning: if healthy animals were not caught soon, or if the vaccination program was unsuccessful, eventually an infected animal would cross the isthmus and introduce the disease into the western population. The Conservancy supported all the suggested actions, a move that was pivotal to the eventual success of the program. Consequently, IWS moved ahead with construction of the captive facility and the initiation of vaccine trials.

7.5 Testing CDV vaccine Vaccinating a large portion of the remaining wild population would require the concurrent implementation of several tasks. First, a vaccine had to be developed that was suitable for use on island foxes. The modified-live CDV vaccine strains that were commonly used on domestic dogs had caused infections and some mortalities in gray foxes (Hallbrooks et al. 1981). This vaccine had also caused mortality in mink (Mustela lutreola; Sutherland-Smith et al. 1997), black-footed ferrets (Carpenter et al. 1976), and the lesser panda (Ailurus fulgens; Bush et al. 1976). An island fox transferred from San Clemente Island to the Hogle Zoo in Utah in 1999 was vaccinated with the modified-live virus vaccine and later contracted CDV and died (L. Munson personal communication). As the ‘killed’ CDV vaccine was no longer available, and provided significantly lower antibody response compared to the modified-live vaccine (Montali et al. 1983), the program was in need of a vaccine that apparently did not yet exist. IWS staff began investigating the possibility of using an experimental CDV vaccine currently being tested and evaluated at the

78 · Disease and decline on Santa Catalina Island Smithsonian National Zoological Park. The vaccine was a Canary Pox virus recombinant canine distemper virus – canine parvovirus (CDVCPV) produced by Merial Limited Company (Duluth, GA). Inclusion of only a segment of the CDV genome into the Canary Pox virus resulted in a vaccine that fails to replicate in mammals, while allowing for initiation of protein replication of the CDV (Timm et al. 2000). Although the vaccine was still in the testing phase, trials suggested that the vaccine would be relatively safe and efficacious even in sensitive species, and Merial agreed to make the vaccine available for testing on island foxes. Testing the vaccine required a protocol that would both determine the efficacy of the vaccine in providing protection to foxes while protecting study animals from infection. It was generally agreed that the risk to foxes from the trials themselves was low, but precautions were necessary to prevent the test animals from either being infected by wild foxes or from cross-contaminating each other, thereby confounding the experimental results. It was also important to ensure that in the unlikely event that a fox did become infected through the vaccine that the animal would not escape and infect other individuals. To hold foxes during the trials, six individual pens were constructed and surrounded by a perimeter fence, both to contain animals if they escaped their individual pens and to prevent contact with wild foxes or other species that might approach the perimeter fence. A skirting was buried below the fence to discourage digging, and a 300 mm overhanging wire was placed to prevent foxes on the outside of the facility from climbing on top. Discussions began with staff and veterinarians from CDFG to explain the testing protocol and obtain permission to proceed. CDFG personnel were understandably concerned about injecting an experimental vaccine into a state-listed threatened species, especially one that had already undergone a catastrophic decline. In a joint decision with the project team, it was determined that the most likely ‘worst case scenario’ would simply be a failure of the vaccine to initiate a response in foxes. Conversely, if the foxes mounted an antibody response, the vaccine would be a success. To begin the vaccine trial, six foxes – three males and three females – were taken from the wild population on the west end. To minimize disruption in the wild population, biologists attempted to trap both members of suspected pairs. On the day of capture blood was drawn from all six foxes to obtain baseline CDV antibody titers, and at the same time four of the six were given 1 ml of the experimental vaccine

7.5 Testing CDV vaccine

· 79

intramuscularly, while two animals selected as controls were given 1 ml of sterile water. On day 14 blood was again drawn and an additional 1 ml of vaccine given to the original four foxes and one of the controls (by accident), with the sixth individual given a sterile water injection. On day 42, an additional blood sample was obtained from all foxes and the sera submitted for analysis. The samples were sent to the Wyoming State Veterinary Laboratory, which was providing the serology testing for Merial Limited Company’s CDV-CPV clinical studies. Pre-vaccination blood samples were also submitted to the Cornell Diagnostic Laboratory (Ithaca, NY) to provide replication and baseline results from separate laboratories (Timm et al. 2000). After each vaccination the foxes were observed for two hours post-injection to monitor for any immediate adverse reaction to the injections (e.g. anaphylactic shock) and were given physical examinations during each handling. The results of the experimental vaccine trial were positive, with all of the foxes exhibiting positive antibody titers (a measure of the ability to produce antibodies) except the one control animal. The test subject that had received only one injection of vaccine had the lowest antibody titer. While the antibody response was relatively low, the experiment demonstrated that the CDV-CPV was likely safe for use in island foxes. However, in the absence of a challenge test (where vaccinated foxes would be exposed to the virus, which no one at the time was willing to do given the current status of the Santa Catalina population), the efficacy of the vaccine in wild populations could not be known. Because the vaccine was apparently safe, at this point it was the best option available to protect the foxes against another CDV outbreak. Field vaccination of wild foxes began in fall 2000, and from 2000– 2003 a total of 244 foxes were vaccinated on Santa Catalina (Schmidt et al. 2004b). Evaluation of the vaccine efficacy in 51 foxes vaccinated in fall 2000 showed that antibody production required two doses of the vaccine and an annual booster (Timm et al. 2000). Since 2003 vaccination programs have been implemented for captive and wild foxes on all islands as part of a vaccination protocol recommended by veterinarians of the Island Fox Conservation Working Group’s fox health group (Schwemm 2008b). Interestingly, more recent serological surveys with a more sensitive test confirm that there is a naturally occurring CDV-like virus to which island foxes on all islands show previous exposure (Clifford et al. 2006). Unlike the CDV strain that swept through Santa Catalina’s foxes in 1999–2000, the naturally occurring strain is not 100% lethal, and survivors gain immunity from exposure (Chapter 11). To allow

80 · Disease and decline on Santa Catalina Island this naturally occurring virus to circulate, not all the foxes on an island are vaccinated against CDV (Schwemm 2008b). The theory that domestic dogs had transmitted CDV to foxes was weakened by the recent discovery of ‘stow-away’ raccoons that have arrived on Santa Catalina (Coonan 2009). From 2007–2009 a total of four raccoons were discovered either on the island or swimming to shore from a boat in Avalon Harbor. Because the CDV strain that decimated the Santa Catalina fox population is more similar to the raccoon strain of CDV than it is to the dog strain (L. Munson, University of California, Davis, unpublished data), it is possible that CDV was transmitted to foxes from raccoons rather than dogs. Encounters between foxes and either domestic dogs or raccoons (or cats) would likely both result in aggressive interactions in which blood or saliva would be exchanged, so both scenarios are plausible. And because it is unlikely that the frequency of interactions between foxes and human-introduced carnivores will decline, the program of trapping and vaccination for island foxes on all the islands will likely continue indefinitely (Schwemm 2008b). Like removal of golden eagles from the northern islands, vaccination of foxes on Santa Catalina represented mitigation of the primary mortality source in support of captive breeding, reintroduction, and translocation.

8

r

Recovery actions Captive breeding of island foxes

Captive breeding and reintroduction, especially on Santa Rosa and San Miguel Islands, was the one recovery action that could not fail if island foxes were to be recovered. If the handful of remaining San Miguel and Santa Rosa animals did not produce young before they died, those subspecies would be lost. Island foxes had never been bred in captivity, so it was unknown how adaptable they would be to captive conditions. As it turned out there were significant challenges to captive breeding, primarily less than ideal reproductive success, injuries to females from aggressive males, and considerable perinatal mortality. Despite such difficulties, captivity protected the northern foxes from golden eagles until eagles could be removed, and productivity was high enough that foxes could be released beginning less than five years after the program began. Once foxes were released the wild populations grew quickly (Chapter 9), and captive breeding ended on Santa Catalina in 2005, on San Miguel and Santa Cruz in 2007, and on Santa Rosa in 2008.

8.1 Captive breeding efforts on the northern islands Developing an island fox captive breeding program was a unique challenge for NPS, which had not previously managed such a program and had minimal fiscal resources with which to begin one. The recommendation from an outside group of experts (the Island Fox Conservation Working Group [IFCWG], Chapter 13) to initiate island fox captive breeding provided the agency with a justification for taking the drastic step of removing all of the remaining individuals from the wild, a very difficult decision to make in rare species conservation (Hendron 1998). Given the possible imminent extinction of two fox subspecies, NPS moved ahead with island fox captive breeding without the normal planning process that would have been invoked for a project of such scope. By acting so swiftly the NPS left itself open to potential criticism and litigation, but little opposition was raised then or at

82 · Captive breeding of island foxes any time during recovery, likely due to the obvious necessity of such actions. With the assistance of the captive breeding sub-group of the IFCWG (Chapter 13), the NPS developed captive breeding goals and objectives as well as standard operating procedures for the captive breeding program (Coonan and Rutz 2001). The overall goal of recovering wild fox populations to viable levels looked beyond the immediate need to provide sanctuary for the few remaining wild foxes and toward the production of animals suitable for reintroduction into appropriate habitat once threats were minimized or eliminated. Captive breeding facilities were established on the islands and not in mainland-based facilities for two reasons. First, the isolation of island foxes meant that foxes were not only vulnerable to canine and feline diseases, but also were free of many parasites commonly found in mainland species. Wildlife veterinarians involved in the project were adamant that captive breeding for reintroduction occur only on the islands, and that to prevent the introduction of novel parasites and disease any island foxes that were brought to the mainland never be returned to the islands (Coonan and Rutz 2001). Consequently the only island fox ever transferred to the mainland from an endangered subspecies (an individual from Santa Rosa with a medical condition that could not be treated on the island) was moved permanently to the Santa Barbara Zoo. The second obstacle to mainland-based captive breeding was simply the lack of available space in mainland zoos for small carnivores (Chapter 12).

8.2 Methodology and techniques Island foxes had never been bred in captivity and there were no established husbandry (caretaking) protocols for the species. Of necessity, island biologists and mainland experts worked collaboratively and quickly to develop appropriate husbandry methods for island foxes (Chapter 13). The fledgling captive breeding program required techniques for facility design and construction, genetic management, pairing and breeding and, ultimately, release techniques and metrics that would measure reintroduction success. 8.2.1 Facility design and construction: San Miguel and Santa Rosa Islands

Addressing the critically low population of foxes on San Miguel was the highest priority for NPS so the first captive facilities were constructed

8.2 Methodology and techniques

· 83

Figure 8.1 Island fox captive breeding pens at the Willow Canyon site, San Miguel Island. Courtesy of Chuck Graham.

there in 1999. (The fact that the population on Santa Rosa was just as low as on San Miguel was not yet known.) The first pens built were modeled after those used for Mexican wolves (Canis lupus baileyi) and were 45–60 m2 in size and constructed of chain-link fencing material attached to tubular frames. Early in the program it was discovered that island foxes are prodigious climbers, so roofs were also required to prevent foxes from escaping over the pen walls. (On Santa Catalina a fox newly introduced to its pen immediately ran up a smooth-surfaced wooden post and began chewing on the netting that was covering the pen. While the fox could not maintain its hold on the beam, it was clear that measures would be needed to prevent foxes from escaping over as well as through enclosure fences.) A ground skirt of fencing material was sunk under the perimeter of each pen to prevent foxes from digging out, and shadecloth affixed to the bottom half of the pen sides to prevent foxes from seeing into adjacent pens (Fig. 8.1). The logistics of construction were difficult. All pen materials had to be transferred to the islands on boats and, although on the larger islands materials could often be transported by truck, on San Miguel there are no roads and all material had to be airlifted by helicopter from the boats to the pen sites. By May 1999, one month after the IFCWG meeting where

84 · Captive breeding of island foxes

Figure 8.2 Island fox captive breeding pens at the Windmill Canyon site, Santa Rosa Island. Courtesy of National Park Service.

the program was approved, two pens had been constructed on San Miguel and one pair of island foxes had been captured and brought into captivity. The remaining pens for the first captive breeding site (Willow Canyon; Fig. 8.1), were completed in August 1999, and 14 of the 15 remaining wild foxes were brought into captivity in the fall of 1999 (one lone female would not be captured until 2003, despite numerous attempts). These 14 animals would now serve as the entire breeding stock for the San Miguel subspecies. As it turned out only four of these individuals were males, a circumstance that would have important impacts for the management of the breeding program and genetics of future populations (Chapters 2 and 10). With the San Miguel facility in operation in 2000, attention turned to Santa Rosa (Chapter 5). NPS constructed captive facilities on Santa Rosa in February and March 2000 (Fig. 8.2), and began capturing foxes in March of that year. NPS staff captured nine foxes, including three pregnant females that subsequently each raised litters in captivity. An additional five wild foxes were captured in the fall of 2000, and the last remaining Santa Rosa wild fox was captured and brought into captivity in May 2001, after which survey and capture efforts yielded no fox sign on the island. For the initial Santa Rosa captive population the sex ratio was nearly 1:2 (five males and nine females).

8.2 Methodology and techniques

· 85

8.2.2 Facility design and construction: Santa Cruz Island

Although predation on Santa Cruz Island foxes was substantial, the population had not declined as steeply as those on San Miguel and Santa Rosa. But to avoid a repeat of those crises, the IFCWG considered captive breeding on Santa Cruz at its 2001 annual meeting: The most urgent threat to a wild island fox population is the continued predation by golden eagles on the Santa Cruz Island fox subspecies, which may be at a critically low population level. The remaining golden eagles need to be removed by October 2001. If they are not removed by that time, or if the island fox population is found to be at a population level which is likely to decline to extinction, then captive breeding is warranted for the Santa Cruz fox population. An October decision deadline is necessary in order to provide enough lead time to establish a facility and bring foxes into captivity before breeding season (March–April 2002). (Coonan and Rutz 2002, p. 29)

By October of 2001, monitoring of radio-collared foxes on Santa Cruz by IWS indicated that mortality from golden eagle predation was significant; eight of 27 collared foxes had died from predation that year, and the island’s fox population was estimated to be only 50–60 adult animals (Dennis et al. 2001). Moreover, at least four golden eagles were known to remain on the island, and NPS and TNC decided to initiate captive breeding for the Santa Cruz subspecies (Coonan and Rutz 2002). In February 2002, 10 breeding pens were built on the isthmus of Santa Cruz Island (Fig. 8.3). Because the facilities on each of the islands were built sequentially, the design of later-built facilities benefited from the experience at earlier sites, and the pens on Santa Cruz incorporated several changes from the original San Miguel design. First, although the pens were the same general size, they were rectangular in shape, not Lshaped as were the San Miguel pens, and were built without the interior walls that had been used on San Miguel and Santa Rosa. This design resulted in more interior room for foxes from the same quantity and cost of pen materials. Second, the pens were widely spaced and in a wooded area, providing foxes (usually housed in pairs) a greater sense of isolation. These changes may have contributed to the better reproductive success and relative lack of injuries in the Santa Cruz captive population compared to Santa Rosa and San Miguel, given later findings that the proximity of other foxes was likely related to aggravated levels of stress and increased aggressive behaviors (Chapter 10). The Santa Cruz pens were completed in February 2002 and, under contract to TNC, IWS immediately began efforts to bring foxes into

86 · Captive breeding of island foxes

Figure 8.3 Island fox captive breeding pens on the isthmus of Santa Cruz Island. Courtesy of Chuck Graham.

captivity. To maximize genetic diversity foxes were trapped from different parts of the island (on San Miguel and Santa Rosa all of the remaining wild animals were caught). Radiotelemetry data enabled IWS staff to trap selectively for specific pairs, and five pairs were brought into captivity in late February–early March. (One of those females was pregnant, and later gave birth to a litter of five pups.) Additional foxes were brought into captivity in the winter of 2002, and by the next breeding season the facility held 10 pairs (Coonan and Rutz 2003).

8.2.3 Facility design and construction: Santa Catalina Island

With very low population densities on the east end of Santa Catalina and unknown prospects for potential outbreaks of disease on the west end (Chapter 7), IWS proposed to the Catalina Conservancy that a captive breeding facility be constructed on the island (Timm et al. 2000). The goals of captive breeding on Santa Catalina would be to produce foxes to help repopulate the eastern portion of the island and to provide a reservoir population in case of further population decline. (A small facility to test the canine distemper vaccine was already in place; Chapter 7.) In 2001 IWS constructed a facility to house 12 pairs of foxes from the island’s

8.2 Methodology and techniques

· 87

west end, where the population was much higher than on the diseaseaffected east end. While island fox breeding facilities already existed on San Miguel and Santa Rosa Islands, success in producing offspring had been limited, and facility design and construction on Santa Catalina incorporated changes intended to address some of the early challenges on the northern islands (Timm et al. 2002). For example, the size of the pens was increased to allow greater movement and development opportunities for pups, and structural diversity in the enclosures was enhanced by allowing more natural vegetation to grow inside the pens and by adding logs, branches, and rock piles to provide foxes with greater climbing and hiding opportunities. To improve denning conditions, the Santa Catalina pens were equipped with two den boxes (instead of one), to allow the male and female to rest separately, as well as provide the female a choice of dens in which to have her pups. The den boxes were also accessible via a door from outside the pen so that project personnel could observe or handle pups if necessary once the female left the den box. Den cameras and video recording equipment were installed at the Santa Catalina facility to observe breeding behaviors and development of the pups. The Santa Catalina facility was powered primarily by batteries charged with solar panels and a smaller generator, but similar systems were difficult to maintain on the smaller islands, and information on what was happening in the den boxes during the neonatal stage was lacking (Chapter 10). On Santa Catalina a total of 72 infrared video cameras were installed, six in each pen, with video cables connected to a trailer housing video monitors and recorders to monitor all aspects of fox behavior, particularly during the breeding season (Fig. 8.4). Another major difference between the Santa Catalina and the northern island programs was the nearly full-time availability of veterinary care on Santa Catalina funded by the Conservancy and IWS. In contrast, on the northern islands veterinary care was provided by contract, by support from other entities (primarily IWS and the Santa Barbara Zoo), and by some volunteer time, because financial and logistical considerations prevented having a veterinarian in residence on those islands. The nature of the threat on Santa Catalina (disease) also underscored the need for greater veterinary presence on that island. Small veterinary hospitals (quickly dubbed ‘foxpitals’) were built on each of the northern islands (Fig. 8.5) and later in the program veterinary visits increased from annual veterinary examinations of all captive foxes to numerous visits to address

88 · Captive breeding of island foxes

Figure 8.4 Island fox breeding facility on Santa Catalina Island, with central trailer housing video monitors. Courtesy of Institute for Wildlife Studies.

specific injuries and ailments. The occurrence of mastitis (an infection of the mammary glands that affected female health and pup survival) in captive facilities in 2005 resulted in greater veterinary scrutiny during subsequent breeding seasons (Chapter 11), as did the considerable intrapair aggression in captivity that resulted in several serious injuries, and neonatal mortality that warranted investigation (Chapter 10).

8.2.4 Selecting the captive population: Santa Catalina

On Santa Catalina and later on Santa Cruz the individuals that would be brought into captivity had to be selected from the wild population. The first foxes to be brought into captivity on Santa Catalina were the six individuals that had been used in the canine distemper vaccine trials (Chapter 7). Trapping from the wild began in late March 2001, with the goal of obtaining existing pairs and, preferably, females that were already pregnant (Timm et al. 2002). Bringing in pregnant females would allow for reproduction during the first year of operation of the facility, and

8.2 Methodology and techniques

· 89

Figure 8.5 Veterinary hospital, or ‘foxpital’, on San Miguel Island. Courtesy of National Park Service.

females who had previously raised a litter in captivity might be more likely to do so in subsequent years (Timm et al. 2002). Mated pairs were identified by examining the trapping history and were assumed to be pairs if they were consistently trapped near each other. During the trapping effort a portable ultrasound device was used to determine if females were pregnant. This was the first time pregnancy had been assessed in island foxes using ultrasound, so determining fetal development and estimating the time of parturition were based on reproductive data from similar-sized domestic dogs and gray foxes. Eight pregnant females were brought into the captive breeding facility over a 10-day period in late March and early April 2001, but there was only time to catch two of the eight presumed mates of the females before the risk of capturing females late in gestation or separating females from newborn pups became too great (Timm et al. 2002). Of the foxes brought in as presumed mates, close observations showed no aggression and a good degree of socialization. The foxes settled well into the breeding pens and some of the females immediately began to use the den boxes. Two females gave birth only two days after being brought

90 · Captive breeding of island foxes into the facility, and three other females gave birth 7–14 days after being released into their pens. Two of the females from the vaccination trials that were introduced to the pens in mid-February gave birth; one produced a litter of three that all survived and one produced a pup that later died. Overall, 18 pups were born to 9 females, of which 12 died and 6 survived (see Section 8.4.1).

8.2.5 Nutrition

In addition to general maintenance, the food provided to captive foxes had to serve two additional purposes. First, it had to be suitable to the age and reproductive condition of each animal. Females would need a different diet during pregnancy and nursing than during other times of the year, and weaned pups would require different foods than mature foxes. Moreover, captivity unfortunately does not provide the same opportunity for activity as does living in the wild, and as foxes aged they tended to gain weight over what was considered normal. Assuring that each animal received the proper foods was sometimes difficult given that two animals were usually housed together (more if they had pups), and once the food was left in the pen it could be difficult to control which animal took which items. Second, because the goal of the program was reintroduction, foxes born in captivity needed to acquire the ability to catch live prey once they were released; providing dry or dead food in a bowl would not teach hunting skills, but providing foxes with live mice was challenging because mice have a tendency to run away when placed in a pen with hungry predators. NPS worked with the National Zoological Park and the Santa Barbara Zoo (which had received island foxes from San Clemente Island in spring 1999) to develop a basic diet (Coonan and Rutz 2001). High-quality dry dog food (25–28% protein, 10–12% fat) formed the basis of the daily diet, and was supplemented with hard-boiled eggs, fruits, and vegetables, and a weekly offering of a live deer mouse or dead Coturnix quail. To increase energetic resources of females during pregnancy, parturition, and nursing, females suspected to be pregnant or with litters were switched to puppy kibble, which had a higher crude protein content and more essential fatty acids (Coonan et al. 2004). Low calcium values detected in breeding females were addressed by increasing the number of calciumrich food items provided in the spring (hard-boiled eggs, mice, and quail). Problems with weight gain in captive foxes prompted a switch

8.2 Methodology and techniques

· 91

to dry dog food with less crude protein and fat in 2004 (Coonan et al. 2005b); however the problem persisted, and a major diet revision was implemented in 2006. The new diet was developed at the Santa Barbara Zoo and provided a greater proportion of calories from supplements amended by insects (mealworms) and frozen mice. On Santa Catalina several modifications were made to the pen design to more effectively remove the handlers from the feeding process and to allow foxes to handle more live prey. Food and water were introduced through sliding drawers placed in the solid walls of each pen that could be opened and closed by handlers without being seen by foxes. There were two drawers approximately 3 m apart in each pen, a design that reduced fighting between foxes over food items. A food ‘chute’ – a PVC pipe with a cap on the end that emptied into a 1.2-m2 wooden box – was also installed to introduce live prey. Prey items (insects, lizards, mice, and birds) could not escape from the box but foxes could get in, thereby allowing them to gain experience with ‘moving food’ (Kohlmann et al. 2003).

8.2.6 Escapees and perimeter fences

Despite being roofed and outfitted with groundskirts under the walls, each of the northern islands had at least one fox escape from captivity during the program. In two cases the pen door was left ajar, and in one case a fox dug out underneath a pen wall. Also of concern was the effect of released foxes on animals residing in captivity. After the initial release of foxes to the wild on Santa Rosa Island in fall 2003, some released foxes returned frequently to the captive pen sites, to the point where aggressive encounters between wild and captive foxes caused injuries to both. Several offending foxes were returned to captivity, and perimeter fences were eventually built around most of the pen sites. The Brooks captive site on San Miguel was too large to fence cost effectively, so individual electric fences were constructed around each pen. Both methods were effective in all but eliminating wild fox interactions with captive foxes. Several adult and juvenile foxes escaped from the breeding pens on Santa Catalina over the course of the program, and in some cases the most frustrating aspect for the IWS staff was determining the ‘means of escape’. The larger size of the Santa Catalina pens required a variety of materials (wood, fence material, and netting) other than chain-link fencing, and resulted in possible exit routes where these materials came

92 · Captive breeding of island foxes together; apparently if a fox could create an opening 75 mm or greater in diameter, it could escape. Also, foxes are such proficient climbers that they could exploit weaknesses on the ground, the walls, or the roof of the pens. The agility of foxes has allowed them to utilize nearly all of the food resources on the islands, but challenged caretakers who endeavored to provide them an environment that was as natural, but as secure, as possible.

8.3 Demographic and genetic objectives of captive breeding The general goals for captive breeding had been set by NPS under guidance from the IFCWG (Coonan and Rutz 2001), but specific demographic objectives were required to structure a program that could bring about recovery of wild populations in a reasonable time frame. Gary Roemer, then of UCLA, led a team convened at the request of NPS to use demographic modeling to develop an appropriate captive breeding program structure. The results (Miller et al. 2001) were appended to the NPS island fox recovery strategy (Coonan 2003) and formed the basis for the captive breeding and reintroduction program of NPS. The authors used the program VORTEX (http://www.vortex9.org/ vortex.html; accessed March 2010) to model life history and population parameters, identify the parameters that were most sensitive to alternative management actions, and evaluate the effects of various management scenarios. The team utilized historic population monitoring data from San Miguel, Santa Cruz and San Clemente Islands to derive age- and island-specific values for reproductive success and mortality, the two factors that most determined population trajectory. Consequently, the team identified a range of pup and adult survival values that would yield an increasing or stable population, as well as a population augmentation (reintroduction) schedule that would build wild populations to a specified target size (200 foxes) with an acceptable risk of extinction (3.6% over 50 years). An evaluation of various reintroduction rates found that adding 12 foxes per year to the wild would result in a population of 55 foxes after five years, 130 after 10, and 200 after 15 years of captive breeding and release (Fig. 8.6). The modeling team estimated that a captive population of 20 pairs would produce the desired augmentation rate, so both the San Miguel and Santa Rosa captive program set a target captive population size of 40 foxes. Actual results were similar to the projected results; the addition of 10–22 foxes per year resulted in a wild population of

8.4 Low reproductive success

· 93

400 6 Annual 6 Semi-annual 12 Annual 12 Semi-annual

Mean population size

350 300 250 200 150 100 50 0

0

10

20

30

40

50

Year of simulation

0

10

20

30

40

50

Year of simulation

Figure 8.6 Predicted wild fox population sizes under various augmentation rates, for a 20-year captive breeding period (left) and a 10-year captive breeding period (right). Adapted from Miller et al. (2001: Fig. 4).

approximately 180 foxes on San Miguel after four years of annual releases (Coonan 2009). In addition to the general programmatic objectives, NPS developed more specific goals and objectives under guidance from the Captive Breeding sub-group of the IFCWG. Besides tying the structure of the program to the results of demographic modeling (above), the group also proposed that no foxes be released until all golden eagles were removed from the islands (a recommendation that was rendered problematic by the ultimate inability to rid the islands of all golden eagles), and that the duration of the captive breeding program would depend upon the survival of released foxes (Coonan and Rutz 2001). Finally, maintaining a captive population for reestablishment of a subspecies with only 15 individuals required deliberate genetic management to avoid inbreeding depression and to maximize genetic variability and retention of founder genomes. A population management plan, or studbook, was established for the island fox captive breeding program, and guided decisions about annual pairings and releases (Chapter 2). Despite the apparent bottlenecks on Santa Rosa and San Miguel, the only symptom of inbreeding expressed in the captive populations may have been the high perinatal mortality observed in 2005–2007 (Ralls et al. 1988, Sovada et al. 2006).

8.4 Low reproductive success When the captive breeding program for island foxes began in 1999, almost nothing was known about the potential of this species to breed

94 · Captive breeding of island foxes Table 8.1 Production of litters and pups by captive island foxes, northern Channel Islands. San Miguel

2000 2001 2002 2003 2004 2005 2006 2007 2008 Total

Santa Rosa

Santa Cruz

Litters/pairs

Pups

Litters/pairs

Pups

Litters/pairs

Pups

1/4 2/5 3/10 5/12 4/14 6/19 3/12 0/7

2 5 8 10 12 10 6 0

3/3 5/8 6/12 4/16 4/18 4/19 5/16 4/14 4/12

10 10 13 11 9 8 9 9 8 88

1/6 5/8 9/9 10/15 8/24 6/11

5 10 19 20 21 10

53

85

out of the wild. The relatively simple pairing technique – in which males and females were paired if they were thought to be mated in the wild or known to be unrelated through microsatellite genotyping (Chapter 2) – was the consensus approach of the IFCWG but was not based on any previous breeding experience with island foxes. Reproductive success in captivity on all the islands (and in the zoo populations; Chapter 12), was ultimately lower than in the wild (see Box 8.1 ‘Reproductive success in captive island foxes’, Table 8.1). Two of the most critical factors affecting reproductive success were the effects of stress on mating behavior, and mastitis.

Box 8.1 Reproductive success in captive island foxes Reproductive success in captive island foxes was variable. Of 189 annual pairings from 1999–2007 on San Miguel and Santa Rosa Islands, 59 (31%) resulted in litters. Logistic regression analysis evaluating the influence of age and place of birth (wild versus captive) as well as number of years paired on reproductive success indicated that female age (p = 0.001) and female place of birth (p < 0.001) were significant factors affecting likelihood of producing a litter (model chi-square = 37.229, df = 5, p < 0.001) and correctly predicted 65% of the outcomes.

8.4 Low reproductive success

· 95

0.7

0.6

Proportion successful

0.5 Wild-born Captive-born

0.4

0.3

0.2

0.1

0 1

2

3

4

5

6

7

8

9

10

11

12

13

Female age

Figure 8.7 Reproductive success of female captive island foxes, San Miguel and Santa Rosa Islands, 1999–2007. Reproductive success is the proportion of annual pairings that produced a litter (total number of annual pairings = 189).

Average reproductive success of captive-born female foxes (16%) was less than that of wild-born captive females (44%, Fig. 8.7). The success of wild-born females, or founder females (i.e. those brought into captivity in 1999–2000) was similar to reproductive success of females in the wild on San Miguel (43%, Coonan et al. 2005b). Reproductive success of captive-born females remained low: below 20% until age 5–6. In contrast, reproductive success of wild-born females in captivity was consistently above 50% until age 7, when it abruptly declined, and no females older than 8 years produced litters.

8.4.1 Stress and behavior

Overall, foxes born in captivity had lower productivity than foxes brought in from the wild, and in particular young captive-born females were less likely to breed than wild-born animals of the same age (see Box 8.1 ‘Reproductive success in captive island foxes’). While behavioral changes in captive animals are not uncommon (Wallace 2000, McPhee 2003,

96 · Captive breeding of island foxes McPhee and Silverman 2004), until late in the program there was no clear link between the possible effects of stress on mating and maternal behavior and the production of live pups (Coonan et al. 2005b). In response to the concerns about productivity, a detailed analysis of breeding success was conducted by Kathy Carlstead of the Honolulu Zoo. The study evaluated the influence of fox behavior, physical pen characteristics, and fox history (wild-born versus captive-born) on litter production, mate aggression, occurrence of mastitis, and individual weight and body condition for the 2005 breeding season (Chapter 10). The overall results indicated that the most successful pairings were those where each mate was wild born, a condition that was becoming more rare each year (Carlstead 2005). The study thus identified behavioral tendencies that were associated with breeding success or failure; however, the direct cause of failure in unsuccessful pairs was still unknown. Stress was apparently also a cause of reproductive failure on Santa Catalina, especially during the first year of captive breeding (2001). Pregnant females were intentionally selected from the wild population (see discussion above). However, 12 of the 18 pups born that first year died soon after birth. Necropsies found that the pups had fully functional lungs, so premature birth was not a factor that had contributed to their deaths. The pups were born only a few days after the females were captured, so gestation was apparently further advanced than was initially thought. The deceased pups had empty stomachs and indications of hypothermia (L. Munson personal communication), and it appeared that milk production in these females had shut down, likely due to the stress of capture and relocation. Had their advanced stage of pregnancy been known, these females would not have been trapped and brought into captivity (Timm et al. 2002). A second reproductive study used observations of breeding foxes (direct or via video camera) to determine the precise point in the reproductive cycle (mating, pregnancy, or postpartum) where reproduction tended to fail (Sovada et al. 2006). Results (Chapter 10) showed that pup loss was apparently occurring late in pregnancy; fetuses were reabsorbed, pups were stillborn, and pups failed to thrive after birth (perhaps because of female inattention or mastitis; Chapter 10). The presence of males was also correlated with reduced pup survival. Overall the losses were significant; in 2006 on Santa Rosa 21 of 30 pups observed on ultrasound were lost to perinatal mortality (Clifford and Vickers 2007, Coonan and Dennis 2007).

