ENDANGERED SPECIES RESEARCHEndang Species Res
Vol. 4: 165–185, 2008doi: 10.3354/esr00063
Printed January 2008Published online January 15, 2007
© Inter-Research 2008 · www.int-res.com*Email: [email protected]
THEME SECTION: REVIEW
Biotelemetry and biologging in endangered species research and animal conservation: relevance
to regional, national, and IUCN Red List threat assessments
Steven J. Cooke*
Fish Ecology and Conservation Physiology Laboratory, Department of Biology and Institute of Environmental Science, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario K1S 5B6, Canada
ABSTRACT: The current biodiversity crisis is characterized by the decline and extinction of numer-ous animal populations and species world-wide. To aid in understanding the threats and causes ofpopulation decline and the assessment of endangerment status of a species, conservation scientistsand practitioners are increasingly relying on remote assessments using biotelemetry (radio telemetry,acoustic telemetry, satellite tracking) and biologging (archival loggers) or hybrid technologies (e.g.pop-up satellite tags). These tools offer increasingly sophisticated means (e.g. large-scale telemetryarrays, fine-scale positioning, and use of physiological and environmental sensors) of evaluating thebehaviour, spatial ecology, energetics, and physiology of free-living animals in their natural environ-ment. Regional, national, and international threat assessments (e.g. the International Union for theConservation of Nature [IUCN] Red List) require basic knowledge of animal distribution, emigration,behaviour, reproductive potential, mortality rates, and habitat use, which in many cases can all beobtained through biotelemetry and biologging studies. Such studies are particularly useful for under-standing the basic biology of animals living in harsh environments (e.g. polar regions, aquatic envi-ronments), for rapidly moving or cryptic animals, and for those that undertake large-scale move-ments/migrations (e.g. birds, insects, marine mammals and fish). The premise of this paper is thatbiotelemetry and biologging have much to offer and should be embraced by the conservation sciencecommunity to aid in assessment of threats and endangerment status. It is crucial that studies onendangered species must not further contribute to species decline or retard recovery. As such, thereare complicated ethical and legal considerations that must be considered prior to implementingtracking studies on endangered wildlife. Furthermore, as many endangered animal species occur indeveloping countries, there is a need to develop capacity (financial support for the research and tech-nical telemetry skills) for designing and conducting tracking studies. To stem the loss of biodiversityand aid in the recovery of endangered animal populations, there is a need for innovative and inter-disciplinary research, monitoring programs and research initiatives to inform decision makers. It isclear that biotelemetry and biologging are not a panacea; however, they are valuable tools availableto conservation practitioners. Used appropriately, biotelemetry and biologging have the potential toprovide data that is often unattainable using other techniques, and can reduce uncertainty in theassignment of conservation status.
KEY WORDS: Biotelemetry · Biologging · Tracking · Conservation · Methods · Endangered species
Resale or republication not permitted without written consent of the publisher
OPENPEN ACCESSCCESS
Endang Species Res 4: 165–185, 2008
INTRODUCTION
The first step to initiating conservation actions forendangered1 organisms is to identify the populationsor species that are in decline (deterministic processes)or are faced with risk of extinction because they aresmall (stochastic processes; Caughley 1994, Brook etal. 2006). Key to this process is the use of objective,quantifiable, and consistent criteria to assess the statusof a species. Included in this analysis is the identifica-tion of threats which are used to inform conservationactions if required. Globally, the Species SurvivalCommission (SSC) of the IUCN World ConservationUnion (IUCN; www.iucn.org) produces the IUCN RedList of threatened species (i.e. the Red List). The RedList classifies globally endangered plant and animaltaxa and is regarded as the most comprehensive andauthoritative list of its kind (Lamoreux et al. 2003,Rodrigues et al. 2006). IUCN has developed a clear andstandardized framework for the assessment of speciesstatus which increasingly relies on rigorous scientificinput (rather than subjective expert opinion) and hasbecome more recognized by the scientific communityas a valuable and necessary tool in biodiversity conser-vation and research (Rodrigues et al. 2006). Nonethe-less, decisions are often made in the face of uncertaintybecause for many species we do not have a completeunderstanding of their natural history, let alone theirdemography (Akçakaya et al. 2000).
A candidate species (or group of species) is evalu-ated relative to a number of criteria which are thenused by the IUCN and their expert panels to assess theneed for designation within formal categories, includ-ing threatened, endangered, critically endangered,and extinct (Mace 1994). Formal thresholds based onpopulation size, population dynamics, geographicrange, connectivity, etc. are used for categorization.Once an animal has been classified as ‘endangered’,recovery plans can be developed and conservationactions implemented (Mace 1995, Collar 1996). Forinstances in which there is insufficient information toassess the status, the phrase ‘data deficient’ is used.Similar assessments also occur at a local, regional, andnational scale, although many rely at least in part onthe IUCN criteria (Gardenfors et al. 2001, Miller et al.
2007). In recent years, the Red List is increasinglybeing used not only as a system for assigning endan-germent status, but also as a means of aiding conserva-tion science, although the utility of this for some groupsis limited (Hayward et al. 2007a). Indeed, Butchart etal. (2005) suggested that Red List indices could be usedto evaluate progress towards meeting biodiversity tar-gets. For the Red List and other related assessments tobe useful in conservation, data used to evaluate andassign endangerment status must be rooted in sound,robust science.
Scientific data that form the basis of threat identifica-tion and endangerment assessments typically comefrom field studies of natural history and populationbiology. The study of animal ecology and demograph-ics is challenging, as many species tend to avoidhuman observers and travel great distances, often inenvironments that present numerous challenges tohumans. As a result, population estimates generatedfor wildlife populations are notoriously fraught withbias and error, which brings uncertainty to threatassessments and the management (see Williams et al.2002). However, improvements in statistical tech-niques and, more critically, innovations in technology,have enabled scientists to generate robust populationestimates and to understand the extent to which differ-ent populations interact (which is linked to the declin-ing population paradigm). In particular, methods suchas biotelemetry and biologging (defined below; hereinbiotelemetry is simply called ‘telemetry’ and biolog-ging ‘logging’) are increasingly being applied to thestudy of animal ecology in the wild because they canprovide detailed information on the fundamental biol-ogy of animals, including assessments of behaviour,survivorship, spatial ecology (i.e. the distribution ofanimals in space and time), energetics, and physiologythat is often unattainable using other techniques(Cooke et al. 2004, Block 2005, Ropert-Coudert & Wil-son 2005, Hooker et al. 2007). Telemetry and loggingare also being used to address more applied questionsassociated with wildlife medicine (Karesh 1999) andwildlife management (Millspaugh & Marzluff 2001).However, only in the last decade or so have these toolsbeen regarded as having utility in studies specificallyrelated to animal conservation.
166
1For this review I have elected to use the word ‘endangered’ when referring to the status of animal populations/species that areregarded as ‘threatened’, ‘endangered’, ‘special concern’, ‘vulnerable’, etc. This term is more generic than IUCN terminology(i.e. my definition includes all categories and terms including those used by regional and national bodies that do not follow theterminology used by the IUCN), but this follows the same approach as the journal ‘Endangered Species Research’ in that theword ‘endangered’ is used in the broadest possible sense and includes organisms of conservation concern. I also adopt theterminology of different authors. For example, if a species is described as ‘critically endangered’, I adopt their terminology whensummarizing their findings. On occasion, I also use the words ‘endangerment status’ or ‘endangerment assessments’ when refer-ring to evaluations of whether or not an animal is endangered (or some other specific status) according to some regional, nationalor international body
Cooke: Biotelemetry and biologging in endangered species research
Although there have been several recent syntheseson telemetry (Cooke et al. 2004) and logging (Block2005, Ropert-Coudert & Wilson 2005), none of thesereviews have explicitly considered or focused on howthese tools could be used to aid in understanding theendangerment status of species and the threats thathave lead to population declines. Hence, the purposeof this paper is to describe and evaluate the extent towhich telemetry and logging could be embraced bythe conservation science community. The focus isintentionally broad, encompassing all animal taxa andhabitats including vertebrates and invertebrates (notethat there are reasonably few studies of invertebratesthat use telemetry, so their coverage is minimal). How-ever, an assessment of the use of telemetry and loggingto identify or assess conservation actions is beyond thescope of this paper. Because of the nearly universalacceptance of the Red List species assessment criteria(Miller et al. 2007), the analysis uses their assessmentcriteria and threat categories as a foundation for theanalysis, based on the assumption that this informationcould easily be adapted to local, regional, or nationalassessments. Considering the current biodiversity cri-sis (Mace 1995) and the extent to which various animaltaxa are in decline or considered endangered, there isa clear need for novel approaches to aid conservationpractitioners in making decisions that will promotethe development of effective conservation strategies(Salafsky et al. 2002) focused on the species that aremost at risk of extinction (Miller et al. 2006).
TOOLS
Overview
Prior to addressing opportunities and applications oftelemetry and logging technology, it is first necessaryto understand the range of tools available and theircapabilities. The characteristics of telemetry and log-ging are similar; both involve the remote monitoringof some behavioural, physiological, or environmentalinformation. However, there are fundamentally differ-ent means of collecting the information. For telemetry,a signal emanating from a device carried by the animal(transmitter) sends the information to a receiver. Attimes, the power for transmission can be derived froman external energy source (e.g. Passive IntegratedTransponders, PIT tags; see Gibbons & Andrews 2004;however, this is not within the scope of this paper). Themost common means of signal transmission is radio fre-quencies. Some devices emit radio signals that can bedetected by satellites (designated satellite telemetryherein), whereby the devices periodically uplink (i.e.send a signal) to orbiting satellites; the location of the
transmitter is then estimated using mathematicalequations (Fancy et al. 1988, Harris et al. 1990). How-ever, in some environments, radio signals are not wellpropagated (e.g. marine environments or deep fresh-waters). Acoustic (or ultrasonic) telemetry is thusfavoured for some aquatic systems but does not workin a terrestrial setting. For logging, the information isrecorded and stored in an animal-borne device(archival logger) and information is downloaded when,and if, the logger is retrieved (Boyd et al. 2004). Todate, logger technologies have primarily been used inaquatic/marine environments (e.g. Block 2005, Hookeret al. 2007) but are becoming increasingly common interrestrial (including avian) applications. There are anincreasing number of techniques that couple these 2technologies, first logging information on board and,when possible, then transmitting the information, usu-ally to a satellite (e.g. pop-up satellite tags which areused with aquatic animals and only transmit to satel-lites when they fall off the animal and float to the sur-face; see Block 2005) or to a receiving station by meansof communicating histogram acoustic transponder(CHAT) tags; see Voegeli et al. 2001). Increasingly,global positioning system (GPS) techniques are beingused where an animal-borne device determines andlogs (or transmits) its geographic position (often within5 m; Rodgers et al. 1996). In this case, the satellitesfunction as a transmitter and the telemetry deviceserves as the receiver. The telemetry device then cal-culates the location of the animal based on the timerequired for the signal to be received, and the data arestored for remote downloading (e.g. via satellite) orsubsequent retrieval.
When using telemetry devices, it is possible to usemanual tracking, deploy fixed telemetry arrays, orcombine the 2 approaches. Increasingly, telemetryarrays have been deployed, enabling far more contin-uous monitoring of animals (Heupel et al. 2006), some-times in 3 dimensions. Some arrays are localized andprovide precise information on animal positions, some-times encompassing multiple species (i.e. communitylevel tracking of predators and prey). For example, theWarner Lake ecological observatory in eastern Ontariouses a whole lake 3D telemetry system to position fishand turtles continuously in 3 dimensions throughoutthe year (including under ice; Cooke et al. 2005). Thereis a comparable terrestrial array in the SmithsonianResearch Project in Panama where massive antennasare used to automatically position animals such asocelots and song birds (Wikelski et al. 2007; seewww.princeton.edu/%7Ewikelski/research/index.htm).Large-scale arrays have been deployed in oceans, andthis network continues to grow (as the Ocean TrackingNetwork; see www.oceantrackingnetwork.org/index.html). Although these systems have not been imple-
167
Endang Species Res 4: 165–185, 2008
mented to specifically study endangered species, thistechnology can be applied to systems that contain ani-mals of conservation concern, to gain better insightinto the species’ fundamental biology, ecology andenvironmental interactions, all of which are relevant tothe assessment of endangerment status. Currently, thePacific Ocean Shelf Tracking (POST) project hasacoustic telemetry arrays deployed from the BeringSea to northern California, as well as into severalinland rivers (e.g. the Fraser River and Columbia River;Welch et al. 2002; see www.postcoml.org/). The POSTproject has already documented unexpected move-ments of endangered white sturgeon Acipenser trans-mountanus tagged in California into the coastal watersof British Columbia, thus expanding its known range(Welch et al. 2006). In addition to the deployment offixed arrays, there is also an increased emphasis onmulti-species and multi-trophic level coordinated tag-ging studies (e.g. the Tagging of Pacific Pelagics[TOPP] project covers sea birds, marine mammals,pelagic fish, and turtles; Block et al. 2002). Such stud-ies provide the opportunity to identify hotspots i.e. notthe classical definition of areas with high levels ofendemism (Myers et al. 2000), but those where multi-ple animals congregate (see Hooker et al. 2007). Con-tinuous tracking studies in marine environments formonitoring highly mobile species or migratory birdshave been replaced with satellite and/or logging tech-nologies that enable the scientists to assess movementsat the scale of entire oceans or continents (e.g. Shafferet al. 2006), and for studies that focus on smaller, oftendiscrete geographic areas, fixed-station telemetryreceiver arrays can be implemented.