8.4 Low reproductive success

· 97

Male aggression increased in the later years of the breeding program, especially on San Miguel and Santa Rosa (Coonan and Dennis 2007). Only one case of aggression was noted on Santa Rosa in 2001, but from 2004–2007 a total of 16 pairs on Santa Rosa had to be separated during the breeding season due to injuries from aggression. The possibility that captive males were also killing newborn pups was suggested by video camera evidence in 2007, when one adult male was observed consuming two dead pups. As a result, for the 2008 breeding season on Santa Rosa, adult males were pulled from breeding pens after ultrasound confirmed that their mates were pregnant. Additional structures and hiding places were also constructed to increase the ability of females to avoid aggressive mates (as they do in the wild). While removing males and improving pen design reduced female injuries, it did not increase productivity; in 2008 12 pairs on Santa Rosa produced only 8 pups. Another cause of reproductive failure was female abandonment of pups, possibly due to increased stress in the females and/or the presence of males during parturition and nursing (Chapter 10). The Santa Catalina program in particular suffered from relatively high rates of pup abandonment during the early years of the project. Den cameras installed on Santa Catalina provided more detailed information on maternal care than could be gained just by waiting for pups and mothers to emerge from den boxes, and it appeared that while most females were very attentive to pups – they nursed regularly and rarely left them for long – others would leave the pups for extended periods. The high level of pup abandonment prompted IWS staff to take a more aggressive approach, which involved both hand-rearing of abandoned pups as well as fostering neglected pups into litters with receptive mothers. For example, in 2003 two females rejected their pups; they would not allow them to nurse, and they carried them frequently in and out of the den box and often left them unattended. IWS staff eventually removed the pups from the dens and placed them in an incubator, providing nourishment via a stomach tube. While the incubator and hand-feeding saved several pups from starvation, IWS staff used a novel approach to return neglected pups to the wild. Orphaned pups were fostered to other females that had pups of a comparable age/size. The procedure involved waiting until the potential foster mother left her own pup(s), then temporarily locking her out of the den box. Using gloves, staff then rubbed the foster pups against the resident pups, and a small amount of urine and feces from the resident pups was placed on the foster pups’ fur. In each case when the females

98 · Captive breeding of island foxes returned to the litter they seemed to take no notice of the new pups and accepted them permanently.

8.4.2 Mastitis

In April 2005, two adult females died in the San Miguel captive breeding facility after giving birth to pups, which also did not survive. Necropsy revealed the cause of both adult deaths to be septicemia (system-wide infection) secondary to mastitis, a bacteria-caused inflammation of the mammary glands. Veterinary examinations showed that 4–6 other adult females in the same facility also had mastitis, and that those females had likely lost pups as a result. No females died from mastitis on Santa Rosa in 2005, but six litters were lost and two of those females showed signs of the infection. Further investigation suggested the high occurrence of mastitis was likely due to environmental conditions combined with stress. Unusually late and intense precipitation in the spring of 2005 had caused the interiors of the den boxes to remain wet throughout the breeding season. The den boxes were 5–7 years old and made of plywood, and high moisture levels in the dens may have led to increased bacterial loads. The susceptibility of females to infection may have been further increased by stress due to male aggression, as discussed above. Several changes were made in the following year to minimize the future occurrence of mastitis. The old den boxes were replaced with smaller ones that would retain less moisture and allow the females to better defend themselves against aggressive males. Additionally, the reproductive diet was adjusted to increase the amount of calories, protein, and fat, and breeding females were given prophylactic doses of antibiotics to prevent re-occurrence of mastitis (contracting mastitis once predisposes animals to subsequent infections). Finally, a more aggressive approach to handling females during the breeding season was adopted that included more frequent examination of captive females to identify both injuries due to aggression and early signs of mastitis.

8.5 Summary Santa Catalina’s captive breeding program produced 37 pups from 2001– 2004, all of which were released to the wild. The pioneering development of the Santa Catalina captive breeding facility produced advances

8.5 Summary

· 99

in pen configuration and construction, husbandry, video monitoring of captive animals, and hand-rearing of pups. From 1999 through 2008 a total of 226 pups was produced in the three captive breeding facilities on the northern islands (Table 8.1), and by 2004 each facility had reached the target captive population goal of 40 foxes (20 pairs), the threshold for beginning releases back to the wild. In the case of the San Miguel and Santa Rosa subspecies, the contribution of captive breeding to population recovery was profound. Each of those subspecies had declined to 15 individuals by 2000, yet had populations exceeding or approaching 400 by 2009, as the consequence of captive breeding and reintroduction.

9

r

Recovery actions Reintroduction and translocation

The intentional movement by humans of plants or animals from one location to another or from captivity back to the wild is an important tool in conservation. In both cases the goal is to augment existing populations (which may be at zero) to increase abundance and stimulate population growth rates (Morrison 2002). For island foxes both translocation (moving from one area to another) and reintroduction (moving from captivity to the wild) were successfully incorporated into recovery. On Santa Catalina Island translocation of foxes from the west end – where the population was stable – to the depleted east end accelerated growth of the east end population. Reintroduction was a critical recovery action on Santa Catalina and especially on the northern islands. The reintroduction of animals from captivity to the wild occurred on four islands, and on Santa Rosa and San Miguel it was the fundamental action responsible for recovery of the wild populations.

9.1 Translocation on Santa Catalina 9.1.1 Methods

Approximately 95% of the foxes remaining on Santa Catalina after the distemper outbreak (Chapter 7) were located on the western, smaller portion of the island (Timm et al. 2009). To promote recovery of the island’s fox population, in late winter 2001 IWS began moving vaccinated wild foxes from the west to the east end. Juveniles (pups born that year) were selected rather than adults in the hope that the natural tendency of juveniles to disperse from adult home ranges would provide those individuals an advantage in unfamiliar surroundings. Juveniles would also be less likely to return to the west end if they had not already established

9.1 Translocation on Santa Catalina

· 101

a home range there. On the east end, release sites were selected in areas that had previously supported high densities of foxes, an indication of high-quality habitat. The sites were also selected because they were near year-round water sources, were relatively distant from busy roads, and had few (if any) resident foxes (Timm et al. 2002). To begin the program, 10 juvenile foxes – 5 males and 5 females – were trapped from sites across the west end. All the captured foxes were given physical examinations, vaccinated against canine distemper virus (CDV), and fitted with radio collars. Because these were wild foxes with developed hunting skills and the capture and release habitats were similar, ‘soft’ release methods (with a transition period and supplemental feeding) were deemed unnecessary. To compare survival of the translocates with animals that were not moved, 10 additional juveniles on the west end were radio-collared and monitored. The translocation program continued in 2002 when an additional 12 juveniles (six males and six females) were moved from the west to the east end using similar methods. 9.1.2 Results

All 10 of the initial translocated foxes were alive 12 months later, and 9 of the 10 survived for 18 months. Conversely, the one-year survival rate of the control animals on the west end was less than 50% (Timm et al. 2002). More importantly, six of the nine surviving translocated juveniles had formed pair bonds that resulted in two confirmed litters and one additional confirmed pregnancy in 2002. In 2003, five of the foxes translocated in 2002 were confirmed to be part of successfully breeding pairs and five more were suspected successful pairs (confirmed denning but no pups observed; Schmidt et al. 2004b). The high survival and reproductive success of translocated foxes on Santa Catalina stands in contrast to translocation programs for other canids. For example, in Canada the survival rate for 33 wild-caught and translocated swift foxes was approximately 50% (Carbyn et al. 1994), and in Kansas the annual mortality rate of translocated juvenile swift foxes was 67% (Sovada et al. 1998). The successful translocations of island foxes demonstrated that – in the absence of novel mortality factors – island fox survival is high and populations grow very rapidly. Later reintroductions, described below, demonstrated the same very rapid population growth.

102 · Reintroduction and translocation

9.2 Reintroduction on Santa Catalina The primary purpose of the captive breeding program on Santa Catalina was to protect foxes from disease and to produce foxes to repopulate the east end (Chapter 8). Successful breeding led to rapid growth of the captive population, and in 2001 the decision was made to release all of the healthy pups produced at the Santa Catalina facility. This would be the first time that captive island foxes were released to the wild. 9.2.1 Methods

Captive-born foxes selected for release required additional measures to prepare them for life outside captivity. For example, released foxes would be more likely to survive if they had adequate hunting skills, so live prey items – including insects, lizards, and mice – were introduced to the pens to provide young foxes with experience in capturing and killing prey. To socialize animals that would be released together, IWS removed all the juveniles from their parents’ pen and placed them together for a few weeks prior to release. They placed enough resting enclosures (airline dog carriers) in the pen so that each juvenile would have a place to escape if they felt threatened by their new pen-mates. Food was provided in several locations in the pen to minimize competitive altercations over resources. Handlers observed newly introduced foxes for several hours and were prepared to separate aggressive individuals, but after a few hours of chasing and posturing the foxes were often found sleeping together in a big ‘dog pile’. Some individuals were less social, but no injuries were detected and the pups generally got along well. The socialization experiment not only allowed the pups to gain experience in dealing with non-siblings, but possibly led to early pair-bonding in the wild. Prior to release, each pup was given a complete physical examination – including blood and fecal sampling – and was fitted with a telemetry collar with a mortality-sensor. Because these were captive and not wild-born pups (in contrast to the translocated animals), a ‘modified hard’ release was developed wherein the animals were not held in transition cages but were provided with supplemental food. As with the translocated juveniles, release sites were selected in areas of adequate food and cover and with low levels of intraspecific competition, and all of the pups were released in one location. Live-capture box traps were wired open and placed around the

9.3 Reintroduction program on the northern islands

· 103

release area, and food (primarily dog kibble) placed in the traps. All released foxes were recaptured two-weeks post-release to weigh them and assess their general condition. 9.2.2 Results

The first group of six juveniles was released from the Santa Catalina facility in November 2001. For the first 30 days post-release the animals remained close to the release site and, in fact, often remained together under the same shrub during daylight hours. The foxes took advantage of the supplemental food in the traps, but scat examination revealed that they were also taking natural food items including insects and fruits. All of the released foxes were recaptured at two weeks post-release and found to be in good health. Additional releases from the Santa Catalina captive facility occurred in fall 2002 (8 animals) and fall 2003 (15 animals; Schmidt et al. 2004b). The one-year survival rate for the 14 pups released in 2001 and 2002 was over 70%, with two deaths and two pups unaccounted for by March 2004. The success of both the translocation and the reintroduction programs on Santa Catalina demonstrated that captive and wild-born foxes could not only survive and reproduce well in the wild, but also that neither group was negatively affected by the experience of being moved to a new location or by the handling process. Providing live prey to young foxes while still in captivity was likely helpful, but their omnivorous habit also eliminated the need to learn complex social hunting skills to take larger prey (such as deer). The development of release protocols on Santa Catalina was an important step in overall island fox recovery, and allowed managers to advance the programs on the northern islands, where reintroduction had to succeed if the northern subspecies were to persist.

9.3 Reintroduction program on the northern islands Captive breeding on the northern islands was sufficiently successful that by 2003 on Santa Cruz and Santa Rosa the captive populations could support releases to the wild. The release program on Santa Catalina had demonstrated the ability of released and translocated pups to thrive, and biologists were confident that food and habitat resources on the northern islands could likewise support foxes in the wild. However,

104 · Reintroduction and translocation unlike on Santa Catalina where released foxes were now protected by vaccination against CDV, foxes released on the northern islands would still be at risk from predation. Four years of golden eagle trapping and removal had resulted in the relocation of 31 birds, but by 2003 it was estimated that as many as 13 eagles remained on Santa Rosa and Santa Cruz (Chapter 6). The original captive breeding plan for the northern islands stated that foxes would be held in captivity until all golden eagles were removed from the islands (Coonan and Rutz 2001). However, the persistence of a relatively large golden eagle population likely supported by mainland dispersers (Chapter 6), and increasingly comprised of wary birds that avoided traps, was leading managers to conclude that complete removal of golden eagles under reasonable funding scenarios might not be possible. Instead, a new recovery strategy might be necessary, one that accepted a low level of mortality on wild foxes (Coonan et al. 2004). Accepting the presence of eagles coincident with reintroduction was difficult for some IFCWG members. To proceed with releases without first eradicating eagles meant that foxes would be introduced into an environment where there was still a risk that individual foxes would die from predation. There was disagreement among the IFCWG (and later the recovery team) over whether such risk was justified. Conventional wisdom regarding reintroductions of rare species holds that the causes of a species’ decline must be eliminated or substantially reduced before captive individuals are released (Kleiman 1989), and predation on released animals has reduced success of other canid reintroduction programs (Ausband and Foresman 2007b). Some members of the IFCWG as well as several outside observers felt strongly that all of the remaining foxes on the northern islands should be kept in captivity until golden eagles were completely removed (Courchamp et al. 2003). To postpone releases, however, would require additional holding sites to accommodate the growing captive populations, and the IFCWG suggested that, if necessary, additional captive facilities could be constructed on the mainland: Given the continuing presence of golden eagles, we recommend for this year [2003] that foxes not be released on the islands. We recommend building temporary holding facilities and developing alternative options for assistance by non-NPS staff in the care of captive-held foxes. We also recommend looking at alternative options for long-term care of captive-held foxes, including mainland facilities. (Coonan et al. 2004, Appendix A)

9.3 Reintroduction program on the northern islands

· 105

Other members of the IFCWG and most NPS managers felt that there was a stronger argument to be made for limited releases (Coonan and Schwemm 2009). Injuries resulting from mate aggression were increasing, possibly due to behavioral changes caused by prolonged captivity (McPhee and Silverman 2004). There were also potential risks to all of the captive populations from disease (Snyder et al. 1996), although there were now two physically separate pen sites on each island. Wildfires were also a threat, and because none of the northern islands have significant on-island fire response capabilities, a fire that started under dangerous conditions (dry and windy) might quickly overrun the captive facilities. (Island personnel were instructed to release captive foxes in the event of an island fire.) Most important, however, was that productivity in the captive populations was decreasing; from 2000 to 2003 the number of pups produced per captive female had declined from 1.0 to 0.4 (Coonan and Dennis 2006). Under these conditions, even if some released individuals were lost to predation, high reproduction in the wild would likely offset those losses, resulting in more rapid population growth in the wild than in captivity (Coonan and Schwemm 2009). Ultimately NPS decided to release foxes on Santa Rosa in the fall of 2003 on an experimental basis. The initial releases would be managed with the following objectives: 1. investigate survivorship of captive island foxes released to the wild under conditions of partial removal of golden eagles; 2. test release methods in relation to short- and long-term survival; 3. begin reestablishment of a wild population on Santa Rosa Island (Coonan and Rutz 2003). On Santa Cruz there were still approximately 100 foxes in the wild in 2003, and captive productivity was high (1.25 pups/female). Fox recovery on Santa Cruz was co-managed by TNC and NPS, and while NPS supported releases, TNC was hesitant given continued predation on wild foxes and the ecologic and public relations complexities imposed by feral pig removal (Chapter 6). By the fall of 2003 TNC and NPS agreed to begin releases on Santa Cruz, but agreed that survival rates of animals reintroduced on each island would be evaluated to determine whether releases would continue in future years (and be implemented on San Miguel), or instead be contingent upon further reduction of eagle numbers (Coonan et al. 2004).

106 · Reintroduction and translocation 9.3.1 Methods

The selection of individuals for release on Santa Rosa used criteria different from those on Santa Catalina. On Santa Catalina, the purpose of releases was to supplement a small extant wild population, while on Santa Rosa – where there were no foxes remaining in the wild – long-term recovery success was dependent on the persistent viability of the captive population. Because maintaining the genetic and demographic integrity of the captive population was of primary importance, and the animals selected for release were those whose potential loss would not compromise the long-term genetic objectives of captive breeding. Because the captive population was relatively small there were not enough surplus animals to release large family groups (which would likely facilitate more rapid reproduction), and individuals were released as pairs, in unrelated groups, or individually (Coonan et al. 2005b). Experiences from reintroduction programs for other species (Carbyn et al. 1994) and the results from Santa Catalina suggested that hard release protocols (no transition period) had a high likelihood of success for island foxes. Consequently foxes were released on Santa Rosa from portable kennels transported directly to the release sites by truck and backpack from the captive facilities (Coonan and Rutz 2003; Fig. 9.1). Feeding stations were maintained near release sites (making this a ‘modified’ hard release), and were kept stocked with food for several weeks to ease the transition into the wild for foxes accustomed to a captive diet. Foxes were released in the fall when they would naturally disperse and form pair bonds, and in previously occupied areas. All released foxes had radio collars with mortality sensors (Fig. 9.2), enabling staff to find carcasses and collect evidence for necropsy. The release protocol for pups on Santa Catalina had included a prerelease period of socialization, wherein unrelated pups were housed together for several weeks to reduce the likelihood of long-distance dispersal after release and to increase the chances that released animals would form pair bonds. Implementation of these methods on Santa Rosa proved unworkable. Three groups, each comprising unrelated animals, were housed together for 10–18 days prior to release (at the main captive facility). Several foxes were injured while housed together, requiring veterinary care that delayed their release. Moreover, the animals that were released together generally did not remain together post-release, and socialization methods were therefore dropped in subsequent years. The failure of pre-release socialization on the northern islands compared

9.3 Reintroduction program on the northern islands

· 107

Figure 9.1 One of the first island foxes released back to the wild, Santa Rosa Island, 2003. Courtesy of National Park Service.

to its success on Santa Catalina may have been partially due to the generally higher levels of aggression observed in Santa Rosa (and San Miguel) foxes compared to other island fox subspecies (National Park Service, unpublished data). 9.3.2 Results

Twelve foxes were released on Santa Rosa and nine on Santa Cruz during the fall and winter of 2003–2004. Eagle predation was initially high; five of the nine foxes released on Santa Cruz were killed by eagles within just a few weeks. All of the foxes released on Santa Cruz were young animals (< 2 years) born in captivity, which may have made them more vulnerable to predation than wild-born animals. Because the release plan for Santa Cruz included a mortality threshold of 50%, the remaining four

108 · Reintroduction and translocation

Figure 9.2 Radio-collared island fox prior to release, San Miguel Island, 2004. Courtesy of National Park Service.

released foxes were recaptured and brought back into captivity in January 2004. On Santa Rosa only one of the released animals was lost to predation and there were no other mortalities. However, several of the released animals returned to the captive facilities and interacted with captive animals, with resulting injuries to both wild and captive individuals. (This situation was later repeated on San Miguel and Santa Cruz, requiring the eventual construction of perimeter fences around all of the captive facilities; Chapter 8.) The five individuals that were frequenting the pen areas on Santa Rosa were recaptured, leaving six released foxes still in the wild. Two of those six paired and produced a litter, and one released fox died from predation in the following year (Coonan et al. 2005b). Due to substantial predation on released foxes in 2003, in 2004 TNC and NPS decided to temporarily halt the release program on Santa Cruz. Because the Santa Cruz captive population had a high growth rate (1.7 pups/female compared to 0.5 on Santa Rosa and 0.8 on San Miguel; Coonan et al. 2005b), a second breeding facility was built to hold the expanding population. Although releases on Santa Cruz were suspended indefinitely, the initial Santa Rosa release was considered successful, and

9.3 Reintroduction program on the northern islands

· 109

in the summer of 2004 NPS developed plans to continue releases on Santa Rosa and to begin them on San Miguel. In the intervening period four of the six subspecies (Santa Catalina, Santa Cruz, Santa Rosa, and San Miguel) had been officially listed as endangered, meaning that all release plans would now need to be reviewed and approved by the USFWS. With listing came the establishment of a formal recovery team, and USFWS now relied heavily on this group to review release plans (Chapter 13). Several of the original IFCWG team who had opposed releases in 2003 were now members of the recovery team and communicated their continuing concerns to USFWS. During the summer of 2004 discussions were held within the USFWS and the recovery team, and ultimately USFWS did not oppose releases. In the fall of 2004, NPS released 10 foxes (4 females and 6 males) on San Miguel and 13 on Santa Rosa. All 10 of the San Miguel foxes survived, and each of the females produced a litter in the spring of 2005, despite the fact that they were juveniles when released. Ten pups were produced in the wild on San Miguel in 2005, resulting in a mean fecundity rate of 2.5 pups/female, much higher than in captivity that same year (0.5 pups/female). On Santa Rosa adult survival was much lower; by May 2005, five of the released animals had died due to predation. Although the threshold for returning foxes to captivity had been reached, USFWS concurred with NPS that to do so would potentially disrupt breeding in the wild, and so all the released animals were left in the wild. No additional predation occurred after this period, and four of the females that would have been brought into captivity produced litters. Releases continued on Santa Rosa and San Miguel from 2004 through 2007, with 10–20 foxes released each year (Fig. 9.3; Table 9.1). Eagles continued to breed on Santa Cruz through 2006, and predation on released Santa Cruz and Santa Rosa foxes was initially high (Coonan and Dennis 2007). Survival rates within the small Santa Rosa population were consequently low, but on Santa Cruz high survival (above 80%) in the remaining wild population compensated for high predation on released foxes. On San Miguel from 2004–2008 predation-caused mortality was low (Coonan et al. 2005b, Coonan and Schwemm 2009), and survival ranged from very high following the first releases (< 90%) to between 80–90% by early 2008. Releases had resumed on Santa Cruz in 2006, where the fox population had increased as eagle abundance declined; the population on Santa Cruz grew from 100 to over 250 between 2004 and 2006 without releases (Schmidt et al. 2007a).

110 · Reintroduction and translocation Table 9.1 Number of island foxes released to the wild on the Channel Islands, 2001–2008. Year 2001 2002 2003 2004 2005 2006 2007 2008 Total

San Miguel

10 22 16 16 62

Santa Rosa

7 13 17 13 12 31 93

Santa Cruz 3 9

Santa Catalina 6 8 15 28

56 31 99

57

Figure 9.3 One of the last island foxes released to the wild on Santa Rosa Island, 2008. Courtesy of Chuck Graham.

9.4 Success of reintroduction

· 111

Island fox populations are naturally characterized by high densities, high survival, stable territories, and little turnover in social structure (Roemer et al. 2001b, Coonan and Schwemm 2009; see Chapter 3), but the remarkably high reproductive rate demonstrated by juvenile foxes on San Miguel had not previously been documented. From 2004–2007 San Miguel foxes were introduced in areas with few or no conspecifics, with the result that nearly all females were able to acquire territories and breed successfully, even in their first year. The potential negative effects of low densities – primarily the challenge of finding a suitable mate – were offset by releasing multiple individuals at each site. Reproductive success was extremely high in the first years post-release on San Miguel, but declined as density increased; as the recovering populations approached carrying capacity, available territories became scarce, and population growth rates declined (see Fig. 3.2). This was apparent on San Miguel and Santa Cruz Islands by 2007. Reproductive success may eventually decline to levels similar to those seen in the early 1990s at previously high densities (Bakker et al. 2009).

9.4 Success of reintroduction High reproduction and survival in the reintroduced populations led to rapid growth on all the islands (for San Miguel, see Fig. 9.4), and by 2009 all four endangered subspecies were each close to or at recovery thresholds (Fig. 9.5; TNC, NPS, and Catalina Island Conservancy, unpublished data; see also Chapter 15). In addition to meeting critical measures of population abundance and growth, the success of reintroduction can be evaluated by the level of reproduction achieved by reintroduced individuals (Boitani et al. 2004). Breeding by the first wild-born generation occurred early on both San Miguel and Santa Rosa Islands, and recruitment exceeded mortality on both islands over a three-year period. The third reintroduction success criterion of Boitani et al. (2004) is the establishment of a self-sustaining population. Current demographic modeling (Bakker et al. 2009) suggests that at the present 90% survivorship, a sustained population size of as few as 150 yields acceptable extinction risk. At current levels of recruitment and survivorship, both the Santa Rosa and San Miguel subspecies had reached or were approaching demographic recovery by 2008. The ultimate indicator of success might be the decision to close captive breeding on the islands, due to the declining contribution of captive

112 · Reintroduction and translocation 450 400

Number of foxes

350 300 Wild Captive

250 200 150 100 50 0

93 994 995 996 997 998 999 000 001 002 003 004 005 006 007 008 009 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1

19

Figure 9.4 Decline and recovery of San Miguel island fox population. Pre-captive breeding adult population estimates are from Coonan et al. (2005b); post-captive breeding estimates are from Coonan (2009) and NPS (unpublished data). 800

Estimated number of adult foxes

700 600 2000

500

2009

400 300 200 100 0 San Miguel

Santa Rosa

Santa Cruz

Santa Catalina

Figure 9.5 Estimated number of adult island foxes for the four island fox subspecies listed as endangered, in 2000 and 2009. Data from Timm et al. (2000), Dennis et al. (2001), Coonan et al. (2005a, 2005b), National Park Service (unpublished data), Catalina Island Conservancy (unpublished data), Vickie Bakker (unpublished data).

9.5 Summary

· 113

releases to wild population growth. By 2007 the San Miguel captive breeding program had produced 53 pups, but no pups were produced in 2007. In contrast, from 2005–2007 the number of pups produced in the wild was over four times the number produced in captivity (T. Coonan, NPS unpublished data), and in 2007 all the remaining healthy captive foxes were released on San Miguel and the breeding facility was closed. Similarly, by 2007 captive breeding was contributing relatively little to growth of the Santa Cruz wild population, which had naturally increased from an estimated 100 foxes in 2003 to 264 in 2006 (Schmidt et al. 2007a), and that facility was closed by NPS and TNC in 2007. For the same reasons (good survival and reproduction in the wild), captive breeding was ceased on Santa Rosa in 2008. The delisting of a species is the final legal acknowledgment of the success of a recovery program. From 2004–2008 USFWS developed a draft island fox recovery plan with recovery criteria derived from demographic modeling (Bakker and Doak 2009, Bakker et al. 2009). The demographic criteria for recovery established values of mortality and population size that minimize the chance of extinction and maximize the probably of persistence over time. At the low mortality rates achieved by northern island fox subspecies (approximately 10%) by 2009, a relatively low total population size (< 200 foxes) is calculated to be necessary to achieve demographic persistence. As a result, by 2009 northern island fox populations were very near demographic recovery, given recent rates of population growth.

9.5 Summary Over the 10-year period of captive breeding and reintroduction, strategic decisions were made regarding population recovery from a longterm perspective. For example, the decision to release foxes on Santa Rosa Island in 2003 was a calculated risk. Some outside observers at the time viewed NPS as being cavalier in its intent to release foxes prior to removal of all golden eagles (Roemer and Donlan 2004, Gibson 2006), and some predicted outright extinction of island foxes if lethal means of eagle removal were not implemented prior to the eradication of feral pigs (Courchamp et al. 2003, Chapter 6). Although eagles were never completely removed, island foxes did not go extinct, because eagle – and pig – removal had worked; eagle presence and predation had been reduced to a level that was negligible, and did not hinder island fox population

114 · Reintroduction and translocation recovery. Had managers not taken a risk and instead kept all island foxes in captivity until eagles were removed, foxes would likely still be in captivity today. By 2009 all four endangered subspecies had in fact increased to near-recovery levels, and on all four islands the captive breeding pens are now empty. They will be left in place until recovery on each island is assured.

10 Reproductive biology, by Cheryl Asa r

An increasing number of endangered species, including several canids, have depended on captive breeding programs as critical components in their recovery. Red wolves (Canis rufus) and Mexican gray wolves (Canis lupus baileyi), for example, became extinct in the wild and existed only in zoos before being reintroduced as wild populations. Despite considerable experience in North American zoos with some canids, each new species brings new challenges. Prior to the recovery effort that began in 1999, island foxes had never before been kept in captivity for breeding purposes. And although their mainland cousin, the gray fox (Urocyon cinereoargenteus), was historically held in small numbers in some zoos, they had never been the focus of a breeding program, so little information existed on their husbandry and breeding requirements. In addition, when establishing a breeding program, knowledge of a species’ mating system and reproductive biology can be keys to success. As captive breeding for recovery tool of island foxes began, very little information on reproduction was available, even though recovery might hinge on a successful captive breeding program. One initial approach when faced with such a challenge is to draw from information about closely related, more common species. In the case of the island fox, the obvious comparison is to the mainland gray fox. With a broad range extending from southern Canada into the northern countries of South America, the gray fox has certainly been studied more than the island fox. However, surprisingly little is known about reproductive processes in gray foxes beyond standard life history information on age of puberty, breeding seasonality and litter size. This is in contrast to the red fox (Vulpes vulpes), which is sympatric to gray foxes in much of the US, and the Arctic fox. Both of these species also exist in semi-domesticated forms as farmed silver and blue foxes, respectively. Intensive commercial breeding of these species has provided considerable detail about their reproductive physiology. However, red and Artic foxes are from different genera than gray and island foxes, and more than a hundred years of

116 · Reproductive biology breeding for fur production may have selected for features not found in their wild counterparts. Although gray wolves and coyotes (Canis latrans) are not bred commercially, they have been studied extensively in research colonies as well as in the wild; this has produced considerable information on reproduction. However, wolves and coyotes are even further removed taxonomically from island foxes than are the other fox species. Finally, while the reproductive physiology of domestic dogs is well described, the effects of domestication – such as independence from seasonal influence – raise questions about its suitability as a model for wild canids.

10.1 Reproductive cycles The modal reproductive pattern emerging from studies of canids characterizes them generally as monestrous (Asa and Valdespino 1998, Asa 1999), which usually means that they can only reproduce once annually. However, the term is also used to describe the reproductive cycle of domestic dogs that is on average only seven to eight months in length, resulting in some individuals actually having two reproductive seasons in a year. Temperate and arctic zone canid species breed seasonally, but for those closer to the equator reproduction may not be restricted to a single period each year. For example, bush dogs (Speothos venaticus) and fennec foxes (Vulpes zerda) can have estrous cycles every four to nine months (Valdespino et al. 2002, DeMatteo et al. 2006). Canids typically ovulate spontaneously during estrus, whether or not they are paired with a male. This contrasts with felids, which require copulatory stimulation to ovulate. An unusual feature of the canid cycle is the obligate pseudopregnancy that follows ovulation if conception does not occur. Most mammalian females can ovulate again within two to three weeks if they do not conceive, but canids have a prolonged period of elevated progesterone, the hormone characteristic of pregnancy, that is roughly equal in length to gestation. Some pseudopregnant females show nest-building, maternal behavior and can even lactate, even though they are not pregnant. Among canid species, pregnancy or pseudopregnancy ranges from around 50 days in the smaller species to about 60 days in the larger ones. A possible adaptive advantage of this prolonged luteal phase – at least in the more social canid species – could be that the non-pregnant subordinate females become hormonally primed to behave maternally and thus contribute to the care of the dominant female’s pups (Asa 1997, Asa and Valdespino 1998).

10.2 Captive breeding

· 117

Canid pairs are commonly described as monogamous (Kleiman 1977), with social units sometimes incorporating young of the previous year. Unlike in felids, canid males are notable for being very good fathers (Kleiman et al. 1981, Asa and Valdespino 2003). In all species studied, males contribute some type of parental care, from guarding and babysitting to bringing food to pups once they can eat solids. Results from field studies of island foxes (Laughrin 1977, Roemer et al. 2001b, Roemer 2004) have not shown them to deviate from these patterns. As a temperate zone species, they have a limited breeding season, with mating typically occurring in February and births in April. Data on home ranges of radio-collared foxes suggested that pairs were socially monogamous, but up to 25% of litters were the result of extra-pair copulations (those between members of different mated pairs; Roemer 1999). However, little more was known about their reproductive processes. Although the precipitous population declines that occurred on four of the islands threatened the survival of those subspecies, the silver lining was the opportunity to study them close at hand. While the foxes were being held in captivity for eventual augmentation of the wild populations, a team with experience in canid captive breeding was brought together to monitor their behavior and collect basic information on their reproductive processes (Coonan and Rutz 2001, 2002).