Beyond simply providing information on the posi-tion of an animal, logging and tracking devices arebecoming increasingly sophisticated and can now beused for the monitoring of biological and environmen-tal variables which can help to define critical habitats(i.e. those specific areas within the geographical areaoccupied by the species on which are found thosephysical or biological features essential to the con-servation of the species). Indeed, logging techniquesall rely on the sensing and logging/recording of in-formation derived from one or more sensor (Boyd etal. 2004). Sensors provide detailed information onhabitat use and environmental relations, such as tem-perature, depth, light levels, salinity, and dissolvedoxygen. In recent years, loggers and telemetrydevices have also been equipped with imagery sen-sors so that scientists can obtain photographs or videofootage from the animals’ vantage point (Marshall1998). Furthermore, some devices focus on the behav-iour (e.g. monitoring sounds to assess chewing/feed-ing activity), energetics (e.g. measuring fin or wingbeats), and physiology (e.g. heart rate or the chem-
istry of body fluids) of the organism being studied.Cooke et al. (2004) provide a detailed overview ofsensor technology and its application to the field ofecology. To date, there have been reasonably fewapplications of sensor technology to understand ani-mal endangerment and conservation status. However,there are many options including the determination ofthe effects of environmental change of phenology(e.g. Cooke et al. 2006) and the thermal ecology ofsensitive species (Parmesan 2006).
As noted above, not all telemetry techniques aresuitable for all environments or species. However,there is a telemetry or logging tool that should work inalmost any environment or on any species (except forsize limitations with smaller animals). Minituarizationof these devices is an ongoing process such that thenumber of species suitable for such research is steadilyincreasing. For each primary habitat type, the opportu-nities and challenges of telemetry were summarizedusing the IUCN authority file on habitat types (IUCNHabitats Authority File Version 2.1; available at:www.iucn.org/themes/ssc/sis/authority.htm) (Table 1).As evident from Table 1, one must have a thoroughunderstanding of the technological choices availableas well as their strengths and weaknesses priorto selecting a technology for a given environment.Similarly, having a basic understanding of the naturalhistory of a given species will also help to ensure thatthe appropriate technology is considered. AlthoughTable 1 provides an overview, it is important for newtelemetry users to consult telemetry engineers andother telemetry practitioners when designing a studythat relies on this technology.
Desirable characteristics
One of the most desirable characteristics of telemetryand logging for the study of endangered species is thatone can study free-living animals in their natural envi-ronment. This is particularly relevant to endangeredspecies where removal of the animals to captivitywould typically only be done as a conservation mea-sure (e.g. to establish a captive breeding species). Innature, animals face a suite of site-specific biotic (e.g.predation) and abiotic (e.g. weather, habitat hetero-geneity) conditions that cannot be adequately repli-cated in captivity and that need to be characterizedand understood in an effort to understand the popula-tion ecology of an endangered animal. The monitoringof unrestrained free-ranging animals in their own envi-ronment eliminates laboratory artifacts but also elimi-nates the need to remove animals with reproductivepotential from an endangered population. Dependingon the technology used, these tools also provide the
168
Cooke: Biotelemetry and biologging in endangered species research 169
Habitat type
(1) Forest
(2) Savanna
(3) Shrubland
(4) Grassland
(5) Wetlands (inland lakesand rivers, terrestrial andaquatic animals)
(6) Rocky areas
(7) Caves and subterraneanhabitats
(8) Desert
(9) Sea
(10) Coastline
(11) Artificial terrestrial
(12) Artificial aquatic
(13) Introduced vegetation
Technology
Radio telemetry (arrays andmanual tracking; e.g. Vilella& Hengstenberg 2006),satellite
Radio telemetry (arrays andmanual tracking; e.g.Loveridge et al. 2007),satellite
As (2), e.g. Haupt et al. (2006)
As (2), e.g. Haupt et al. (2006)
Radio (arrays and manualtracking; Zurstadt & Stephan2004) and acoustic (arraysand manual tracking; Cookeet al. 2005); satellite forterrestrial organisms(McCulloch et al. 2003)
Radio telemetry (arrays andmanual tracking; Dickson &Beier 2007), satellite(McCarthy et al. 2005)
Radio telemetry (arrays andmanual; Clark et al. 1993)
As (2)
Acoustic telemetry (arraysand manual; Zeller 1997),satellite tags (conventional;Call et al. 2007; pop-up andother loggers; Block et al.2001), radio (manualcombined with othertechniques; Ries et al. 1998,Deutsch et al. 1998)
As (9), Starr et al. (2005)
As (2), Gehrt (2005)
As (5), Szedlmayer &Schroepfer (2005)
As (1) or (2) depending ondensity of foliage
Benefits and opportunities
Trees can be used forestablishing large-scale radiotelemetry arrays that reachfrom the ground to thecanopy; antennas can bedeployed at strategic points(e.g. wildlife trails, nestboxes)
Open environments areconducive to satellitetracking; radio telemetryeffective in open environ-ments (manual plane, truck,foot, or arrays)
As (2)
As (2)
Manual tracking can be doneon foot, by boat, canoe, orplane
Manual tracking can be doneon foot, by helicopter, andplane
Radio telemetry arrays couldbe deployed at entrances andchoke points
As (2); watering holes mayserve as an appropriate sitefor arrays
Acoustic telemetry extremelyeffective in marine environ-ments; arrays frequently usedto track movements; radiotracking can be used to locateanimals (e.g. marine mam-mals) during periods wherethey are on the surface orshore
As (9), acoustic telemetryarrays useful for monitoringcoastal movements
As (2)
As (5)
As (1) or (2)
Challenges
Dense foliage can impedeaerial radio tracking fromplanes and can interfere withsatellite tracking; manualtracking typically restricted tofoot or helicopter
Manual tracking likely haslimited utility
Dense shrub growth canmake manual trackingdifficult
As (2)
When tracking aquaticorganisms, acoustic telemetrymay not work well in shallowor noisy (river) environments;conversely, radio telemetrydoes not work well in deepwater
May be difficult to determineprecise locations; radio signalreflections from rock facescan make pin-pointinganimals difficult
Radio signal reflections; lackof room for radio telemetryantennas
As (2)
Difficult to deploy acousticarrays in deep environments;satellite transmissions limitedto periods when animals (ortransmitters; i.e. pop-up) areat the surface; loggers mustbe retrieved for downloading
Nearshore regions can benoisy, making acoustictelemetry challenging
Electrical interference fromhuman infrastructure canmake radio tracking difficult;use of some manual trackingmethods difficult; potentialfor vandalism to arrays
As (5)
As (1) or (2)
Table 1. Relevance of biotelemetry and biologging techniques to different habitat types. Habitat types based on the IUCN authority file, available at: www.iucn.org/themes/ssc/sis/authority.htm
Endang Species Res 4: 165–185, 2008
opportunity to focus on animal behaviour across a vari-ety of scales. For example, to identify the seasonal crit-ical habitats and geographic range of a species, tele-metry or logging could be used at a spatial (e.g. site,regional, continental) and temporal (e.g. hours, days,years) scale that coincides with the biology of theanimal.
Another benefit of telemetry and logging technologyis that they can produce continuous data streams(through use of arrays, loggers, or satellites) that elim-inate data gaps during periods when animals are notmonitored manually by research team members. Long-term and continuous records of behaviour facilitate thedetection of trends through time in terms of spatialecology and phenology. Indeed, data can be collectedday and night and in harsh environmental conditionsfor extended periods without requiring continuoushuman support. Such an approach is particularlyimportant for organisms that inhabit large ranges,exhibit rapid movement, or occupy habitats that aredifficult to study. These tools also enable a researcherto characterize the variation among individuals and torecognize the plasticity of the responses. Individualvariation in behaviour is increasingly being recognizedas important for the conservation of biodiversity, as thevariation can provide a better idea of the extent towhich animals will differ from a ‘mean’ response (e.g.how far will they range from their ‘mean’ home range).Telemetry is also an ideal tool for linking individualbehaviour with physiology and energy status (Wikelski& Cooke 2006), information that is fundamentalfor conservation. This integration can be achievedthrough the use of sensors (discussed in ‘Overview’above) or by obtaining non-lethal biopsies (e.g. bloodsamples). Energetic analyses are particularly useful inconservation, as energy is the common currency inecology and is essential for inferring the bioenergeticcosts of different behaviours or exposure to differentstressors.
Limitations and challenges
A detailed evaluation of the limitations and chal-lenges of telemetry and logging is presented in Cookeet al. (2004). Here, a similar framework is used tounderstand the specific limitations and challenges inthe context of endangered species research. One of theprimary limitations and challenges to the study of ani-mals, but in particular endangered species, is the needto minimize the burden of the device and associatedattachment/implantation on the animal (Wilson &McMahon 2006). The trade-off between battery sizeand longevity continues to limit research on smallorganisms or long-term monitoring and also limits the
complexity (mass) of the required circuitry. The typesof transmitter and sensor, and the biology of the organ-ism of interest will determine the type of attachmentand implantation procedure that is possible and appro-priate. Godfrey & Bryant (2003) reviewed the literatureon telemetry studies and determined that only 10.4%of 836 studies in the 1990s directly addressed the effectof (radio) tags on their bearers. Alarmingly, theyreported that conservation-oriented studies were lesslikely to assess effects compared with studies on non-endangered animals. The authors attributed this pat-tern to the use of tags for conservation studies thatwere better designed to avoid adverse effects, and to apublication bias whereby deleterious effects were sim-ply not reported. Clearly, there is a need for considera-tion of tagging effects as well as more transparentreporting of tagging effects on all studies involvingendangered species. To this end, when dealing withvertebrates or any endangered animals, consultationwith veterinarians early in the study design is essential(see Hutchins et al. 1993). Furthermore, any surgicalimplantation will require specialized training that onlyveterinarians might have (Mulcahy 2003). However,such assistance may not be required for externalattachments (e.g. dart tags on marine pelagics, tape onbirds). Surgical and handling practice is needed andthis can be obtained by working on similar species(e.g. congenerics or confamilials) that are not endan-gered. Each species, environment, and technology hasspecific challenges that must be considered whendetermining which techniques are appropriate for agiven objective. For some taxa, there are publishedguidelines on the maximum size (mass) of device thatcan be deployed on an organism of a specific size (e.g.Phillips et al. 2003) or recommendations for the physi-cal placement of the device (e.g. Bannasch et al. 1994).
Another challenge is that many researchers (as wellas conservation practitioners) are unfamiliar with thetelemetric techniques that are now available to moni-tor the behavioural, physiological and microenviron-mental variables that would most effectively addresstheir research question. A further problem may be alack of commercial suppliers for much of the telemetryapparatus; the inability to purchase an existing deviceand obtain adequate technical support would impederesearchers from adopting these techniques. In devel-oping countries the capacity to implement a telemetrystudy may not exist. To this end, there is a need tobuild capacity within the local research communitiesand ensure effective transfer of technology (Marmulla& Bénech 1999). There have been some attempts todevelop training materials and hold telemetry work-shops in developing countries (e.g. Baras et al. 2002)although this is still not widespread. Information andtraining can also be obtained through attendance at
170
Cooke: Biotelemetry and biologging in endangered species research
relevant conferences or symposia and some confer-ences include specific workshops on technology andtagging techniques that are often part of continuingeducation and professional certification initiativesthrough scientific societies (e.g. American FisheriesSociety).
Finally, there is a perception that the cost of teleme-try is rather high. Indeed, there can be some initialcapital expenses, and devices themselves can beexpensive (in some cases leading to low sample sizes).However, this must be contrasted with the benefitderived from data that cannot be collected using othertechniques and should not be the sole reason for notconducting a sorely needed telemetry study on anendangered species. It is important to include systemmaintenance as well as data processing and archivingservices (large volumes of data can be generated) inany budgets for a telemetry study. Franco et al. (2007)conducted one of the first assessments of the cost effec-tiveness of different techniques for assessing habitatselection of lesser kestrels Falco naumanni with a focuson radio telemetry and transect surveys. The authorsreported that, in general, both techniques producedsimilar findings, with the telemetry technique beingmore sensitive to detecting differences in habitatselection. When the authors expressed the costs, radiotelemetry data (€ 312 per statistically significant differ-ence) was more costly to obtain than transect data(€ 82.5). The authors concede that their results arequite specific to their study species and the focus of thestudy. As such, it is not possible to directly extrapolatetheir findings to other studies. The authors also empha-sized that comparisons between techniques are rarelysimple because the selection of the best method willdepend on the time, budget and equipment, andhuman resources available.
Ethical and legal perspectives
The paradox is that field research activities that usetelemetry have generated valuable information whichinforms conservation efforts, yet there are also poten-tial negative impacts on individuals (i.e. welfare) andpopulations (Cooper & Carling 1995). Governmentagencies around the world as well as peer reviewedpublication outlets (e.g. Animal Behaviour, Conserva-tion Biology) are increasingly asking for researchers toaccount for their potential impacts. Although welfareissues were once restricted to ‘higher vertebrates’ (i.e.birds, mammals), there is now recognition that welfareof all vertebrates such as fish, reptiles, and amphib-ians, must be considered. In some jurisdictions, permitsare required for the scientific collection of wildlife andstudy of endangered wildlife even if the animals are to
be released with a telemetry transmitter (e.g. Peck &Simmonds 1995). Failure to obtain permits is not onlywrong on ethical grounds, but could lead to prosecu-tion and a halt even to well intended research. At theleast, all relevant levels of government (from local tonational) as well as the Convention on the Trade ofEndangered Species (CITES) should be consulted ifany tissue samples associated with tagging (e.g. bloodsample, feather, scale) are kept and transported acrossinternational borders.