10.2 Captive breeding When foxes were first brought into captivity on San Miguel Island just before the 2000 breeding season, mated pairs were established by placing together males and females that appeared likely to have been paired in the wild, based on information available at capture. Because there were more females than males, some females were housed together. In subsequent years as pups were born and reached maturity, they were assigned mates based on a computer program designed to prevent inbreeding and maximize genetic heterozygosity in the population (Lynch 2005a, 2005b; Chapter 2). The reproductive success rate the first season was low; only one pair of four produced pups. However, considering their recent capture and confinement, even one litter of pups was a sign of success. Breeding improved slightly in the second year, in the 2001 breeding season, when two of five pairs produced litters. In subsequent years an increasing number of pairings was made possible by the birth of male pups that were then paired with the excess females. Nevertheless, through 2005,

118 · Reproductive biology 100

Percent of pairs with pups

80

San Miguel Island Santa Rosa Island Santa Cruz Island

60

40

20

0 2000

2001

2002

2003

2004

2005

2006

2007

Year

Figure 10.1 Reproductive success rates for captive island fox pairs in successive years. Data collection began for Santa Rosa in 2001 and for Santa Cruz in 2003. The zero value for San Miguel Island in 2007 reflects the fact that no pups were produced by the pairs that year.

with the number of pairs reaching as many as 19, the percent producing litters never rose above 42% (see Fig. 3.1). In 2000 and 2002, respectively, captive colonies were also established on Santa Rosa and Santa Cruz Islands, but they were considerably more successful in producing pups, at least in the first two breeding seasons (Fig. 10.1). Despite this higher rate of success, the fact remained that not all pairs were reproducing. The primary objective of the captive breeding program was to produce as many pups as possible for reintroduction, in order to recover wild populations to viable levels (Coonan and Rutz 2001, Coonan 2003). Of concern, however, was which individuals reproduced and thus contributed genes to the next generation. After four breeding seasons some of the originally captured foxes – the potential genetic founders of the captive population – had not produced offspring. This observation prompted an investigation of possible causes for this failure. The first step was a multi-factorial analysis of various factors that might be related to reproductive failure (Carlstead 2005): whether

10.2 Captive breeding

· 119

a fox had been born in the wild or in captivity, the relative ages of a fox pair, history of aggression, aspects of husbandry, and features of the enclosures such as layout or exposure to wind. The foxes were also rated by keepers on a scale from Tense to Calm and on features reflecting Compatibility. The results suggested possible pair incompatibility based on age differences (e.g. a yearling male with an older female) or history in captivity (i.e. born in captivity versus in the wild). Recommendations included adding wind-breaks and natural furnishings to the pens, and increasing exposure to keepers to reduce fear responses. The behavioral assessments were also related to breeding success, in that male foxes showing mate-directed aggression were more likely to be captive-born and were reported to be more competitive over food and to maintain more distance from their female partners. Carlstead also found that captiveborn females were more often rated as Tense than were their wild-born counterparts. These results were used in subsequent pair assignments, and increased monitoring helped determine their effectiveness in overcoming breeding problems. 10.2.1 Hormone studies

Through fecal hormone data from females on San Miguel and Santa Rosa Islands, the team confirmed that, in all cases but two, each female paired with a male had sustained elevated progesterone concentrations indicative of ovulation, beginning in February, the typical month for estrus and mating. More importantly, levels of progesterone – a hormone necessary for pregnancy maintenance – did not differ in females that produced pups and those that did not. These results indicated not only that even the non-reproductive females were ovulating – which of course is required for conception – but also that they appeared hormonally normal. Thus, in these females, lack of breeding success was not due to failure to ovulate. Analysis of hormone patterns of females not paired with males assessed their potential as future breeders. Surprisingly, none of them had sustained progesterone elevations comparable to the paired females, suggesting that at least contact with a male and perhaps copulation was required to induce ovulation (Asa et al. 2007a). Although common among carnivore species such as felids, mustelids, and ursids, such induced ovulation had never been documented in a canid. The adaptive value and evolutionary history of induced versus spontaneous ovulation have not been resolved. Complicating the interpretation for island foxes is that we also

120 · Reproductive biology do not know which mode characterizes their closest relative, the mainland gray fox. If indeed they are also induced ovulators, the explanation may lie in gray foxes being one of the most, if not the most, ancestral of the canids (Wayne et al. 1997). That is, older canid species such as gray foxes might be more likely to share characteristics such as induced ovulation with the other carnivores, whereas spontaneous ovulation may be a feature of the more recently derived canid species. Unfortunately, the mode of ovulation is not known for most canid species. This mechanism of ovulation – that is, the requirement for copulation – may explain the lone case in which a mature female paired with a yearling male failed to ovulate. She may have refused to mate with him due to his sexual inexperience and immaturity and hence failed to ovulate. That case occurred during the first breeding season after capture, so the lack of ovulation could be attributed to the stress of capture and confinement. In the subsequent year that female did ovulate, suggesting that she acclimated to captivity and accepted her male partner. After confirming that ovulation was occurring in females, attention turned to potential problems with males. Recent work with endangered Mexican wolves had documented poor sperm quality in some males, especially those that had become inbred (Asa et al. 2007b). The restricted size of the island populations of foxes might also have resulted in some degree of inbreeding, which – as with the Mexican wolves – could affect sperm quality and thus their fertility. To evaluate this possibility the team collected semen in February 2004 from equal numbers of reproductively successful and unsuccessful males on both San Miguel and Santa Rosa Islands. Unlike Mexican wolves, there was no measurable difference in sperm quality between these groups, and most of the abnormalities found were relatively minor. These results suggested that the reproductive failures were not due to male infertility. 10.2.2 Investigation of possible causes of reproductive failure

Although the above results revealed no problem with female ovulation or male fertility, they did not explain why reproduction was failing in so many pairs, so the investigation was widened. The female reproductive process comprises many sequential steps at which failure could occur; interference at any point halts the sequence and prevents reproduction. Using captive pairs on all three islands, methods were devised for monitoring as many aspects of the sequence as possible (Table 10.1), in hopes of identifying and treating or correcting any problems detected.

10.2 Captive breeding

· 121

Table 10.1 Sequential steps in the process of female reproduction and monitoring methods. Reproductive feature

Monitoring methods

Estrus Courtship and mating Ovulation

Estradiol rise detected by fecal hormone analysis Direct observation or videotaping for later review Sustained progesterone elevation detected by fecal hormone analysis Ultrasound exam and blood sample for relaxin assay at mid-gestation Video monitoring and pup checks Video monitoring and pup checks Video monitoring and pup checks

Pregnancy Birth of live pups Health of neonates Parental care

The first physiological phase to be evaluated was the period of estrus, characterized by maturing follicles that secrete estradiol, the female hormone responsible for stimulating myriad physiological, morphological, and behavioral effects. Estradiol and other steroid hormones are metabolized and excreted in feces as well as urine. Fecal samples are easier to collect, but when animals are housed together – for example, in pairs as were the foxes – it becomes necessary to identify each sample so the hormone results can be attributed to the correct animal. In most cases inert colored beads – one color for each animal – can be presented in a food treat the day before a sample is collected. The beads pass through the digestive tract, along with the hormones, in about 24 to 36 hours. The samples are then sealed in plastic bags and frozen for shipment to the lab for assay. During estrus, females are attractive to courting males and receptive to mating, which may be seen either by direct observation or, where possible, by digital video recording, which allows observations to continue 24 hours per day. Video has the further advantage of being unobtrusive, something that can be important when studying animals not acclimated to human presence. For direct observations, foxes needed to be habituated to a blind constructed nearby that people regularly entered and exited. Behavioral observations became even more important as the study progressed and aggression between members of some pairs was reported. This was surprising, since male canids are known to be solicitous of their mates. The aggression possibly resulted from the method by which mates

122 · Reproductive biology were assigned. Individuals had been placed in pairs based on genetics, but this did not guarantee social or sexual compatibility. Especially for a monogamous species, mate choice might be a crucial component of reproductive success. Another problem that may have resulted from mate assignment (rather than mate choice) was the lower reproductive success of foxes born in captivity compared to that of wild-born animals. The lower reproductive success of captive-born foxes was attributed, at first, to their young age and inexperience. Indeed, reproductive rate is usually lower for animals that have just reached puberty and was documented for the wild population of island foxes on Santa Cruz, San Miguel, and Santa Catalina Islands (Roemer et al. 2001a, Coonan et al. 2005b, Clifford et al. 2007). Of course, some wild-caught foxes that were captured either together – or near enough to each other to suggest they were already a pair – may actually have been paired in the wild. Thus, those foxes were likely to have chosen their partners, whereas the captive-born foxes were assigned theirs. However, as young foxes matured, reproductive success remained lower for those born in captivity, which was completely unexpected given the experience with other canid species. For example, gray wolves captured from the wild and brought into captivity (because they were caught predating domestic livestock) seldom reproduced, whereas virtually all the captive-born wolves produced pups each year (C. Asa, unpublished data). The island fox team approached the question of compatibility and stress in two ways. First, pairs were observed for signs of compatibility as well as evidence of aggression. Mech and Knick (1978) found that newly formed pairs of gray wolves spend considerable time in proximity and can often be identified in the field by their close sleeping distance. Thus, recording whether and for how long pairs of foxes spent time resting in contact could be a potential measure of compatibility. A second method that could indirectly assess their adjustment to captivity and to their mates was to measure cortisol in fecal samples. Cortisol and other glucocorticoids are involved primarily in mobilization and metabolism of glucose as an energy source, but glucocorticoids are also elevated during times of physiological and psychological stress. Chronically high levels can interfere with secretion of reproductive hormones and, hence, with reproductive processes. To determine whether reproductive failure was occurring before or after pregnancy, methods were needed to detect pregnancy in island fox females. Island fox fecal samples were already being collected and analyzed

10.3 Results of the monitoring study

· 123

for progesterone to document ovulation. In many species, progesterone concentrations that remain elevated beyond the typical ovulatory cycle are often sufficient to diagnose pregnancy. But this is not the case for canids, because hormonal pseudopregnancy with elevated progesterone follows ovulation even in females that fail to conceive. Therefore two other methods were used to detect pregnancy in captive island foxes. First, ultrasound was used to examine females in mid-March in the middle of the expected period of pregnancy, based on data on births in wild foxes and records of births in the captive colony. The other method involved measuring relaxin, the only hormone that distinguishes pregnancy from pseudopregnancy in domestic dogs (Bauman et al. 2008). Both methods proved accurate, although ultrasound could not visualize embryos before about day 15 of gestation (Clifford et al. 2007), and the assay could not detect relaxin until at least day 25 (Bauman et al. 2008). Early pregnancy loss within the first few weeks of gestation is relatively common in mammals, and can be typically determined via assays for early pregnancy factors. However, no early pregnancy factor has been identified in canids, so there is no way to detect pregnancy during the first two to three weeks. Thus, the true conception rate may be underestimated. Video surveillance and direct observation were used to determine if reproductive failure was occurring after pregnancy, starting with whether pups were actually born to females that had been diagnosed pregnant. Although late-term abortion is less likely than embryo loss early in pregnancy, it can be induced by certain infections, stressful events, and even by insufficient levels of thyroid hormones. If pups were born, the next question was whether they were alive and healthy or possibly stillborn, again probably due to infection. Without videomonitoring, documentation of the presence of pups must rely on direct observation, which can disturb the new parents, perhaps jeopardizing the neonates. It is not uncommon for animals to kill and consume young when they perceive that something is wrong, either with the health of the neonates themselves or with the conditions; thus they can cut their losses, when the probability of successfully raising the young is low (Hrdy 1979).

10.3 Results of the monitoring study Collecting sufficient samples and observational data presented many methodological and logistical challenges. Foxes sometimes refused the food treat with colored beads for marking fecal samples, or one fox

124 · Reproductive biology 3000

Fecal cortisol (ng/g)

2500

2000

1500

1000 Mated 500

0 January

February

Figure 10.2 Fecal glucocorticoid levels in a fox housed in the veterinary facility in early January, represented by the solid bar, and following return to her home enclosure.

would sometimes eat both samples, so its partner got nothing. Unfortunately, absence of beads in a sample does not guarantee that that animal got no beads. These missing data points proved fatal for documenting the estrous phase, since it is relatively short: it is typically only one to three days. In contrast, the long period of progesterone elevation following ovulation was evident, even when the sample set was incomplete. However, without regular, frequent sampling, cortisol patterns can be difficult to interpret. In particular, without establishing a clear baseline during undisturbed, unstressed periods, responses to stress may not be evident. Recognizing this limitation, nonetheless, cortisol levels did not appear elevated except briefly in specific cases. For example, when a fox needed to be moved to a veterinary holding area, its cortisol levels spiked and remained high until it was returned to its home enclosure (Fig. 10.2). On San Miguel Island during the first breeding season following capture, fecal cortisol levels were higher than those measured during their second year in captivity, indicating that the foxes were acclimating to their new surroundings. Although stress could not be completely ruled out, there

10.3 Results of the monitoring study

· 125

Table 10.2 Summary of reproductive success rates for pairs of captive foxes. San Miguel

Number of pairs Percent pregnant Percent with pups

Santa Rosa

Santa Cruz

2006

2007

2006

2007

2006

2007

12 33 25

7 71 0

16 75 31

14 64 29

29 38 24

11 73.0 54.5

was no clear evidence of elevated cortisol in the females that failed to produce pups. As in previous years, patterns of progesterone elevation again indicated that virtually all paired females had ovulated, and had physically mated. This was surprising since video review revealed bouts of aggression in all pairs, yet they also spent considerable time resting in contact, which is considered a sign of compatibility (Mech and Knick 1978). In some of the pairs, aggression at times escalated, resulting in injuries requiring veterinary treatment. This serious aggression was even documented during pregnancy, a time when it was least expected, especially for a monogamous species in which males have been seen providing paternal care (Garcelon et al. 1999). As previously mentioned, males of other canid species are notably solicitous of the female partners, even in captivity. Video monitoring presented further challenges. Because of budget limitations and the logistical challenges of operating solar-powered video on remote islands, not all pairs could be observed via video. Each enclosure had at least two den boxes, so that each female would have a choice and could move pups if she felt the need; however, the provision of an additional den box further reduced the chances of observing a birth and subsequent parental care. There was thus a trade-off between providing optimal conditions for breeding and providing conditions that would maximize collection of data. During the two seasons of the monitoring study (2006 and 2007) the possible points of reproductive failure were narrowed. Ultrasound examination at mid-gestation showed that almost all the females were ovulating and that many were also conceiving (Table 10.2). This was a conservative measure of pregnancy rates, since early embryo loss during the period before embryo implantation (e.g. red and gray foxes: Layne and MaKeon 1956) is quite common among mammals, with rates ranging from 20% to as high as 60% of pregnancies being lost in some species

126 · Reproductive biology (Wilmut et al. 1986). It was also the case that during 2006 many of the foxes were juveniles (first-year animals). Results from studies of other canids (e.g. coyotes: Green et al. 2002) have shown that reproductive rates can be considerably lower for juveniles, and in wild foxes on San Miguel and Santa Cruz Islands only 20% of first-year females produced pups, compared to 60% of older females (Roemer et al. 2001a, Coonan et al. 2005b). However, explaining pup loss following ultrasound confirmation of embryos was not simple. Although embryo loss may be common very early in pregnancy, it is much less so during later stages, when it would more properly be called spontaneous abortion. Causes for late-term abortion in domestic dogs include infections, toxicants, low concentrations of thyroid hormone, trauma, and stress (Romagnoli 2002, Verstegen et al. 2008). Veterinarians could rule out the particular infections associated with canine abortions, and it is likely that their enclosures were free of toxicants. Mastitis was detected in some females in 2005, which may have contributed to mortality in their pups, but subsequent treatment with appropriate antibiotics in advance of parturition reduced that threat (Chapters 8 and 11). Although aggression might appear a likely factor, the incidence of aggression did not appear any greater in pairs that failed to produce pups than in those that had healthy litters. As mentioned previously, the relationship between stress and cortisol in this species is incompletely known, but the cortisol concentrations in the available samples were not different in successful and unsuccessful pairs. To further complicate matters, pregnant females can lose pups through mechanisms other than abortion. Due to incomplete den box video coverage, the birth process could not be observed, so it was impossible to distinguish between pre-term abortion, full-term stillbirth, or neonatal death. The possible causes for stillbirth are the same as those for abortion, but neonatal death could be due to congenital defects intrinsic to the pup – causing failure to thrive – or to parental neglect or abuse. Unfortunately, the mere absence of pups was not very informative. As mentioned previously, animals may kill and consume their young if they perceive breeding success has been compromised, and it is also common for females to consume young that die naturally. The occurrence of mastitis in 2005 was an obvious source of pup mortality, and even adult female mortality. Mastitis is a bacterial infection of the mammary glands that can prevent pups from getting adequate nourishment. Because those infections can be treated or even prevented by antibiotic treatment, prophylactic treatment of potentially pregnant

10.4 Summary

· 127

captive females began in 2006. But there were still females without infections and without pups, which raised the concern about the possibility of infanticide. If indeed that was happening, the males were the more likely culprits. In fact, video review revealed some sequences in which males might be killing or consuming pups, but the images lacked sufficient detail to be certain. Paternal provisioning of pups has been documented for wild island foxes (Garcelon et al. 1999), suggesting this species is similar to other canids in showing biparental care. In such a parenting system males should not pose any danger to their pups, but there was no evidence that fathers were necessary for pup survival in captivity. The situation might well be different in the wild, where males might provide protection and bring food to the mother and pups. But in captivity, protection as well as unlimited food were provided, so there seemed to be no compelling reason to keep males with the females after pregnancy was established if there was a chance they were killing pups. For the 2008 breeding season, the males were removed after ultrasound examination confirmed pregnancies, and the effort to catch births on video was resumed. This, however, did not improve reproductive success. In that year 9 of 12 paired females became pregnant, but only four litters were produced. The discovery of fetuses in two of the pens confirmed abortion as immediate cause of pup loss. The possibility of abortion cannot be ruled out in the other litters, since the mothers could have consumed the fetuses. One of the possible causes of late-term abortion is low thyroid hormone levels, and this cause remains a possibility for island fox reproductive failure. In the few island foxes that have been tested, thyroid hormone levels have been lower than what would be considered normal for the domestic dog, the most closely related species for which there are published normal values. The effect of apparently low thyroid hormone levels on island fox reproduction could be investigated by comparing thyroid function in both reproductively successful and unsuccessful foxes. If fox reproduction is hindered by low thyroid function, then reproduction might be improved, at least in captive animals, by thyroid hormone supplements.

10.4 Summary Captive breeding was a necessary – and ultimately successful – recovery action for island foxes (Chapter 8), and it also provided an opportunity to study island fox reproductive biology, which was previously unknown.

128 · Reproductive biology As a result, we now know that island foxes have induced estrus and ovulation, which is unique among canids, and this difference may be related to other features of their mating system that makes them distinct. Island foxes differ substantially from wolves, the species perhaps best known for its complex social system, strong pair bonds, and solicitous parental care. All the canid species appear to be socially monogamous, but from there they diverge in terms of group size, degree of sociality, and age of dispersal of young. For example, male aggression toward females – something that appeared rather common even in the reproductively successful island foxes – is virtually unknown for other canid species, in which males tend to be very solicitous to females, especially during estrus and pregnancy. However, it is unclear whether this behavior is characteristic of free-ranging island foxes or merely an artifact of captive conditions. In another contrast with other canids, the reproductive rate was lower for captive-born island foxes, the opposite of what is usually observed in other canid species. This suggests that something, such as social experience during the juvenile period, might have been missing in the captive environment, something that apparently is less important to other canid species. Island foxes, as their name indicates, are island species that have lived for thousands of years in isolated populations that are, by necessity, small in comparison to mainland canids. The lack of predators on the islands has no doubt also shaped their behavior and social relationships. Small populations also may have resulted in inbreeding depression, perhaps manifest in impaired thyroid function in some individuals, jeopardizing pregnancy and causing late-term abortions. The occurrence and influence of these factors in free-ranging island foxes remains to be studied.

11 Diseases of island foxes, by Linda Munson r

Diseases occur in all wildlife populations to some degree, but usually go unnoticed unless an epidemic occurs or a species approaches extinction. When a species becomes endangered, the major factors contributing to morbidity and mortality need to be identified, the disease risks assessed, and appropriate interventions implemented to prevent extinction and ultimately to ensure recovery. When island fox populations declined precipitously on San Miguel, Santa Rosa, Santa Cruz, and Santa Catalina Islands in the late 1990s, disease was considered among the possible causes (Chapters 5 and 7). Similar population declines had occurred on San Nicolas Island in the 1970s and 1980s but the cause was not identified (Laughrin 1977, Kovach and Dow 1985). Few carcasses were retrieved during the catastrophic declines of the 1990s, and little historic information was available on the incidence of disease in island foxes; as a result it was difficult to determine if disease had played a role in these declines. The near extinction of four subspecies and need for information to guide recovery efforts prompted the initiation of a comprehensive disease surveillance program in 1998 (Coonan and Rutz 2001; see Chapter 5). The goal of the program was to determine the principal causes of death as well as the prevalence of underlying diseases that might impact foxes through reduced fertility and/or survival. For this survey, a comprehensive necropsy was conducted on all deceased island foxes to characterize their diseases and parasites, as well as to determine causes of death. Data from this surveillance, ongoing as of 2009, are the principal source of information for this chapter. Although this survey represents only a recent snapshot of island fox disease, it portrays the status of the current population that will be the founders of all future island fox populations.

130 · Diseases of island foxes

11.1 Disease in island populations The ecology of infectious diseases in insular populations differs from that in populations residing in larger complex ecosystems where new pathogens are commonly introduced and may have a variety of hosts to facilitate circulation and maintenance of pathogenic strains. Hosts in isolated ecosystems either acquire resistance to their pathogens or the hosts, and pathogens co-evolve to allow maintenance of low pathogenic strains. The loss of genetic diversity that typically occurs in insular populations can lead to loss of plasticity when populations are challenged with new pathogens, as well as to an increased expression of deleterious traits that would otherwise be diluted in populations with constant immigration/emigration. Because few mammalian species reside on the Channel Islands, and the island fox is the only native carnivore except for spotted skunks on Santa Cruz and Santa Rosa, pathogens affecting island fox populations would only persist through circulation within those fox populations or their environment. Most other mammals on the Channel Islands are either rodents (deer mice, and introduced black rats) or herbivores (mule deer and bison [Bison bison] on Santa Catalina and deer and elk on Santa Rosa), neither of which are known to host carnivore-specific pathogens. Domestic and feral cats and domestic dogs also reside on some islands and may contribute to carnivore pathogen ecology. The viruses, bacteria, protozoa, helminths, and arthropods currently circulating in island fox populations may have originated from the population founders that came from the mainland, or from other carnivores such as spotted skunks or domestic dogs and cats, historically and currently present on some islands. Multiple introductions of domestic dogs have been recorded, the first with Native Americans who settled the islands as early as 13,000 BP (Rick et al. 2008), and subsequently through ranchers, other permanent residents, and visitors during the twentieth century. Regardless of the source, co-evolution of the fox and these pathogens in island isolation should have minimized the impact of disease over time and led to the fox becoming a maintenance host, as has happened for many canid viruses (Garcelon et al. 1992, Clifford et al. 2006). Yet co-evolution with a limited number of pathogens carries the risk of losing diversity in genes that determine disease resistance, making that population potentially more susceptible to new pathogens. Since the 1960s the risk of introducing pathogens capable of severely impacting island fox populations has increased through intensification of human

11.2 Could viral disease explain the population declines?

· 131

activities, such as ranching, military operations, and tourism. The extent and nature of the population declines in the four island fox subspecies during the 1990s raised concerns that an introduced pathogen may have been the cause.

11.2 Could viral disease explain the population declines? Some canine viruses, when introduced into a naive population (one lacking immunity from previous non-lethal exposure) can cause widespread epidemics with extensive mortalities, similar in character to the declines in the San Miguel, Santa Rosa, Santa Cruz, and Santa Catalina island fox populations. Among such potentially lethal viruses are rabies, canine distemper virus (CDV), canine adenovirus (CAV) and canine parvovirus (CPV) (Barker and Parrish 2001, Williams 2001, Woods 2001). Similar catastrophic declines occurred in black-footed ferrets (Mustela nigripes) (Williams et al. 1988) and African lions (Roelke-Parker et al. 1996) from CDV and in Ethiopian wolves (Canis simensis) from rabies (Randall et al. 2004). Viral infections in wild populations are typically monitored through analysis of blood serum for evidence of antibodies against specific viruses. Although these antibodies indicate that an animal has been infected by a virus (historically or currently) and survived infection, they do not provide evidence that the virus caused disease or was lethal in this species. Establishing causation in an epidemic requires identifying lesions characteristic of viral infection in carcasses and confirming the presence of the virus in those lesions through molecular or immunohistochemical methods. What evidence of viral exposure was present following the decline of the four island fox subspecies in the 1990s? On Santa Catalina, the single fox carcass retrieved during the decline had lesions typical of CDV infection, and sequence analysis of viral DNA obtained from the lung by reverse transcriptase PCR indicated that the virus was CDV (Timm et al. 2009). Other Santa Catalina foxes were observed with clinical signs compatible with canine distemper and, overall, 95% of the Santa Catalina fox population was lost from their eastern range, an area representing 87% of the island (Chapter 7). Several east end foxes that survived the epidemic had high titers against CDV. Taken together, these data suggested that CDV infection was the principal cause of the Santa Catalina fox decline. On the other hand, there was no evidence that disease caused the population declines on the northern islands. The seven

132 · Diseases of island foxes carcasses recovered during the San Miguel decline died from a variety of causes (Coonan et al. 2005b), indicating that a disease epidemic was not occurring. Rather, necropsies confirmed that golden eagle predation (see Fig. 5.4) was the mortality cause for the majority of foxes on San Miguel, Santa Rosa, and Santa Cruz Islands (Chapter 5). The identification of CDV as the cause of the Santa Catalina fox decline was confounded by the unexpected finding of CDV antibodies in the Santa Catalina fox population on the west end of the island, where the population had remained stable (Timm et al. 2009). Those antibodies could have resulted from the current disease epidemic, or a previous one. Surprisingly, an all-island serosurvey conducted from 2001–2003 determined that numerous foxes on all islands had antibodies against CDV, including juvenile foxes that must have been recently exposed (Clifford et al. 2006). Yet no CDV-related mortalities had been detected in any island fox subspecies after the 1999 epidemic on Santa Catalina, despite close monitoring of all populations and necropsy of all dead foxes. Furthermore, analysis of fox serum collected in 1988, a full decade before the declines, detected CDV antibodies as well (Clifford et al. 2006). Together these findings indicated that CDV exposure was widespread among all six subspecies before and after the epidemic, and suggests that CDV is endemic among island foxes. It therefore remains unclear why CDV was so lethal during the Santa Catalina epidemic. Greater subspecies susceptibility of Santa Catalina foxes was not likely because healthy juvenile Santa Catalina foxes were among the CDV antibodypositive animals detected in the 2001–2003 survey (Clifford et al. 2006). Other factors that could account for the extent of mortalities on Santa Catalina include: 1. introduction of a more virulent CDV strain; 2. emergence of a more virulent strain from the CDV present on Santa Catalina; and 3. co-infection of foxes with CDV and another pathogen. Phylogenetic and evolutionary analyses of CDV isolated from the single fox sampled during the 1999 epidemic indicated that the virus was most closely related to mainland raccoon (Procyon lotor) CDV (Timm et al. 2009), although raccoons do not occur on the Channel Islands. Subsequent to the epidemic, several raccoons were found to have been unintentionally transported from the mainland on boats (J. King, Catalina Island Conservancy, personal communication), underscoring

11.3 Do other viruses infect island foxes?

· 133

the possibility that a more virulent and lethal strain of CDV was introduced into the Santa Catalina fox population in 1999. Co-infection with CDV and other pathogens could also account for the high mortality rate. The one dead fox available from the 1999 CDV epidemic also had systemic infection with the protozoan Toxoplasma gondii, a common and often fatal occurrence in other species due to the immunosuppressive effects of CDV infection (Williams 2001). Similar circumstances contributed to the extent of mortalities in African lions during the 1993 CDV epidemic, where fatalities occurred in lions coinfected with CDV and high burdens of the protozoa, Babesia (Munson et al. 2008).

11.3 Do other viruses infect island foxes? Serosurveys to detect antibodies against (and therefore to detect previous infection with) canine pathogens were conducted in 1988 and in 2001– 2003 in all six island fox subspecies (Garcelon et al. 1992, Clifford et al. 2006). In addition to CDV antibodies, these surveys detected antibodies against CPV in a large proportion of foxes from all islands and a high prevalence of CAV antibodies in all subspecies except Santa Catalina foxes. Both CPV and CAV can potentially cause significant mortality in young foxes (Woods 2001), although no deaths were attributable to these viruses during the pathology surveillance period (1998–2008). Antibodies were also detected against canine corona virus (CCV) in San Clemente and Santa Catalina foxes and against canine herpes virus (CHV) in Santa Rosa, Santa Cruz, and San Clemente foxes. Both CCV and CHV usually cause only mild disease, and no evidence of infection was detected in any necropsied fox. Overall, these surveys indicate that island foxes are exposed to many canid viruses, but disease is not the usual outcome of exposure. There is no evidence of rabies virus infection among island foxes on any of the Channel Islands. A small serosurvey of Santa Catalina foxes after the 1999 epidemic did not detect any antibodies against rabies virus (Timm et al. 2009), and there have been no typical signs of rabies in island foxes or any other susceptible species. However, should rabies be introduced from the mainland, it would pose a serious threat to the island fox due to its usual fatal clinical course in carnivores. The proportion of each subspecies having antibodies against specific canid viruses differed between 1988 and 2001–2003 (Garcelon et al. 1992, Clifford et al. 2006), suggesting that these viruses are circulating

134 · Diseases of island foxes periodically through the population. Alternatively, these variations may be due to population sampling methods. For viruses with low seroprevalence (CCV and CHV), failure to detect antibodies in a population at one time point may be due to sampling error, rather than indicating that these viruses are periodically eradicated or introduced into these ecosystems. The high prevalence of antibodies against CPV and CAV across all age classes and populations without evidence of associated mortalities in these closely monitored populations indicates that these circulating viruses are endemic and of low pathogenicity in the island fox, likely due to co-evolution from continued exposure.

11.4 Do non-viral pathogens infect island foxes? The 1988 and 2001–2003 serosurveys also investigated island fox exposure to the bacteria, Leptospira interrogans serovars canicola, icterohaemorrhagiae, bratislava, and pomona, and the protozoa, T. gondii. Antibodies against L. interrogans serovar icterohaemorrhagiae were found only in Santa Cruz foxes, serovar bratislava in Santa Rosa, Santa Cruz, Santa Catalina, and San Clemente foxes, and serovar pomona in Santa Rosa foxes only. Antibodies against T. gondii were detected in all subspecies except San Miguel foxes. Except for the one Santa Catalina fox that died from CDV and toxoplasmosis, there has been no evidence of disease or mortalities from either T. gondii or Leptospira in island foxes. A survey for antibodies against Bartonella vinsonii ssp. berkhoffii (Bvb) and Bartonella clarridgeiae (Bc) – bacteria associated with endocarditis and liver disease in humans and dogs – was conducted between 2001–2004 across all islands (Namekata et al. 2008). Antibodies against both Bartonella species were found in a quarter of the foxes tested and in all subspecies. No disease characteristic of Bartonella infection has been detected to date, suggesting that the fox may be a reservoir for this organism, and not a susceptible host.

11.5 Parasites infecting island foxes When foxes first colonized the islands (8,000–16,000 BP; Chapter 2), they likely brought with them a diverse fauna of canid endoparasites (internal parasites). This endoparasitic fauna has probably been diversified by several introductions of domestic dogs and cats since then, as well as by accidental introduction of other mainland wild mammals. Pathology and fecal parasite surveys indicate that endoparasitism is widespread in island foxes, but in only a few cases have these parasites caused significant lesions, reflecting acquired tolerance from a history of co-evolution.

11.5 Parasites infecting island foxes

· 135

Figure 11.1 The colon of an island fox opened to reveal parasite nodules that are present in the wall and project into the lumen. Two nodules have been incised to expose the nematode, Spirocerca sp.