The ethical considerations of tagging endangeredanimals is a complex issue, as one of the assumptionsof telemetry is that the tagging and presence of thedevice do not deleteriously affect the individual (Wil-son & McMahon 2006). However, sample sizes arerelatively low (relative to other methods) and animalscan be studied in their natural environment. Severalexplanations have been proposed to account for a per-ceived lack of public ethical discourse among fieldscientists (reviewed in Farnsworth & Rosovsky 1993).Of particular relevance to telemetry studies is theassumption that the relative benefits of the researchtechnique outweigh potential short-term costs to thestudy organism or population (i.e. increased knowl-edge may inform and promote its long-term conserva-tion; Farnsworth & Rosovsky 1993). Institutional ani-mal care committees usually require researchers toconsider the impacts of their tagging activities onpopulations, and this is coupled with the developmentand testing of tagging techniques. There has been anexplosion of studies that compare and contrast differ-ent tagging techniques with the purpose of trying toidentify techniques that minimize the impact on theanimal. Indeed, data derived from telemetry and log-ging studies would not be useful if the observationsgenerated were not genuine. A number of authorshave proposed that ethical considerations must beconsidered when conducting research on all animalspecies (particularly those that are endangered) andwhen developing conservation measures (Farnsworth& Rosovsky 1993, Putman 1995, Wilson & McMahon2006). In many cases, the burden still lies on thetelemetry practitioner (Minteer & Collins 2005), as notall countries (or institutions) regulate or require ethi-cal approval to conduct research on wild animals(Peck & Simmonds 1995). In such cases, it would beworthwhile to obtain external peer review from ex-perts in the field (including a veterinarian) prior toembarking on research on endangered species. Typi-cally, if animal care approvals are needed by a re-searcher’s home country/jurisdiction, the permit mustbe obtained there, even if the research is to be con-ducted elsewhere. In some cases, this means obtain-ing approvals from 2 jurisdictions (home institutionand study site).
171
Endang Species Res 4: 165–185, 2008
Understanding IUCN threat categories
Papers located from the ISI Web of Science(see http://scientific.thomson.com/products/wos/), withsearch terms including ‘conserv* or endanger* orimperil* or threatened or IUCN’ and ‘tracking orbiotelemetry or telemetry or archival or loggers orsatellite or biologging’, were qualitatively examinedto identify how telemetry and logging can be used toassess conservation status. The IUCN uses a hierar-chical classification system on the causes of speciesdecline and requires assessors to indicate the threatsthat triggered a listing of the taxon concerned at thefinest level possible. The IUCN and the ConservationMeasures Partnership are currently updating andstandardizing the direct threats classification system.For the purpose of this paper, the forthcoming classi-fication system (see www.iucn.org/themes/ssc/redlists/classification.htm) was used as a framework for list-ing all of the possible threats to species decline andmaking a critical assessment of the potential fortelemetry and logging to contribute to understandingthe respective threats and their consequences onendangered vertebrates.
This exercise revealed that every single threat listedcould be identified or better understood through theuse of telemetry or logging (Table 2). The most commonapplication would be to document the spatial ecology ofanimals relative to different threats. For example, in thecase of residential and commercial development pres-sures, telemetry could be used to understand the dis-placement of animals, their interactions with humansand human infrastructure, and associated altered habi-tats. Telemetry and logging can also be used to evalu-ate mortality-specific threats (e.g. bird strikes, bycatchof sea turtles, birds, marine mammals and fish). Knowl-edge of the specific threats and their impact on animalpopulations is essential for the development and imple-mentation of conservation actions, although this goesbeyond the scope of the present review.
IUCN RED LIST STATUS: OPPORTUNITIES FORBIOTELEMETRY AND BIOLOGGING
The IUCN uses a series of consistent and defensiblecriteria (with thresholds) to objectively assess endan-germent status. I contend that telemetry has the poten-tial to yield important information needed to assess theendangerment category into which a given specieswould be placed. For example, for a taxon to be classi-fied as critically endangered, endangered, or vulnera-ble, they must meet a number of well-defined criteriaas outlined by the IUCN (www.iucn.org/themes/ssc/redlists/RLcategories2000.html). Core to the classi-
fication of species by the IUCN is information on thepopulation trends and mortality rates of a given spe-cies (Lamoreux et al. 2003). It is also obvious that in thecase of a ‘data deficient’ categorization (not a threatcategory per se, as listing of taxa in this category indi-cates that more information is required), telemetrycould provide information such as described abovethat would be needed to make a more informed classi-fication. A taxon in this category may be well studied,and its natural history well understood, but appropri-ate data on abundance and/or distribution are lacking.Many assessments are conducted at a regional ornational level and the following discussion is equallyrelevant on a regional, national and internationallevel).
Mortality rates and causes
For some species, telemetry and logging provide theonly tools for reliably assessing mortality rates in wildpopulations. This is particularly true for wide-ranging(e.g. whale sharks Rhincodon typus; Gifford et al.2007) or cryptic species (e.g. flying-foxes Acerodonjubatus; Mildenstein et al. 2005) as well as those thatoccupy environments that are poorly accessible byhumans (e.g. loggerhead turtles Caretta caretta inmarine pelagic systems; Chaloupka et al. 2004, andsnow leopards Uncia uncia in steep and rugged terrainof Mongolia; McCarthy et al. 2005). There are a num-ber of studies that have been conducted to documentmortality rates and identify the causes of mortality forendangered species (or to be used to determine if aspecies is endangered). In fact, there are many speciesfor which demographic models are routinely used, yetthere is no information on life-stage specific mortalityrates except from studies recently conducted withtelemetry.
McIntyre et al. (2006) used satellite telemetry to esti-mate the probability of first-year survival for migratorygolden eagles Aquila chrysaetos raised in DenaliNational Park and Preserve, Alaska, USA. The authorsrevealed that first year survival was low, with 0.34survival probability for an 11 mo period for the 1997cohort and 0.19 for an 11 mo period for the 1999 cohort.The majority of mortalities focused on the autumn mi-gration and early winter period and were attributed tostarvation, electrocution, and poaching. It was foundthat low first-year survival may limit recruitment, andthe causal factors of mortality that could be addressedthrough conservation actions were identified. In Aus-tralia, Hayward et al. (2005) used radio telemetry todetermine the survival rates and causes of mortality fora threatened macropodid marsupial, the quokka Se-tonix brachyurus. Predation was determined as the
172
Cooke: Biotelemetry and biologging in endangered species research 173
Op
por
tun
itie
s an
d a
pp
lica
tion
s
Un
der
stan
din
g t
he
dis
pla
cem
ent
of a
nim
als
or t
hei
r sp
atia
l ec
olog
y in
urb
an e
nvi
ron
-m
ents
; eva
luat
ion
of
hu
man
–w
ild
life
in
tera
ctio
ns
incl
ud
ing
pot
enti
al f
or d
isea
setr
ansm
issi
on; e
valu
atio
n o
f b
ird
beh
avio
ur
in r
elat
ion
to
win
dow
s an
d b
uil
din
gs;
doc
um
enta
tion
of
surv
ival
rat
es a
nd
em
igra
tion
(in
clu
din
g d
isp
ersa
l) r
elat
ive
to d
evel
op-
men
t in
ten
sity
; eva
luat
ion
of
pat
ch d
ynam
ics
and
gap
cro
ssin
g i
n u
rban
sys
tem
s
As
(1.1
) w
ith
gre
ater
em
ph
asis
on
in
du
stri
al e
colo
gy
rath
er t
han
urb
an e
colo
gy
Un
der
stan
din
g t
he
dis
pla
cem
ent
of a
nim
als
or t
hei
r sp
atia
l ec
olog
y in
or
adja
cen
t to
recr
eati
onal
en
viro
nm
ents
su
ch a
s g
olf
cou
rses
or
ski
hil
ls; d
ocu
men
tati
on o
f su
rviv
alra
tes
and
em
igra
tion
rel
ativ
e to
ch
ang
es i
n l
and
scap
e; e
valu
atio
n o
f p
atch
dyn
amic
s an
dg
ap c
ross
ing
Det
erm
inat
ion
of
spat
ial
ecol
ogy
rela
tive
to
agri
cult
ura
l cr
ops
(in
clu
din
g m
onoc
ult
ure
)an
d a
ssoc
iate
d f
arm
in
fras
tru
ctu
re; u
nd
erst
and
ing
of
hu
man
–w
ild
life
in
tera
ctio
ns
tore
du
ce p
oten
tial
of
cull
ing
(e.
g. d
eer,
ele
ph
ants
). D
ocu
men
tati
on o
f su
rviv
al r
ates
rela
tive
to
dif
fere
nt
agri
cult
ura
l la
nd
scap
es
Eva
luat
ion
of
shif
ts i
n s
pec
ies
dis
trib
uti
on a
ssoc
iate
d w
ith
har
vest
; un
der
stan
din
g s
pat
ial
ecol
ogy
of a
nim
als
and
pre
dat
or-p
rey
dyn
amic
s in
mon
ocu
ltu
res;
exp
erim
enta
l ev
alu
a-ti
on o
f an
imal
su
rviv
al r
elat
ive
to p
olyc
ult
ure
sce
nar
ios
Un
der
stan
din
g d
irec
t in
tera
ctio
ns
wit
h w
ild
life
; eva
luat
ion
of
chan
ges
in
sp
atia
l ec
olog
yre
lati
ve t
o fe
nce
s an
d l
and
scap
e cl
eari
ng
Eva
luat
ion
of
spat
ial
ecol
ogy
rela
tive
to
alte
rati
on i
n h
abit
ats
asso
ciat
ed w
ith
pro
du
ctio
n(e
.g. f
lood
ing
of
man
gro
ves
for
shri
mp
pro
du
ctio
n; s
ea c
ages
); d
eter
min
atio
n o
f th
e fa
teof
rel
ease
d o
rgan
ism
s
Doc
um
enta
tion
of
the
resp
onse
s of
an
imal
s to
dri
llin
g a
ctiv
ity;
eva
luat
ion
of
foot
pri
nt
imp
acts
an
d p
oten
tial
sp
ills
at
sou
rce
on a
nim
al d
istr
ibu
tion
Un
der
stan
din
g h
abit
at s
hif
ts a
nd
sp
atia
l ec
olog
y re
lati
ve t
o ex
trac
tion
of
surf
ace-
bas
edor
aq
uat
ic/m
arin
e ag
gre
gat
e re
sou
rces
Un
der
stan
din
g a
nim
al i
nte
ract
ion
s w
ith
pow
er p
rod
uct
ion
in
fras
tru
ctu
re (
e.g
. bir
ds
and
win
dp
ower
; mar
ine
mam
mal
s an
d t
idal
pow
er; n
ote:
exc
lud
es h
ydro
pow
er);
doc
um
enta
-ti
on o
f th
e sp
atia
l ec
olog
y of
an
imal
s re
lati
ve t
o en
erg
y p
rod
uct
ion
an
d i
nfr
astr
uct
ure
Eva
luat
ion
of
pat
ch d
ynam
ics
and
gap
cro
ssin
g; d
eter
min
atio
n o
f an
imal
col
lisi
onfr
equ
ency
an
d a
ssoc
iate
d m
orta
lity
; det
erm
inat
ion
of
anim
al b
ehav
iou
r re
lati
ve t
o ro
ads
and
rai
lroa
ds
wit
h s
pec
ific
ref
eren
ce t
o cr
ossi
ng
Un
der
stan
din
g t
he
con
seq
uen
ces
of u
tili
ty a
nd
ser
vice
lin
es o
n s
pat
ial
ecol
ogy
ofan
imal
s
Un
der
stan
din
g t
he
imp
acts
of
dre
dg
ing
of
ben
thic
an
d p
elag
ic o
rgan
ism
s; u
nd
erst
and
-in
g s
pat
ial
ecol
ogy
and
beh
avio
ur
of a
nim
als
rela
tive
to
vess
el t
raff
ic (
e.g
. mar
ine
mam
mal
s); d
ocu
men
tati
on o
f m
arin
e an
imal
an
d v
esse
l co
llis
ion
rat
es; u
nd
erst
and
ing
wav
e-in
du
ced
dis
turb
ance
on
an
imal
dis
trib
uti
on, e
ner
get
ics,
an
d s
tres
s
Tab
le 2
. Th
reat
s to
an
imal
sp
ecie
s d
ecli
ne
and
op
por
tun
itie
s fo
r te
lem
etry
to
enh
ance
th
e u
nd
erst
and
ing
of
the
thre
at a
nd
its
im
pac
t on
th
e th
reat
ened
sp
ecie
s. T
hre
ats
wer
e id
enti
fied
fro
m t
he
IUC
N-C
onse
rvat
ion
Mea
sure
s P
rog
ram
(C
MP
)u
nif
ied
cla
ssif
icat
ion
of
dir
ect
thre
ats
doc
um
ent
(Ver
sion
1.0
; 200
6) a
nd
lis
ted
in
th
e sa
me
ord
erin
th
e ta
ble
. B
eyon
d i
den
tify
ing
op
por
tun
itie
s, t
he
per
ceiv
ed c
urr
ent
stat
us
of t
he
use
of
tele
met
ry a
s w
ell
as t
he
pot
enti
al o
pp
ortu
nit
y fo
r te
lem
etry
for
eac
h t
hre
at a
rera
nk
ed a
s ‘l
ow’,
‘m
oder
ate’
, or
‘h
igh
’ b
ased
on
au
thor
ass
essm
ent.
Not
e th
at t
he
IUC
N-C
MP
doc
um
ent
wil
l so
on r
epla
ce t
he
Th
reat
s A
uth
orit
y F
ile
(Ver
sion
2.1
) th
at i
scu
rren
tly
use
d b
y th
e IU
CN
(se
e w
ww
.iu
cnre
dli
st.o
rg/i
nfo
/maj
or_t
hre
ats
and
ww
w.i
ucn
.org
/th
emes
/ssc
/sis
/au
thor
ity.
htm
). P
erce
ived
cu
rren
t st
atu
s is
bas
ed o
n t
he
au-
thor
’s r
esea
rch
ass
ocia
ted
wit
h t
he
pre
sen
t st
ud
yan
d t
he
freq
uen
cy w
ith
wh
ich
bio
tele
met
ry o
r b
iolo
gg
ing
tech
niq
ues
wer
e u
sed
to
asse
ss a
sp
ecif
ic t
hre
at.
Pot
enti
alop
por
tun
ity
is b
ased
on
th
e au
thor
’s a
sses
smen
t of
th
e p
oten
tial
for
bio
tele
met
ry a
nd
bio
log
gin
g t
o b
e u
sed
for
th
e as
sess
men
t of
a g
iven
th
reat
Gen
eral
th
reat
(1)0
Res
iden
tial
an
dco
mm
erci
ald
evel
opm
ent
(2)0
Ag
ricu
ltu
re a
nd
aqu
acu
ltu
re
(3)0
En
erg
y p
rod
uct
ion
and
min
ing
(4)0
Tra
nsp
orta
tion
and
ser
vice
corr
idor
s
Sp
ecif
ic t
hre
at
(1.1
)0H
ousi
ng
an
du
rban
are
as
(1.2
)0C
omm
erci
al a
nd
ind
ust
rial
are
as
(1.3
)0T
ouri
sm a
nd
recr
eati
on a
reas
(2.1
)0A
nn
ual
an
dp
eren
nia
l n
on-
tim
ber
cro
ps
(2.2
)0W
ood
an
d p
ulp
pla
nta
tion
s
(2.3
)0L
ives
tock
farm
ing
an
dra
nch
ing
(2.4
)0M
arin
e an
dfr
esh
wat
eraq
uac
ult
ure
(3.1
)0O
il a
nd
gas
dri
llin
g
(3.2
)0M
inin
g a
nd
qu
arry
ing
(3.3
)0R
enew
able
ener
gy
(4.1
)0R
oad
s an
dra
ilro
ads
(4.2
)0U
tili
ty a
nd
serv
ice
lin
es
(4.3
)0S
hip
pin
g l
anes
Per
ceiv
edcu
rren
t st
atu
s
Mod
erat
e
Mod
erat
e
Low
Mod
erat
e
Low
Low
Mod
erat
e
Low
Low
Low
Hig
h
Low
Mod
erat
e
Pot
enti
alop
por
tun
ity
Hig
h
Hig
h
Hig
h
Hig
h
Mod
erat
e
Mod
erat
e
Mod
erat
e
Hig
h
Mod
erat
e
Hig
h
Hig
h
Mod
erat
e
Hig
h
Endang Species Res 4: 165–185, 2008174T
able
2 (
con
tin
ued
)
Gen
eral
th
reat
(5)0
Bio
log
ical
reso
urc
e u
se
(6)0
Hu
man
in
tru
sion
san
d d
istu
rban
ce
(7)0
Nat
ura
l sy
stem
mod
ific
atio
ns
(8)0
Inva
sive
an
d o
ther
pro
ble
mat
icsp
ecie
s an
d g
enes
(9)0
Pol
luti
on
Op
por
tun
itie
s an
d a
pp
lica
tion
s
Un
der
stan
din
g s
pat
ial
ecol
ogy
and
beh
avio
ur
of a
nim
als
(i.e
. bir
ds)
rel
ativ
e to
air
tra
ffic
;d
eter
min
atio
n o
f an
imal
col
lisi
on f
req
uen
cy a
nd
ass
ocia
ted
mor
tali
ty
Un
der
stan
din
g a
nim
al d
istr
ibu
tion
rel
ativ
e to
hu
nti
ng
an
d c
olle
ctin
g p
ract
ices
an
dsp
ecif
ic h
abit
ats
(in
clu
des
doc
um
enti
ng
hu
man
beh
avio
ur
such
as
du
rin
g b
ush
mea
th
arve
st);
un
der
stan
din
g s
ize-
sele
ctiv
e or
sex
-sp
ecif
ic h
arve
st p
atte
rns;
doc
um
enta
tion
of
over
all
har
vest
an
d n
atu
ral
mor
tali
ty r
ates
Det
erm
inat
ion
of
the
con
seq
uen
ces
of t
erre
stri
al p
lan
t h
arve
st o
n t
he
spat
ial
ecol
ogy
ofan
imal
s (e
.g. r
emov
al o
f cr
itic
al h
abit
ats
or f
ood
res
ourc
es)
Det
erm
inat
ion
of
the
con
seq
uen
ces
of d
iffe
ren
t h
arve
st s
trat
egie
s on
an
imal
dis
trib
uti
on,
beh
avio
ur,
an
d s
urv
ival
; doc
um
enta
tion
of
anim
al u
se o
f cl
eare
d h
abit
ats
du
rin
gre
gen
erat
ion
/rep
lan
tin
g
Un
der
stan
din
g t
he
beh
avio
ur
and
fat
e of
an
imal
s th
at a
re c
aptu
red
(in
ten
tion
ally
or
un
inte
nti
onal
ly)
and
rel
ease
d (
com
mer
cial
or
recr
eati
onal
); d
ocu
men
tati
on o
f th
e sp
atia
lec
olog
y of
an
imal
s re
lati
ve t
o h
arve
stin
g a
ctiv
itie
s an
d m
anag
emen
t ju
risd
icti
ons
Doc
um
enta
tion
of
anim
als’
beh
avio
ur
rela
tive
to
hu
man
dis
turb
ance
; eva
luat
ion
of
the
ener
get
ic a
nd
ph
ysio
log
ical
con
seq
uen
ces
of r
ecre
atio
nal
act
ivit
ies
on a
nim
als
Eva
luat
ion
of
anim
al r
esp
onse
s to
alt
ered
hab
itat
s; d
eter
min
atio
n o
f th
e en
erg
etic
an
dp
hys
iolo
gic
al c
onse
qu
ence
s of
mil
itar
y ac
tivi
ty o
n a
nim
als
As
(6.1
); d
eter
min
atio
n o
f an
imal
s’ r
esp
onse
to
rese
arch
col
lect
ion
, han
dli
ng
, an
d r
elea
se
Eva
luat
ion
of
anim
al m
ovem
ent
rela
tive
to
hu
man
-in
du
ced
fir
es; d
ocu
men
tati
on o
fh
abit
at u
se p
ost
fire
s; c
omp
aris
on o
f an
imal
res
pon
ses
to f
ire
in r
egio
ns
wit
h d
iffe
ren
tfi
re r
egim
es
Doc
um
enta
tion
of
mor
tali
ty a
ssoc
iate
d w
ith
pas
sag
e of
an
imal
s th
rou
gh
hyd
roel
ectr
icin
fras
tru
ctu
re; i
den
tifi
cati
on o
f b
arri
ers
to r
iver
con
nec
tivi
ty a
nd
th
eir
imp
act
on a
nim
alb
ehav
iou
r, e
ner
get
ics,
an
d s
urv
ival
; un
der
stan
din
g t
he
resp
onse
of
anim
als
to d
iffe
ren
tw
ater
man
agem
ent
acti
viti
es (
e.g
. wat
er w
ith
dra
wal
s, p
eak
ing
flo
ws)
; com
par
ison
of
anim
al b
ehav
iou
r in
sys
tem
s b
efor
e an
d a
fter
dam
con
stru
ctio
n o
r re
mov
al
Eva
luat
ion
of
anim
al m
ovem
ents
an
d h
abit
at u
se r
elat
ive
to h
um
an a
lter
atio
ns
of h
abit
at
Det
erm
inat
ion
of
dir
ect
inte
ract
ion
s b
etw
een
th
reat
ened
an
imal
s an
d i
nva
sive
non
-n
ativ
e or
ali
en s
pec
ies
(in
clu
din
g d
omes
tic
anim
als)
; id
enti
fica
tion
of
spat
ial
ecol
ogy,
pop
ula
tion
dyn
amic
s, a
nd
en
erg
etic
s in
an
imal
pop
ula
tion
s in
du
ced
by
inva
sive
sp
ecie
s(e
.g. z
ebra
mu
ssel
s ch
ang
ing
wat
er c
lari
ty, m
oun
tain
pin
e b
eetl
e al
teri
ng
for
est
cove
r);
det
erm
inat
ion
of
the
beh
avio
ur
and
su
rviv
al o
f an
imal
s re
lati
ve t
o d
isea
se o
r p
aras
ites
Det
erm
inat
ion
of
the
beh
avio
ura
l an
d e
ner
get
ic r
esp
onse
of
anim
als
to a
lter
ed h
abit
ats,
dis
ease
, com
pet
itio
n, o
r p
red
atio
n f
rom
pro
ble
mat
ic s
pec
ies
Doc
um
enta
tion
of
inte
ract
ion
s b
etw
een
an
imal
s an
d t
he
intr
odu
ced
org
anis
m
Det
erm
inat
ion
of
the
spat
ial
ecol
ogy
and
exp
osu
re p
oten
tial
of
anim
als
rela
tive
to
sew
age
and
was
te w
ater
; eva
luat
ion
of
anim
als’
su
rviv
al f
ollo
win
g e
xpos
ure
to
sew
age
and
was
te w
ater
Sp
ecif
ic t
hre
at
(4.4
)0F
lig
ht
pat
hs
(5.1
)0H
un
tin
g a
nd
coll
ecti
ng
terr
estr
ial
anim
als
(5.2
)0G
ath
erin
gte
rres
tria
l p
lan
ts
(5.3
)0L
ogg
ing
an
dw
ood
har
vest
ing
(5.4
)0F
ish
ing
an
dh
arve
stin
gaq
uat
ic r
esou
rces
(6.1
)0R
ecre
atio
nal
acti
viti
es
(6.2
)0W
ar, c
ivil
un
-re
stan
d m
ilit
ary
exer
cise
s
(6.3
)0W
ork
an
d o
ther
acti
viti
es
(7.1
)0F
ire
and
fir
esu
pp
ress
ion
(7.2
)0D
ams
and
wat
erm
anag
emen
t/u
se
(7.3
)0O
ther
eco
syst
emm
odif
icat
ion
s
(8.1
)0In
vasi
ve n
on-
nat
ive/
alie
nsp
ecie
s
(8.2
)0P
rob
lem
atic
nat
ive
spec
ies
(8.3
) 0In
trod
uce
dg
enet
ic m
ater
ial
(9.1
)0H
ouse
hol
dse
wag
e an
du
rban
was
tew
ater
Per
ceiv
edcu
rren
t st
atu
s
Mod
erat
e
Low
Low
Mod
erat
e
Mod
erat
e
Mod
erat
e
Low
Low
Low
Hig
h
Low
Low
Low
Low
Low
Pot
enti
alop
por
tun
ity
Hig
h
Hig
h
Low
Hig
h
Hig
h
Hig
h
Mod
erat
e
Mod
erat
e
Low
Hig
h
Mod
erat
e
Mod
erat
e
Mod
erat
e
Mod
erat
e
Mod
erat
e
Cooke: Biotelemetry and biologging in endangered species research
major cause of death, followedby road kills, with individualshaving an 81% chance of sur-viving for 1 yr. The authorsfound no evidence of differ-ences in survivorship betweenadults (of both sexes) and juve-niles. Bright & Hervert (2005)used radio collars and weeklytracking by fixed-wing airplaneto document adult mortality andfawn recruitment of a popula-tion of endangered Sonoranpronghorn Antilocapra ameri-cana sonoriensis in Arizona.Once mortalities were detectedduring flights, sites were visitedin an effort to determine thecause of mortality. Over the sev-eral years of study, annual adultmortality rates varied from 11 to83%, with mortality associatedwith predation and drought,whereas for fawns, drought wasthe primary source of mortality.Collectively, all of these studiesprovided information that couldbe used in demographic mod-els, and also data on the timing(season- and life-stage specific)and causes of mortality, en-abling the development of ap-propriate conservation actions.Although the analysis of sur-vival/mortality data from tele-metry and logging studies is notstraightforward (White & Gar-rott 1990), the data sets arisingfrom these technologies typi-cally have many advantagesrelative to conventional tech-niques.