Angiocaulus gubernaculatus – a lung and intravascular parasite whose usual hosts are mustelids – was first identified in two San Miguel foxes found dead toward the end of the decline (Faulkner et al. 2001, Coonan et al. 2005b). The extensive lesions from this parasite in the lung of one fox likely contributed to its death. Angiocaulus infections were subsequently detected by fecal egg identification in approximately one third of San Miguel foxes (Sohn and Thomas 2005) but without clinical signs or significant lesions at death. Angiocaulus has not been found in any of the other subspecies of island fox. Uncinaria stenocephala – a common intestinal hookworm of dogs and cats – is present in most San Miguel foxes and has been identified in some foxes on Santa Rosa and Santa Cruz, but not in subspecies on the southern islands. Although one San Miguel fox that died near the end of the decline had extensive infestation that likely contributed to death (Coonan et al. 2005b) most foxes harbor subclinical infections. A common and potentially pathogenic parasite in island foxes is Spirocerca. This unique Spirocerca resides in the wall of the colon causing thick fibrous plaques (Fig. 11.1). Because of this consistent location, this parasite does not appear to be Spirocerca lupi of domestic and wild canids, which usually resides in the esophagus and aorta. A similar Spirocerca

136 · Diseases of island foxes has been noted in gray foxes from northern California (Lavopierre et al. 1986), and so presumably the island fox Spirocerca originated from founder mainland foxes that settled the islands. A high prevalence of infection with Spirocerca occurs in San Miguel, Santa Rosa, Santa Cruz, and San Nicolas fox populations, but only a few infected foxes have been found on San Clemente and Santa Catalina. Most infections appear to be incidental findings at necropsy, but some animals have had intestinal obstruction or perforation from the parasite. Infections are generally more severe in San Nicolas foxes, which may reflect less genetic resistance in this homogenic population (however, see Aguilar et al. 2004, Hedrick 2004). However, the fact that the San Nicolas population continues to be robust (Garcelon and Hudgens 2008) suggests that this parasite appears to have little impact at the population level. The very low prevalence in San Clemente and Santa Catalina fox populations may reflect either genetic resistance or low densities of an intermediate host essential to maintain this multi-host parasite in the environment. The most prevalent parasite of island foxes is Mesocestoides, a tapeworm found in all island fox subspecies. Only adult intestinal forms of Mesocestoides have been identified in the island fox, indicating they are a definitive host. None of the harmful metacestodes forms have been found in the body cavities of any island fox. Deer mice, western fence lizards (Sceloporus occidentalis), and ants (Lasius niger and Tapinoma sessile) are intermediate hosts for this parasite and the likely sources of infection for island foxes (Padgett and Boyce 2005). Large tapeworm burdens have been noted in otherwise healthy animals, so the impact of this parasite on island foxes is presumed to be minimal. Other parasites rarely found in island foxes include Toxascaris (the common intestinal round worm of canids), an unidentified lung nematode, an unidentified pancreatic trematode, and various protozoa including Giardia, Isospora, and Sarcocystis (Sohn and Thomas 2005). No adverse health effects have been noted in island foxes with any of these infections. Dirofilaria immitis – the common carnivore heartworm – was suspected to infect island foxes, based on positive results for Dirofilaria antigen in R (IDEXX Laboratories, Westthe serum of many foxes using PetChek  R brook, Maine, USA) and DiroCHEK (Synbiotics Corporation, San Diego, California, USA) (Crooks et al. 2001b, Roemer et al. 2001b). The prevalence of antigen-positive foxes on San Miguel, Santa Rosa, Santa Cruz, and San Nicolas ranged from 25–100%, whereas San Clemente and Santa Catalina foxes were negative. However, no heartworm infections have been found in any of the more than 500 foxes necropsied over the last 10 years, suggesting that these tests for Dirofilaria antigen

11.5 Parasites infecting island foxes

· 137

Figure 11.2 View through an otoscope of the ear canal from a Santa Catalina Island fox with moderate ear mite infection. Courtesy of Winston Vickers.

are not specific and likely cross-react with another parasite antigen. The geographic distribution of Dirofilaria antigen-positive foxes parallels that of Spirocerca infections, so cross-reaction with Spirocerca antigens by these tests is most likely. This illustrates the importance of using caution when interpreting the results of commercial tests in species for which the tests were not validated. Ectoparasites have also been found on all island fox subspecies. A survey of Santa Cruz foxes found fleas (Pulex irritans), lice (Neotrichodectes mephitidis), and ticks (Ixodes pacificus) (Crooks et al. 2001a). Heaviest burdens of lice and fleas have been recorded on Santa Catalina and Santa Cruz during monitoring of wild fox populations in the dry season on all six islands (D. Garcelon, T. Coonan, and W. Vickers personal communication). Heavy infestations with lice have been noted at necropsy in several foxes that were emaciated, but it is difficult to distinguish cause from consequence in these cases. Ear mites (Otodectes sp.) are common in foxes from all the southern islands (Fig. 11.2), but absent from northern island foxes. In many Santa Catalina foxes and some San Nicolas foxes, there is a marked

138 · Diseases of island foxes

Figure 11.3 Kidney of an island fox with amyloidosis. The medulla (center) of the kidney is swollen from deposition of amyloid, and the papilla (lower center) is necrotic due to loss of circulation from amyloid deposition.

inflammatory response to the mites that may cause considerable debility and compromise hearing (W. Vickers personal communication). Ear mites only occur in foxes on islands that also have feral cats, so an historical introduction of this parasite to the islands by cats is suspected. However, feral cats on the islands now only rarely harbor ear mites (D. Clifford and W. Vickers personal communication), so the island fox is now likely the principal host for this Otodectes.

11.6 Non-infectious diseases in island foxes Several non-infectious disease conditions have contributed to island fox mortalities and therefore may be of significance to ongoing population recovery and survival. These degenerative and neoplastic conditions include amyloidosis, ceruminous gland cancer, systemic mineralization, and thyroid disease. Amyloidosis (Fig. 11.3) is the deposition of abnormal proteins in organs and tissues leading to tissue damage, organ failure, and sometimes

11.6 Non-infectious diseases in island foxes

· 139

Figure 11.4 Cross section through the head at the level of the ear canal of an island fox with ceruminous gland carcinoma. The cancer has invaded the muscles and bone on the right side and extended into the brain and oral cavities.

death. In island foxes, amyloid deposits have been found in the oropharynx, heart, kidney, lung, intestines, spleen, liver, and skin. Affected foxes usually become emaciated from kidney failure or malabsorption secondary to amyloid deposition. Some foxes developed difficulty breathing due to obstruction of their larynx from amyloid deposits. The cause of amyloidosis in all species is not known, but there is a genetic predisposition in some species (Rideout et al. 1989, DiBartola et al. 1990, Papendick et al. 1997). Among island foxes, amyloidosis has been identified in all subspecies, but is more prevalent and severe in San Clemente foxes, suggesting a genetic influence. Additionally, amyloidosis tended to be more severe in captive than wild island foxes, possibly reflecting the exacerbation of amyloidosis by stress, as has been noted in other species (Hoffman and Leighton 1985, Lipman et al. 1993). Ceruminous gland carcinomas (Fig. 11.4) are cancers of the ear canal glands that were first noted in Santa Catalina foxes in 2001 and are now found in a large proportion of adult foxes. The prevalence in foxes greater that 3 years old is approximately 35% (W. Vickers and L. Munson, unpublished data), one of the highest cancer rates reported in any wildlife population. Foxes have had significant negative health effects from this

140 · Diseases of island foxes cancer, including morbidity and mortality from local tissue invasion, metastasis, or secondary infections (W. Vickers personal communication, L. Munson unpublished data). Since intensive study of this cancer began in 2003–2004, many tumors have been detected in an early stage and removed (W. Vickers personal communication); this action may reduce the morbidity and mortality from this cancer in the wild population. This cancer has not been found in any other subspecies of island fox. Ceruminous gland cancers occur in Santa Catalina foxes with severe inflammation of the ear canal, a reaction that includes damage to the ceruminous glands that may lead to cancer. Why Santa Catalina foxes are more prone to develop more severe inflammation than other subspecies is under investigation. A common lesion found in all island fox subspecies is small thyroid glands with scant reserves of thyroid hormone (thyroid atrophy). Thyroid gland tumors also occur in island foxes and may be an outcome of continued stimulation of the thyroid gland to produce more thyroid hormone. Thyroid hormone is essential for metabolic homeostasis and reproduction, so low thyroid hormone levels can have subtle effects on well-being. Whether these atrophied glands affect island fox health is currently not known. The cause of thyroid hormone depletion is also uncertain. Similar lesions occur in other species exposed to environmental toxins, such as perchlorate and polychlorinated biphenyls, which are compounds that can interfere with thyroid hormone production (Rolland 2000). These contaminants are present in the Channel Island environment. A genetic basis for thyroid atrophy is also possible, as small thyroid glands may have evolved along with island miniaturization.

11.7 Genetic diversity versus disease resistance All island fox subspecies are genetically depauperate with San Nicolas foxes demonstrating the least diversity, having no variation in hypervariable markers (Gilbert et al. 1990; see Chapter 2). This absence of heterogeneity may influence the susceptibility of island foxes to disease, both infectious and non-infectious. The major histocompatibility complex (MHC) Class II loci encode proteins that regulate an animal’s response to pathogens and therefore contribute to the evolution of pathogen resistance. Assessment of genetic variability of MHC Class II loci in island foxes demonstrated considerably more variation than noted in neutral loci, including San Nicolas foxes (Aguilar et al. 2004). Therefore, genetic variation may have been retained in critical loci that increase fitness

11.8 Disease as a cause of death

· 141

to island fox populations. Diversity of MHC Class II proteins within the population should theoretically promote survival of some individuals when infected with a pathogen. It is unclear whether small populations such as the island fox have sufficient numbers to survive lethal pathogens even with MHC diversity. The high prevalence of foxes with antibodies against potentially lethal pathogens such as CAV, CPV, and CDV implies that they have evolved the ability to be infected with these pathogens without extensive mortalities. This tolerance of endemic pathogens does not preclude their succumbing to a more lethal variant, should one be introduced or evolve from existing strains via mutation. Of greater interest and possible importance to island fox health is the apparent predisposition of subspecies to develop some diseases. Examples include amyloidosis in San Clemente foxes, more severe Spirocerca lesions in San Nicolas foxes, and the notably more severe otitis in Santa Catalina foxes that likely predisposes them to develop ceruminous gland cancer. This propensity for Santa Catalina foxes to develop severe otitis may now be a trait that is fixed in the population after the recent genetic bottleneck following the 1999 decline. In an insular genetically depauparate species such as the island fox (Gilbert et al. 1990), sublethal genetic defects may be more common, but lethal defects are usually eliminated. Surprisingly few congenital defects have been identified in island foxes to date. Defects have included cryptorchidism (intra-abdominal testis), fused tail vertebrae, and a heart defect. The genetic basis for these defects has not been established.

11.8 Disease as a cause of death Disease is a natural part of any wild population and usually contributes to mortality in aged individuals that have already contributed reproductively to the population. Mortality rates from natural causes in wild populations are usually sustainable in the absence of an epidemic. However, anthropogenic influences can greatly alter that balance by removing reproductively active animals from the population. Island fox populations have been impacted by several anthropogenic factors, the greatest being death from vehicular trauma. Vehicular trauma has been the major cause of death of necropsied foxes from San Clemente and San Nicolas Islands. In spite of the number of trauma-related deaths sustained by the San Nicolas and San Clemente foxes, those populations remain robust. Vehicular trauma has also been a major cause of death in the Santa Catalina population. This capacity to

142 · Diseases of island foxes maintain stable populations despite continued removal of reproductive adults from the population due to unnatural causes bodes well for persistence of island foxes. Other anthropogenic causes of mortality have been R and entrapment accidental poisoning with the rodenticide Quintox or drowning in structures around human habitations. On the northern islands, golden eagle predation has been the major cause of death in the 1990s and 2000s (Roemer et al. 2001a, Chapter 5). The extent to which disease has contributed to island fox mortality has been difficult to determine because mortalities in foxes not monitored through radio-tracking usually go undetected. Deaths in wild foxes have occurred from septicemia secondary to traumatic wounds (possibly from intraspecific aggression), ceruminous gland cancer, and emaciation secondary to amyloidosis. Disease was the principal cause of death in captive animals, including bacterial septicemia from fight wounds or mastitis, ceruminous gland cancer, amyloidosis, and neonatal deaths from maternal neglect.

11.9 Overall health of the island fox populations As of 2009 disease did not appear to be a significant threat to island fox recovery. The decision to vaccinate foxes against CDV and rabies after the epidemic on Santa Catalina provided a safety net against the most serious threats to recovery and sustainability (Chapter 7). Strategies in recovering populations on San Miguel, Santa Rosa, Santa Cruz, and Santa Catalina included vaccinating all foxes released from captivity or trapped for other reasons until a core group of animals is vaccinated (Schwemm 2008b). As the populations recover on those islands, the core vaccinated group will become a smaller proportion of the population, providing opportunity for continued evolution of disease resistance. On San Nicolas and San Clemente, a core number of foxes have been vaccinated and some radiocollared foxes left unvaccinated to serve as sentinels for early detection of an epidemic. Long-term disease monitoring of all fox subspecies is necessary in order to detect emerging health issues or shifts in prevalence of existing diseases in the populations. The Fox Health Group of the Island Fox Conservation Working Group has developed the long-term strategies for health surveillance and protocols for an epidemic response, removal of introduced species, quarantine of foxes or dogs moved between the islands and the mainland, and collection of biomaterials (Schwemm 2008b). Monitoring in live foxes will include opportunistic collection and

11.9 Overall health of the island fox populations

· 143

archiving of biomaterials (whole blood, serum, and feces) and physical assessment of all foxes handled for any reason. Additionally, all deceased foxes will continue to have complete necropsies to determine cause of death and to detect underlying diseases. Island foxes will always be vulnerable to canid and other diseases. It is likely, therefore, that a vaccinated core of foxes will be maintained on each island, and that foxes will be monitored for the foreseeable future. Such vigilance is required to avoid a repeat of the disease-caused collapse of Santa Catalina Island foxes in 2000.

12 Zoos, education, and public participation r

In some ways protected islands are exceedingly good places to implement rare species recovery; restricted human access nearly eliminates direct impacts (e.g. poaching or illegal hunting), while the absence of adjacent lands reduces inputs that degrade habitat or hinder population growth (Janzen 1986). However, species isolation also limits opportunities for most people to see and interact with plants and animals, weakening the emotional connection that leads to support for recovery programs (Solomon 1998, Restani and Marzluff 2002, Rabb and Saunders 2005). Until recently relatively few people had ever seen an island fox or were familiar with the species (K. Dearborn, Friends of the Island Fox, personal communication). When it was apparent that several subspecies of island foxes faced extinction, the need to increase public awareness and gain support for risky and sometimes controversial recovery actions became a priority. The goal of successfully educating the public in a relatively short period of time fell to citizen advocates, zoos, and environmental educators, and a chronicle of recovery would be incomplete if it did not include the important contributions made to incorporate the public as much as possible within the island fox recovery program.

12.1 Public advocacy The first efforts to increase public awareness of island foxes were undertaken by a unique environmental group organized and run by children. In 1999 Alexandra Morris was a third grader whose father was an archeologist at Channel Islands National Park. The captive breeding program for island foxes had begun, and as Alexandra heard more about the fox situation she became increasingly concerned. Over the Christmas holidays that year she and her family spent five days on San Miguel, caring for captive foxes during staff leave. Alexandra returned from the island

12.1 Public advocacy

· 145

determined to become active in fox conservation, and in the ensuing months organized two school events to introduce her teachers and peers to the plight of the foxes. The response was so positive that Alexandra and four of her friends formed an organization called Save Our Species (SOS) to begin small fundraising and outreach efforts. With parental encouragement and support (particularly from Alexandra’s mother, Susan), the group expanded their activities from the school outward to the local community. SOS met with local city officials, who responded in part by declaring ‘Island Fox Days’ in several cities, and the group was recognized by the California State Legislature, the US Congress, and NPS for their efforts. The most lasting effort undertaken by SOS began in 2001 with the establishment of an annual Island Fox Festival at the Santa Barbara Zoo. These events were organized by the students in conjunction with zoo staff and focused on family activities that introduced thousands of zoo visitors to island fox and island conservation issues. The visibility generated by the festivals and other SOS activities ultimately led Dr. Jane Goodall to meet with SOS on several occasions, and the group elected to participate in her Roots and Shoots program, giving the children and the island fox international recognition. The once third graders are now in high school, but during the group’s tenure SOS cumulatively raised over $10,000 to support island fox captive breeding and recovery efforts, and these young adults will always know that they contributed to recovery of an endangered species. There was also a need in the public advocacy arena for a more formal and permanent entity to advocate for island fox conservation. In 2005 a group of volunteers formed Friends of the Island Fox (FIF), a nonprofit organization whose primary mission was to educate the public about island foxes and the conservation efforts needed to protect foxes and their habitats. FIF was extremely effective in increasing awareness of island foxes both through direct educational experiences and through its website. Much of FIF’s success is due to the commitment of several of its founding members, and as is characteristic of many of the collaborators in the recovery effort, its close partnerships with agencies and NGOs (Clark and Brunner 1996). FIF has participated for many years in the annual island fox working group meetings (Chapter 13), and developed a formal cooperative agreement with NPS. These inclusive actions by the regulatory agencies illustrate the mutual respect between managers and advocates, and serve to further strengthen public support for recovery efforts.

146 · Zoos, education, and public participation

12.2 The role of zoos Providing a connection between people and wild animal species has long been the unique role of zoos (Rabb and Saunders 2005). With the ever-increasing loss of biodiversity has come the recognition that zoos also have the ability (and some would say the responsibility) to expand their mission to ecological conservation and endangered species recovery (Hutchins and Conway 1995, Baker 2007, Sterling et al. 2007). Over the last several decades the worldwide zoo and aquarium community has been successful in coordinating ex situ (‘out of place’, in this case meaning away from a species’ natural range) rare species breeding and reintroduction programs (Hutchins 2003), but the participation by the Santa Barbara Zoo and other zoos in island fox recovery vividly illustrated how zoos can also be a critically important component of in situ (in a species’ range) conservation programs. It is in fact difficult to imagine how island fox recovery success could have been so quickly achieved without their help. 12.2.1 Mainland exhibits

The first contribution of local zoos to island fox conservation was to facilitate the public display of island foxes for the first time. In 1999 island foxes were preying heavily on endangered San Clemente Island shrikes, and a lethal fox removal program to protect the birds was planned (Chapter 13). To reduce the number of foxes to be killed, IWS worked through the Association of Zoos and Aquariums (AZA) to contact zoos that might be interested in taking foxes into their collections. The Santa Barbara Zoo agreed to take four of the 14 foxes removed from San Clemente Island. This is not something that a zoo would normally do, given the general scarcity of space and funds for housing animals that do not have a specific educational or breeding function (A. Varsik personal communication). But the timing coincided with the zoo’s desire to increase their focus on regional species and conservation issues. The remainder of the foxes were distributed among several California facilities and the Hogle Zoo in Utah, and were eventually put on display at each site, providing the first opportunity for the public to view island foxes on the mainland. In addition to providing sanctuary, the Santa Barbara Zoo and other zoos holding foxes began using their animals to communicate the increasing threats to the species and the need for conservation and protection.

12.2 The role of zoos

· 147

For example, during an Island Fox Day at the Los Angeles Zoo, volunteers used radio collars and receivers identical to those used on the islands to demonstrate how biologists were tracking and counting foxes in the wild; children enjoyed following the ‘lost’ docent, and parents were occasionally persuaded to contribute to the purchase of more radio collars. Activities like this not only promoted island fox conservation, but in a larger sense hopefully contributed to an increased awareness of environmental responsibility in the children who attended the programs (Price et al. 2008). 12.2.2 Husbandry practices and collaboration

The Santa Barbara Zoo also played a critical role in developing husbandry practices for captive foxes. In 2001 the Santa Barbara Zoo organized the first of several island fox husbandry meetings, bringing together small canid biologists, zoo caretakers and island staff to determine how best to care for foxes and facilitate breeding. The series of husbandry meetings held across several years provided information that aided island managers in caring for captive foxes, and led to the development of the first island fox husbandry manual. A secondary, but extremely valuable, result of these meetings was the professional relationships that developed between zoo staff and island caretakers. The perceived distance between the islands and the mainland is often larger than the true distance, and can sometimes hinder cooperative efforts between island and mainland workers. The husbandry meetings provided a common venue to discuss emerging caretaking concerns, thus avoiding knowledge gaps that can occur in rapidly developing programs (Smith et al. 2007, Coonan et al. 2010). Finally, the collaboration led to trips by zoo staff to the islands to assist with fox care, support that was greatly appreciated by island fox managers. It also provided zoo staff with the experience of working in fox habitats. 12.2.3 Species management

Genetic management of the captive populations required the development of a studbook (Species Survival Plan [SSP]; see also Chapter 2), which the Santa Barbara Zoo staff volunteered to maintain. In partnership with island managers, staff from the Santa Barbara Zoo and the Association of Zoos and Aquariums (AZA) used the studbook to determine

148 · Zoos, education, and public participation captive pairings and to select individuals for release. The Santa Barbara Zoo staff also currently coordinates all efforts in support of island fox conservation by AZA institutions. Captive breeding is no longer needed for island fox conservation, but should captive breeding be required in the future or a research population be established, the SSP will be critical for oversight and coordination. 12.2.4 Future role for mainland populations

Generally there are three primary reasons to maintain captive populations of rare species: education and public contact, scientific research, and long-term management of breeding populations to provide animals for reintroduction (Baker 2007). The education opportunities provided by mainland captive foxes have been extremely successful; island fox festivals still draw many visitors, and the Santa Barbara Zoo has made island foxes one of the primary attractions in a new exhibit entitled ‘California Trails’ that includes several endangered species native to California including condors (Gymnogyps californianus), desert tortoises (Gopherus agassizii), and island foxes. The reproductive success of the mainland populations is still low, however, and at present their future depends on the infusion of animals from wild populations. This effort is fully supported by the IFCWG, and a long-term protocol for transferring foxes from the islands is being developed that will hopefully support mainland education populations for many years to come. There are many questions that remain regarding island fox and general small canid biology that could be addressed by a research population of foxes on the mainland. A research population could be used to both address general questions on canid biology, as well as issues directly related to island fox conservation, specifically reproductive biology and disease (C. Asa, Saint Louis Zoo, personal communication). As with the educational efforts, the future of a mainland research population is dependent on the availability of funding and facilities, as well as stable island populations from which animals could be removed without diminishing the viability of wild populations. There is little disagreement regarding the benefits of a mainland research effort, although the financial and oversight challenges would need to be addressed. The need for a mainland breeding population that could be used to produce animals for reintroduction was one of the more challenging topics addressed by the recovery team. Ex situ populations of rare species

12.3 Tachi and Finnegan

· 149

are often critical to recovery, and have been incorporated successfully in many endangered species recovery efforts (the approach is, however, often debated, see Hutchins and Conway 1995, Balmford et al. 1996). Island foxes present unique challenges for incorporating mainland captive breeding into recovery, specifically the goal to preserve all subspecies and the biological difficulties in transferring animals back to the islands. If one island subspecies were to go extinct, the IFCWG has agreed that translocating a subspecies from another island would clearly be ecologically preferable to having no foxes, but the source of those animals could be either a wild population or a mainland captive population. For example it could be argued that the San Miguel population has the highest risk of extinction because it is the smallest, and thus most susceptible to stochastic variability leading to critically low population numbers. Although it might make sense to maintain a large breeding population of San Miguel foxes on the mainland, there would also be a risk to removing animals from San Miguel Island precisely because of the small population size. Regardless of which subspecies was selected, maintaining a separate mainland population with sufficient genetic diversity to serve as a source population would require a large number of animals with associated space and funding needs. Consequently, while the draft recovery plan supports the maintenance of a mainland population (of one or at most two subspecies; Schwemm 2007) for reintroduction if needed, this is not a required action, and it is generally agreed that if in the future an island sub-population were to go extinct, reintroduction would occur from one of the other islands.

12.3 Tachi and Finnegan The use of ‘ambassador’ animals – captive individuals that are purposely managed to come in frequent and close contact with humans – have an ambiguous role in wildlife conservation. Usually maintained by zoos and wildlife rescue organizations, domesticated wild animals allow the public, particularly children, to personally interact with otherwise elusive creatures. At the same time, the fact that someone can ‘pet’ a tiger or a dolphin has the potential to change the perception of wildness that many people feel should be associated with wildlife. There were several instances when island fox pups born in captivity on the islands had to be removed from their mother and hand-reared due to parental neglect. Those pups eventually became highly tame and have since been used as ‘ambassadors’ for island fox conservation (Fig. 12.1 and Fig. 12.2).

150 · Zoos, education, and public participation

Figure 12.1 The animal ambassador island fox ‘Tachi’ on Santa Catalina Island, with veterinarian Winston Vickers. Courtesy of Catalina Island Conservancy.

Tachi (‘our little girl of hope’ in Chumash) was born in the captive facility on Santa Catalina Island in 2003. She was neglected by her mother and while being held for fostering to a new mother in the captive breeding facility, she required treatment for an eye infection. Though she was eventually fostered to a new mother successfully, during her brief stay in the IWS clinic she apparently imprinted on humans. This was evident when caretakers entered her pen to clean. At such times Tachi would run up to the humans, whereas her litter mates would all run to the far end of the pen. IWS determined that it would not be appropriate to release her to the wild and was granted a permit by the Department of Fish and Game to use her for public education. A special pen was constructed separate from the captive breeding facility and this is visible to the public. Since then she has been handled frequently by a small group of caretakers, and is shown often at public events that promote island fox and habitat conservation efforts. She has never been taken off the island, but Tachi interacts with hundreds of island visitors and residents each year, serving as a focal point for communicating actions needed to protect foxes, such as restraining dogs and cats and driving slowly on island roads (J. King personal communication). At the Santa Barbara Zoo, a male pup was born in 2007 to an older female who likewise neglected him. No foster mothers were available,

12.3 Tachi and Finnegan

· 151

Figure 12.2 The animal ambassador island fox ‘Finnegan’ at the Santa Barbara Zoo. Courtesy of Catherin Schwemm.

and at only a few days of age he was also removed from his mother and hand-reared. Caretakers soon realized that the pup (eventually named Finnegan) had an extremely calm but inquisitive personality, and the zoo decided to try him as an ambassador. ‘Finn’ has adapted easily to his role, and in only two years has come in contact with thousands of zoo visitors, certainly more than any other island fox in history. While the sight of Tachi and Finnegan in the arms of their handlers with leashes attached certainly conveys the message that these individuals are no longer wild, most people understand the difference between an animal who has by necessity been cared for by humans and the remainder of their species who truly are wild. The experience of seeing and interacting with animals that represent threatened populations can greatly increase a person’s interest in conservation and – especially for children – enhances their desire to participate in activities that will help protect animals and their habitats (Price et al. 2008). If Tachi and Finnegan have encouraged the children and adults with whom they have interacted to be more active in island fox protection and conservation, then their loss of wildness should be considered an investment in the future of their species, and one that they both seem content to have made.

152 · Zoos, education, and public participation

12.4 Environmental education Few entities have more experience in communicating the importance of protecting nature than does NPS, and the ability of the interpretive staff at Channel Islands National Park to immediately focus on the island fox situation once the threats were identified was critical to generating publicity. For example, very soon after the golden eagle–fox connection was established, an exhibit in the park’s visitor center presented a prepared specimen of an island fox next to one of a golden eagle. The comparison of sizes of prey and predator dramatically illustrated the danger to foxes from eagles, something that even children not yet old enough to read could easily understand. Another program that was particularly successful at overcoming the challenges of island isolation linked elementary students with field biologists via radio communications. Prior to school visits, NPS interpretive staff communicated with field biologists to determine if they (the biologists) would be available during a specified period. If they were, the educator would then call the biologist on the park radio while the students were listening and ask the biologist to describe how he or she was tracking foxes using radio collars. A demonstration collar was available in the classroom for students to see, and in some cases students could even see a picture of the individual fox being followed. This educational project required full support of NPS management; for example, park radio frequencies are normally used only for emergencies and business matters, and these transmissions were obviously neither of those. But this innovative approach increased not only the appreciation of the students for island fox conservation and the mission of the NPS, but also indirectly improved the working relationships between NPS staff in various management roles (biologists, educators, and managers). The fox recovery program benefited enormously for nearly 15 years from the expertise and experience of NPS educators.

12.5 Summary Long-term protection of endangered species and their habitats requires not only the expertise of scientists and managers, but also the consistent support of the general public. The Endangered Species Act (‘Act’ or ESA) has come under strong pressure over the last decade, with periodic efforts put forth to repeal or limit its scope (Shogren and Tschirhart 2001). Studies have shown, however, that the support of the public to

12.5 Summary

· 153

the principles included in the Act remains high, a factor that translates directly to continued political support in Congress (Restani and Marzluff 2002). It is therefore not a luxury but a necessity to have public support for conservation efforts, especially those conducted on public lands at taxpayer expense. Because of the work of educators, zoos, and volunteers, the public strongly supported island fox conservation efforts, including feral pig removal from Santa Cruz Island (which required the killing of thousands of animals). But perhaps it was the animals themselves that contributed the most to public perception and support. Humans, intentionally or not, assign value to species in direct relation to their appearance. Although most biologists would hesitate to make such a statement in public, island foxes are extremely cute, and the simple visual attraction of a species, while probably irrelevant in an ecological sense, directly affects the willingness of human beings to protect it (Metrick and Weitzman 1996, Walpole and Leader-Williams 2002). Certainly there is a risk in marketing endangered species based on their appeal, but ultimately it is the ecological value of island foxes that needs protection. And if the risk of less action is the permanent loss of foxes and the island systems they enhance, then bringing all tools to bear is the only acceptable strategy.

13 Managing recovery r

Cooperative conservation, politics, and the Endangered Species Act Recovery of island foxes occurred relatively quickly: within 10 years populations of all four endangered subspecies grew from levels nearing extinction to almost recovered, and the primary threats to the wild populations had been mitigated. The effectiveness of the recovery actions was enhanced by the cooperation among management agencies, scientists, and the public in developing and implementing recovery strategies. The most tangible example of this cooperation was the consistent work of a voluntary group, the Island Fox Conservation Working Group (IFCWG), a collaboration of representatives from agency, academic, and non-profit organizations. The IFCWG began meeting before island foxes were officially listed as endangered, and was instrumental in the development and evaluation of island fox recovery actions. Listing of island foxes came relatively late in recovery, and was accompanied by the incorporation of the IFCWG into a somewhat uniquely constructed US Fish and Wildlife Service (USFWS) recovery team. This chapter discusses the importance of working groups and collaboration in endangered species recovery, the costs and benefits of formal Endangered Species Act (ESA) listing to the recovery efforts, and the overall implications of listing on island fox recovery.

13.1 Stakeholders and recovery The existing conservation and land protection mandates in place provided managers and biologists the relative freedom to implement various recovery actions with nearly full support of all of the landowners within the entire range of the species. This circumstance, fortuitous for island fox recovery, is likely the exception rather than the rule for conservation of endangered species. Four of the six island fox subspecies (the four that were listed as endangered) occur on lands owned and/or managed by the National Park Service (NPS) and two private conservancies, The Nature Conservancy (Santa Cruz Island) and the Catalina Island Conservancy.

13.1 Stakeholders and recovery

· 155

The mission of all three entities is strongly conservation oriented, and each made recovery of island foxes a high priority within their organization. No island fox range was owned by entities wishing to develop the lands or use the lands in such a way as to negatively impact island foxes. Because of this, there were few groups opposed to island fox recovery actions or to listing the species as endangered. The two island fox subspecies that were not listed as endangered occur on islands owned and managed by the US Navy (USN), whose primary mission is national defense, but which does implement conservation programs and has responsibility for conservation under state and national law. The USN has funded several monitoring and research conservation efforts for island foxes, projects which provided important information for management of the endangered subspecies (see, for example, Spencer et al. 2006). Ironically, the USN managed one of the most challenging endangered species issues when recovery actions for the endangered San Clemente loggerhead shrike (Lanius ludovicianus mearnsi) at one point included killing island foxes that threatened shrike nests (see Box 13.1 ‘Island foxes versus endangered shrikes’). Box 13.1 Island foxes versus endangered shrikes At one point island foxes were being killed by the US Navy (USN) on San Clemente Island to protect nests of an endangered bird species. The entire population of loggerhead shrikes had declined to 12 birds by 1998, and USFWS had required USN to protect the species (Cooper et al. 2005). Shrikes were the subject of an intense captive breeding and reintroduction program (Heath et al. 2008), and video monitoring of nest sites had documented successful nest predation by island foxes (Cooper et al. 2005). Initially a system was deployed that involved affixing shock collars to foxes living near shrike territories. If a fox approached the immediate area of a shrike nest it would receive a shock. The USFWS later required USN to trap island foxes that were within 300 m of shrike nest sites active within the last year and transfer them to mainland zoos, or euthanize them. As a result, 14 foxes were transferred to mainland zoos in spring 1999, and 15 were killed. The USN eventually stopped killing and removing foxes and utilized other methods, including continued use of shock collars, and holding foxes in captivity for the duration of the shrike-breeding season. The lethal control program implemented to protect shrikes from island foxes elicited strong opinions about the perils of managing

156 · Managing recovery one rare species at the expense of another. In particular, the importance of the genetic distinctiveness of the island shrike and hence its conservation value was questioned in relation to the genetic and conservation importance of island foxes (Roemer and Wayne 2003). A different subspecies of loggerhead shrike (Lanius ludovicianus anthonyi) inhabits several of the northern Channel Islands, where it exists at low population levels. Landscape level changes due to the cessation of grazing on these islands may be shifting habitat away from that favored by shrikes (open grassland and edge) to types not favored by shrikes (thicker shrub communities), decreasing long-term chances of persistence for these island shrike subspecies. However, these putative island subspecies of loggerhead shrike may not be genetically distinct, and may have become established fairly recently, when habitat was more open as a result of the impacts of grazing by introduced animals.

13.2 A model for management: the Island Fox Conservation Working Group In 1999 when the Santa Rosa and San Miguel island fox populations were rapidly approaching extinction, NPS decided to bring all of the remaining foxes on those islands into captivity. In only a limited number of other cases – e.g. California condors (Crawford 1985) and black-footed ferrets (Clark 1987) – had all of the remaining individuals of a taxon been taken from the wild, and NPS as an agency had little experience in managing captive breeding programs for rare species. NPS sought input (and support) for such drastic action by convening a group of island fox and rare species conservation experts. The group, which met first in 1999, included nearly all the biologists who had previously studied or monitored island foxes, as well as experts in canid conservation, genetics, carnivore pathology and disease, carnivore reproductive biology, and raptor biology and management. The group also included representatives of the pertinent regulatory agencies (USFWS and CDFG), whose participation early in recovery led to an easier transition when foxes became listed and outside agency oversight was necessary. From 1999–2009 the group ranged in size from 50–100 people, and met annually each June for three days to summarize and share information. At the conclusion of its first meeting the group recognized the precarious status of island foxes as well as the threat from golden eagles.