Population size
Biologists are continually insearch of tools or approachesthat enable cost-effective, sim-ple, and accurate estimates ofpopulation size with minimalassumptions (Boyce 1992, Brooket al. 2000). Population viabilityanalysis (PVA; Boyce 1992) andother metrics for evaluating the
175
Tab
le 2
(co
nti
nu
ed)
Gen
eral
th
reat
(10)
Geo
log
ical
eve
nts
(11)
Cli
mat
e ch
ang
ean
d s
ever
ew
eath
er
Op
por
tun
itie
s an
d a
pp
lica
tion
s
As
(9.1
); d
eter
min
atio
n o
f th
e m
ovem
ent
of p
ollu
tan
ts b
etw
een
hab
itat
s as
tra
nsp
orte
db
y an
imal
s; s
pat
ial
ecol
ogy
of a
nim
als
rela
tive
to
pol
lute
d s
ites
an
d h
abit
ats;
det
erm
ina-
tion
of
pot
enti
al f
ood
ch
ain
bio
mag
nif
icat
ion
th
rou
gh
stu
die
s on
an
imal
sp
atia
l ec
olog
y
Det
erm
inat
ion
of
anim
al r
esp
onse
s (i
ncl
ud
ing
ch
ang
es i
n b
ehav
iou
r an
d h
abit
at u
se)
toef
flu
ents
an
d a
lter
ed n
utr
ien
t d
ynam
ics
Eva
luat
ion
of
anim
al b
ehav
iou
r an
d s
urv
ival
rel
ativ
e to
en
tan
gle
men
t in
deb
ris/
tras
h;
doc
um
enta
tion
of
anim
al m
ovem
ent
and
hab
itat
use
rel
ativ
e to
gar
bag
e; d
eter
min
atio
nof
ch
ang
es i
n f
orag
ing
act
ivit
y re
lati
ve t
o g
arb
age
site
s
Det
erm
inat
ion
of
chan
ges
in
hab
itat
use
or
beh
avio
ur
rela
tive
to
chan
ges
in
hab
itat
(e.
g.
acid
rai
n, s
mog
); d
eter
min
atio
n o
f th
e sp
atia
l ec
olog
y an
d e
xpos
ure
pot
enti
al o
f an
imal
s
Det
erm
inat
ion
of
the
spat
ial
ecol
ogy
and
exp
osu
re p
oten
tial
of
anim
als
rela
tive
to
pol
luti
on; d
eter
min
atio
n o
f th
e en
erg
etic
con
seq
uen
ces
of e
xpos
ure
to
pol
luta
nts
(esp
ecia
lly
ther
mal
pol
luti
on a
nd
ass
ocia
ted
th
erm
oreg
ula
tory
beh
avio
ur)
; det
erm
inat
ion
of b
ehav
iou
ral
dis
turb
ance
rel
ativ
e to
pol
luti
on (
e.g
. noi
se f
rom
hig
hw
ays
or p
lan
es,
son
ar n
oise
an
d i
mp
acts
on
mar
ine
mam
mal
s, l
igh
t p
ollu
tion
fro
m u
rban
en
viro
nm
ents
)
Det
erm
inat
ion
of
anim
al r
esp
onse
s (b
ehav
iou
r, m
ovem
ent,
su
rviv
al, a
nd
hab
itat
use
)re
lati
ve t
o g
eolo
gic
eve
nts
an
d a
ssoc
iate
d i
mp
acts
on
hab
itat
an
d d
irec
tly
on t
he
anim
al
As
(10.
1)
As
(10.
1)
Eva
luat
ion
of
anim
al h
abit
at u
se d
uri
ng
an
d a
fter
hab
itat
alt
erat
ion
s (e
.g. f
rom
cor
alb
leac
hin
g, t
un
dra
th
awin
g);
det
erm
inat
ion
of
anim
al e
mig
rati
on o
r m
orta
lity
rel
ativ
e to
hab
itat
alt
erat
ion
s
As
(11.
1); d
eter
min
atio
n o
f es
tiva
tion
per
iod
s an
d h
abit
ats
As
(11.
1); d
ocu
men
tati
on o
f ch
ang
es i
n a
nim
al m
igra
tion
pat
tern
s re
lati
ve t
o th
erm
alen
viro
nm
ents
(e.
g. l
oss
of g
laci
ers,
hea
t w
aves
); e
valu
atio
n o
f th
erm
al e
colo
gy
of a
nim
als
As
(11.
1); d
eter
min
atio
n o
f an
imal
dis
pla
cem
ent
Sp
ecif
ic t
hre
at
(9.2
)0In
du
stri
al a
nd
mil
itar
y ef
flu
ents
(9.3
)0A
gri
cult
ure
an
dfo
rest
ry e
fflu
ents
(9.4
)0G
arb
age
and
soli
d w
aste
(9.5
)0A
ir-b
orn
ep
ollu
tan
ts
(9.6
)0E
xces
s en
erg
y
(10.
1) V
olca
noe
s
(10.
2) E
arth
qu
akes
/ts
un
amis
(10.
3) A
vala
nch
es/
lan
dsl
ides
(11.
1) H
abit
at s
hif
tin
gan
d a
lter
atio
n
(11.
2) D
rou
gh
ts
(11.
3) T
emp
erat
ure
extr
emes
(11.
4) S
torm
s an
dfl
ood
ing
Per
ceiv
edcu
rren
t st
atu
s
Low
Low
Low
Low
Mod
erat
e
Low
Low
Low
Low
Low
Low
Low
Pot
enti
alop
por
tun
ity
Hig
h
Mod
erat
e
Mod
erat
e
Low
Hig
h
Low
Mod
erat
e
Low
Hig
h
Hig
h
Hig
h
Hig
h
Endang Species Res 4: 165–185, 2008
population biology of animals all depend on some esti-mate of population size and are routinely used for theassessment of conservation status (Brook et al. 2000).Many studies attempting to estimate population sizehave included telemetry and logging data, but rarelyare data from those sources used in isolation to esti-mate population size. Instead, telemetry and loggingare coupled with other techniques such as photo-counts, line-transects, scat counts, mark-resight, andmark-recapture. In fact, many studies involve compar-ing multiple techniques in the quest for the bestmethod. For example, Fisher et al. (2000) studied thecritically endangered bridled nailtail wallaby Ony-chogalea fraenata in central Queensland, Australia.Because of its small size and its nocturnal, solitary, andcryptic behaviour, evaluation of population size wasdifficult. Using mark-recapture, mark-resight, radio-tagging, and line-transect methods, the researchersassessed biases and the value of each method for man-agement. Interestingly, the projected value of lambda(finite rate of increase) based on radio-tagging datawas most sensitive to adult survival (see ‘Mortalityrates and causes’), and line-transect estimation wasfound to be the most appropriate long-term monitoringmethod. In many studies, data from telemetry is con-sidered to be the ‘gold standard’ to which other metricsare compared (e.g. in a study of endangered white-crowned pigeon Columba leucocephala in Florida,USA; Strong et al. 1994, and in a study of Lower Keysmarsh rabbit Sylvilagus palustris hefneri on BocaChica Key, Florida; Forys & Humphrey 1997).
Reproductive biology and potential
Knowledge of the reproductive biology of animals iscritical to understanding population dynamics, particu-larly in the case of endangered species. For manyendangered species, there is a rudimentary under-standing of basic natural history information related toreproduction, including the reproductive timing andoutput, which is critical to the understanding of endan-germent risk and status. One of the unique character-istics of telemetry and technology is that it enables thesame individuals to be monitored throughout multipleperiods of their life cycle.
When an animal engages in reproduction, additionalinformation can be obtained with respect to differen-tial reproductive success, age at maturation, andreproductive output. For example, Litzgus & Mousseau(2006) used radio telemetry to study the reproductivebiology of spotted turtle Clemmys guttata in SouthCarolina, USA. They documented the timing ofcourtship, the proportion of females that were gravidin each year, the timing duration of the nesting period,
nesting times (nocturnal) and habitats, and clutch size.Palomares et al. (2005) used radio tracking over a 9 yrperiod to study the reproductive biology of the Iberianlynx Lynx pardinus, the most endangered felid in theworld, in a population in southwestern Spain. Theauthors found that the potential breeding subpopula-tion was usually composed of 3 adult females (whichwere tracked for almost their complete reproductivelife) with a lifetime reproductive output of between 11and 19 cubs. However, mortality rates for young (pre-dispersal) cubs were sufficiently high that the authorsproposed the extraction of cubs from a mother with alow survival probability.
In some cases, telemetry can be used to locate repro-ductive sites, enabling researchers to collect data onreproductive potential. For example, Fox et al. (2000)used both acoustic and radio telemetry to monitor themovements of endangered adult Gulf sturgeonAcipenser oxyrinchus desotoi as they moved betweenChoctawhatchee Bay and the Choctawhatchee Riversystem. Telemetry results coupled with egg samplingwere used to identify Gulf sturgeon spawning sites, thetiming of reproduction, and sex-specific behaviour.Results from histology and their telemetry data sup-ported the hypothesis that male Gulf sturgeon mayspawn annually, whereas females require more than1 yr between spawning events. By combining teleme-try with other approaches (e.g. histology, in the aboveexample) conservation scientists can elucidate the sub-tle mechanisms of reproductive biology to improveconservation efforts.
Geographic range
Another metric used to assess endangerment statusis a decline in the area of occupancy, extent of occur-rence and/or quality of habitat. Telemetry and loggingare typically the most reliable means of determiningthe spatial ecology of animals, particularly those thatare migratory. The occupancy of a small geographicalrange is one of the potential justifications for cate-gorizing species (or populations). The criteria includespecific spatial distribution thresholds (e.g. from10 000 km2 to 10 km2). Telemetry can be used to iden-tify the exact boundaries of an animal’s range and isparticularly effective in habitats where visual observa-tions may be difficult (e.g. ocean, turbid freshwaters,dense forest canopy). In general, satellite (Fedak et al.2002) and logger (Block 2005) technologies tend to bebest suited to animals with broad geographic distri-butions.
Many of the studies on the geographic range or dis-tribution of animals are conducted within politicalboundaries (e.g. states/provinces, countries) or park/
176
Cooke: Biotelemetry and biologging in endangered species research
reserve environments. With most conservation actionsimplemented at the regional and national scale, stud-ies focused on this level are extremely valuable andnecessary. For example, Raum-Suryan et al. (2004)studied Stellar sea lions Eumetopias jubatus in westernAlaska to understand coastal distribution in Alaskanwaters, although the entire range of this speciesextends southward to California. Mills & Gorman(1997) used conventional and satellite radio telemetryto track endangered African wild dogs Lycaon pictusin the Kruger National Park, South Africa. Althoughthese animals have a geographic range that extendsbeyond the park boundary, there was a need to deter-mine the distribution of the dogs within the park. Inanother application, Wittmer et al. (2005) used radiotelemetry to study woodland caribou Rangifer taran-dus caribou across the entire distribution of an endan-gered mountain ecotype (i.e. a terrestrial landscapeunit) in British Columbia, Canada. In this case, distrib-ution was assessed relative to an endangered habitattype in a specific province.
Relatively few studies evaluate distribution acrossbroader scales, although there are some fascinatingexceptions that would not have been possible withouttelemetry. For example, Tuck et al. (1999) used geo-location (from data loggers) to determine the geo-graphic range of wandering albatrosses Diomedeaexulans. After tagging in South Georgia and theCrozet Islands, the birds made extensive journeysfrom southern Africa across the Indian Ocean tosoutheastern Australia and the east of New Zealand.Such broad movements of seabirds have not previ-ously been documented and will assist assessments ofrisk, such as those related to seabird by-catch bylong-line fisheries (e.g. Weimerskirch et al. 2000).Weng et al. (2005) used satellite tags attached to thedorsal fins of salmon sharks Lamna ditropis in thePacific Ocean and revealed that they have a subarcticto subtropical range, much larger than previouslydocumented. In another example, Cheng (2000) usedsatellite transmitters to track green turtles Cheloniamydas from nesting sites in the Peng Hu Archipelago,Taiwan to the continental shelf east of mainlandChina, including both trans-oceanic and coastal legs.The authors concluded that the distribution andvagility of the species requires regional and inter-national cooperation. Indeed, as a result of many tele-metry and logging studies on marine turtles, inter-national cooperation has now been recognized asfundamental to turtle conservation (Blumenthal et al.2006). In another study, a single endangered NorthPacific right whale Eubalaena japonica was taggedwith a satellite radio transmitter (Wade et al. 2006).The animal was tracked for a 40 d monitoring periodduring which it moved throughout a large part of the
southeast Bering Sea shelf and traversed regions ofthe outer-shelf where right whales have not beenseen in decades, thus expanding the known rangeof the species. Although relatively few studies haveused telemetry to explicitly determine the globalgeographic range of a species, information from tele-metry and logging studies often is used to supplementother information to determine the full geographicdistribution.
There has been much research on endangered spe-cies at the edges of their geographic distribution (e.g.Hoving et al. 1994 studied the threatened [according tothe United States Endangered Species Act, US ESA]Canada lynx Lynx canadensis in the United States withradio telemetry; Galois et al. (2002) used radio teleme-try to study threatened spiny softshell turtle Apalonespinifera in northern Lake Champlain [Quebec,Canada and Vermont, USA] at the northern limit of itsrange) and it is likely that this research will grow,given the recent finding that those populations/indi-viduals may be particularly important (Hampe & Petit2005). Furthermore, research at the edges of geo-graphic ranges is also relevant to the prediction andmonitoring of climate change impacts on animal distri-butions (e.g. Hampe & Petit 2005).
Habitat associations
Identification of habitat association is essential todetermine the habitats which an animal can occupy aswell as to understand their spatial ecology. When criti-cal habitats (e.g. hibernation sites, reproductive sites,feeding areas) are identified, it is possible to determinethe extent to which those habitats are threatened.Identification of critical habitats is also a prerequisitefor conservation actions, such as habitat restoration orprotection. For example, Prior & Weatherhead (1996)used radio telemetry to identify the hibernacula ofblack rat snakes Elaphe obsoleta obsoleta in Ontario atthe northern edge of their range, where winter protec-tion is essential. The authors revealed that rat snakehibernacula could not be predictably located bysearching for surface habitat features and suggestedthat radio telemetry be used to identify and protectadditional hibernacula as well as to preserve baskingtrees at known hibernacula. In another example,Sedgley & O’Donnell (1999) identified the factors thatinfluenced the selection of roost cavities by the threat-ened New Zealand long-tailed bat Chalinolobus tuber-culatus in a rainforest in Fiordland, New Zealand.Using radio telemetry, they identified that all day roostsites were in tree cavities and compared those thatwere actually used to those available. Factors such asdistance to the nearest vegetation, cavity condition
177
Endang Species Res 4: 165–185, 2008
(wet or dry inside), and height from the ground ex-plained 97% of the variance between used and avail-able roosts.
A number of studies have used telemetry to identifykey reproductive sites or sites traversed or used asstopovers during migration. For example, Paragamianet al. (2002) studied the endangered (US ESA) Koote-nai River white sturgeon using radio and acoustictelemetry and identified critical spawning habitats.They identified at least 5 primary spawning locations,most in the vicinity of outside bends, which tend tohave sandy substrate. Due to low recruitment, it issuspected that the animals are spawning in unsuit-able habitat, which may indicate that this is the siteof historic spawning sites that have been degraded.Hayward et al. (2007b) determined that adequateprey resources were critical for lions Panthera leo byusing telemetry to test predictions of their diet atreintroduction sites. Kanai et al. (2002) used satellitetransmitters to track critically endangered Siberiancranes Grits leucogeranus from breeding grounds innortheastern Siberia to wintering sites in China.Throughout the migration, several key wetland stop-over sites were identified in China. Collectively, theidentification of critical habitats is fundamental toidentifying the threats facing an organism and inidentifying potential conservation actions. Such stud-ies also have the potential to generate political con-flicts when long-distance migrants cross politicalborders and are exposed to different levels of pro-tection.
Connectivity of subpopulations
Several assessment criteria deal with the extent towhich the species range (or habitat) is fragmented.Indeed, a fundamental challenge in conservation biol-ogy is delineating discrete population units, especiallyin highly mobile animals with large geographicranges. Telemetry and logging provide insight into themovement of animals between fragmented habitats(i.e. vagility, dispersal, and emigration) and can definethe extent to which these animals are isolated or formsubpopulations. Telemetry and logging tools, particu-larly when combined with other techniques such asstable isotope and genetic analyses, provide the oppor-tunity to determine the connectivity of populations ofanimals even when they are highly mobile and migratevast distances (Webster et al. 2002). For example, Selo-nen et al. (2005) used genetic tools (microsatellites)and complementary radio telemetry data to study geneflow of declining Siberian flying squirrel Pteromysvolans in Finland. Radio telemetry studies indicatedthat the flying squirrels had good dispersal abilities,
but the high level of genetic differentiation betweensampling sites indicated that the actual gene flow overlarge distances was low.
Some studies have used telemetry tools withoutgenetic tools to assess connectivity. For example,Iverson & Esler (2006) studied the demographic con-nectivity among population segments of Harlequinducks Histrionicus histrionicus during the non-breed-ing season in Prince William Sound, Alaska. Theyradio-tagged 434 ind. over a 6 yr period and trackedthem by aircraft. Home range analyses indicatedrestricted movement of individuals (mean 95% kernelhome range estimates, 11.5 ± 2.2 km2). The authorsthen developed a demographic model, which incor-porated estimates for population size, survival, andmovement rates (all obtained from telemetry data), toinfer the degree of independence among populationsegments. In another study, Tyus (1990) radio-trackedendangered Colorado squawfish Ptychochelius luciusin the Green River basin of Colorado and Utah, USAfrom 1980 to 1988. A high proportion (63%) of indi-viduals was highly mobile. However, a number oftelemetered individuals spawned at the same site formore than 1 yr after migrating from both upstreamand downstream areas, indicating potential for sub-populations.
Telemetry can also be used to assess movement ofanimals between fragmented habitats. For example,Riley et al. (2006) studied 2 highly mobile carnivores(i.e. coyotes Canis latrans and bobcats Felis rufus)across the Ventura Freeway near Los Angeles, Califor-nia. Combining radio telemetry data and geneticallybased assignments to identify individuals, revealedthat although there were moderate levels of migration,populations on either side of the freeway were geneti-cally differentiated. Hence, the authors inferred thatthe individuals that cross the freeway rarely repro-duce. Smyth & Pavey (2001) used radio telemetry tostudy the movement of endangered black-breastedbutton-quail Turnix melanogaster in 13 rainforestpatches of an agricultural landscape in eastern Aus-tralia. No movement was observed between patches,with animals only resident in the 3 largest patches.Telemetry has also been used to detect the presence ofmetapopulations of endangered species. QuokkasSetonix brachyurus are restricted to isolated habitatpatches, which led researchers to conclude the speciesoriginally occurred as a natural metapopulation (Hay-ward et al. 2003). However, restriction to those patchestoday suggests the metapopulation has collapsed andregional extinction is imminent (Hayward et al. 2004).Collectively, telemetry and logging have muchpromise for providing information on the connectivityof populations, which should enhance the evaluationand assignment of endangerment status.
178
Cooke: Biotelemetry and biologging in endangered species research
REDUCTION OF UNCERTAINTY
One of the challenges that face the assessors ofendangerment status is how to make consistent classi-fications when faced with uncertainty. Even thoughthere are clear decision rules based on establishedthresholds, the data for these parameters are often esti-mated. Hence, uncertainty can arise from measure-ment error and natural biological variation (Akçakayaet al. 2000). Considering that the interpretation ofuncertain data by different assessors may lead toinconsistent classifications which could ultimatelyaffect conservation actions and the fate of a species,there is a clear need to reduce uncertainty (Burgman etal. 1999, Akçakaya et al. 2000). However, an alterna-tive and longer-term approach is to reduce uncertaintyby selecting an appropriate means of collecting thenecessary data and reducing measurement error. Theother primary source of uncertainty, natural biologicalvariability (both spatially and temporally) in demo-graphy and distribution is fundamentally more difficultto incorporate into decision making because it requiresa probabilistic approach (Burgman et al. 1999). How-ever, as mentioned above, understanding individualvariation is an important element of conservation sci-ence. Telemetry and logging, the focus of this paper,both have the potential to illuminate these 2 primarysources of uncertainty and aid in the reliable and con-sistent determination of endangerment status.
Measurement error
Relative to other study methods such as externalmarking, telemetry techniques enable the researcherto locate individual organisms through time. Externalmarking and associated mark-recapture methodsrequire that significant time is invested in recapturingor resighting the animals or that no data is provided.Essentially, mark-recapture techniques are biasedagainst the detection of movement and can lead toerroneous conclusions regarding population demo-graphics or the vagility and spatial distribution of aspecies (Gowan et al. 1994). Conversely, telemetryenables one to locate individuals across broader spatialscales, provided that the technology (e.g. satellite,telemetry array) or tracking protocols are compatiblewith the scale in question. In a review of techniques forestimating animal populations, Seber (1992) proposedthat telemetry could improve population estimates byproviding information on emigration, immigration, andspatial distribution relative to the focal area for whichthe population estimate is required. Nonetheless, thereare a number of challenges when dealing with teleme-try and logging data (White & Garrott 1990).
A tenet of most mark-recapture studies is that thesystem is closed, meaning that it is particularly difficultto assess ‘edge’ individuals (i.e. those individuals thathave home ranges that overlap with the region wherethe population estimate is being conducted; Seber1992). Techniques now exist for using telemetry data todocument the extent to which this closed populationassumption is violated and to correct for this bias(Eberhardt 1990). A combined analysis of recapture/resighting data coupled with telemetry data allowsseparate estimates of true survival and emigrationrates to be generated (Nasution et al. 2002). When onlyusing recapture/resighting data, only apparent sur-vival can be estimated. The combined estimates canbe more precise. For example, Powell et al. (2000)assessed the survival and movement of wood thrushesHylocichla mustelina in central Georgia, USA, by com-bining mark-recapture and radio telemetry tech-niques. The combined model that used both datasources resulted in more precise estimates of move-ment and recapture rates than separate estimation.However, the authors also identified that there wereminimum sample sizes required (for both marked andtelemetered individuals) to generate reliable estimatesin this system. Other authors have now developed opti-mal allocation models to determine the number oftelemetry devices required relative to standard mark-ing techniques (e.g. Nasution et al. 2002). An addi-tional assumption of population estimates, particularlywhen using an index, is that the population index isdirectly proportional to the population density. How-ever, in practice, probability of detection variesthrough space and time. Pollock et al. (2002) advocatea measure of detection probability that could be builtinto the monitoring programs through a double sam-pling approach that could use telemetry.
In some cases, the measurement error is associatedwith experimental design flaws. For example, Maehret al. (2004) drew attention to a model of Florida pan-ther Puma concolor coryi habitat that erred by arbitrar-ily creating buffers around radio locations collectedduring daylight hours, on the assumption that studyanimals were only at rest during these times. Theauthors claimed that this error could lead to the im-pression that unfragmented forest cover is unimpor-tant to panther conservation, and could encourageinaccurate characterizations of panther habitat. Hence,although telemetry data provided important data onhabitat use, failure to include activity during night orcrepuscular periods lead to the suggestion of conserva-tion actions that were not supported by scientificallyderived data. Although this is not truly a measurementerror, it serves as an important lesson that telemetryfindings are only as good as the experimental designand data quality. In this case, the use of logging (rather
179
Endang Species Res 4: 165–185, 2008
than telemetry) could provide a continuous time-bud-get over a 24 h period and help to describe pantheractivity at night.
On occasion, IUCN specialist groups develop andpublish resolutions intended to advance the study orconservation of a particular species or group of species.The IUCN specialist groups also have a history ofdeveloping resolutions that recognize the need fordevelopment of techniques that can aid in conserva-tion and management of species in decline. For exam-ple, in 1981 the IUCN Polar Bear (Ursus maritimus)Specialist Group made a resolution (Res#4-1981) onthe development and use of telemetry techniqueswhich stated,
The IUCN Polar Bear Specialist Group, recognizing thatconventional and satellite telemetry are effective tech-niques to study ecology of polar bears; and, recognizingthat existing systems of attaching transmitters to polarbears are not sufficiently reliable; and, recognizing thatexisting technology for satellite tracking needs signifi-cant improvement; therefore urges use of conventionaland satellite telemetry to study polar bears and coopera-tive efforts to improve telemetry techniques.
Such a resolution recognized the potential for study-ing species such as polar bears that are spatially dif-fuse and the fact that telemetry had the potential toilluminate the movements of this species. Furthermore,the resolution identified that technological improve-ments were needed to reduce uncertainty andenhance the reliability of the data.
Since this time, there have been a number of satellitetelemetry studies on polar bears focused on under-standing their spatial ecology (Messier et al. 1992,Mauritzen et al. 2002) and population structure(Bethke et al. 1996, Mauritzen et al. 2002). In a subse-quent resolution published in 1985 (Res#5-1985) theIUCN Polar Bear Specialist Group stated
that research be directed at improving the cost-effective-ness and statistical reliability of the mark and recapturestudies by developing and testing new research designs,and by using movement data gathered from telemetry tounderstand and develop corrections for capture biases inthe mark and recapture data.
In this case, telemetry was suggested as a tool forimproving the reliability of mark and recapture data,which is the standard technique for assessing polarbear population dynamics. This example demonstratesthe potential of telemetry techniques to reduce uncer-tainty in the threat assessment process.
It is also worth mentioning that telemetry locationsare themselves subject to error. The accuracy of theposition estimate can be impacted by signal bounce ormultipath (see Cooke et al. 2005). In some systems andfor some taxa, it is possible to visually confirm the loca-tion of the animal. Trackers should be experienced and
trained in a standardized manner that involves fre-quent assessment using hidden tags to assess the accu-racy of the positions. In some habitats (e.g. under ice,dense forest, caves) animals spend significant time inareas where they can not be located, and this can alsointroduce bias. Substantial bias can also occur whenusing light-based geolocation tags (Teo et al. 2004) formarine organisms and birds or when using satellitetracking (Hays et al. 2001). Although there are sophis-ticated algorithms for assigning latitude and longitudepositions, the error can at times be on the order of tensto hundreds of kilometers (indeed, the definition ofgeolocation is 2 points per day with errors around theequinoxes). Understanding the capabilities of the tech-nology is critical to determining the extent to whichtelemetry can address the study objectives and providereliable data that reduces uncertainty regarding thebiology or status of a species (Hays et al. 2001).
Natural biological variability
Immense variation is inherent in biological systems.Although biologically important and of fundamentalinterest, attempts to assess threats to populations,determine the conservation status of a species, anddevelop and implement management plans in caseswhere variation among individuals and populationsoccurs, pose some unique challenges to the conserva-tion practitioner. The idea of intra-specific variation isnot, however, new in biology (e.g. Bennett 1987). Indi-viduals and populations can vary in such basic attrib-utes as physiology, behaviour, morphology, and lifehistory. Telemetry and logging have recently providedunique insight into the magnitude and biological con-sequences of natural variability. For example, recentwork on sooty shearwater Puffinus griseus using log-gers has revealed that different individuals use differ-ent migratory routes, thus exposing them to differentthreats (Shaffer et al. 2006). Cooke et al. (2006)revealed that a segment of the sockeye salmonOncorhynchus nerka populations that initiated migra-tion early relative to the historical norm, tended tohave much reduced survivorship as well as a uniquephysiological signature. As noted above (see ‘Mortalityrates and causes of mortality’), different sexes and dif-ferent life stages can exhibit substantially differentpatterns in mortality and emigration, all data thatshould be embraced, not ‘averaged’ and consideredstatistical noise (Bennett 1987). Natural biological vari-ability (both spatial and temporal) in demography anddistribution is difficult to both quantify and predict(Burgman et al. 1999), leading to uncertainty. How-ever, a greater focus on individual and intra-specificvariation in organismal biology that is inherent in
180
Cooke: Biotelemetry and biologging in endangered species research
telemetry and logging studies will help to providegreater context to the divergent patterns observed innature and help to reduce uncertainty in decision mak-ing in applied conservation questions. Integration oftelemetry and logging studies with physiological andcondition-related measures will be useful in under-standing the consequences of individual variability onpopulation level phenomena (O’Connor et al. 2006).