13.2 A model for management

· 157

The group recommended the immediate capture of all remaining wild foxes on San Miguel and Santa Rosa, and the initiation of live-capture efforts to remove golden eagles (Coonan and Rutz 2001). The team also recommended several longer-term actions including the development of a comprehensive recovery plan, the removal of feral pigs, and the reintroduction of bald eagles to the northern islands. The recommendations from this group of experts provided NPS with the support needed to proceed with captive breeding and golden eagle trapping, and to pursue within-agency funding for recovery actions. The group convened in 1999 was not an official USFWS recovery team (island foxes were not yet listed), but for convenience it was referred to as such by NPS and others. To minimize confusion, in 2001 USFWS suggested the group be called the Island Fox Conservation Working Group (IFCWG; Coonan and Rutz 2002). The IFCWG functioned as an ad hoc recovery team until island foxes were officially listed in 2004, at which time USFWS created a formal recovery team (described below). Beyond the immediate need for emergency recovery actions, the managing agencies also sought guidance for captive breeding methods and practices. Caring for and promoting breeding of captive foxes would require the development of husbandry techniques specific to island foxes, including diet, enclosure design, mating and breeding methods, genetic management, and veterinary care (Chapter 8). An important feature of the IFCWG was the establishment of sub-groups. One of the subgroups worked with local zoos, particularly the Santa Barbara Zoo, to standardize captive breeding practices across the islands. Another important sub-group focused on health and disease issues; in collaboration with additional outside experts, it developed methods and techniques for maintaining the health of the captive animals while addressing information gaps. For example, these collaborations resulted in a parasite survey of all six island fox subspecies (US Geological Survey, Sohn and Thomas 2005), and an investigation of captive reproductive success (Sovada et al. 2006). Conservation and management priorities were dynamic from year to year, and the annual meetings facilitated buy-in from members in an adaptive fashion. The group came to consensus on most issues until 2003, when NPS proposed releasing foxes from captivity amid the continued presence of a small number of golden eagles (Chapter 9). The high level of communication in early years resulted in good working relationships that became critical when management decisions became more difficult. The merit of this voluntary, coalition-of-the-willing approach was proven by

158 · Managing recovery the inclusion of the entire IFCWG into the official USFWS island fox recovery team in 2004 (see Section 13.4.2).

13.3 Listing the island fox as endangered Species become listed as endangered or threatened under the US Endangered Species Act (ESA) either through the candidate process, an internal USFWS procedure, or the petition process, which is driven by external advocacy (Nicholopoulos 1999). Under the candidate process USFWS identifies species that may warrant listing and essentially places them in priority ranking for further consideration. In actuality the candidate list has become a sort of ‘paperwork purgatory’ where listing proposals can languish for years. For example, the 286 species on the candidate list in 2005 had been on that list an average of 17 years, and many had been on the list for over 25 years (Greenwald and Suckling 2005). Island foxes had been on the candidate list since 1982, first as a full species (US Fish and Wildlife Service 1982), and then as six separate subspecies (US Fish and Wildlife Service 1985). Island foxes were designated as a ‘Category 2’ candidate species, meaning that listing was probably appropriate but that ‘persuasive evidence on biological vulnerability and threat’ was lacking (US Fish and Wildlife Service 1985). In 1994 the USFWS considered island foxes stable, but retained them on the Category 2 list (US Fish and Wildlife Service 1994). The long delay in listing species through the candidate process has led to the increased importance of the petitioning process, wherein USFWS (or the National Marine Fisheries Service for marine species) considers formal requests (‘petitions’) from outside groups to list a species. Upon receiving a petition, the USFWS has 90 days to determine whether there is adequate information to make a decision on listing and, if there is, one year to determine whether listing is warranted. By 1999–2000 the low numbers of individuals in each island fox subspecies clearly warranted listing; four subspecies had populations of less than 100, and two of those had less than 20 individuals remaining. However, there were no indications that USFWS intended to proceed with the listing process for island foxes, so in 2000 IWS and the Center for Biological Diversity (CBD) formally petitioned the USFWS to list four of the six island fox subspecies as endangered (all but the two on the USN-owned islands of San Clemente and San Nicolas). Both groups felt that in addition to being clearly biologically warranted, listing would also help increase funding for the ongoing but expensive recovery actions that had been recommended, particularly captive breeding and golden eagle translocation. It was also

13.4 Changes in island fox management due to listing

· 159

thought that the listing process would increase the notoriety of island foxes, and facilitate fund-raising and greater awareness of the island fox situation by the public (Chapter 13). After receiving the listing petition USFWS responded to the petitioners that they would not be able to issue a 90-day finding due to higher priority actions within the agency. In November 2000 USFWS further stated that the petition would not be addressed that year because of their existing workload and the lack of additional funding. The CBD subsequently sued USFWS for their failure to respond to the petition, and USFWS settled the lawsuit, in combination with several others, by agreeing to expedite protection for a total of 29 taxa, including island foxes. In December 2001 USFWS proposed listing the four island fox subspecies as endangered (US Fish and Wildlife Service 2001a, 2001b), but in April 2003 CBD sued USFWS again over their failure to finalize the listing package. In October 2003 USFWS finally agreed to place the four island fox subspecies on the endangered species list, and Santa Catalina, Santa Cruz, Santa Rosa, and San Miguel island foxes became officially endangered in March 2004 (US Fish and Wildlife Service 2004). Although ESA listing was clearly warranted, island foxes would not have been listed had IWS and CBD not sued USFWS twice.

13.4 Changes in island fox management due to listing 13.4.1 Permitting and USFWS oversight

Listing of the four island fox subspecies as endangered had substantial legal and management consequences. One of the immediate effects was the oversight by USFWS for all aspects of island fox management and research for the listed subspecies. All activities, even those that pre-dated listing, now required a permit from USFWS. Although the decision to list was handled at higher levels of the USFWS, management of island fox permits fell to the local USFWS office in Ventura, California, which had a history of close cooperation with NPS staff from Channel Islands National Park. USFWS immediately placed NPS and IWS (which were implementing island fox monitoring and management on San Miguel, Santa Rosa, Santa Catalina, and Santa Cruz Islands) on their general permit. USFWS worked closely with both agencies to develop their own permits, deferring to NPS and IWS on development of terms and conditions for captive breeding (all four subspecies were being captive bred), husbandry, veterinary care, trapping, handling, vaccination, and release from captivity. Existing relationships streamlined the permitting

160 · Managing recovery Technical advice request Recommendation/ analysis

USFWS

Recovery coordination group

Resource managers

Technical advisory group

Management question/issue Task order

Figure 13.1 Process by which the island fox recovery team provided technical assistance to land management agencies for island fox management.

process so that there were minimal breaks in the implementation of recovery actions.

13.4.2 Recovery team and recovery plan

In addition to increased oversight, ESA listing requires the establishment of an official recovery team, whose primary task is to produce a recovery plan. When island foxes were listed, USFWS considered identifying the entire IFCWG as the island fox recovery team. But a team of that size would have been unprecedented, and USFWS opted for a smaller recovery team called the Island Fox Recovery Coordination Group (RCG, initially seven people). The RCG was designed to work closely with the IFCWG, and the latter was recast as the Integrated Island Fox Recovery Team (IRT). The RCG could utilize members of the IRT via Technical Expertise Groups (TEGs, described below), but within the legal structure of the ESA the RCG was the official recovery team. In addition to their task of writing a recovery plan, the RCG and TEGs were incorporated into recovery efforts in ways somewhat atypical of the standard recovery team model. For example, the groups were asked by the land managers to provide technical assistance for island fox recovery actions. An innovative process was developed (Fig. 13.1) in which the land management agencies posed technical questions (Technical Analysis Requests: TARs) to the RCG, who then assigned them to the appropriate TEGs. Very specific TEGs were created to respond to each request and to report their findings back through the RCG to USFWS and ultimately to managers. From 2004 through 2006 the TEG process was utilized to address the following issues:

13.4 Changes in island fox management due to listing

r r r r r

· 161

decision-making and protocols for releasing captive foxes to the wild; development of husbandry standards; investigation of reduced reproductive success in captivity; management of golden eagles; and development of population viability models to determine recovery criteria (see Box 13.2 ‘Recovery criteria derived from demographic modeling’).

Box 13.2 Recovery criteria derived from demographic modeling Although recovery plans typically identify recovery criteria that are statistical targets for the focal population, the parameters identified may not be directly linked to population persistence/extinction. The USFWS island fox recovery team instead opted for recovery criteria that were directly tied to extinction risk and were readily monitored and developed through a process called population viability management (PVM; Bakker and Doak 2009). The recovery standards were an outcome of the substantial demographic modeling effort conducted for island foxes (Bakker et al. 2009; see Section 3.5). That effort incorporated an unprecedented body of data and incorporated ecological drivers (predation, density dependence, and climate) and uncertainty to evaluate extinction risk under a wide range of conditions. The PVM in turn evaluated choice of monitoring schemes and resultant management decisions on probability of extinction, which the authors termed the ‘common currency’ for comparing management options. The final choice of parameters included annual mortality, monitored via radiotelemetry, and population density, for which a new monitoring approach was recommended (Rubin et al. 2007). Both parameters have direct influence on population persistence, and the authors provided a framework (Fig. 13.2) that was easy for managers to understand and into which annual monitoring results could be entered. For each subspecies to be considered eligible for delisting, three-year averages of mortality rate and population size, including 80% confidence intervals, must fall below a line representing 5% extinction risk over 50 years. As of 2009 the San Miguel subspecies – with an adult population estimate of > 300 and high survival since reintroduction began in 2004 – had attained biological recovery according to this model.

162 · Managing recovery 0.4

5

70 30 50

Adult mortality rate (3-year mean)

70

0.3

50

30

30

10 10

0.2

10 5

5

5

0.1

2008 2009

0 0

50 100 150 200 250 Adult population size (3-year mean)

300

Figure 13.2 Risk of quasi-extinction in 50 years for San Miguel Island foxes under three-year averages for adult mortality and population size, showing improvement from 2008 to 2009 (T. Coonan and V. Bakker, unpublished data).

The technical analyses also resulted in advances in methods and decision-making processes for many aspects of island fox recovery. Advances in mark–resight survey design and data analysis facilitated development of novel monitoring techniques for estimating population density (Rubin et al. 2007), and the thorough TEG analysis of the benefits, costs, and logistics behind moving captive breeding to the mainland or establishing a ‘redundant’ mainland captive fox population led to several stimulating discussions within the group on the role of captive breeding and zoos in species conservation (Chapter 12). Because USFWS did not have in-house island fox expertise, they also used the RCG to review proposed management actions by land managers (a process that was sometimes challenging for the managers). For instance, an RCG-requested review of the population viability analysis used to support closing the Santa Catalina captive breeding program (Kohlmann et al. 2005) resulted in the delay of releases on Santa Catalina for several months. When NPS proposed continuing the releases of foxes on Santa Rosa in 2004 and beginning releases on San Miguel (Chapter 9), the

13.5 The benefits and challenges of ESA listing

· 163

RCG decided against the recommendations of the assigned TEG and instead recommended that no releases occur that year. USFWS ultimately approved the NPS release plan with recommended contingencies and the releases were successful, but the decision-making process was difficult for the larger group (Chapter 9). Toward the end of the recovery program the heavy use by USFWS of the RCG in reviewing and providing recommendations on management actions was perhaps less productive than the earlier collaboration between the agencies and the IFCWG. The USFWS also used the RCG and the annual meetings to develop a draft recovery plan for the four listed subspecies. Because the plan was essentially developed within the larger group (the RCG and the TEGs), USFWS was able to vet elements of the draft plan before releasing them to the public, resulting in a draft that was comprehensive and that had previously been reviewed by most of the relevant experts. Recovery criteria for the plan were developed as an outcome of the demographic modeling TEG (Bakker and Doak 2009, Bakker et al. 2009), and the clarity and biological relevance of those criteria were a strength of the draft plan and of the recovery team’s approach (see Box 13.2 ‘Recovery criteria derived from demographic modeling’ on pp. 161–162). As of 2010 the draft recovery plan for island foxes had not been released to the public. The RCG–IFCWG meetings in 2004–2006 were dominated by recovery-related issues and the development of the recovery plan, and were run as formal recovery team meetings by USFWS. With the completion of the RCG contribution to the draft recovery plan in 2006 and the continuing success of recovery efforts, momentum of the official recovery team waned. By 2007 annual fox survival was near 90% for all endangered subspecies, golden eagles no longer bred on the islands, captive breeding had been successful and was ending, and populations of all four endangered subspecies had increased substantially. In 2007 the annual island fox meeting reverted to the all-volunteer effort it had been prior to listing.

13.5 The benefits and challenges of ESA listing Listing a species as endangered under the ESA is at once a recognition of the species’ plight, and a statement of the federal government’s intent to recover that species. One of the most immediate benefits of ESA listing is the increased legal protection afforded to a species. Section 7 of the ESA requires federal agencies to review their proposed activities to insure they do not result in jeopardy to listed species, and to consult with

164 · Managing recovery USFWS on potential impacts. For island foxes there was little concern by USFWS that the management activities of NPS (the only federal agency involved with management of endangered foxes) would place island foxes at increased risk. Likewise, non-federal entities are not permitted to allow ‘take’ (harm) of endangered species under the ESA. Listed fox populations on non-federal lands were the responsibility of the Catalina Conservancy and TNC, both of which had a strong conservation-oriented mission. Increased protection from management activities was therefore not a primary benefit of ESA listing for island foxes. Another advantage of ESA listing can be increased access to funding for recovery efforts. But funding does not come automatically with listing, and funding for recovery varies dramatically by species. Some recovery programs are extremely well funded by USFWS (e.g. gray wolves and black-footed ferrets; Restani and Marzluff 2002), but when island foxes were listed in 2004 there was little USFWS funding available to implement recovery actions. All of the recovery actions for island foxes, including the annual recovery team meetings, were largely funded by the land management agencies themselves. Official listing did increase the funding allocated to island fox management from NPS, where the within-agency funding process gives higher priority to projects with listed or proposed species. USFWS also directed conservation funding through the state of California to TNC and the Catalina Conservancy, who could then leverage this support with contributions from private donors to contribute to fox recovery. So, although listing of island foxes did not lead to direct USFWS funding, it did increase the availability of funding from other sources. The primary benefit of listing for island foxes was to coordinate recovery efforts within the structure of a sanctioned recovery team, described above, and to bring in additional diverse expertise to the recovery effort (Clark et al. 2002). Though all the major actions (captive breeding, reintroduction, translocation, trapping, and relocation of golden eagles) had concluded before a recovery plan was released, the use of demographic modeling to construct biological recovery criteria was a particularly important contribution. Effectively managing organizational structure is critical for successful endangered species recovery (Reading and Miller 1994), and without facilitation and direction, recovery teams – usually composed of experienced professionals who may or may not have worked together previously – can become mired in conflicts that divide the group and distract participants from focusing on recovery goals (Clark et al. 1994).

13.6 Conclusion

· 165

Stronger participation by USFWS at island fox recovery team meetings, as well as the use of a trained facilitator, might have enabled more productive discussions and reduced conflict, especially for consideration of contentious issues such as reintroductions to the wild while some eagles still remained. In the same vein more direction from USFWS, and a clearer definition of the role of the RCG, could have hastened the complex and often time-consuming process by which USFWS used the RCG to review proposed recovery actions.

13.6 Conclusion The value of ESA listing for island foxes can be judged by the extent to which it ultimately facilitated recovery. Although the critical status of four island fox subspecies probably made their listing inevitable, whether listing supported recovery or in some ways made it more difficult is equivocal. ESA listing gained island foxes national notoriety, access to technical expertise, and advantages for both public and private fund-raising. A positive benefit of listing was the incorporation of the established IFCWG into the official USFWS recovery team, and as of 2009 the group has advised agencies on island fox recovery actions and issues for over a decade and remains cohesive. Such voluntary conservation groups may become increasingly necessary as the number of species in peril greatly exceeds the capacity of governments to address their conservation effectively. However, the majority of factors that contributed to the success of island fox recovery were unrelated to ESA listing. For example, several critical recovery actions (captive breeding, eagle removal, and CDV vaccination development) were implemented and well underway before the species was officially listed. Two additional factors that greatly aided recovery – the lack of landowner impediments (no development or other economic activity was threatened by island fox listing or recovery) and ownership of island fox habitat by conservation entities – likewise were unaffected by ESA status. The IFCWG was also organized and effective for many years before foxes were officially endangered, and while the formal recovery team provided important expertise, there were challenges of organization and design with its inception. Some benefits that could have accrued from listing (for example, a more timely investigation, and possibly resolution, of the low reproductive success in captivity) were likely lost due to the four-year period between petitioning and listing. Minimal funding ensued from listing, and the land management agencies largely raised their own funds to support recovery.

166 · Managing recovery We do not suggest that the ESA is not critically important, and for many species it is the tool that has prevented extinction. But for island fox recovery, the elements that led to recovery were the commitment of the IFCWG, a group of people that functioned effectively independent of the ESA, and the fortunate situation where all island fox habitat was under conservation-oriented management. While the location of endangered species habitats are beyond our ability to manipulate, we suggest that with or without ESA protection, species and habitat conservation will benefit most from productive and sincere human collaborations.

14 The ecological role of island foxes r

14.1 Introduction The conservation of rare species is based on influencing population parameters of abundance, distribution, and rate of change, and in most cases recovery criteria are established that quantify success based solely on the population dynamics of the focal species. Rarely addressed (for good reason) are the ecological impacts that the rareness or absence of a species has on the community from which it has been lost and whether the community responds in any observable way when the species is returned. Models and theories abound on the ecologic value of species (Kareiva and Levin 2003), but examinations of community response to true species extinctions are rare (Doak and Marvier 2003). The extirpation of island foxes from Santa Rosa and San Miguel was a unique occurrence in which the top and only mammalian carnivore was completely eliminated from two ecosystems while community responses were observed. This chapter describes community changes that occurred in response to the loss of foxes and discusses the importance of considering ecosystem impacts as a component of species recovery programs.

14.2 Background Interactions between species affect ecological communities in a myriad of ways (Menge and Sutherland 1987, Thebault and Loreau 2006). Species relationships influence the abundance and diversity of organisms as well as ecosystem processes such as succession, nutrient cycling, and resilience to perturbation (Morin 1999, Brown et al. 2001). However, while interactions between species can be readily observed and occasionally quantified (Berlow et al. 1999), the roles of individual species remain elusive, and the following question persists: Does the extinction of a species change the ecosystem from which it has been lost? (Kareiva and Levin 2003). To investigate the role of individual species and species groups, empirical and theoretical removal experiments have been conducted wherein

168 · The ecological role of island foxes selected community attributes are measured with a species present and again after it has been removed (Sih et al. 1985). To maintain experimental rigor, the majority of removal studies have been conducted in laboratories or in very controlled natural systems (Schmitz 2003). For example, field experiments have often focused on intertidal communities where species diversity and environmental variability are high but organisms are sessile and easily counted, or in closed aquatic systems such as lakes (Mittelbach et al. 1995, Pace et al. 1999). There are few published studies where entire populations of mammals, especially carnivores, were removed in a controlled manner from a system to test hypotheses regarding species interactions and function (Shurin et al. 2002, Vucetich and Peterson 2004; although see also Brown and Munger 1985, Meserve et al. 2003). It is not a lack of interest that has limited top carnivore removal studies, but the associated logistical and ethical obstacles (Crooks and Soule 1999); purposely removing an entire population of native mammals from any system is not only time and resource intensive, but fundamentally undesirable. Consequently, as Shurin et al. (2002, p. 787) suggest, ‘The lack of experiments [that remove] vertebrate predators and herbivores in terrestrial ecosystems highlights a major gap in the ecological literature.’ For a period of eight years on San Miguel Island and seven years on Santa Rosa Island the top carnivore was completely eliminated from the ecosystem. This chapter describes community changes that were either measured or observed in response to the complete absence of foxes on Santa Rosa and San Miguel, and briefly discusses observations related to extremely low numbers of foxes on Santa Cruz. Although the loss (removal) of foxes due to golden eagle predation might be called a natural ‘experiment’, the conditions surrounding the events were not controlled and our observations are not results; the ecosystem manipulation (fox removal) was unintentional and undesired, and the response variables (primarily the population dynamics of other species) were generally measured only within existing monitoring programs, or as time permitted within the larger effort to recover foxes. However, given that intentional or experimental removals of top native mammalian carnivores from biological communities are unlikely to occur, we suggest that additional information on the functional role of carnivores will come from two scenarios: r unanticipated extinctions or near-extinctions where ecological moni-

toring has previously been implemented; and

r intentional removal programs for non-native carnivores (i.e. feral cats).

14.3 Prey response

· 169

Several factors related to the loss of island foxes provide unique and valuable information to the role of top carnivores in terrestrial systems. They include the spatial scale involved (several medium-sized islands), the fact that these were nearly closed terrestrial systems, and the descriptive community data that were collected prior to, during, and following the event, provide unique and valuable information on the role of top carnivores in terrestrial systems. Much of the discussion focuses on San Miguel Island due to the establishment by the National Park Service (NPS) of a terrestrial monitoring program there in 1993. The fox decline on San Miguel Island began in 1995, and ecological monitoring prior to the decline and then during the periods of fox absence and recovery has provided a consistent record of island ecosystem response to fox loss. We evaluate trend and behavioral data for deer mice, ground-nesting birds, black rats, and raptors during this period to elucidate important predatory and competitive relationships between foxes and these groups. Cumulatively, quantitative data and personal observations reveal community responses to fox absence at multiple trophic levels, as well as indirect community effects that may ultimately have important long-term consequences. Island foxes are a critical element of island food webs, with relationships entangled throughout island communities. Without foxes these systems would not only be structurally different but possibly less resistant to future impacts of perturbation and climate change. The community responses observed during the absence of foxes emphasize the importance of maintaining top predators in natural ecosystems, and the critical need for ongoing monitoring and research in anticipation of future events.

14.3 Prey response 14.3.1 Deer mice

Deer mice (Peromyscus maniculatus ssp.) are the primary vertebrate prey for island foxes (Chapter 4). Mouse populations on the islands fluctuate between spring lows following winter mortality events, and fall maxima driven by spring and summer reproduction (Drost and Fellers 1991, Schwemm 2008a). On San Miguel Island mice are sampled semi-annually (spring and fall) using mark–recapture methods on three grids in grassland, mixed scrub, and lupine shrub habitats (Fellers et al. 1988). Prior to 1995, during a period when fox populations are thought to have been

170 · The ecological role of island foxes

800

Spring mice

450

Fall mice

400

700 Summer foxes

Mice/ha

600

350 300

500

250 400

200

300

150

200

100

100

50

Number of adult foxes

500

900

0

97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07

96

19

19

19

19

94 95

0

Figure 14.1 Change in deer mouse population dynamics with removal of island foxes from the wild on San Miguel Island. Density estimates are averages of grids sampled.

near carrying capacity (Roemer et al. 2001a), deer mouse abundance on San Miguel Island varied fairly consistently between spring densities of 100–200/ha and fall densities of 300–500/ha. The rapid decline of foxes beginning in 1996 coincided with a dramatic shift in mouse abundance patterns (Fig. 14.1). Fall mouse densities on all three San Miguel Island grids nearly tripled between 1997 and 1998 (just prior to total fox absence), reaching levels in the fall of 1998 not previously recorded on the island (> 800/ha). Fall densities fell to pre-fox decline levels in 1999 and 2000, but on at least one grid (grassland) in 2001 were again over 800/ha. Spring densities remained at levels similar to those observed when foxes were present, until the spring of 2007 when on two grids (grassland and mixed scrub) they were lower than previously recorded (< 100/ha; Schwemm 2008a, Drost et al. in preparation). Small mammal dynamics are often driven by annual weather events (Ernest et al. 2000, Meserve et al. 2003), but on San Miguel Island only extreme precipitation events appear to significantly influence mouse densities (Schwemm 2008a). Increased volatility in mouse dynamics in the absence of foxes suggests the strong moderating influence of predation (Brown et al. 2001, Meserve et al. 2003); opportunistic foxes prey heavily on mice when mouse densities are high and mice are relatively easy to

14.3 Prey response

· 171

find, thereby preventing irruptions and ensuing crashes in mouse populations (Morin 1999). The effect of reduced predation pressure on mouse populations resulting from fox absence may have been moderated by an increase in predation by owls and harriers, and an even greater increase in mouse abundance might have been observed at least temporarily had raptors not partially filled the top predator niche as foxes declined (Pace et al. 1999). In addition to the numerical response of mouse populations to fox absence, a behavioral response was also apparent. Orrock (in review) found that mouse foraging behavior changed as foxes were reintroduced to the San Miguel Island system. Using giving-up densities (GUD; in this case the amount of seed that is left when a mouse stops foraging due to perceived risk), Orrock suggested that in the absence of foxes mice foraged equally between ‘safe’ and ‘risky’ habitats (with and without cover). As fox densities increased mice changed their behavior, and the difference between their use of safe and risky sites became significant, indicating a change in risk perception in response to fox presence (Orrock et al. 2004, Orrock in review). Foraging decisions by consumers are made by balancing the risk of predation against food preferences (‘the ecology of fear’, Lima and Dill 1990, Phelan and Baker 1992), and these results suggest that mice altered their foraging patterns in response to fox abundance. Alternatively, greater foraging activity outside of safe sites may have resulted from proportionally higher densities of mice in those areas as mouse populations increased coincident with the fox decline (C. Drost personal communication). Whether altered behavior or simply more mice were responsible for increased foraging in open areas, seed predation likely increased in those sites. Foraging decisions by seed predators can have important impacts on seed bank diversity and resulting plant recruitment (Schmitz et al. 2000, Crawley 2002), and future vegetation analysis may provide evidence of indirect impacts of fox absence on vegetation composition as mediated by mouse behavior (Crawley 2002). 14.3.2 Landbirds

Landbirds comprise approximately 5% of fox diets on San Miguel Island, although this proportion varies seasonally and between years (Collins 1979a, Crowell 2001). The dynamics of two island endemic bird subspecies – San Miguel Island song sparrows (Melospiza melodia micronyx) and orange-crowned warblers (Vermivora celata sordida) – changed dramatically on San Miguel Island as foxes were removed from the system

172 · The ecological role of island foxes 3

500 Island foxes

450

Song sparrows

400

Orange-crowned warblers

350

2 Birds/ha

300 1.5

250 200

1

150

Number of adult foxes

2.5

100

0.5

50 0

0 1993

1995

1997

1999

2001

2003

2005

2007

Figure 14.2 Population increase of two endemic landbirds (song sparrow and orange-crowned warbler) during absence of island foxes on San Miguel Island.

(Fig. 14.2). Song sparrow densities increased from an average of 1.4/ha from 1993–1997 (with a declining trend during the period), to 1.8/ha from 1998–2007 (NPS unpublished data, Fig. 14.2). Orange-crowned warblers showed an even more substantial increase, with an average of 0.5/ha from 1993–1997, and 1.2/ha from 1999–2007. Song sparrows and orange-crowned warblers have similar nesting habits; nests are constructed in shrubs and on the ground. Although concealed, they are generally accessible to ground predators (Collins 1979a, Peluc et al. 2008). Shrub communities, in particular those dominated by giant coreopsis (Coreopsis gigantea) have been increasing on San Miguel Island since the removal of non-native grazing animals (Corry 2006), and increases in shrub communities would conceivably favor both species (Collins 1979a). However, shrub densities have been steadily increasing for several decades while sparrow and warbler abundance increased only when foxes were gone from the system. Consequently a connection between habitat change and landbird abundance alone seems unlikely. Multi-year differences in abundance also increased for song sparrows, with both higher and lower densities in the absence of foxes. Many species of Peromyscus are known to be important nest predators (Schmidt et al.

14.3 Prey response 450

· 173 3

400 2.5

Prior spring mice

350

Song sparrows

Mice/ha

250 1.5 200

Birds/ha

2

300

1

150 100

0.5 50 0

0

93

19

19

94

95

19

19

96

19

97

19

98

19

99

20

00

20

01

20

02

20

03

20

04

20

05

20

06

07

20

Figure 14.3 Relationship of song sparrow density to deer mouse density on San Miguel Island.

2001), and changes in spring song sparrow abundance without foxes may suggest an indirect effect of fox absence; higher spring mouse populations result in greater egg loss, which is then reflected in fewer adult birds the following spring, as was found in temperate forests by Schmidt and Ostfeld (2003). The data from San Miguel Island suggest such a pattern, particularly for the years 1999–2000 (Fig. 14.3). A sharp rise in mouse densities in the spring of 1999 (likely in response to extremely high rains in the winter of 1997–1998, which provided abundant food resources in the absence of predation), was followed by a steep decline in song sparrow observations in the spring of 2000. The relationship between mice and sparrows may indicate a direct numerical response, where more mice simply eat more eggs. However, there could also be a behavioral component, whereby a reduction in perceived risk by mice leads to foraging behavior that increases pressure on eggs (Schmidt et al. 2001). 14.3.3 Black rats

Black rats were unintentionally introduced to San Miguel Island likely via landed ships and shipwrecks, possibly as long ago as the 1800s (Collins 1979b). A survey of San Miguel Island by Collins (1979b) found rats

174 · The ecological role of island foxes distributed along most of the northwest beaches, at Point Bennett on the far west end, and in the Cardwell Point area on the east end of the island. However, neither Collins nor any other observers recorded rats at interior areas of the island. In 2002, when foxes had been completely absent for three years, two rats were caught in traps used for deer mouse monitoring on the terrace west of Nidever Canyon, approximately 0.2 km from the nearest beach. At least one rat was photographed in the same year by a camera trap near the Dry Lakebed, far inland on the western portion of the island (NPS unpublished data). In a brief 2004 survey one rat was caught near the campground close to Nidever Canyon, but no other observations of rats were made (NPS unpublished data). Since 2004 when fox reintroductions began, no rats have been trapped during deer mouse monitoring and no rat sign has been noted in areas away from beaches (directed islandwide surveys for rats have not been conducted, however). Foxes likely prey on rats opportunistically, and at normal densities foxes apparently keep rats on San Miguel Island at low levels and localized.

14.4 Competitors 14.4.1 Spotted skunks

Island spotted skunks are an endemic subspecies that occurs on Santa Rosa and Santa Cruz Islands. Skunk abundance on both islands increased dramatically from the mid 1990s until the mid 2000s (Jones et al. 2008b, NPS unpublished data), almost certainly in response to fox loss. Spotted skunks are smaller than foxes and, while foxes and skunks utilize similar resources, foxes are almost certainly stronger competitors (Roemer et al. 2002). Jones et al. (2008b) found the diet of skunks on Santa Cruz from 1992–2004 (during low fox densities) shifted from carnivory to slight omnivory, and the proportion of lizards and snakes consumed increased in relation to the proportion of mice. A shift in prey selection by skunks suggests that either available prey increased in the absence of foxes and skunks prefer reptiles over rodents, or that with reduced competition skunks utilized spatial and temporal niches that were previously unavailable. The absence of foxes may also have reduced the importance of competition as a population-regulating factor for skunks. Jones et al. (2008b) suggest that by 2004–2005 skunks may have been close to carrying capacity, evidenced by a decrease in average weight and reproductive effort. It is difficult to know whether the substantial increase in skunk abundance on Santa Cruz and Santa Rosa Islands was due solely to

14.4 Competitors

· 175

competitive release from foxes, or if habitat changes resulting from feral animal removal that occurred during the same period also contributed (Jones et al. 2008b). Still, it seems unlikely that the response to habitat changes by skunk populations would have been as substantial as it was had normal densities of foxes been present. Finally, evidence suggests that the increase in skunks on Santa Cruz has resulted in a substantial increase in predation on seabirds. Ashy storm petrels (Oceanodroma homochroa) are small seabirds that breed on islands from Mendocino County in northern California to Baja, Mexico. The species is of special concern in California due to its restricted breeding range (Carter et al. 2008), and one of the largest known breeding colonies of ashy storm petrels is found in a complex of sea caves on the east end of Santa Cruz. As far as is known these caves are inaccessible to foxes, but beginning in 2005 large numbers of dead ashy storm petrels were discovered in caves along with skunk sign. At least 70 carcasses were found in 2005, and in 2008, 32 carcasses were found in an adjacent cave, again in association with skunk sign (Carter personal communication, McIver et al. 2009). At least one skunk was subsequently trapped in a cave by NPS, confirming skunk presence and their ability to access the sites. The conjecture that seabird predation by skunks has increased coincident with declining fox numbers alludes to a relationship whereby foxes at normal densities somehow prevent skunks from accessing sea caves. Groundnesting birds are always at risk from terrestrial predators, which can easily prey on eggs of bird species that lack adequate defenses. For this reason islands with mammalian predators rarely support large seabird populations. But offshore rocks and sea caves of islands are often free of predators and provide seabirds with safe nesting areas near food resources. When these sites for some reason become accessible to predators, there are often disastrous ecosystem consequences (Croll et al. 2005). Almost nothing is known about the physical abilities of foxes and skunks to utilize the steep cliffs of the islands, but it is conceivable that skunks are physically capable of utilizing cliff sites while foxes are not. When fox densities are near normal, they likely fully occupy habitats near cliffs (as they do most habitats), and either actively or passively prevent skunks from accessing cliff and cave sites. 14.4.2 Intraguild predation response

A relationship where two species compete for resources and one also preys on the other is termed intraguild predation (IGP). IGP is illustrated on

176 · The ecological role of island foxes the Channel Islands by the interaction between island foxes and small- to medium-sized raptors: northern harriers (Circus cyaneus), barn owls (Tyto alba), and short-eared owls (Asio flammeus). Predatory birds compete with foxes primarily for deer mice, but eggs and chicks of avian predators can also be prey for foxes. There are no records of harriers nesting on the northern Channel Islands prior to the late 1990s, but in the fall of 1998 five harriers were seen displaying territorial behavior on San Miguel (Drost et al. in preparation). By 2002 there were at least three breeding pairs of harriers on San Miguel, and breeding behavior was also noted for the first time on Santa Cruz (CS personal observation). Up to six established pairs have been observed numerous times on San Miguel in recent years, and it remains to be seen whether breeding continues as fox densities increase. Likewise short-eared owls, which were previously rarely seen on San Miguel, apparently became resident for several years from at least 1999–2005, with suspected breeding in 2002 (Drost et al. in preparation).