CONCLUSIONS AND OUTLOOK FOR THE FUTURE
Telemetry and logging have much to offer endan-gered species research and conservation. Indeed,there have already been many advances in ecologyowing to these technologies (e.g. Cooke et al. 2004,Block 2005). This paper has revealed that there areclear opportunities to obtain the novel data needed toassess IUCN or regional population status and to iden-tify the causes of population declines. There are stillchallenges with respect to the application of telemetryand logging data to conservation and managementactions, as the data generated by these techniques arealways retrospective and rarely probabilistic (Amstrupet al. 2004). Nonetheless, the focus of the present paperis on using telemetry and logging to inform threatassessments and conservation status evaluations, atask that tends to be retrospective. Telemetry andlogging are particularly powerful techniques whencombined with other assessment methods such aspoint-counts or mark-recapture enabling the crosscalibration and validation of different population esti-mate tools (e.g. Amstrup et al. 2004). Telemetry andlogging can also be linked with other tools and disci-plines to quantify animal responses to stress (e.g. con-servation physiology, Wikelski & Cooke 2006; fieldphysiology, Goldstein & Pinshow 2006), identifyingdiscrete populations (e.g. through genetic analyses;Fernando & Lande 2000), or assessing animal nutri-tional/energetic condition (e.g. stable isotope analysis,Cunjak et al. 2005; non-lethal energetic assessment,Cooke et al. 2006).
Despite the fact that telemetry and logging havemuch promise for endangered species research, somechallenges remain. Fundamental to all tagging studiesis the need to attach or implant a device on or in an ani-mal. Therefore, the premise of all studies should bethat the population does not suffer any harm and thatthe welfare status of individuals is taken into consider-ation. Essentially, this means considering the ethics oftagging and telemetry studies when dealing withendangered species. This does not mean that teleme-try should be avoided. Instead, when a telemetry orlogging study is determined to be the best method forachieving a desired objective, all efforts must be taken
to minimize the burden of the transmitter and theattachment/implantation on the individuals. There arenow a number of syntheses that provide telemetrypractitioners with guidance for minimizing the impactof telemetry techniques on animals. However, evenwith ethical guidelines and governmental regulations,the burden is often left on the practitioners to ensurethat their research techniques do not contribute to spe-cies declines. Interestingly, an important study thatoutlined the research needs for enhancing the assess-ment and management of species at risk (e.g. Mace etal. 2001), failed to list telemetry as a tool to fill researchgaps, perhaps emphasizing the need for a study suchas the present one to highlight opportunities.
Given the limited funds available for conservationresearch (Myers et al. 2000, Halpern et al. 2006), it isessential that the most cost-effective research methodis used. Considering the perceived high cost of teleme-try, it would be useful for authors to include data on thecost of their studies as well as the resources required(staff, equipment, time) in published works; this wouldmake it possible to conduct formal evidence-basedassessments to determine the effectiveness of differentresearch techniques (e.g. comparing telemetry to moretraditional marking and monitoring techniques) rela-tive to their cost (cost-effectiveness; Sutherland et al.2004). At present, there are few data available for aglobal assessment (but see Franco et al. 2007). Asmany endangered species occur in developing coun-tries, it is important to build the capacity (financial andtechnical training) to do such studies if needed. It maybe possible for research teams from comparativelywealthy institutions or countries to share their devicesand expertise (via a retribution such as co-authorship)with researchers from other institutions and countries.Telemetry and logging are not a panacea and as withany tool, should only be used after considering alterna-tives and determining what is the best approach toachieve the desired objective — and which techniqueswill have the least impact on the animal. Indeed, thereare several comparative studies that have determinedthat telemetry techniques were not the most appropri-ate for their given conservation application (e.g.McCann et al. 2001). In summary, telemetry and log-ging can provide conservation practitioners with datathat is unattainable using other techniques. However,it is important to only use these technologies whenthey are determined to be the best means of achievinga specific conservation objective. Telemetry and log-ging, as well as other innovative research, assessment,and monitoring tools are needed in order to informdecision makers and thus achieve biodiversity targets(e.g. the Convention of Biological Diversity 2010 tar-gets; see Balmford et al. 2005) and reverse the appar-ent global decline of many animal species.
181
Endang Species Res 4: 165–185, 2008
Acknowledgements. This work was supported by the NaturalSciences and Engineering Research Council of Canada, theCanada Foundation for Innovation, the Ontario ResearchFund, the Ontario Ministry of Research and Innovation (EarlyResearcher Award), and in particular, Carleton University.I acknowledge my many mentors and colleagues for theirsupport and sharing of expertise on biotelemetry and biolog-ging, including Scott Hinch, Tony Farrell, David Philipp,David Wahl, Patrick Weatherhead, Chris Bunt, Tony Beddow,Gary Anderson, Richard Brown, Jason Schreer, Martin Wikel-ski, George Niezgoda, Glenn Wagner, Andy Danylchuk, CorySuski, and Martyn Lucas. I also thank Yan Ropert-Coudert,Matt Hayward, Rory Wilson, Michael Donaldson, CalebHasler, Karen Murchie, and Kyle Hanson, and an anonymousreferee for providing comments on the manuscript. AmandaO’Toole assisted with final preparation of the manuscript.
LITERATURE CITED
Akçakaya HR, Ferson S, Burgman MA, Keith DA, Mace GM,Todd CR (2000) Making consistent IUCN classificationsunder uncertainty. Conserv Biol 14:1001–1013
Amstrup SC, McDonald TL, Durner GM (2004) Using satelliteradiotelemetry data to delineate and manage wildlife pop-ulations. Wildl Soc Bull 32:661–679
Balmford A, Bennun L, 10 Brink B, Cooper D and others (2005)The convention on biological diversity’s 2010 target.Science 307:212–213
Bannasch R, Wilson RP, Culik BM (1994) Hydrodynamicaspects of design and attachment of a back-mounted devicein penguins. J Exp Biol 194:83–96
Baras E, Benech V, Marmulla G (2002) Outcomes of a pilotfish telemetry workshop for developing countries. Hydro-biologia 483:9–11
Bennett AF (1987) Interindividual variability: an underutilizedresource. In: Feder ME, Bennett AF, Burggren WW, HueyRB (eds) New directions in ecological physiology. Cam-bridge University Press, Cambridge p 147–169
Bethke R, Taylor MK, Amstrup SC, Messier F (1996) Popula-tion delineation of polar bears using satellite collar data.Ecol Appl 6:311–317
Block BA (2005) Physiological ecology in the 21st century:Advancements in biologging science. Integr Comp Biol 45:305–320
Block BA, Dewar H, Blackwell SB, Williams TD and others(2001) Migratory movements, depth preferences, andthermal biology of Atlantic bluefin tuna. Science 293:1310–1314
Block BA, Costa DP, Boehlert GW, Kochevar RE (2002)Revealing pelagic habitat use: the tagging of Pacificpelagics program. Oceanol Acta 25:255–266
Blumenthal JM, Solomon JL, Bell CD, Austin TJ and others(2006) Satellite tracking highlights the need for inter-national cooperation in marine turtle management.Endang Species Res 2:51–61
Boyce MS (1992) Population viability analysis. Annu Rev EcolSyst 23:481–506
Boyd IL, Kato A, Ropert-Coudert Y (2004) Bio-logging science:sensing beyond the boundaries. Mem Nat Inst Polar Res 58:1–15
Bright JL, Hervert JJ (2005) Adult and fawn mortality a Sono-ran pronghorn. Wildl Soc Bull 33:43–50
Brook BW, O’Grady JJ, Chapman AP, Burgman MA,Akcakaya HR, Frankham R (2000) Predictive accuracy ofpopulation viability analysis in conservation biology.Nature 404:385–387
Brook BW, Traill LW, Bradshaw CJA (2006) Minimum viablepopulation size and global extinction risk are unrelated.Ecol Lett 9:375–382
Burgman MA, Keith DA, Rohlf FJ, Todd CR (1999) Probabilis-tic classification rules for setting conservation priorities.Biol Conserv 89:227–231
Butchart SHM, Stattersfield AJ, Baillie J, Bennun LA and oth-ers (2005) Using Red List indices to measure progresstowards the 2010 target and beyond. Phil Trans R SocLond Ser B Biol Sci 360:255–268
Call KA, Fadely BS, Greig A, Rehberg MJ (2007) At-sea andon-shore cycles of juvenile Steller sea lions (Eumetopiasjubatus) derived from satellite dive recorders: a compari-son between declining and increasing populations. Deep-Sea Res II Top Stud Oceanogr 54:298–310
Caughley G (1994) Directions in conservation biology. J AnimEcol 63:215–244
Chaloupka M, Parker D, Balazs G (2004) Modelling post-release mortality of loggerhead sea turtles exposed to theHawaii-based pelagic longline fishery. Mar Ecol Prog Ser280:285–293
Cheng IJ (2000) Post-nesting migrations of green turtles(Chelonia mydas) at Wan-An Island, Penghu Archipelago,Taiwan. Mar Biol 137:747–754
Clark BS, Leslie DM, Carter TS (1993) Foraging activity ofadult female ozark big-eared bats (Plecotus townsendiiingens) in summer. J Mammal 74:422–427
Collar NJ (1996) The reasons for red data books. Oryx 30:121–130
Cooke SJ, Hinch SG, Wikelski M, Andrews RD, Wolcott TG,Butler PJ (2004) Biotelemetry: a mechanistic approach toecology. Trends Ecol Evol 19:334–343
Cooke SJ, Niezgoda GH, Hanson KC, Suski CD, Phelan FJS,Tinline R, Philipp DP (2005) Use of CDMA acoustictelemetry to document 3-D positions of fish: relevance tothe design and monitoring of aquatic protected areas. MarTechnol Soc J 39:31–41
Cooke SJ, Hinch SG, Crossin GT, Patterson DA and others(2006) Physiology of individual late-run Fraser River sock-eye salmon (Oncorhynchus nerka) sampled in the oceancorrelates with fate during spawning migration. Can JFish Aquat Sci 63:1469–1480
Cooper N, Carling B (1995) Ecologists and ethical judge-ments. Biodivers Conserv 4:783–785
Cunjak RA, Roussel JM, Gray MA, Dietrich JP, Cartwright DF,Munkittrick KR, Jardine TD (2005) Using stable isotopeanalysis with telemetry or mark-recapture data to identifyfish movement and foraging. Oecologia 144:636–646
Deutsch CJ, Bonde RK, Reid JP (1998) Radio-tracking mana-tees from land and space: Tag design, implementation,and lessons learned from long-term study. Mar TechnolSoc J 32:18–29
Dickson BG, Beier P (2007) Quantifying the influence of topo-graphic position on cougar (Puma concolor) movement insouthern California, USA. J Zool 271:270–277
Eberhardt LL (1990) Using radio-telemetry for mark-recap-ture studies with edge effects. J Appl Ecol 27:259–271
Fancy SG, Pank LF, Douglas DC, Curby CH, Garner GW(1988) Satellite telemetry: a new tool for wildlife researchand management. US Fish and Wildlife Service, Wash-ington, DC
Farnsworth EJ, Rosovsky J (1993) The ethics of ecologicalfield experimentation. Conserv Biol 7:463–472
Fedak M, Lovell P, McConnell B, Hunter C (2002) Overcom-ing the constraints of long range radio telemetry from ani-mals: getting more useful data from smaller packages.Integr Comp Biol 42:3–10
182
Cooke: Biotelemetry and biologging in endangered species research
Fernando P, Lande R (2000) Molecular genetic and behavioralanalyses of social organization in the Asian elephant.Behav Ecol Sociobiol 48:84–91
Fisher DO, Hoyle SD, Blomberg SP (2000) Population dynam-ics and survival of an endangered wallaby: a comparisonof four methods. Ecol Appl 10:901–910
Forys EA, Humphrey SR (1997) Comparison of two methodsto estimate density of an endangered lagomorph. J WildlManag 61:86–92
Fox DA, Hightower JE, Paruka FM (2000) Gulf sturgeonspawning migration and habitat in the ChoctawhatcheeRiver system, Alabama–Florida. Trans Am Fish Soc 129:811–826
Franco AMA, Palmeirim JM, Sutherland WJ (2007) A methodfor comparing effectiveness of research techniques in con-servation and applied ecology. Biol Conserv 134:96–105