14.5 Island communities without foxes A large body of experimental and theoretical work has addressed the question of how changes in biodiversity – specifically the loss of species – affect ecosystems and communities (McCann 2000, Kareiva and Levin 2003, Hooper et al. 2005). Response measures that have particular value to ecosystem management (Srivastava 2002) and conservation (Doak and Marvier 2003) are stability and resilience, i.e. how well a community responds to perturbation (Hooper et al. 2005). Given the nature of our observations and data we can only speculate, but suggest two important roles of foxes in island systems. First, as generalists that prey on a range of plant and animal materials, foxes help maintain consumer diversity. For example, foxes may indirectly facilitate higher songbird populations by preying on mice (thereby reducing nest predation). By preying on rats, foxes may also limit rat impacts on intertidal and nearshore species. Rats have had devastating impacts in many ecosystems, especially on islands (Atkinson 1985, Croll et al. 2005), and it is all but certain that without strong pressure by foxes rat populations on San Miguel would expand across the island. Second, the competitive role of foxes may also serve to maintain diversity. The decline of foxes on Santa Cruz and Santa Rosa likely resulted in substantial population increases and associated range expansions of skunks on those islands, with associated impacts on seabirds and reptiles

14.6 Implications for research and management

· 177

(Jones et al. 2008b). A more complex situation exists on San Nicolas, San Clemente and Santa Catalina Islands where foxes compete in unknown ways with feral cats (Felis catus). While cats certainly prey on native vertebrate species, including the threatened island night lizard (Xantusia riversiana), there is no evidence at present that cats have any populationlevel impacts on prey species (although, for a contrary perspective, see US Fish and Wildlife Service 2009). Further, there is no direct evidence that cats outcompete foxes for resources; the cat population on San Nicolas has remained stable for several decades (US Fish and Wildlife Service 2009), and foxes on San Nicolas exist at consistently higher densities than on any other island (Chapter 3). Although the nature of the competitive relationship between foxes and cats is unknown, the presence of foxes on San Nicolas may well be limiting cat abundance and reducing cat impacts on prey species. If this were not the case, cats on San Nicolas (and Santa Catalina) would likely have had the same level of impacts as cat introductions on islands that were previously predator free (McChesney and Tershey 1998). If foxes were permanently extirpated from island communities, it is highly unlikely that the top trophic level would be filled by a functionally equivalent species (Doak and Marvier 2003). Natural dispersal to islands is extremely rare (Simberloff 1974), and in no scenario can managers envision a repeat colonization by a mainland species of a generalist predator that would fill the role that island foxes do now. Instead this niche would be filled in some manner from either an increase in abundance of one or more avian predators, or by human-assisted colonization by a non-native predator (for example, cats), which in the absence of foxes would almost certainly become invasive. If we had been (or ultimately are) unsuccessful in returning foxes to island systems, there is no doubt that island communities will be permanently altered.

14.6 Implications for research and management The largest controversy encountered by the island fox recovery team centered on the timing and manner of reintroduction. By the early 2000s the threat from golden eagles was still present, yet the risks to foxes in captivity were also increasing (Chapter 9). The decision-basis for releasing foxes was equivocal; some released foxes would probably be killed by eagles, but foxes in captivity were at risk from aggression and disease, and productivity in the captive population was declining for unknown reasons (Chapter 10). Further, it was increasingly apparent that the continued absence of foxes was resulting in ecological changes to

178 · The ecological role of island foxes island communities. If the risks to fox populations truly were equal in and out of captivity (a presumption generally but not wholly supported), then managers, and NPS in particular, had a responsibility to not only protect foxes, but also to protect ‘the systems upon which they depend’. Ultimately the final decisions on whether or not to release foxes were not predicated on the need for protecting ecological communities. This argument did, however, have some persuasive value, and we propose that the ecological benefits of rapid reintroductions should carry greater weight in future recovery planning than they have historically (Snyder et al. 1996). A paramount lesson learned from the near extinction of island foxes is the critical value of long-term baseline ecosystem monitoring (Brown et al. 2001). This endeavor has unfortunately been supplanted in recent years by a focus on effects and response monitoring, an approach that forsakes community and ‘surveillance’ monitoring unless there is a specific project or impact to be assessed (Nichols and Williams 2006). NPS established vegetation monitoring on San Miguel Island in the 1980s, followed by landbird, deer mouse, and island fox monitoring in the early 1990s. The terrestrial monitoring program was initiated not because there were specific threats to the resources or ongoing management actions that might have impacts (both valid reasons for monitoring), but because these resources were identified as important ecosystem elements for which the NPS had stewardship responsibility (Davis et al. 1994). If the fox monitoring program had not been in place, it is all but certain – given the very high rate at which animals were dying – that the island fox subspecies on San Miguel Island would have become extinct. Unfortunately a similar monitoring program, which would have started coincidentally on Santa Rosa Island, was eliminated for political reasons, resulting in the irrecoverable loss of valuable information describing Santa Rosa Island communities pre- and post-fox decline. Given that there was no monitoring at all on Santa Rosa, the fox decline there would likely have gone completely undocumented had it not been for the critical information being obtained coincidentally on San Miguel and Santa Cruz Islands. Finally, without the surveillance monitoring that was already in place on San Miguel, the demands of the recovery program would likely have prevented the initiation of any new community monitoring, and few if any ecological responses would have been documented (Simberloff 2003). Just as the catastrophic decline of island fox populations was wholly unanticipated, future events will surely occur that threaten island species and their habitats. The data reported here and additional observations

14.6 Implications for research and management

· 179

suggest that the removal of foxes from island systems may in fact have caused greater changes in a shorter amount of time then were originally thought, although in hindsight many of these changes might have been predicted (Doak and Marvier 2003). A management approach supporting and supported by research and monitoring should address long-term strategies and, in future, endeavors should consider: 1. that ecosystem monitoring programs should be considered critically important in all protected natural areas, even in the absence of immediate threats; and 2. other things being equal, the reintroduction of captively managed species to their historic range should be given not only equal but higher priority than continued, long-term management of the species in captivity.

15 Conclusion r

15.1 Status By 2008 all four of the captive breeding programs for island foxes had ended, and by 2009 each of the six island fox subspecies, including the four listed as endangered in 2004, were either stable or increasing. Both major mortality factors that had driven the declines of the 1990s – golden eagle predation and canine distemper virus (CDV) – had been successfully mitigated. Population monitoring in 2009 showed that the estimated population of foxes on San Miguel was over 320, while on Santa Rosa there were close to 400 foxes (NPS unpublished data). On Santa Cruz and Santa Catalina populations had grown even more rapidly, with a 2008 adult population estimate of 740 on Santa Cruz (V. Bakker, Arizona State University, unpublished data), and over 700 on Santa Catalina (J. King, Catalina Island Conservancy, unpublished data). The two nonlisted subspecies on San Nicolas and San Clemente had stable populations of at least 500 adults. The existence of stable or upward trends for all subspecies in 2009 is testament to the effectiveness of the major recovery actions implemented during the previous decade. From 1999 to 2006, the relocation of 44 golden eagles from the northern islands coincided with an increase in fox survival. As of 2009 there were occasional visits to the islands by transient golden eagles that preyed on foxes, but golden eagles no longer bred on the islands, and recent predation can be viewed more as isolated incidents rather than having an impact on fox populations, all of which had annual survival of 80–95% in 2008 and 2009. Captive breeding was surprisingly successful, especially for the San Miguel and Santa Rosa subspecies. Although the number of founders was small (eight on San Miguel and 13 on Santa Rosa), 140 pups were born in captivity on the two islands and cumulatively over 150 foxes were released to the wild. Concerns about the effects of captivity on fox behavior were allayed by the quick reproductive success of released individuals. On Santa

15.2 Future

· 181

Catalina, translocation of foxes from the west end to the disease-ravaged east end proved extremely effective at initiating recovery of the entire Santa Catalina population, and the threat of CDV was mitigated by development and application of a vaccine administered to a large proportion of the population. Alhough all four endangered subspecies are likely to be biologically recovered within 5–7 years of listing (Chapter 13), each will always exist at naturally low population levels, and each will be forever vulnerable to extinction from human impacts and novel mortality factors. Even at carrying capacity, island fox populations will be low compared to those of other carnivores. The available habitat for island foxes will never increase, so even under the best management scenarios the populations of foxes on the larger islands will always be < 1,500, and < 500 on San Miguel. While demographic modeling suggests that island fox populations are resistant to extinction at modest population sizes, there is no doubt that the reemergence of predation or disease, or previously unidentified impacts, could have immediate and devastating effects on these small populations. Therefore, coordinated and intensive population monitoring programs have been adopted by all agencies and implemented for all six subspecies. Permanent, long-term monitoring plans will be an essential component of the final USFWS recovery plan, and any decision to delist the species (Chapter 12).

15.2 Future Efforts to recover island foxes clearly demonstrate the vulnerability of isolated island species and ecosystems to the impacts – both direct and indirect – of introduced species. The ability of alien herbivores to support novel predators, which also preyed on native species, was unanticipated 20 years ago. But the scope of the resulting impacts illustrates the importance of protecting whole ecosystems, particularly simple ones where introductions can have even greater consequences. Conversely, removing vulnerable native species from the wild, even if only temporarily, has consequences for entire communities, and is another reason to improve reintroduction success. Island fox biology reflects adaptation to their island environment, and facilitated much of the recovery success. While the species’ vulnerability to novel predators and pathogens was nearly their undoing, characteristics such as their relatively simple mating, the young age at which they first reproduce, density-dependent

182 · Conclusion reproductive responses, and high survival rates contributed greatly to recovery once threats were mitigated. The greatest future threat to island fox populations may be that posed by global climate change. As generalist predators, island foxes thrive in the Mediterranean climate and diverse habitats characteristic of the Channel Islands. Current models include high levels of uncertainty regarding the impacts of climate change on southern California coastal zones, but increased temperatures and changes in precipitation regimes are likely (Hayhoe et al. 2004), and may affect foxes in two ways. First, as generalist predators, island foxes rely on the relatively simple island communities for food. Many studies have demonstrated the importance of precipitation in structuring Mediterranean plant communities, particularly in regards to the impacts of invasive species. Reduced winter precipitation (caused either by lower precipitation overall or by changes in precipitation patterns that shift periodicity) will likely reduce recruitment of native plants and have mixed effects on invasive plant species. The result for island fox food resources in terms of available fruits is unknown, as are the greater trophic implications; though deer mice are the most common vertebrate prey for foxes, they are also the primary seed eaters on the islands. The other potential impact of climate change is changes in pathogen abundance and distribution. Diseases and parasites, as well as their vectors and hosts, are in many cases island specific, with each island fox subspecies having a unique history of pathogen exposure (Chapter 10). Changes in precipitation and/or temperature could bring a subspecies into contact with novel parasites or pathogens. For example, warmer temperatures may allow the mosquito species that harbor West Nile Virus (WNV) to colonize Santa Cruz Island, thus threatening the endemic island scrub jay (Aphelocoma insularis). Managers are particularly concerned because the island scrub jay’s closest relative, the mainland western scrub jay (A. californica) is among those bird species most susceptible to WNV (Reisen et al. 2005). To address this concern The Nature Conservancy is experimentally vaccinating island scrub jays against WNV (S. Morrison, The Nature Conservancy, personal communication). Given the intrinsic vulnerability of island fox populations, the recent investment in island fox conservation, and the uncertain impacts of global climate change, it is clear that monitoring (and hopefully research) of island fox populations will continue for the foreseeable future. What

15.2 Future

· 183

is less clear is whether the island fox has now become a ‘conservationreliant’ species, forever dependent upon human actions, or at least eternal vigilance, for its persistence into subsequent centuries. For those wishing to ‘save all the parts’, as Aldo Leopold said, it is disconcerting to think that there will never come a time when island foxes do not need to be closely watched, like a well-intentioned but unruly child with a tendency toward trouble. For example, many managers have concluded that the only way to mitigate possible future outbreaks of CDV in island fox populations is to vaccinate a portion of wild foxes annually against it, while others worry about the effects of vaccination on the natural immunity conferred by the apparently endemic strain of CDV in island fox populations (Chapter 11). Does full protection of this wild carnivore require a permanent vaccination program? If so, it is important that managers and scientists fully understand the level of commitment that may be necessary for conservation of island foxes, and for other species deemed conservation-reliant. The potential scope of such conservation work is daunting, but many crises might be avoided by judicious use of ecological monitoring, as was the case for island foxes. Island fox monitoring will continue for the foreseeable future, and we argue that it should be accompanied by monitoring of other ecosystem elements and processes. Such an approach would allow identification of community changes that might affect foxes, such as changes in prey, pathogen, competitor, and/or predator abundance and diversity. Environmental monitoring, particularly of weather and climate will provide evidence of the possible onset of climate change effects, and research into questions regarding island fox and/or community ecology will provide more information useful for protection and conservation. Perhaps the obvious need for a permanent island fox monitoring program is portentous for other conservation efforts in this era of spiraling human population growth, and pervasive human influence in all environments. If nothing else, the story of the island fox decline underscores the unpredictable effects of human actions. Mitigation of the primary mortality factors allowed other recovery actions – such as captive breeding – to work. Significant progress in mid- to long-range, indirect recovery actions (reintroduction of bald eagles, removal of feral pigs and introduced ungulates) began to restore balance to the altered island ecosystems, and tipped the ecological balance toward island foxes and away from continued golden eagle use of the islands. Finally, the cooperative

184 · Conclusion conservation effort marshaled for island fox recovery facilitated a consensus approach to development and implementation of recovery actions, with buy-in from stakeholders. Such an approach may become increasingly necessary to save other species, well before their plight is recognized through official listing as endangered.

References

Adler, G.H. and R. Levins. 1994. The island syndrome in rodent populations. Quarterly Review of Biology 69(4):473–490. Agenbroad, L.D. 2002. New localities, chronology, and comparisons for the pygmy mammoth (Mammuthus exilis): 1994–1998. Pages 518–524 in D. Brown, K. Mitchell, and H. Chaney, eds., Proceedings of the Fifth California Islands Symposium. USDI Minerals Management Service and The Santa Barbara Museum of Natural History, Santa Barbara, CA. Agenbroad, L.D., J.R. Johnson, D. Morris, and T.W. Stafford, Jr. 2005. Mammoths and humans as late Pleistocene contemporaries on Santa Rosa island. Pages 3–7 in D.K. Garcelon and C.A. Schwemm, eds., Proceedings of the Sixth California Islands Symposium. National Park Service Technical Publication CHIS-05-01, Institute for Wildlife Studies, Arcata, CA. Aguilar, A., G. Roemer, S. Deenham, M. Binns, D. Garcelon, and R.K. Wayne. 2004. High MHC diversity maintained by balancing selection in an otherwise genetically monomorphic mammal. Proceedings of the National Academy of Sciences of the United States of America 101(10):3490–3494. Alexander, K.A., L.A. Munson, A. Kalake, and M.J.G. Appel. 1996. Canine distemper-related mortality among wild dogs (Lycaon pictus) in Chobe National Park, Botswana. Journal of Zoo and Wildlife Medicine 26:201–206. Appel, M.J.G. and B.A. Summers. 1995. Pathogenicity of morbilliviruses for terrestrial carnivores. Veterinary Microbiology 44:187–191. Asa, C.S. 1997. Hormonal and experiential factors in the expression of social and parental behavior in canids. Pages 129–149 in N.G. Solomon and J.A. French, eds., Cooperative Breeding in Mammals. Cambridge University Press, Cambridge. Asa, C.S. 1999. Dogs (Canidae). Pages 80–87 in E. Knobil and J.D. Neill, eds., Encyclopedia of Reproduction. Academic Press, New York, NY. Asa, C.S. and C. Valdespino. 1998. Canid reproductive biology: integration of proximate mechanisms and ultimate causes. American Zoologist 38:251– 259. Asa, C.S. and C. Valdespino. 2003. A review of small canid reproduction. Pages 117– 123 in M.A. Sovada and L. Carbyn, eds., The Swift Fox: Ecology and Conservation of Swift Foxes in a Changing World. Canadian Plains Research Center, Regina, Saskatchewan, Canada. Asa, C.S., Bauman, J.E., Coonan, T.J., and M.M. Gray. 2007a. Evidence for induced estrus or ovulation in a canid, the island fox (Urocyon littoralis). Journal of Mammalogy 88:436–440.

186 · References Asa, C., P. Miller, M. Agnew, et al. 2007b. Relationship of inbreeding with sperm quality and reproductive success in Mexican gray wolves. Animal Conservation 10:326–331. Atkinson, A.E. 1985. The spread of commensal species of Rattus to oceanic islands and their effects on island avifauna. Pages 35–81 in P.J. Moors, ed., Conservation of Island Birds. Technical Publication 3. International Council for Bird Preservation, Cambridge, MA. Ausband, D.E. and K.R. Foresman. 2007a. Dispersal, survival, and reproduction of wild-born, yearling swift foxes in a reintroduced population. Canadian Journal of Zoology 85:185–189. Ausband, D.E. and K.R. Foresman. 2007b. Swift fox reintroductions on the Blackfeet Indian Reservation, Montana, USA. Biological Conservation 136:423–430. Baird, S.F. 1857. Mammals: general report upon the zoology of the several Pacific Railroad Routes. Page 148 in Reports of Explorations and Surveys to Ascertain the Most Practicable and Economical Route for a Railroad Across the Mississippi River to the Pacific Ocean. Volume 8. Washington, DC. Baker, A. 2007. Animal ambassadors: an analysis of the effectiveness and conservation impact of ex situ breeding efforts. Chapter 10 in A. Zimmerman, M. Hatchwell, L. Dickie, and C. West, eds., Zoos in the 21st Century: Catalysts for Conservation? Cambridge University Press, Cambridge. Bakker, V.J. and D.F. Doak. 2009. Population viability management: ecological standards to guide adaptive management for rare species. Frontiers in Ecology and the Environment 7:158–165. Bakker, V.J., D.F. Doak, G.W. Roemer, et al. 2009. Estimating and incorporating ecological drivers and parameter uncertainty into a demographic population viability analysis for the Island Fox (Urocyon littoralis). Ecological Monographs 79(1):77–108. Ballou, J.D. and T.J. Foose. 1996. Demographic and genetic management of captive populations. Pages 263–283 in D.G. Kleiman, M.E. Allen, K.V. Thompson, and S. Lumpkin, eds., Wild Mammals in Captivity: Principles and Techniques. University of Chicago Press, Chicago, IL. Balmford, A., G.M. Mace, and N. Leader-Williams. 1996. Designing the ark: setting priorities for captive breeding. Conservation Biology 10:719–727. Barker, I.K. and C.R. Parrish. 2001. Parvovirus infections. Pages 131–146 in E.S. Williams and I.K. Barker, eds., Infectious Diseases of Wild Mammals, 3rd edition. Iowa State University Press, Ames, IA Iowa. Bauman, J.E., D.L. Clifford, and C.S. Asa. 2008. Pregnancy diagnosis in wild canids using a commercially available relaxin assay. Zoo Biology 27:406–413. Berlow, E.L., S.A. Navarrete, C.J. Briggs, M.E. Power, and B.E. Menge. 1999. Quantifying variation in the strengths of species interactions. Ecology 80:2206– 2224. Boitani, L., C.S. Asa, and A. Moerhenschlager. 2004. Tools. Pages 143–163 in D.W. MacDonald and C. Sillero-Zubiri, eds., Biology and Conservation of Wild Canids. Oxford University Press, Oxford. Brown, J. and J. Munger. 1985. Experimental manipulation of desert rodent communities: food addition and species removal. Ecology 66:1545–1563.

References

· 187

Brown, J.H., T.G. Whitham, S.K.M. Ernest, and C.A. Gehring. 2001. Complex species interactions and the dynamics of ecological systems: long-term experiments. Science 293:643–650. Bush, M.R., J. Montali, D. Brownstein, A.E. James, Jr., and M.J.G. Appel. 1976. Vaccine-induced canine distemper in a lesser panda. Journal of the American Veterinary Medical Association 169:959–960. Carbyn, L.N., H.J. Armbruster, and C. Mamo. 1994. The swift fox reintroduction program in Canada from 1983 to 1992. Pages 247–272 in M.L Bowles and C.J. Whelan, eds., Restoration of Endangered Species: Conceptual Issues, Planning and Implementation. Cambridge University Press, Cambridge. Carlstead, K. 2005. Factors affecting the captive breeding success of the Channel Island foxes. Unpublished manuscript on file at park headquarters. Channel Islands National Park, Ventura, CA. Carpenter, J.W., M.J.G. Appell, R.C. Erickson, and M.N. Novilla. 1976. Fatal vaccine-induced canine distemper virus infection in black-footed ferrets. Journal of the American Veterinary Medical Association 169:961–964. Carroll, R., C. Augsperger, A. Dobson, J. Franklin, G. Orians, W. Reid, R. Tracy, D. Wilcove, and J. Wilson. 1996. Strengthening the use of science in achieving the goals of the Endangered Species Act: an assessment by the Ecological Society of America. Ecological Applications 6:1–11. Carter, H.R., W.R. McIver, and G.J. McChesney. 2008. Ashy Storm-petrel (Oceanodroma homochroa). Pages 117–124 in Shuford, D.W. and T. Gardali, eds., California Bird Species of Special Concern: A Ranked Assessment of Species, Subspecies and Distinct Populations of Birds of Immediate Conservation Concern in California. Studies in Western Birds 1. Western Field Ornithologists, Camarillo, CA, and California Department of Fish and Game, Sacramento, CA. Clark J.A., J.M. Hoekstra, P.D. Boersma, P. Kareiva 2002. Improving US Endangered Species Act recovery plans: key findings and recommendations of the SCB recovery plan project. Conservation Biology 16:1510–1519. Clark, R.A., W.L. Halvorson, A.A. Sawdo, and K.C. Danielsen. 1990. Plant communities of Santa Rosa Island, Channel Islands National Park. Cooperative Park Studies Unit, University of California, Davis, CA. Technical Report 42. Clark, T.A. 1987. Black-footed ferret recovery: a progress report. Conservation Biology 1:8–11. Clark, T.A., R.P. Reading, and A.L. Clarke, eds., 1994. Endangered Species Recovery: Finding the Lesions, Improving the Process. Island Press, Washington, DC. Clark, T.W. and R.D. Brunner. 1996. Making partnerships work in endangered species conservation: an introduction to the decision process. Endangered Species Update 13:1–5. Clifford, D. and T.W. Vickers. 2007. 2007 northern island fox reproductive health examination report. Unpublished manuscript on file at park headquarters. Channel Islands National Park, Ventura, CA. Clifford, D.L., J.A.K. Mazet, E.J. Dubovi, et al. 2006. Pathogen exposure in endangered island fox (Urocyon littoralis) populations: implications for conservation management. Biological Conservation 131:230–243.

188 · References Clifford, D.L., D.K. Garcelon, S.F. Timm, and J.A.K. Mazet. 2007. Using the pregnancy rates and perinatal mortality to evaluate the success of recovery strategies for endangered island foxes. Animal Conservation 10:442–451. Clobert, J., E. Danchin, A.A. Dhondt, and J.D. Nichols. 2001. Dispersal. Oxford University Press, Oxford. Coblentz, B.E. 1977. Some range relationships of feral goats on Santa Catalina Island, California. Journal of Range Management 30:415–419. Collins, P.W. 1979a. Vertebrate zoology: the San Miguel Island song sparrow. Pages 10.21–10.24 in D.M. Power, ed., Natural Resources Study of the Channel Islands National Monument. Santa Barbara Museum of Natural History, Santa Barbara, CA. Collins, P.W. 1979b. Vertebrate zoology: the biology of introduced black rats on Anacap and San Miguel Islands. Pages 14.1–14.56 in D.M. Power, ed., Natural Resources Study of the Channel Islands National Monument. Santa Barbara Museum of Natural History, Santa Barbara, CA. Collins, P.W. 1980. Food habits of the island fox (Urocyon littoralis littoralis) and San Miguel Island, California. Pages 152–164 in Proceedings of the second conference on scientific research in the national parks. Vol 12: Terrestrial Biology and Zoology. NTIS PB81-100133. Collins, P.W. 1982. Origin and differentiation of the island fox: a study of evolution in insular populations. Unpublished MA thesis, University of California, Santa Barbara, CA. Collins, P.W. 1991a. Interaction between island foxes (Urocyon littoralis) and Indians on islands off the coast of Southern California: I. Morphologic and archeological evidence of human assisted dispersal. Journal of Ethnobiology 11: 51–81. Collins, P.W. 1991b. Interaction between island foxes (Urocyon littoralis) and Indians on islands off the coast of Southern California: II. Ethnographic, archaeological and historical evidence. Journal of Ethnobiology 11:205–229. Collins, P.W. 1993. Taxonomic and biogeographic relationships of the island fox (Urocyon littoralis) and gray fox (Urocyon cinereoargenteus) from western North America. Pages 351–390 in F.G. Hochberg, ed., Proceedings of the Third California Islands Symposium: Recent Advances in Research on the California Islands. Santa Barbara Museum of Natural History, Santa Barbara, CA. Collins, P.W., and L.L. Laughrin. 1979. Vertebrate zoology: the island fox on San Miguel Island. Pages 12.1–12.52 in D.M. Power, ed., Natural Resources Study of the Channel Islands National Monument, California. Prepared for the National Park Service, Denver Service Center by the Santa Barbara Museum of Natural History, Santa Barbara, CA. Collins, P.W., D.A. Guthrie, T.C. Rick, and J.M. Erlandson. 2005. Analysis of prey remains excavated from an historic Bald Eagle nest site on San Miguel Island, California. Pages 103–120 in D.K Garcelon and C.A. Schwemm, eds., Proceedings of the Sixth California Islands Symposium, Ventura, CA, December 2003. National Park Service Technical Publication CHIS-05-01, Institute for Wildlife Studies, Arcata, CA. Collins, P.W., B.C. Latta, and G.W. Roemer. 2009. Does the order of invasive species removal matter? The case of the eagle and the pig. PLOS One 4:1–6.

References

· 189

Conway, W. 2007. Entering the 21st century. Chapter 2 in Zimmerman, A., M. Hatchwell, L. Dickie, and C. West, Eds. Zoos in the 21st Century: Catalysts for Conservation? Cambridge University Press, Cambridge. Coonan, T.J. 2003. Recovery strategy for island foxes (Urocyon littoralis) on the northern Channel Islands. Unpublished report on file at headquarters. Channel Islands National Park, Ventura, CA. Coonan, T.J. 2009. Eleventh annual meeting, island fox working group, summary report. Unpublished report on file at headquarters. Channel Islands National Park, Ventura, CA. Coonan, T.J. and M. Dennis. 2006. Island fox recovery program, 2005 annual report. Channel Islands National Park Technical Report 06-02. National Park Service, Ventura, CA. Coonan, T.J. and M. Dennis. 2007. Island fox recovery program, San Miguel and Santa Rosa Islands, 2006 annual report. Channel Islands National Park Technical Report 07-04. National Park Service, Ventura, CA. Coonan, T.J., and K. Rutz. 2001. Island fox captive breeding program, 1999– 2000 annual report. Technical Report 01-01. National Park Service, Ventura, CA. Coonan, T.J. and K. Rutz. 2002. Island fox captive breeding program, 2001 annual report. Technical Report 02-01. National Park Service, Ventura, CA. Coonan, T.J., and K. Rutz. 2003. Island fox captive breeding program, 2002 annual report. Technical Report 03-01. National Park Service, Ventura, CA. Coonan, T.J. and C.A. Schwemm. 2009. Factors contributing to success of island fox reintroductions on San Miguel and Santa Rosa Islands, California. Pages 363–376 in C.C. Damiani and D.K. Garcelon, ed., Proceedings of the Seventh California Islands Symposium, Oxnard, CA, February 2008. Institute for Wildlife Studies, Arcata, CA. Coonan, T.J., G. Austin, and C. Schwemm. 1998. Status and trend of island fox, San Miguel Island, Channel Islands National Park. Technical Report 98-01. National Park Service, Ventura, CA. Coonan, T.J., C.A. Schwemm, G.W. Roemer, and G. Austin. 2000. Population decline of island foxes on San Miguel Island. Pages 289–297 in D.K. Browne, K.L. Mitchell, and H.W. Chaney, eds., Proceedings of the Fifth California Islands Symposium. US Department of the Interior, Minerals Management Service, Pacific OCS Region, Camarillo, CA. Coonan, T., P. Dratch, C. Schwemm, D. Graber, and B. Fahey. 2003. Efforts to reduce golden eagle predation on island foxes at Channel Islands National Park. Unpublished report on file at park headquarters. Channel Islands National Park, Ventura, CA. Coonan, T.J., K.A. Rutz, K. McCurdy, D.K. Garcelon, B.C. Latta, and L. Munson. 2004. Island fox recovery program, 2003 annual report. Technical Report 04-02. National Park Service, Ventura, CA. Coonan, T.J., K.A. Rutz, D.K. Garcelon, B.C. Latta, M.M. Gray, and E.T. Aschehoug. 2005a. Progress in island fox recovery efforts on the northern Channel Islands. Pages 263–273 in D.K Garcelon and C.A. Schwemm eds., Proceedings of the Sixth California Islands Symposium. National Park Service Technical Publication CHIS-05-01, Institute for Wildlife Studies, Arcata, CA.

190 · References Coonan, T.J., C.A. Schwemm, G.W. Roemer, D.K. Garcelon, and L. Munson. 2005b. Decline of an island fox subspecies to near extinction. Southwestern Naturalist 50(1):32–41. Coonan, T.J., A. Varsik, C. Lynch, and C.A. Schwemm. 2010. Cooperative conservation: zoos and in-situ captive breeding for the endangered island fox. International Zoo Yearbook 44:165–172. Cooper, D.M., E.L. Kershner, and D.K. Garcelon. 2005. The use of shock collars to prevent island fox (Urocyon littoralis) predation on the endangered San Clemente loggerhead shrike (Lanius ludovicianus mearnsi). Pages 287–297 in Garcelon, D.K. and C.A. Schwemm, eds., Proceedings of the Sixth California Islands symposium. National Park Service Technical Publication CHIS-05-01, Institute for Wildlife Studies, Arcata, CA. Corry, P.M. 2006. Vegetation dynamics following grazing cessation on the Channel Islands, California. Unpublished Ph.D. Dissertation. University of North Carolina, Chapel Hill, NC. Courchamp, F., M. Langlais, and G. Sugihara. 1999. Control of rabbits to protect island birds from cat predation. Biological Conservation 89:219–225. Courchamp, F., R. Woodroffe, and G. Roemer. 2003. Removing protected populations to save endangered species. Science 302:1532. Crawford, M. 1985. The last days of the wild condor?: a debate is raging over whether to capture the few remaining wild California condors for breeding in captivity. Science 229:844–845. Crawley, M.J. 2002. Seed predators and plant population dynamics. Pages 167–183 in M. Fenner, ed., Seeds: The Ecology of Regeneration in Plant Communities. CAB International, New York. Croll, P.A., J.L. Maron, J.A. Estes, E.M. Danner, and G.V. Byrd. 2005. Introduced predators transform subarctic islands from grassland to tundra. Science 307:1959– 1961. Crooks, K.R. 1994. Comparative ecology of the island spotted skunk and the island fox of Santa Cruz Island, California. Unpublished master’s thesis. University of California, Davis, CA. Crooks, K.R. and M.E. Soule. 1999. Mesopredator release and avifaunal extinctions in a fragmented system. Nature 400:563–566. Crooks, K.R. and D. Van Vuren. 1995. Resource utilization by two insular endemic mammalian carnivores, the island fox and the island spotted skunk. Oecologia 104:301–307. Crooks, K.R. and D. Van Vuren. 1996. Spatial organization of the island fox (Urocyon littoralis) on Santa Cruz Island, California. Journal of Mammalogy 77:801–806. Crooks, K.R., C.A. Scott, L. Angeloni, L. Bowen, R.B. Kimsey, and D.H. Van Vuren. 2001a. Ectoparasites of the island fox on Santa Cruz Island. Journal of Wildlife Diseases 37:189–193. Crooks, K.R., C.A. Scott, and D.H. Van Vuren. 2001b. Exotic disease and an insular endemic carnivore, the island fox. Biological Conservation 98:55–60. Crowell, H.E.D. 2001. Food habits and prey availability of the threatened San Miguel Island fox, Urocyon littoralis littoralis, during a decline in population, 1993–1995. Unpublished master’s thesis. California Polytechnic State University, San Luis Obispo, CA.

References

· 191

Cypher, B.L., G.D. Warrick, J.R.M. Otten, et al. 2000. Population dynamics of San Joaquin kit foxes at the Naval petroleum reserves in California. Wildlife Monograph 145:1–43. Davis, G.E., K.R. Faulkner, and W.L. Halvorson. 1994. Ecological monitoring in Channel Islands National Park. Pages 465–482 in W.L. Halvorson and G.J. Maender, eds., The Fourth California Islands Symposium: Update on the Status of Resources. Santa Barbara Museum of Natural History, Santa Barbara, CA. DeMatteo, K.E., I.L. Porton, D.G. Kleiman and C.S. Asa. 2006. The effect of the male bush dog (Speothos venaticus) on the female reproductive cycle. Journal of Mammalogy 87:723–732. Demetrius, L. 2000. Directionality theory and the evolution of body size. Proceedings of the Royal Society of London Series B: Biological Sciences 267:2385–2391. Dennis, M., K. Randall, G. Schmidt, and D. Garcelon. 2001. Island fox (Urocyon littoralis santacruzae) distribution, abundance and survival on Santa Cruz Island, California. Progress report: May through October 2001. Institute for Wildlife Studies, Arcata, CA. DiBartola, S.P., M.J. Tarr, D.M. Webb and U. Giger. 1990. Familial renal amyloidosis in Chinese Shar Pei dogs. Journal of the American Veterinary Medical Association 197:483–487. Doak, D.F. and M.A. Marvier. 2003. Idiosyncratic effects of species loss in models of community stability. Pages 140–160 in P. Karevia and S. Levin, eds., The Importance of Species: Perspectives on Expendability and Triage. Princeton Press, Princeton, NJ. Dooley, J.A., P.B. Sharpe, and D.K. Garcelon. 2005. Movements, foraging, and survival of bald eagles reintroduced in the northern Channel Islands, California. Pages 313–322 in D.K. Garcelon and C.A. Schwemm, eds., Proceedings of the Sixth California Islands Symposium. National Park Service publication number CHIS-05-01, Institute for Wildlife Studies, Arcata, CA. Dratch, P., T. Coonan and D. Graber. 2004. Predators and prey in the Channel Islands. Science 305:777. Drost, C.A. and G.M. Fellers. 1991. Density cycles in an island population of deer mice, Peromyscus maniculatus (Wagner). Oikos 60:351–364. Drost, C., C.A. Schwemm, T. Coonan, and G. Fellers. In preparation. A large-scale natural experiment on generalist and specialist predators and small mammal populations: extirpation of foxes on San Miguel Island, California. Egoscue, H.J. 1975. Population dynamics of the kit fox in western Utah. Bulletin of the Southern California Academy of Sciences 74:122–127. Ellis, D.H., K.R. Clegg, J.C. Lewis, and E. Spaulding. 1999. Golden eagle predation on experimental sandhill and whooping cranes. Condor 101:664–666. Erlandson, J.E., T.C. Rick, J.A. Estes, et al. 2005. Sea otters, shellfish, and humans: 10,000 years of ecological interactions on San Miguel Island, California. Pages 9–21 in D.K Garcelon and C.A. Schwemm, eds., Proceedings of the Sixth California Islands Symposium. National Park Service Technical Publication CHIS05-01, Institute for Wildlife Studies, Arcata, CA. Ernest, S.K.M., J.H. Brown, and R.R. Parmenter. 2000. Rodents, plants, and precipitation: spatial and temporal dynamics of consumers and resources. Oikos 88:470–482.

192 · References Farias, V., Fuller, T.K., Wayne, R.K., and Sauvajot, R.M. 2005. Survival and causespecific mortality of gray foxes (Urocyon cinereoargenteus) in southern California. Journal of the Zoological Society of London, 266:249–254. Faulkner, C.T., S. Patton, L. Munson, E.M. Johnson, and T.J. Coonan. 2001. Angiocaulus gubernaculatus in the island fox (Urocyon littoralis) from the California Channel Islands and comments on the diagnosis of Angiostrongylidae nematodes in canid and mustelid hosts. Journal of Parasitology 87(5):1174–1176. Fausett, L.L. 1993. Activity and movement of the island fox, Urocyon littoralis, on Santa Cruz Island, California. Pages 391–404 in F.G. Hochberg, ed., The Third California Islands Symposium: Recent Advances in Research on the California Islands. Santa Barbara Museum of Natural History, Santa Barbara, CA. Fellers, G.M., C.A. Drost, and B.W. Arnold. 1988. Terrestrial vertebrates monitoring handbook. Unpublished report on file at park headquarters, Channel Islands National Park, Ventura, CA. Fiedel, S. and G. Haynes. 2004. A premature burial: comments on Grayson’s and Meltzer’s ‘Requiem for overkill’. Journal of Archaeological Science 31:121–131. Foin, T.C., S.P.D. Riley, A.L. Pawley, et al. 1998. Improving recovery planning for threatened and endangered species. BioScience 48:177–184. Foster, J.B. 1964. The evolution of mammals on islands. Nature 202:234–235. Frankham, R., J.D. Ballou, and D.A. Briscoe. 2002. Introduction to Conservation Genetics. Cambridge University Press, Cambridge. Garcelon, D.K. 1999. Island fox population analysis and management recommendations. Unpublished report. Southwest Division, Naval Facilities Engineering Command, San Diego, CA. Garcelon, D.K. and B.R. Hudgens. 2008. Island fox monitoring and demography on San Nicolas Island, 2007. Unpublished report prepared by the Institute for Wildlife Studies, Arcata, CA. Garcelon, D.K. and G.A. Schmidt. 2005. Island fox monitoring and demography on San Nicolas Island, 2004. Unpublished report prepared by the Institute for Wildlife Studies, Arcata, CA. Garcelon, D.K., G.W. Roemer, and G.C. Brundige. 1991. The biology of the island fox on San Clemente and Santa Catalina islands, California. Report to Wildlife Management Branch. California Department of Fish and Game, Contract Number FG-9402. Garcelon, D.K., R.K. Wayne, and B.J. Gonzales. 1992. A serologic survey of the island fox (Urocyon littoralis) on the Channel Islands, California. Journal of Wildlife Diseases 28:223–229. Garcelon, D.K., G.W. Roemer, R.B. Phillips, and T.J. Coonan. 1999. Food provisioning by island foxes, Urocyon littoralis, to conspecifics caught in traps. Southwestern Naturalist 44:83–86. Garcelon, D.K., K.P. Ryan, and P.T. Schuyler. 2005. Application of techniques for feral pig eradication on Santa Catalina Island, California. Pages 331–340 in D.K Garcelon and C.A. Schwemm, eds., Proceedings of the Sixth California Islands Symposium. National Park Service Technical Publication CHIS-05-01. Institute for Wildlife Studies, Arcata, CA. George, S.B. and R.K. Wayne. 1991. Island foxes: a model for conservation genetics. Terra 30:18–23.

References

· 193

Gibson, L. 2006. The role of lethal control in managing the effects of apparent competition on endangered prey species. Wildlife Society Bulletin 34: 1220–1224. Gilbert, D.A., N. Lehman, S.J. O’Brien, and R.K. Wayne. 1990. Genetic fingerprinting reflects population differentiation in the California Channel Island fox. Nature 344:764–767. Gilpin, M.E. and Soule, M.E. 1986. Minimum viable populations: processes of species extinction. Pages 19–34 in M.E. Soule, ed., Conservation Biology: The Science of Scarcity and Diversity. Sinauer, Sunderland, MA. Goldstein, D.B., G.W. Roemer, D.A. Smith, et al. 1999. The use of microsatellite variation to infer population structure and demographic history in a natural model system. Genetics 151:797–801. Goltsman, M., E.P. Kruchenkova, S. Sergeev, I. Volodin, and D.W. Macdonald. 2005. ‘Island syndrome’ in a population of Arctic foxes (Alopex lagopus) from Mednyi Island. Journal of Zoology 267:405–418. Goszczynski, J. 1989. Population dynamics of the red fox in central Poland. Acta Theologica 34:141–154. Gotelli, N.J. 2001. A Primer of Ecology. Sinauer, Sunderland, MA. Gray, M.M. 2002. Impact of a bottleneck on genetic diversity in the Channel Island fox (Urocyon littoralis) and genetic assessment of relatedness among individuals in a captive breeding program. Unpublished master’s thesis. California State University, Los Angeles, CA. Gray, M.M. 2003. Relatedness of Santa Cruz captive population and female 07061 within the Santa Rosa population. Unpublished report on file at headquarters. Channel Islands National Park, Ventura, CA. Gray, M.M., G.W. Roemer, and E. Torres. 2001. Genetic assessment of relatedness among individuals in the island fox (Urocyon littoralis) captive breeding program. Unpublished report on file at Park headquarters. Channel Islands National Park, Ventura, CA. Green, J.S., F.F. Knowlton, and W.C. Pitt, 2002. Reproduction in captive wildcaught coyotes (Canis latrans). Journal of Mammalogy 83:501–506. Greenwald, D.N. and K.F. Suckling. 2005. Progress or extinction? A systematic review of the US Fish and Wildlife Service’s Endangered Species Act Listing Program 1974–2004. Unpublished report available online at http://www.biologicaldiversity.org/publications/papers/esareport-revised.pdf (accessed March 2010). Grinnell, J., J.S. Dixon, and J.M. Linsdale. 1937. Fur-Bearing Mammals of California: Their Natural History, Systematic Status, and Relations to Man. University of California Press, Berkeley, CA. Guthrie, D. 1980. Analysis of avifaunal and bat remains from midden sites on San Miguel Island. Pages 689–702 in D.M. Power, ed., The California Islands: Proceedings of A Multidisciplinary Symposium. Santa Barbara Museum of Natural History, Santa Barbara, CA. Guthrie, D.A. 1993. New information on the prehistoric fauna of San Miguel Island, California. Pages 405–416 in F.G. Hochberg, ed., Proceedings of the Third California Islands Symposium: Recent Advances in Research on the California Islands. Santa Barbara Museum of Natural History, Santa Barbara, CA.

194 · References Guthrie, D.A. 2005. Distribution and provenance of fossil avifauna on San Miguel Island. Pages 35–39 in D.K Garcelon and C.A. Schwemm, eds., Proceedings of the Sixth California Islands Symposium. National Park Service Technical Publication CHIS-05-01, Institute for Wildlife Studies, Arcata, CA. Hallbrooks, R.D., L.J. Swango, P.R. Schnurrenberger, F.E. Mitchell, and E.P. Hill. 1981. Response of gray foxes to modified live-virus canine distemper vaccines. Journal of the American Veterinary Medical Association 179:1170–1174. Hale, A. and R.A. Salls. 2000. The canine ceremony: dog and fox burials of San Clemente Island. Pacific Coast Archaeological Society Quarterly 36:80–94. Harestad, A.S. and R.L. Bunnell. 1979. Home range and body weight: a reevaluation. Ecology 60:389–402. Harrison, R.L. 2003. Swift fox demography, movements, denning, and diet in New Mexico. The Southwestern Naturalist 48(2):261–273. Harrison, R.L., D.J. Barr, and J.W. Dragoo. 2002. A comparison of population survey techniques for swift foxes (Vulpes velox) in New Mexico. American Midland Naturalist 148:320–337. Hayhoe, K., D. Cayan, C.B. Field, et al. 2004. Emissions pathways, climate change and impacts on California. Proceedings of the National Academy of Sciences of the United States of America 101:12422–12427. Heath, S.R., E.L. Kershner, D.M. Cooper, et al. 2008. Rodent control and food supplementation increase productivity of endangered San Clemente Loggerhead Shrikes (Lanius ludovicianus mearnsi). Biological Conservation 141: 2506–2515. Hedrick, P.W. 2004. Foxy MHC selection story. Heredity 93:237–238. Hendron, J. 1998. Condors soar toward recovery. Endangered Species Bulletin 23:26– 29. Hoffman A.M. and F.A. Leighton. 1985. Hemograms and microscopic lesions of herring gulls in captivity. Journal of the American Veterinary Medical Association 187:1125–1128. Holmgren, M., M. Scheffer, E. Excurra, J.R. Gutierrez, and G.M.J. Mohren. 2001. El Ni˜no effects on the dynamics of terrestrial ecosystems. Trends in Ecology and Evolution 16:89–94. Holsinger, K.E. 2000. Demography and extinction in small populations. Pages 55– 74 in A.G. Young and G.M. Clarke, eds. Genetics, Demography and Viability of Fragmented Populations. Cambridge University Press, Cambridge. Holt, R.D. 1977. Predation, apparent competition, and the structure of prey communities. Theoretical Population Biology 12:197–229. Hooper, D.U., F.S. Chapin III, J.J. Ewel, et al. 2005. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecological Monographs 75:3–35. Hrdy, S.B. 1979. Infanticide among animals: a review, classification and examination of the implications for the reproductive strategies of females. Ethology and Sociobiology 1:13–40. Hudgens, B.R., and D.K. Garcelon. In preparation. Predator-induced changes in daily activity patterns of the Channel Islands fox. Hutchins, M. 2003. Zoo and aquarium animal management and conservation: current trends and future challenges. International Zoo Yearbook 38:14–28.

References

· 195

Hutchins, M. and W.G. Conway. 1995. Beyond Noah’s ark: the evolving role of modern zoological parks and aquariums in field conservation. International Zoo Yearbook 34:117–130. Ingvarsson, P.K. 2002. Lone wolf to the rescue. Nature 420:472. Inman, D.L., and S.A. Jenkins. 1999. Climate change and the episodicity of sediment flux of small California rivers. Journal of Geology 107:251–270. Institute for Wildlife Studies. 2006. Population status and golden eagle removal efforts on Santa Cruz and Santa Rosa Islands, 2005–2006. Unpublished report submitted to The Nature Conservancy. On file at headquarters, Channel Islands National Park, Ventura, CA. Jackman, R.E., W.G. Hunt, D.E. Driscoll, and J.F. Lapsansky. 1994. Refinements to selective trapping techniques: a radio-controlled bow net and power snare for bald and golden eagles. Journal of Raptor Research 28:268–273. Janzen, D.H. 1986. The eternal external threat. Chapter 13 in M.E. Soule, ed., Conservation Biology: The Science of Scarcity and Diversity. Sinauer, Sunderland, MA. Johnson, D.L. 1980. Episodic vegetation stripping, soil erosion, and landscape modification in prehistoric and recent historic time, San Miguel Island, California. Pages 103–122 in D.M. Power, ed., The California Islands: Proceedings of a Multidisciplinary Symposium. Santa Barbara Museum of Natural History, Santa Barbara, CA. Johnson, D.L. 1983. The California continental borderland: landbridges, watergaps and biotic dispersals. Pages 481–527 in P.M. Masters and N.C. Fleming, eds., Quaternary Coastlines and Marine Archaeology: Towards the Prehistory of Land Bridges and Continental Shelves. Academic Press, London. Johnson, R.R., T.W. Stafford, Jr., H.O. Ajie, and D.P. Morris. 2002. Arlington Springs revisited. Pages 541–545 in D. Browne, K. Mitchell, and H. Chaney, eds., Proceedings of the Fifth California Islands Symposium. Santa Barbara Museum of Natural History, Santa Barbara, CA. Jones, T.L., J.F. Porcasi, J.M. Erlandson, et al. 2008a. The protracted Holocene extinction of California’s flightless sea duck (Chendytes lawi) and its implications for the Pleistocene overkill hypothesis. Proceedings of the National Academy of Sciences 105:4105–4108. Jones, K.L., D.H. Van Vuren, and K.R. Crooks. 2008b. Sudden increase in a rare endemic carnivore: ecology of the island spotted skunk. Journal of Mammalogy 89:75–86. Kamler, J.F., W.B. Ballard, E.M. Gese, R.L. Harrison, and S.M. Karki. 2004. Dispersal characteristics of swift foxes. Canadian Journal of Zoology 82:1837–1842. Kareiva, P. and S.A. Levin. 2003. The Importance of Species: Perspectives on Expendability and Triage. Princeton University Press, Princeton, NJ. Keller, L.F., and D.M. Waller. 2002. Inbreeding effects in wild populations. Trends in Ecology and Evolution 17:230–241. Kiff, L.F. 1980. Historical changes in resident populations of California islands raptors. Pages 651–673 in D.M. Power, ed., The California Islands: Proceedings of a Multidisciplinary Symposium. Santa Barbara Museum of Natural History, Santa Barbara, CA. Kleiman, D.G. 1977. Monogamy in mammals. Quarterly Review of Biology 52:39–69.

196 · References Kleiman, D. 1989. Reintroduction of captive mammals for conservation. BioScience 39:152–161. Kleiman, D.G., J.R. Malcolm, and P.H. Klopfer. 1981. The evolution of male parental investment in mammals. Pages 347–387 in D.J. Gubernick, ed., Parental Care in Mammals. Plenum Press, New York. Klinger, R.C., P. Schuyler, and D.D. Sterner. 2002. The response of herbaceous vegetation and endemic plant species to the removal of feral sheep from Santa Cruz Island, California. Pages 141–154 in C.R Veitch and M.N. Clout, eds., Turning the Tide: The Eradication of Invasive Species: Proceedings of the International Conference on Eradication of Island Invasives. IUCN, Cambridge. Kochert, M.N., K. Steenhof, C.L. McIntyre, and E.H. Craig. 2002. Golden eagle (Aquila chrysaetos). In A. Poole and F. Gill, eds., The Birds of North America No. 684. The Birds of North America, Inc., Philadelphia, PA. Kohler, S.L. and W.K. Hoiland. 2001. Population regulation in an aquatic insect: the role of disease. Ecology 82:2294–2305. Kohlmann, S.G., G.A. Schmidt, R.C. Wolstenholme, and D.K. Garcelon. 2003. Island fox recovery effort on Santa Catalina Island, California, October 2001– October 2002, Annual Report. Unpublished report by the Institute for Wildlife Studies, Arcata, CA. Kohlmann, S.G., G.A. Schmidt, and D.K. Garcelon. 2005. A population viability analysis for the island fox on Santa Catalina Island. Ecological Modeling 183:77– 94. Kovach, S.D. and R.J. Dow. 1981. Status and Ecology of the Island Fox on San Nicolas Island, 1980. Department of the Navy, Pacific Missile Test Center, Point Mugu, CA. Kovach, S.D. and R.J. Dow. 1985. Island Fox Research, San Nicolas Island, Ventura County, California. 1984 Annual Report. US Navy, Pacific Missile Test Center, Point Mugu, CA. Latta, B.C., D.D. Driscoll, J.L. Linthicum, R.E. Jackman, and G. Doney. 2005. Capture and translocation of golden eagles from the California Channel Islands to mitigate depredation of endemic island foxes. Pages 341–350 in D.K Garcelon and C.A. Schwemm, eds., Proceedings of the Sixth California Islands Symposium. National Park Service Technical Publication CHIS-05-01, Institute for Wildlife Studies, Arcata, CA. Laughrin, L. 1973. California Island Fox Survey. Wildlife Management Branch Administrative Report 73-3. California Department of Fish and Game, Sacramento, CA. Laughrin, L.L. 1977. The island fox: a field study of its behavior and ecology. Unpublished Ph.D. dissertation. University of California, Santa Barbara, CA. Laughrin, L.L. 1980. Population and status of the island fox. Pages 745–750 in D.M. Power, ed., The California Islands: Proceedings of a Multidisciplinary Symposium. Santa Barbara Museum of Natural History, Santa Barbara, CA. Lavopierre, M.M., T.W. Graham, L.L. Walters, and J.A. Howarth. 1986. The occurrence of two nematodes, Spirocerca lupi and Dirofiliria immitis, in wild canids of the Lake Berryessa area, Northern California. California Fish and Game 72:78– 82. Lawlor, T.E. 1982. The evolution of body size in mammals: evidence from insular populations in Mexico. American Naturalist 119:54–72.

References

· 197

Layne, J.N. and W.H. McKeon. 1956. Some aspects of red fox and gray fox reproduction in New York. New York Fish and Game Journal 3:44–74. Lima, S.L. and L.M. Dill. 1990. Behavioral decisions made under the risk of predation: a review and prospectus Canadian Journal of Zoology 68:619–640. Lipman, R.D., E.T. Gaillard, D.E. Harrison, and R.T. Bronson. 1993. Husbandry factors and the prevalence of age-related amyloidosis in mice. Laboratory Animal Science 43:439–444. Lisk, R. and D.W. MacDonald. 2003. Home range and habitat use of the kit fox (Vulpes macrotis) in a prairie dog (Cynomys ludovicianus) complex. Journal of Zoology 259:1–5. Lombardo, C.A. and K.R. Faulkner. 2000. Eradication of feral pigs (Sus scrofa) from Santa Rosa Island, Channel Islands National Park, California. Pages 300–306 in D.K. Browne, K.L. Mitchell, and H.W. Chaney, eds., Proceedings of the Fifth California Islands Symposium. US Department of the Interior, Minerals Management Service, Pacific OCS Region, Camarillo, CA. Lomolino, M.V. 1985. Body size of mammals on islands: the island rule reexamined. American Naturalist 125:310–316. Lynch, C. 2004. Analysis and Population Management Plan: Island Fox Urocyon Littoralis spp. Population Management Plan. American Zoo and Aquarium Association Population Management Center, Chicago, IL. Lynch, C. 2005a. Analysis and Population Management Plan: San Miguel Island Fox Urocyon littoralis littoralis Population Management Plan. American Zoo and Aquarium Association Population Management Center, Chicago, IL. Lynch, C. 2005b. Analysis and Population Management Plan: San Miguel Island Fox Urocyon Littoralis Littoralis Population Management Plan. American Zoo and Aquarium Association Population Management Center, Chicago, IL. Macpherson, A.H. 1969. The Dynamics of Canadian Arctic Fox Populations. Canadian Wildlife Service Report Series 8. Canadian Wildlife Service, Ottawa. McCann, K.S. 2000. The diversity–stability debate. Nature 405:228–233. McChesney, G.J. and B.R. Tershey. 1998. History and status of introduced mammals and impacts to breeding seabirds on the California Channel and northwestern Baja California islands. Colonial Waterbirds 21:335–347. McIver, W.R., A.L. Harvey, and H.R. Carter. 2009. Monitoring and restoration of Ashy Storm-Petrels at Santa Cruz Island, California in 2008. Unpublished report, US Fish and Wildlife Service, Arcata, California; Channel Islands National Park, Ventura, CA; and Carter Biological Consulting, Victoria, British Columbia. McNab, B.K. 1989. Basal rates of metabolism, body size and food habits in the Order Carnivora. Chapter 12 in J.L. Gittleman, ed. Carnivore Behavior, Ecology and Evolution. Cornell University Press, Ithaca, NY. McPhee, M.E. 2003. Generations in captivity increases behavioral variance: considerations for captive breeding and reintroduction programs. Biological Conservation 115:71–77. McPhee, M.E. and E.D. Silverman. 2004. Increased behavioral variation and the calculation of release numbers for reintroduction programs. Conservation Biology 18:705–715. Mech, L.D. and S.T. Knick. 1978. Sleeping distances in wolf pairs in relation to the breeding season. Behavioral Biology 23:521–525.

198 · References Meiri, S., T. Dayan, and D. Simberloff. 2004. Body size of insular carnivores: little support for the island rule. American Naturalist 163:E469–E479. Menge, B.A. and J.P. Sutherland. 1987. Community regulation: variation in disturbance, competition and predation in relation to environmental stress and recruitment. American Naturalist 130:730–757. Merriam, C.H. 1903. Eight new mammals from the United States. Proceedings of the Biological Society of Washington 16:73–78. Meserve, P.L., D.A. Kelt, B. Milstead, and J.R. Gutierrez. 2003. Thirteen years of shifting top-down and bottom-up control. BioScience 53:633–646. Metrick, A. and M.L. Weitzman. 1996. Patterns of behavior in endangered species protection. Land Economics, 72:1–16. Miller, P.S., G.W. Roemer, J. Laake, C. Wilcox, and T.J. Coonan. 2001. Population viability assessment for selected Channel Island fox wild populations. Unpublished report on file at headquarters. Channel Islands National Park, Ventura, CA. Mittelbach, G.G., A.M. Turner, D.J. Hall, J.E. Rettig, and C.W. Osenberg. 1995. Perturbation and resilience: a long-term, whole-lake study of predator extinction and reintroduction. Ecology 76:2347–2360. Montali, R.J., C.R. Bartz, J.A. Teare, et al. 1983. Clinical trials with canine distemper vaccines in exotic carnivores. Journal of the American Veterinary Medical Association 183:1163–1167. Moore, C.M. and P.W. Collins. 1995. Urocyon littoralis. Mammalian Species 489:1–7. Morin, P.J. 1999. Community Ecology. Blackwell Science, Malden, MA. Morris, W.F., P.L. Bloch, B.R. Hudgens, L.C. Moyle, and J.R. Stinchcombe. 2002. Population viability analysis in endangered species recovery plans: past use and future improvements. Ecological Applications 12:708–712. Morrison, M.L. 2002. Wildlife Restoration: Techniques for Habitat Analysis And Animal Monitoring. Island Press, Washington, DC. Morrison, S.A., N. Macdonald, K. Walker, L. Lozier, and M.R. Shaw. 2007. Facing the dilemma at eradication’s end: uncertainty of absence and the Lazarus effect. Frontiers in Ecology and the Environment 5:271–276. Munson, L. 1999. Unpublished letter to David K. Garcelon and Timothy J. Coonan: January 16, 1999. On file at park headquarters, Channel Islands National Park, Ventura, CA. Munson, L., K.A. Terio, R. Kock, et al. 2008. Climate extremes promote fatal co-infections during canine distemper epidemics in African lions. PLoS ONE 3(6):2545. Available online at http://www.plosone.org/article/info:doi/ 10.1371/journal.pone.0002545 (accessed March 2010). Namekata, M.S., D.L. Clifford, R.W. Kasten, et al. 2009. Seroprevalence of Bartonella spp. in the endangered island fox (Urocyon littoralis). Veterinary Microbiology 136:184–187. National Park Service. 1985. General Management Plan for Channel Islands National Park, Volume 1. National Park Service, Ventura, CA. National Park Service. 2002. Santa Cruz Island Primary Restoration Plan and Final Environmental Impact Statement. National Park Service, Ventura, CA. Natural Resources Trustees. 2006. Executive summary of the Montrose Settlements Restoration Program Restoration Plan, Programmatic

References

· 199

Environmental Impact Statement, and Programmatic Environmental Impact Report. Available online at http://www.darrp.noaa.gov/southwest/montrose/ pdf/msrp_rp_execsumm.pdf (accessed March 2010). Nicholopoulos, J. 1999. The endangered species listing program. Endangered Species Bulletin 24:6–9. Nichols, J.D. and B.K. Williams. 2006. Monitoring for conservation. Trends in Ecology and Evolution 21:668–673. O’Gara, B.W. and D.C. Getz. 1986. Capturing golden eagles using a helicopter and net-gun. Wildlife Society Bulletin 14:400–402. Olson, T.L. and F.G. Lindzey. 2002. Swift fox survival and production in southeastern Wyoming. Journal of Mammalogy 83:199–206. Orr, P.C. 1968. Prehistory of Santa Rosa Island. Santa Barbara Museum of Natural History, Santa Barbara, CA. Orrock, J.L., B.J. Danielsen, and R.J. Brinkerhoff. 2004. Rodent foraging is affected by indirect, but not by direct, cues of predation risk. Behavioral Ecology 15:433– 437. Pace, M.L., J.J. Cole, S.R. Carpenter, and J.F. Kitchell. 1999. Trophic cascades revealed in diverse ecosystems. TREE 14:484–488. Padgett, K.A. and W.M. Boyce. 2005. Ants as first intermediate hosts of Mesocestoides on San Miguel Island, USA. Journal of Helminthology 79:67–73. Papendick, R.E., L. Munson, T.D. O’Brien, and K.H. Johnson. 1997. Systemic AA amyloidosis in captive cheetahs (Acinonyx jubatus). Veterinary Pathology 34:549– 556. Peluc, S.I., T.S. Sillett, J.T. Rotenberry, and C.K. Ghalambor. 2008. Adaptive phenotypic plasticity in an island songbird exposed to a novel predation risk. Behavioral Ecology 19:830–835. Phelan, J.P. and R.H. Baker. 1992. Optimal foraging in Peromyscus polionotus: the influence of item-size and predation risk. Behaviour 121:95–109. Phillips, R.B., C.S. Winchell, and R.H. Schmidt. 2007. Dietary overlap of an alien and native carnivore on San Clemente Island, California. Journal of Mammalogy 88:173–180. Phillips, R.L., J.L. Cummings, and J.D. Berry. 1991. Response of breeding golden eagles to relocation. Wildlife Society Bulletin 19:430–434. Price, E.A., J. Vining, and C.D. Saunders. 2008. Intrinsic and extrinsic rewards in a nonformal environmental education program. Zoo Biology 28: 361–376. Propst, B. 1975. A population survey of the Santa Catalina Island Fox. A report to the California Department of Fish and Game, Nongame Wildlife Investigation Project W-54-R-8, Job I-1.10. Rabb, G.B. and C.D. Saunders. 2005. The future of zoos and aquariums: conservation and caring. International Zoo Yearbook 39:1–26. Ralls, K., and P.J. White. 1995. Predation on San Joaquin kit foxes by larger canids. Journal of Mammalogy 76:723–729. Ralls, K., J.D. Ballou, and A. Templeton. 1988. Estimates of lethal equivalents and the cost of inbreeding in mammals. Conservation Biology 2:185–193. Randall, D.A., S.D. Williams, I.V. Kuzmin, et al. 2004. Rabies in endangered Ethiopian wolves. Emerging Infectious Diseases 10:2214–2217.

200 · References Ranta, E., P. Lundberg, and V. Kaitala. 2006. Ecology of Populations. Cambridge University Press, Cambridge. Rasmussen, E.M. and J.M. Wallace. 1983. Meteorological aspects of the El Ni˜no/Southern Oscillation. Science 222:1195–1202. Reading, R.P. and B.J. Miller. 1994. The black-footed ferret recovery program: unmasking professional and organizational weakness. Pages 73–100 in T.A. Clark, R.P. Reading, and A.L. Clarke, eds. Endangered Species Recovery: Finding the Lesions, Improving the Process. Island Press, Washington, DC. Reisen, W.K., Y. Fang, and V.M. Martinez. 2005. Avian host and mosquito (Diptera: Culicidae) vector competence determine the efficiency of West Nile and St. Louis Encephalitis Virus transmission. Journal of Medical Entomology 42(3):367– 375. Restani, M. and J.M. Marzluff. 2002. Funding extinction? Biological needs and political realities in the allocation of resources to endangered species recovery. BioScience 52:169–177. Rick, T.C., P.L. Walker, L.M. Willis, et al. 2008. Dogs, humans, and island ecosystems: the distribution, antiquity and ecology of domestic dogs (Canis familiaris) on California’s Channel Islands, USA. The Holocene 18:1077–1087. Rick, T.C., J.M. Erlandson, R.L. Vellanoweth, et al. 2009. The origins and antiquity of the island fox (Urocyon littoralis): AMS 14C dates from California’s Channel Islands. Quaternary Research 71:93–98. Rideout, B.A., R.J. Montali, R.S. Wallace, et al. 1989. Renal medullary amyloidosis in Dorcas gazelles. Veterinary Pathology 26:129–135. Roelke-Parker, M.E., L. Munson, C. Packer, et al. 1996. A canine distemper virus epidemic in Serengeti lions (Panthera leo). Nature 379:441–445. Roemer, G.W. 1999. The ecology and conservation of the island fox (Urocyon littoralis). Unpublished Ph.D. dissertation. University of California, Los Angeles, CA. Roemer, G.W. 2004. Island foxes: evolution, behavioural ecology, and conservation of island foxes. Pages 173–198 in D.W MacDonald and C. Sillero-Zubiri, eds., Biology and Conservation of Wild Canids. Oxford University Press, Oxford. Roemer, G.W. and C.J. Donlan. 2004. Biology, policy and law in endangered species conservation I: the case history of the island fox on the northern Channel Islands. Endangered Species Update 21:23–31. Roemer, G.W. and R.K. Wayne. 2003. Conservation in conflict: the tale of two endangered species. Conservation Biology 17:1251–1260. Roemer, G.W., D.K. Garcelon, T.J. Coonan, and C. Schwemm. 1994. The use of capture-recapture methods for estimating, monitoring, and conserving island fox populations. Pages 387–400 in W.L. Halvorson and G.J. Maender, eds. The Fourth California Channel Islands Symposium: Update on the Status of Resources. Santa Barbara Museum of Natural History, Santa Barbara, CA. Roemer, G.W., T.J. Coonan, D.K. Garcelon, C.H. Starbird, and J.W. McCall. 2000. Spatial and temporal variation in the seroprevalence of canine heartworm antigen in the island fox. Journal of Wildlife Diseases. 36:723–728. Roemer, G.W., T.J. Coonan, D.K. Garcelon, J. Bascompte, and L. Laughrin. 2001a. Feral pigs facilitate hyperpredation by golden eagles and indirectly cause the decline of the island fox. Animal Conservation 4:307–318.

References

· 201

Roemer, G.W., D.A. Smith, D.K. Garcelon, and R.K. Wayne. 2001b. The behavioral ecology of the island fox (Urocyon littoralis). Journal of the Zoological Society of London 255:1–14. Roemer, G.W., C.J. Donlan, and F. Courchamp. 2002. Golden eagles, feral pigs, and insular carnivores: how exotic species turn native predators into prey. Proceedings of the National Academy of Sciences 99:791–796. Roemer, G.W., T.J. Coonan, L. Munson, and R.K. Wayne. 2004. Island fox. Pages 97–105 in C. Sillero-Zubiri, M. Hoffmann, and D.W. Macdonald, eds., Canids: Foxes, Wolves, Jackals and Dogs. IUCN, Gland, Switzerland. Rolland, R.M. 2000. A review of chemically-induced alteration in thyroid and vitamin A status from field studies of wildlife and fish. Journal of Wildlife Diseases 36:615–635. Romagnoli, S. 2002. Complications in canine pregnancy and their clinical approach. Pages 159–162 in Proceedings of the Veterinary Sciences Congress. SPCV, Oeiras. Roth, V.L. 1993. Dwarfism and variability in the Santa Rosa Island mammoth (Mammuthus exilis): an interspecific comparison on limb-bone sizes and shapes in elephants. Pages 433–442 in F.G. Hochberg, ed., Third California Islands Symposium: Recent Advances in Research on the California Islands: Proceedings of the Third California Islands Symposium. Santa Barbara Museum of Natural History, Santa Barbara, CA. Rubin, E.S., V.J. Bakker, M.G. Efford, et al. 2007. A population monitoring framework for five subspecies of island fox (Urocyon littoralis). Prepared by the Conservation Biology Institute and The Nature Conservancy for the Recovery Coordination Group of the Island Fox Integrated Recovery Team. US Fish and Wildlife Service, Ventura, CA. Schauster, E.R., E.M. Gese, and A.M. Kitchen. 2002a. An evaluation of survey methods for monitoring swift fox abundance. Wildlife Society Bulletin 30:464– 477. Schauster, E.R., E.M. Gese, and A.M. Kitchen. 2002b. Population ecology of swift foxes (Vulpes velox) in southeastern Colorado. Canadian Journal of Zoology 80:307–319. Schmidt, G.A. and D.K. Garcelon. 2003. Island fox monitoring and demography on San Nicolas Island: 2002. Unpublished report prepared by the Institute for Wildlife Studies, Arcata, CA. Schmidt, K.A. and R.S. Ostfeld. 2003. Songbird populations in fluctuating environments: predator responses to pulsed resources. Ecology 84:406–415. Schmidt, K.A., J.R. Goheen, and R. Naumann. 2001. Incidental nest predation in songbirds: behavioral indicators detect ecological scales and processes. Ecology 82:2937–2947. Schmidt, G.A., C.M. Steele, T.W. Vickers, and D.K. Garcelon. 2004a. Island fox recovery efforts on Santa Catalina Island, California, October 2002–October 2003, Annual Report. Unpublished report by the Institute for Wildlife Studies, Arcata, California for the Ecological Restoration Department, Catalina Island Conservancy, Avalon, CA. Schmidt, G.A., B.J. Willson, and D.K. Garcelon. 2004b. Island fox monitoring and research on naval auxiliary landing field San Clemente Island, California. Institute for Wildlife Studies, Arcata, CA.

202 · References Schmidt, G.A., T.W. Vickers, and D.K. Garcelon. 2005. Island fox recovery efforts on Santa Catalina Island, California, 1 April 2004–30 April 2005. Unpublished report by the Institute for Wildlife Studies. Arcata, California for the Ecological Restoration Department, Catalina Island Conservancy, Avalon, CA. Schmidt, G.A., J. Fox, and D.K. Garcelon. 2007a. Island fox recovery on Santa Cruz Island, California: December 2000–January 2007. Unpublished report by the Institute for Wildlife Studies, Arcata, CA. Schmidt, G.A., B.R. Hudgens, and D.K. Garcelon. 2007b. Island fox monitoring and demography on San Nicolas Island, 2006. Unpublished report prepared by the Institute for Wildlife Studies, Arcata, CA. Schmitz, O.J. 2003. Top predator control of plant biodiversity and productivity in an old-field ecosystem. Ecological Letters 6:156–163. Schmitz, O.J., P.A. Hamb¨ack, and A.P. Beckerman. 2000. Trophic cascades in terrestrial systems: a review of the effects of carnivore removals on plants. American Naturalist 155:141–153. Schwemm, C.A. 2007. Ninth annual meeting, island fox working group, summary report. Unpublished report on file at headquarters. Channel Islands National Park, Ventura, CA. Schwemm, C.A. 2008a. Establishment limitations and population recovery of giant coreopsis (Coreopsis gigantea) on the California Channel Islands. Unpublished Ph.D. Dissertation. University of California, Santa Barbara, CA. Schwemm, C.A. 2008b. Tenth annual meeting, island fox working group, summary report. Available online at http://www.nps.gov/chis/naturescience/ upload/IF_2008MeetingReport_FinalC.pdf (accessed March 2010). Scott, J.M., D.D. Goble, J.A. Wiens, et al. 2005. Recovery of imperiled species under the Endangered Species Act: the need for a new approach. Frontiers in Ecology and the Environment 3:383–389. Scrivner, J.H., T.P. O’Farrell, and T.T. Kato. 1987. Dispersal of San Joaquin kit foxes, Vulpes macrotis mutica, on Naval Petroleum Reserve, Volume 1, Kern County, California. EGG 10282–2190. EG&G Energy Measurements, Goleta, CA. Sharpe, P.B., and D.K. Garcelon. 2005. Restoring and monitoring bald eagles in southern California: the legacy of DDT. Pages 323–330 in D.K. Garcelon and C.A. Schwemm, eds., Proceedings of the Sixth California Islands Symposium. National Park Service publication number CHIS-05-01. Institute for Wildlife Studies, Arcata, California. Shogren, J.F. and J. Tschirhart. 2001. Protecting Endangered Species in the United States. Cambridge University Press, Cambridge. Shurin, J.B., E.T. Borer, E.W. Seabloom, et al. 2002. A cross-ecosystem comparison of the strength of trophic cascades. Ecological Letters 5:785–791. Sih, A., P. Crowley, M. McPeek, J. Petranka, and K. Strohmeier. 1985. Predation, competition and prey communities: a review of field experiments. Annual Review of Ecology and Systematics. 16:269–311. Simberloff, D.S. 1974. Equilibrium theory of island biogeography and ecology. Annual Review of Ecology and Systematics 5:161–182. Simberloff, D. 2003. How much information on population biology is needed to manage introduced species? Conservation Biology 17:83–92.

References

· 203

Smith, K.N., J.H. Shaw, T. Bettinger, B. Caniglia, and T. Carter. 2007. Conservation partnerships between zoos and aquariums, federal and state agencies, and nongovernmental organizations. Zoo Biology 26:471–486. Snyder, N.F., S.R. Derrickson, S.R. Beissinger, et al. 1996. Limitations of captive breeding in endangered species recovery. Conservation Biology 10:338–348. Sohn, R.A. and N. Thomas. 2005. Captive island fox parasite assessment and control on San Miguel, Santa Rosa and Santa Cruz Islands. Final report to the National Park Service. Unpublished report on file at headquarters. Channel Islands National Park, Ventura, CA. Solomon, B.D. 1998. Impending recovery of Kirtland’s warbler: case study in the effectiveness of the Endangered Species Act. Environmental Management 22:9– 17. Sovada, M.A., C.C. Roy, J.B. Bright, and J.R. Gillis. 1998. Causes and rates of mortality of swift foxes in western Kansas. Journal of Wildlife Management 62:1300– 1306. Sovada, M.A., C. Asa, K. Bauman, and D. Clifford. 2006. Identification of factors influencing breeding success and productivity in captive populations of the endangered Channel Island Fox. Unpublished report submitted to USFWS. On file at headquarters, Channel Islands National Park, Ventura, CA. Spencer, W., E. Rubin, J. Stallcup, et al. 2006. Framework monitoring plan for the San Clemente Island fox with specific recommendations for the 2006 field season. Conservation Biology Institute, Encinitas, CA. Srivastava, D.S. 2002. The role of conservation in expanding biodiversity research. Oikos 98:351–360. Stamps, J.A. and M. Buechner. 1985. The territorial defense hypothesis and the ecology of insular vertebrates. The Quarterly Review of Biology 60:155–181. Sterling, E., J. Lee, and T. Wood. 2007. Conservation education in zoos: an emphasis on behavioral change. Chapter 4 in A. Zimmerman, M. Hatchwell, L. Dickie, and C. West, Eds. Zoos in the 21st Century: Catalysts for Conservation? Cambridge University Press, Cambridge. Sullivan, T.P. 1977. Demography and dispersal in island and mainland populations of the deer mouse, Peromyscus maniculatus. Ecology 58:964–978. Sumner, E.L., Jr. and R.M. Bond. 1939. An investigation of Santa Barbara, Anacapa, and San Miguel islands, California. Unpublished report on file at headquarters. Channel Islands National Park, Ventura, CA. Sutherland-Smith, M.R., B.A. Rideout, A.B. Mikolon, et al. 1997. Vaccine-induced canine distemper in European mink, Mustela lutreola. Journal of Zoo and Wildlife Medicine 28:312–318. Swarts, H.M., K.R. Crooks, N. Willits, and R. Woodroffe. 2009. Possible contemporary evolution in an endangered species, the Santa Cruz Island fox. Animal Conservation 12:120–127. Sweitzer, R.A., J.M. Constible, D.H. Van Vuren, P.T. Schuyler, and F.R. Starkey. 2005. History, habitat use and management of bison on Catalina Island, California. Pages 231–248 in D.K Garcelon and C.A. Schwemm, eds., Proceedings of the sixth California Islands symposium. National Park Service Technical Publication CHIS-05-01, Institute for Wildlife Studies, Arcata, CA.

204 · References Thebault, E. and M. Loreau. 2006. The relationship between biodiversity and ecosystem function in food webs. Ecological Research 21:17–25. Timm, S.F., J.M. Stokely, T.B. Gehr, R.L. Peebles, and D.K. Garcelon. 2000. Investigation into the decline of island foxes on Santa Catalina Island. Unpublished report. Institute for Wildlife Studies, Arcata, CA. Timm, S.F., W.D. Barker, S.A. Johnson, et al. 2002. Island fox recovery effort on Santa Catalina Island, California, September 2000–October 2001, Annual Report. Institute for Wildlife Studies, Arcata, CA. Timm, S.F., L. Munson, B.A. Summers, et al. 2009. A suspected canine distemper epidemic as the cause of a catastrophic decline in Santa Catalina Island foxes (Urocyon littoralis catalinae). Journal of Wildlife Diseases 45:333–343. Urios, V., A. Soutullo, P. Lopez-Lopez, et al. 2007. The first case of successful breeding of a golden eagle, Aquila chrysaetos, tracked from birth by satellite telemetry. Acta Ornithologica 42:205–209. US Fish and Wildlife Service. 1982. Endangered and threatened wildlife and plants: review of vertebrate wildlife for listing as endangered or threatened species. Federal Register 47(251):58454–58461. US Fish and Wildlife Service. 1985. Endangered and threatened wildlife and plants: review of vertebrate wildlife. Federal Register 50(181):37958–37968. US Fish and Wildlife Service. 1994. Endangered and threatened wildlife and plants: animal candidate review for listing as endangered or threatened species. Federal Register 59(219):58982–59129. US Fish and Wildlife Service. 2001a. Endangered and threatened wildlife and plants: review of plant and animal species that are candidates or proposed for listing as endangered or threatened, annual notice of findings on recycled petitions, and annual description of progress on listing actions; proposed rule. Federal Register 66(210):54808–54832. US Fish and Wildlife Service. 2001b. Listing of the San Miguel island fox, Santa Rosa island fox, Santa Cruz island fox, and Santa Catalina island fox as endangered; final rule. Federal Register 66(237):63654–63666. US Fish and Wildlife Service. 2004. Listing of the San Miguel island fox, Santa Rosa island Fox, Santa Cruz island fox, and Santa Catalina island fox as endangered; final rule. Federal Register 69(44):10335–10353. US Fish and Wildlife Service. 2009. Final Environmental Assessment for the Restoration of San Nicolas Island’s Seabirds and Protection of Other Fauna by Removing Feral Cats. US Fish and Wildlife Service, Sacramento, CA. Valdespino, C., C.S. Asa, and J.E. Bauman. 2002. Estrous cycles, copulation, and pregnancy in the fennec fox (Vulpes zerda). Journal of Mammalogy 83:99–109. Van Valen, L. 1973. Body size and numbers of plants and animals. Evolution 27: 27–35. Vellanoweth, R.L. 1998. Earliest island fox remains on the Southern Channel Islands: evidence from San Nicolas Island, California. Journal of California and Great Basin Anthropology 20:100–108. Verstegen, J., G. Dhaliwal, and K. Verstegen-Onclin. 2008. Canine and feline pregnancy loss due to viral and non-infectious causes: a review. Theriogenology 70:304–319.

References

· 205

Vucetich, J.A. and R.O. Peterson. 2004. The influence of top-down, bottom-up and abiotic factors on the moose (Alces alces) population of Isle Royale. Proceedings of the Royal Society Biological Sciences 271:183–189. Wallace, M.P. 2000. Retaining natural behavior in captivity for re-introduction programmes. Pages 300–314 in L.M Gosling and W.J. Sutherland, eds. Behaviour and Conservation. Cambridge University Press, Cambridge. Walpole, M.J. and N. Leader-Williams. 2002. Tourism and flagship species in conservation. Biodiversity and Conservation 11:543–547. Waser, P.M. 1996. Patterns and consequences of dispersal in gregarious carnivores. Pages 267–295 in J.L. Gittleman, ed., Carnivore Behavior, Ecology and Evolution, Volume 2. Cornell University Press, Ithaca, NY. Watson, J. 1997. The Golden Eagle. Princeton University Press, Princeton, NJ. Wayne, R.K., S.B. George, D. Gilbert, and P.W. Collins. 1991a. The Channel Island fox (Urocyon littoralis) as a model of genetic change in small population. Pages 639–649 in E.C. Dudley, ed., The Unity of Evolutionary Biology: Proceedings of the Fourth International Congress of Systematic and Evolutionary Biology. Dioscorides Press, Portland, OR. Wayne, R.K., S.B. George, D. Gilbert, et al. 1991b. A morphologic and genetic study of the island fox, Urocyon littoralis. Evolution 45:1849–1868. Wayne, R.K., E. Geffen, D.J. Girman, et al. 1997. Molecular systematics of the Canidae. Systematic Biology 46:622–653. Wayne, R.K., E. Geffen, and C. Vila. 2004. Population genetics: population and conservation genetics of canids. Pages 55–84 in D.W. MacDonald and C. Sillero-Zubiri, eds., The Biology and Conservation of Wild Canids. Oxford University Press, Oxford. Wenner, A.M., and D.L. Johnson. 1980. Land vertebrates in the California Channel Islands: sweepstakes or bridges? Pages 497–530 in D.M. Power, ed., The California Islands: Proceedings of a Multi-disciplinary Symposium. Santa Barbara Museum of Natural History, Santa Barbara, CA. White, G.C., D.R. Anderson, K.P. Burnham, and D.L. Otis. 1982. Capture– recapture and removal methods for sampling closed populations. Los Alamos National Laboratory Report LA-8787-NERP, Los Alamos, New Mexico. White, P.J., and K. Ralls. 1993. Reproduction and spacing patterns of kit foxes relative to changing prey availability. Journal of Wildlife Management 57:861– 867. White, P.J., C.A. Vanderbilt White, and K. Ralls. 1996. Functional and numerical responses of kit foxes to a short-term decline in mammalian prey. Journal of Mammalogy 77:370–376. White, P.J., W.H. Berry, J.J. Eliason, and M.T. Hanson. 2000. Catastrophic decrease in an isolated population of kit foxes. Southwestern Naturalist 45:204–211. Williams, E.S. 2001. Canine distemper. Pages 50–59 in E.S Williams and I.K. Barker, eds., Infectious Diseases of Wild Mammals, 3. Iowa State University Press, Ames, IA. Williams, E.S., E.T. Thorne, M.J.G. Appel, and D.W. Belitsky. 1988. Canine distemper in black-footed ferrets (Mustela nigripes) from Wyoming. Journal of Wildlife Diseases 24:385–398.

206 · References Wilmut, I., D.I. Sales, and C.J. Ashworth. 1986. Maternal and embryonic factors associated with prenatal losses in mammals. Journal of Reproduction and Fertility 76:851–864. Woodroffe, R., S. Cleaveland, O. Courtenay, K.M. Laurenson, and M. Artois. 2004. Infectious disease. Pages 123–142 in D.W MacDonald and C. Sillero-Zubiri, eds., Biology and Conservation of Wild Canids. Oxford University Press, Oxford. Woods, L.W. 2001. Adenoviral diseases. Pages 202–212 in E.S. Williams and I.K. Barker, eds., Infectious Diseases of Wild Mammals. Iowa State University Press, Ames, IA. Zabel, C.J. and S.J. Taggart. 1989. Shift in red fox, Vulpes vulpes, mating system associated with El Ni˜no in the Bering Sea. Animal Behaviour 38:830–838.

Index

Acinonyx jubatus, 19 activity patterns of island foxes, 41–42, 56 African lion, 19, 76, 131 African wild dog, 19, 76 aggression in captive island foxes, 96, 97, 121, 125 Ailurus fulgens, 77 Aldo Leopold, 183 Alectoris chukar, 36 Amazona vittata, 17 ambassador animals in environmental education, 149–151 amyloidosis, 138, 138, 141, 142 Anas platyrhyncos, 53 Anas strepera, 53 ancient murrelet, 13 Angiocaulus gubernaculatus, 135 ants, 136 Aphelocoma californica, 182 Aphelocoma insularis, 182 Arabian oryx, 17 Arctic fox, 30, 49, 115 Ashy storm petrel, 175 Asio flammeus, 176 Association of Zoos and Aquariums (AZA), 18, 147 Atriplex sp., 37 Avena, 41 AZA. See Association of Zoos and Aquariums Babesia, 133 Baird, Spencer F., 5 Bald and Golden Eagle Protection Act, 54, 63 bald eagle, 53, 70, 71 diet, 70 extirpation, 55 restoration, 3, 71

barn owl, 53, 176 bird response to island fox removal, 171–173 black rat, 35, 130, 173–174 black-footed ferret, 17, 77, 131, 164 bownet trap, 59 Buteo jamaicensis, 56 California condor, 17, 148 California Department of Fish and Game (CDFG), 64, 76, 78, 156 California ground squirrel, 35 Callipela californica, 53 Canid Specialist Group, 64 canine adenovirus (CAV), 133 canine distemper virus (CDV), 30 as cause of decline on Santa Catalina Island, 73–80, 131 as mortality factor, 3, 21, 75–76 endemic strain, 12, 132 field vaccination of wild foxes for, 80 testing of a vaccine for, 77–79 canine parvovirus (CPV), 133 Canis latrans, 116 Canis lupus, 19 Canis lupus baileyi, 19, 83, 115 Canis rufus, 115 Canis simensis, 19, 131 captive breeding, 3, 81–99 aggression in captive island foxes, 96, 125 demographic modeling for, 92, 93 diet of captive foxes, 90–91 facility design and construction, 82–88 fostering of neglected captive pups, 97 genetic management of captive island foxes, 15–19, 93

208 · Index captive breeding (cont.) mastitis in captive island foxes, 96, 98, 126 methods, 92 on Santa Catalina Island, 86–90 on the northern Channel Islands, 81–86 reproductive success in captive island foxes, 93–95, 95, 99, 118, 125, 128 stress in captive island foxes, 96 Carlstead, Kathy, 96 Carpobrotus aequilaterus, 36 carrying capacities, 28 Cassin’s auklet, 13 cat. See feral cat Catalina Island Conservancy, 73, 76, 154, 164 CAV. See canine adenovirus CDFG. See California Department of Fish and Game CDV. See canine distemper virus Center for Biological Diversity, 158 ceruminous gland carcinoma, 139, 139, 141, 142 Cervus elephas, 54 Channel Islands, 2 chaparral, 40, 56 cheetah, 19 Chendytes lawi, 13 chukar, 36 Chumash culture, 9, 12 Circus cyaneus, 176 climate change, and island foxes, 182 coastal sage scrub, 56 coastal scrub, 40 Collins, Paul, 53 colonization of islands by foxes, 8–11 Coluber constrictor mormon, 35 Columbian mammoth, 8 common raven, 51, 53 competitors of island foxes, 174–177 feral cats, 177 island spotted skunks, 35, 174–175 raptors, 175–176 Coreopsis gigantea, 172 cormorant, 51, 53 cortisol, 122, 124, 124 Corvus corax, 51, 53 CPV. See canine parvovirus crickets, Jerusalem, 36 Crotalus viridis, 13

DDE. See dichloro-diphenyldichloroethylene (DDE) DDT. See dichloro-diphenyl-trichloroethane (DDT) deer mouse, 7, 34, 130, 170, 172 in island fox diet, 34–35, 36–37 response to removal of island foxes, 169–171 demographic modeling, 32, 111, 162 and development of recovery criteria, 161 for captive breeding, 92 den sites of island foxes, 41 density of island fox populations, 28, 29–30 of other fox species, 29 density dependence in island fox populations, 31, 31–32 desert tortoise, 148 dho-gaza trap, 61 dichloro-diphenyldichloro-ethylene (DDE), 55, 69 dichloro-diphenyl-trichloroethane (DDT), 55, 69 diet of bald eagles, 70 of captive island foxes, 90–91 of golden eagles, 51, 52–53, 54–55, 59–60 of wild island foxes, 34–37 Dirofilaria immitis, 136 disease, 129–143 amyloidosis, 138, 138, 142 and genetic diversity, 140–141 Babesia, 133 cancer. See ceruminous gland carcinoma in island populations, 130–131 Leptospira interrogans, 134 mastitis, 96, 98, 126, 142 otitis, 141 septicemia, 142 thyroid atrophy, 140 viral, 131–134, see also canine adenovirus, canine distemper virus, canine parvovirus, rabies dispersal in island foxes, 39 in other fox species, 38 diurnal behavior in island foxes, 41–42 dog, domestic, 130 and native Americans, 12

Index as mortality source, 26 as vector of disease, 73, 76 Dow’s puffin, 13 dwarfism, 7–8 ear mite, 137, 137 ecological communities, 167–169 response to removal of island foxes, 169–177 El Ni˜no/Southern Oscillations (ENSO), 9, 30 elk, 54 Endangered Species Act (ESA), 158, 163–166 listing process, 158–159 permitting recovery actions, 159–160 recovery team and recovery plan, 160–163 ENSO. See El Ni˜no/Southern Oscillation environmental education, and island fox conservation, 152 Eremophila alpestris, 35 ESA. See Endangered Species Act estradiol, 121 estrus, 116, 121 Ethiopian wolf, 19, 131 evolution of island foxes, 8–12 extinction vortex, 16 feeding behavior of island foxes, 37–38 Felis catus, 177 Felis concolor, 19 Felis leo, 19, 76 feral cat, 130, 177 feral pig in golden eagle diet, 51, 53, 54, 59, 68 removal of, 3, 66–68 Finnegan, ambassador animal, 151, 151 fleas, 137 flightless sea duck, 13 Florida panther, 16 fossil evidence of island foxes, 8, 9 fostering of neglected captive pups, 97 founder effect, 14, 16 founders number of in island fox captive breeding program, 16, 18, 19 in other captive breeding programs, 17

· 209

Fratercula dowi, 13 Friends of the Island Fox (FIF), 145 gadwall, 53 garden snails, 36 gene diversity retention for captive island foxes, 18 genetics bottlenecks, 14, 15 founder effect, 14, 16 gene diversity retention, 18 genetic divesity and disease, 140–141 genetic management of captive island foxes, 15–19, 93 genetic status of island foxes, 13–15, 19 genetic variation, 15, 17 in captive island foxes, 17–18 measurement of, 14 heterozygosity, 17, 19 inbreeding, 16, 19 inbreeding depression, 16 major histocompatibility complex, 15 relatedness values for captive island foxes, 17, 18 status of San Nicolas Island fox, 13 giant coreopsis, 172 giant island deer mouse, 8, 13 giant vole, 8, 13 gigantism, 8 giving up density (GUD), 171 golden eagle, 60 breeding, 51–52 colonization of islands, 50–56 diet, 51, 52–53, 54–55, 59–60 predation by, 2, 21, 43–57 removal of, 3, 58–66 Goodall, Jane, 145 Gopherus agassizii, 148 Graber, David, 63 grassland, 41, 57 gray fox, 5, 6, 7, 9, 25, 57, 115, 120 gray wolf, 19, 164 grazing impacts on habitat, 29, 56–57 ground squirrel, California, 35 GUD. See giving up density gull, 51, 53 Gymnogyps californianus, 17, 148 habitat use by island foxes, 40–41 Haliaeetus leucocephalus, 53 heartworm, 136

210 · Index Helix aspersa, 36 Heteromeles arbutifolia, 36 Holocene epoch, 8, 10 home range size in island foxes, 39–40 hookworm, 135 horned lark, 35 house mouse, 35 hyperpredation, 51–52, 53 IFCWG. See Island Fox Conservation Working Group inbreeding, 16, 19 inbreeding depression, 16 induced ovulation in island foxes, 119 insects in island fox diet, 36 Institute for Wildlife Studies (IWS), 49, 61, 69, 73, 76, 77, 85, 86, 146, 158 intraguild predation (IGP), 175–176 island cherry, 38 Island Fox Conservation Working Group (IFCWG), 18, 66, 72, 81, 83, 104, 142, 156–157 incorporation into recovery team, 64 island night lizard, 177 island scrub-jay, 182 island spotted skunk, 35, 52, 54, 60, 130, 174–175 island syndrome, 7 IWS. See Institute for Wildlife Studies Ixodes pacificus, 137 Jerusalem crickets, 36 kit fox, 25, 38, 49, 57 Lanius ludovicianus mearnsi, 36, 155 Larus spp., 51, 53 Lasius niger, 136 Latta, Brian, 53 Leptospira interrogans, 134 lesser panda, 77 Lester, Herbert, 44 lice, 137 listing process, 158–159 litter size in island foxes, 24 loggerhead shrike. See San Clemente loggerhead shrike luteal phase, 116 Lycaon pictus, 19, 76

major histocompatibility complex (MHC), 15, 140 mallard, 53 mammoth Columbian, 8 pygmy, 8 Mammuthus columbi, 8 Mammuthus exilis, 8 mark–recapture methods, 27, 44 mastitis in captive island foxes, 96, 98, 126, 142 Melospiza melodia micronyx, 171 Mesocestoides, 136 Mexican wolf, 19, 83 MHC. See major histocompatibility complex Microtus miguelensis, 8 mink, 77 monitoring detection of population decline, 44 ecosystem, 178–179 for survival and mortality cause, 50 standardized population monitoring, 44 monogamy in canids, 117 Montrose Chemical Company, 69 Montrose Trustee Council, 69 Morris, Alexandra, 144 mortality and survivorship in island foxes, 25–27 mortality sources for island foxes, 24 mule deer, 68 in golden eagle diet, 53, 54, 68 removal, 69 Mus musculus, 35 Mustela lutreola, 77 Mustela nigripes, 17, 131 National Park Service (NPS), 46, 48, 50, 64, 65, 66–68, 81, 82, 85, 105, 108, 154, 164, 169 Nature Conservancy, The (TNC), 62, 64, 65, 66–68, 85, 105, 108, 154, 164, 182 Neotrichodectes mephitidis, 137 net-gunning, by helicopter, 62 northern harrier, 176 NPS. See National Park Service oak, 38 oak woodland, 56 Oceanodroma homochroa, 175

Index Odocoileus hemionus, 54 Opuntia sp., 36 orange-crowned warbler, 171, 172 Oryx leucoryx, 17 otitis, 141 Otodectes sp., 137 Panthera tigris altaica, 17 parasites, 134–138 Angiocaulus gubernaculatus, 135 Dirofilaria immitis, 136 Ixodes pacificus, 137 Mesocestoides, 136 Neotrichodectes mephitidis, 137 Otodectes sp., 137 Pulex irritans, 137 Spirocerca, 135 Toxascaris, 136 Uncinaria stenocephala, 135 passive integrated transponder (‘PIT’) tags, 27 permitting recovery activities under the Endangered Species Act, 159–160 Peromyscus maniculatus, 34 Peromyscus nesodytes, 8 Phalacrocorax spp., 51, 53 pine woodland, 56 plants in island fox diet, 36–37 Pleistocene epoch, 8, 10, 11 polar bear, 19 population size/status of island foxes, 2, 27–29, 46, 47, 75, 112 population viability management, 161 predation by golden eagles, 2, 21, 43–57 on released foxes, 107–109 by red-tailed hawks, 56 prickly-pear cactus, 36 Procyon lotor, 73, 132 progesterone, 119, 125 Prunus ilicifoia, 38 pseudopregnancy in canids, 116 Ptychoramphus aleuticus, 13 public advocacy, and island fox conservation, 144–145 Puerto Rican parrot, 17 Pulex irritans, 137 puma, 19 Puma concolor coryi, 16 pygmy mammoth, 8, 13

· 211

Quercus spp., 38 rabies, 133 raccoon, 73 as vector of disease, 80, 132 radiotelemetry to detect mortality cause, 47, 50 Rattus rattus, 35, 130 Recovery Coordination Group (RCG), 163 recovery team and recovery plan, 160–163 red fox, 49, 115 red-tailed hawk, 56 reintroduction of island foxes, 102–114 decision-making, 105, 177–178 methods, 102–103, 106–107 on Santa Catalina Island, 102–103 on the northern Channel Islands, 103–114 predation on released foxes, 107–109 relatedness values for captive island foxes, 17, 18 reproduction, 23–24, 115–128 reproductive cycle of island foxes and other canids, 116–117 reproductive failure, likely causes of, 120–128 reproductive success in captive gray wolves, 122 in captive island foxes, 95, 93–95, 99, 118, 117–128 riparian areas, 40 Roemer, Gary, 21, 47–48, 92 saltbush, 37 San Clemente island foxes, and San Clemente loggerhead shrikes, 146, 155–156 San Clemente loggerhead shrike, 36, 38, 146 San Miguel Island song sparrow, 171–173, 172, 173 San Nicolas island fox genetics, 15 sand dunes, 40 Santa Barbara Zoo, 90, 145, 146–151 Santa Catalina Island, 74 Santa Clara River, 9 Santa Cruz Predatory Bird Research Group (SCPBRG), 58, 61 Santarosae, 10 and island fox evolution, 11–12 separation of, 10

212 · Index Save Our Species (SOS), 145 Sceloporus occidentalis, 136 SCPBRG. See Santa Cruz Predatory Bird Research Group sea fig, 36 septicemia, 142 short-eared owl, 176 shrike. See San Clemente loggerhead shrike Siberian tiger, 17 skunk, island spotted, 35, 52, 60, 130, 174–175 Smithsonian National Zoological Park, 78, 90 snails, garden, in island fox diet, 36 snow leopard, 17 social structure of island foxes, 22–23 southern Pacific rattlesnake, 13 SOS. See Save Our Species sperm quality in island foxes, 120 Spermophilus beechyi, 35 Spilogale gracilis amphiala, 35, 52 Spirocerca, 135, 135, 141 Stenopelmatus fuscus, 36 stress in captive island foxes, 95–96, 122 Sturnella neglecta, 35, 53 survival rates in island foxes, 25, 65 in other canids, 25 Sus scrofa, 51 swift fox, 25, 38, 57 Synthliboramphus antiquus, 13 Tachi, ambassador animal, 149–151, 150 Tapinoma sessile, 136 thyroid atrophy, 140 thyroid hormone levels in island foxes, 127 ticks, 137 TNC. See Nature Conservancy, The (TNC) Tongva culture, 9, 12 Toxascaris, 136 Toxoplasma gondii, 133 toyon, 36, 38 translocation of island foxes on Santa Catalina Island, 100–101 Trowbridge, William P., 5 Tyto alba, 53, 176

ultrasound examination for pregnancy, 123, 125 Uncia uncia, 17 Uncinaria stenocephala, 135 Urocyon cinereoargenteus, 5, 115 Ursus maritimus, 19 US Fish and Wildlife Service (USFWS), 64, 109, 157 and development of recovery team and recovery plan, 64, 160–163 and listing of island fox, 158–159 and permitting under the Endangered Species Act, 159–160 vaccination field vaccination of wild foxes, 80, 183 of island foxes, 3 testing of vaccine, 77–79 vehicular trauma, as mortality source, 26, 141 Ventura River, 9 Vermivora celata sordida, 171 veterinary care for captive island foxes, 87 viral diseases, 131–134 Vulpes macrotis, 25 Vulpes velox, 25 Vulpes vulpes, 115 Walton, Brian, 58 West Nile Virus (WNV), 182 western fence lizard, 136 western meadowlark, 35, 53 western scrub jay, 182 western yellowbellied racer, 35 WNV. See West Nile Virus woodlands, 40 Xantusia riversiana, 177 zoos and development of island fox husbandry methods, 90–91, 147 and island fox conservation, 146–151 Association of Zoos and Aquariums, 18, 147 Santa Barbara Zoo, 90, 145, 146–151