Galois P, Leveille M, Bouthillier L, Daigle C, Parren S (2002)Movement patterns, activity and home range of the easternspiny softshell turtle (Apalone spinifera) in northern LakeChamplain, Quebec, Vermont. J Herpetol 36:402–411
Gardenfors U, Hilton-Taylor C, Mace GM, Rodriguez JP(2001) The application of IUCN Red List criteria atregional levels. Conserv Biol 15:1206–1212
Gehrt SD (2005) Seasonal survival and cause-specific mortal-ity of urban and rural striped skunks in the absence ofrabies. J Mammal 86:1164–1170
Gibbons JW, Andrews KM (2004) PIT tagging: simple tech-nology at its best. Bioscience 54:447–454
Gifford A, Compagno LJV, Levine M, Antoniou A (2007) Satel-lite tracking of whale sharks using tethered tags. Fish Res84:17–24
Godfrey JD, Bryant DM (2003) Effects of radio transmitters:review of recent radio tracking studies. Sci Conserv 214:83–95
Goldstein DL, Pinshow B (2006) Taking physiology to the field:using physiological methods to answer questions aboutanimals in their environments. Physiol Biochem Zool 79:237–241
Gowan C, Young MK, Fausch KD, Riley SC (1994) Restrictedmovement in resident stream salmonids: a paradigm lost?Can J Fish Aquat Sci 51:2626–2637
Haines AM, Grassman LI, Tewes ME, Janecka JE (2006) Firstocelot (Leopardus pardalis) monitored with GPS tele-metry. Eur J Wildl Res 52:216–218
Halpern BS, Pyke CR, Fox HE, Haney JC, Schlaepfer MA,Zaradic P (2006) Gaps and mismatches between globalconservation priorities and spending. Conserv Biol 20:56–64
Hampe A, Petit RJ (2005) Conserving biodiversity under cli-mate change: the rear edge matters. Ecol Lett 8:461–467
Harris S, Cresswell WJ, Forde PG, Trewhella WJ, Woollard T,Wray S (1990) Home-range analysis using radio-trackingdata — a review of problems and techniques particularly asapplied to the study of mammals. Mammal Rev 20:97–123
Haupt M, Menzler S, Schmidt S (2006) Flexibility of habitatuse in Eptesicus nilssonii: does the species profit fromanthropogenically altered habitats? J Mammal 87:351–361
Hays GC, Åkesson S, Godley BJ, Luschi P, Santidrian P (2001)The implications of location accuracy for the interpretationof satellite tracking data. Anim Behav 61:1035–1040
Hayward MW, de Tores PJ, Dillon MJ, Fox BJ (2003) Localpopulation structure of a naturally occurring metapopula-tion of the quokka (Setonix brachyurus, Macropodidae:Marsupialia). Biol Conserv 110:343–355
Hayward MW, de Tores PJ, Augee ML, Fox BJ, Banks PB(2004) Home range and movements of the quokka,
Setonix brachyurus (Macropodidae: Marsupialia), and itsimpact on the viability of the metapopulation on the Aus-tralian mainland. J Zool 263:219–228
Hayward MW, de Tores PJ, Augee ML, Banks PB (2005) Mor-tality and survivorship of the quokka (Setonix brachyurus)(Macropodidae: Marsupialia) in the northern jarrah forestof Western Australia. Wildl Res 32:715–722
Hayward MW, Adendorff J, O’Brien J, Sholto-Douglas A andothers (2007a) The reintroduction of large carnivores tothe Eastern Cape Province, South Africa: an assessment.Oryx 41:205–214
Hayward MW, Hofmeyr M, O’Brien J, Kerle GIH (2007b)Testing predictions of the prey of the lion (Panthera leo)derived from modelled prey preferences. J Wildl Manag71:1567–1575
Heupel MR, Semmens JM, Hobday AJ (2006) Automatedacoustic tracking of aquatic animals: scales, design anddeployment of listening station arrays. Mar Freshw Res 57:1–13
Hooker S, Biuw M, McConnell BJ, Miller P, Sparling C (2007)Bio-logging science: logging and relaying physical andbiological data using animal-attached tags. Deep-Sea ResII 54:177–182
Hoving CL, Harrison DJ, Krohn WB, Jakubas WJ, McCol-lough MA (1994) Canada lynx Lynx canadensis habitatand forest succession in northern Maine, USA. Wildl Biol10:285–294
Hutchins M, Foose T, Seal US (1993) The role of veterinarymedicine in endangered species conservation. J Zoo WildlMed 22:277–281
Iverson SA, Esler D (2006) Site fidelity and the demographicimplications of winter movements by a migratory bird, theharlequin duck Histrionicus histrionicus. J Avian Biol 37:219–228
Kanai Y, Ueta M, Germogenov N, Nagendran M, Mita N,Higuchi H (2002) Migration routes and important restingareas of Siberian cranes (Grus leucogeranus) betweennortheastern Siberia and China as revealed by satellitetracking. Biol Conserv 106:339–346
Karesh WB (1999) Applications of biotelemetry in wildlifemedicine. In: Fowler ME, Miller RE (eds) Zoo and wildanimal medicine: current therapy, 4th edn. W.B. SaundersCompany, Philadelphia, PA, p 7–13
Lamoreux J, Akcakaya HR, Bennun L, Collar NJ and others(2003) Value of the IUCN Red List. Trends Ecol Evol 18:214–215
Litzgus JD, Mousseau TA (2006) Geographic variation inreproduction in a freshwater turtle (Clemmys guttata).Herpetologica 62:132–140
Loveridge AJ, Searle AW, Murindagomo F, Macdonald DW(2007) The impact of sport-hunting on the populationdynamics of an African lion population in a protected area.Biol Conserv 134:548–558
Mace GM (1994) An investigation into methods for categoris-ing the conservation status of species. In: Edwards PJ,May RM, Webb NR (eds) Large scale ecology and conser-vation Biology. Blackwell Scientific Publications, Oxford,p 293–312
Mace GM (1995) Classification of threatened species and itsrole in conservation planning. In: Lawton JH, May RM(eds) Extinction rates. Oxford University Press, Oxford,p 197–213
Mace GM, Baillie JEM, Redford KH, Beissinger SR (2001)Assessment and management of species at risk. In: SouleME, Orians GH (eds) Conservation biology: research pri-orities for the next decade. Island Press, New York,p 11–29
183
Endang Species Res 4: 165–185, 2008
Maehr DS, Larkin JL, Cox JJ (2004) Shopping centers as pan-ther habitat: Inferring animal locations from models. Ecol-ogy and Society 9(2):9. Available at: www.ecologyandsociety.org/vol9/iss2/art9/
Marmulla G, Bénech V (1999) Necessity for an informationexchange network for fisheries projects using bioteleme-try in developing countries. In: Moore A, Russell I (eds)Advances in fish telemetry. Proc 3rd Conf on FishTelemetry in Europe, Norwich, UK, 20–25 June 1999,p 197–205
Marshall GJ (1998) Crittercam: an animal borne imaging anddata logging system. Mar Technol Soc J 32:11–17
Mauritzen M, Derocher AE, Wiig O, Belikov SE, Boltunov AN,Hansen E, Garner GW (2002) Using satellite telemetry to de-fine spatial population structure in polar bears in the Norwe-gian and western Russian Arctic. J Appl Ecol 39: 79–90
McCann KI, Shaw K, Anderson MD, Morrison K (2001) Tech-niques for determining movement patterns of blue andwattled cranes in South Africa — colour-ringing versussatellite telemetry. Ostrich 15:104–108
McCarthy TM, Fuller TK, Munkhtsog B (2005) Movementsand activities of snow leopards in Southwestern Mongolia.Biol Conserv 124:527–537
McCulloch G, Aebischer A, Irvine K (2003) Satellite trackingof flamingos in southern Africa: the importance of smallwetlands for management and conservation. Oryx 37:480–483
McIntyre CL, Collopy MW, Lindberg MS (2006) Survivalprobability and mortality of migratory juvenile goldeneagles from interior Alaska. J Wildl Manag 70:717–722
Messier F, Taylor MK, Ramsay MA (1992) Seasonal activitypatterns of female polar bears (Ursus maritimus) in theCanadian Arctic as revealed by satellite telemetry. J Zool226:219–229
Mildenstein TL, Stier SC, Nuevo-Diego CE, Mills LS (2005)Habitat selection of endangered and endemic large fly-ing-foxes in Subic Bay, Philippines. Biol Conserv 126:93–102
Miller RM, Rodriguez JP, Aniskowicz-Fowler T, Bambara-deniya C and others (2006) Extinction risk and conserva-tion priorities. Science 313:441a
Miller RM, Rodriguez JP, Aniskowicz-Fowler T, Bamba-radeniya C and others (2007) National threatened specieslisting based on IUCN criteria and regional guidelines:current status and future perspectives. Conserv Biol 21:684–696
Mills MGL, Gorman ML (1997) Factors affecting the densityand distribution of wild dogs in the Kruger National Park.Conserv Biol 11:1397–1406
Millspaugh JJ, Marzluff JM (2001) Radio-tracking and animalpopulations. Academic Press, San Diego, CA
Minteer BA, Collins JP (2005) Ecological ethics: building anew tool kit for ecologists and biodiversity managers.Conserv Biol 19:1803–1812
Mulcahy DM (2003) Surgical implantation of transmitters intofish. ILAR J 44:295–306
Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GAB,Kent J (2000) Biodiversity hotspots for conservation priori-ties. Nature 403:853–858
Nasution MD, Brownie C, Pollock KH (2002) Optimal alloca-tion of sample sizes between regular banding and radio-tagging for estimating annual survival and emigrationrates. J Appl Stat 29:443–457
Palomares F, Revilla E, Calzada J, Fernandez N, Delibes M(2005) Reproduction and pre-dispersal survival of Iberianlynx in a subpopulation of the Donana National Park. BiolConserv 122:53–59
Paragamian VL, Wakkinen VD, Kruse G (2002) Spawninglocations and movement of Kootenai River white sturgeon.J Appl Ichthyol 18:608–616
Parmesan C (2006) Ecological and evolutionary responses torecent climate change. Annu Rev Ecol Evol Syst 37:637–669
Peck FR, Simmonds RC (1995) Understanding animal researchregulations: obligations of wildlife departments and fieldresearchers. Wildl Soc Bull 23:279–282
Phillips RA, Xavier JC, Croxall JP (2003) Effects of satellitetransmitters on albatrosses and petrels. Auk 120:1082–1090
Pollock KH, Nichols JD, Simons TR, Farnsworth GL, BaileyLL, Sauer JR (2002) Large scale wildlife monitoring stud-ies: statistical methods for design and analysis. Environ-metrics 13:105–119
Powell LA, Conroy MJ, Hines JE, Nichols JD, Krementz DG(2000) Simultaneous use of mark-recapture and radiotelemetry to estimate survival, movement, and capturerates. J Wildl Manag 64:302–313
Prior KA, Weatherhead PJ (1996) Habitat features of black ratsnake hibernacula in Ontario. J Herpetol 30:211–218
Putman RJ (1995) Ethical considerations and animal welfarein ecological field studies. Biodivers Conserv 4:903–915
Raum-Suryan KL, Rehberg MJ, Pendleton GW, Pitcher KW,Gelatt TS (2004) Development of dispersal, movementpatterns, and haul-out use by pup and juvenile Steller sealions (Eumetopias jubatus) in Alaska. Mar Mamm Sci20:823–850
Ries EH, Hiby LR, Reijnders PJH (1998) Maximum likelihoodpopulation size estimation of harbour seals in the DutchWadden Sea based on a mark-recapture experiment.J Appl Ecol 35:332–339
Riley SPD, Pollinger JP, Sauvajot RM, York EC, Bromley C,Fuller TK, Wayne RK (2006) A southern California free-way is a physical and social barrier to gene flow in carni-vores. Mol Ecol 15:1733–1741
Rodgers AR, Rempel RS, Abraham KF (1996) A GPS-basedtelemetry system. Wildl Soc Bull 24:559–566
Rodrigues ASL, Pilgrim JD, Lamoreux JF, Hoffman M, BrooksTM (2006) The value of the IUCN Red List for conserva-tion. Trends Ecol Evol 21:71–76
Ropert-Coudert Y, Wilson RP (2005) Trends and perspectivesin animal-attached remote-sensing. Front Ecol Environ 3:437–444
Salafsky N, Margoluis R, Redford KH, Robinson JG (2002)Improving the practice of conservation: a conceptualframework and research agenda for conservation science.Conserv Biol 16:1469–1479
Seber GAF (1992) A review of estimating animal abundance.II. Int Stat Rev 60:129–166
Sedgeley J, O’Donnell CFJ (1999) Roost selection by the long-tailed bat, Chalinolobus tuberculatus, in temperate NewZealand rainforest and its implications for the conserva-tion of bats in managed forests. Biol Conserv 88:261–276
Selonen V, Painter JN, Hanski PK (2005) Microsatellite varia-tion in the Siberian flying squirrel in Finland. Ann ZoolFenn 42:505–511
Shaffer SA, Tremblay Y, Weimerskirch H, Scott D and others(2006) Migratory shearwaters integrate oceanic resourcesacross the Pacific Ocean in an endless summer. Proc NatlAcad Sci USA 103:12799–12802
Smyth AK, Pavey CR (2001) Foraging by the endangeredblack-breasted button-quail (Turnix melanogaster) withinfragmented rainforest of an agricultural landscape. BiolConserv 98:149–157
Starr RM, O’Connell V, Ralston S, Breaker L (2005) Use ofacoustic tags to estimate natural mortality, spillover, and
184
Cooke: Biotelemetry and biologging in endangered species research
movements of lingcod (Ophiodon elongatzu) in a marinereserve. Mar Technol Soc J 39:19–30
Strong AM, Sawocki RJ, Bancroft GT (1994) Estimatingwhite-crowned pigeon population-size from flight-linecounts. J Wildl Manag 58:156–162
Sutherland WJ, Pullin AS, Dolman PM, Knight TM (2004) Theneed for evidence-based conservation. Trends Ecol Evol 19:305–308
Szedlmayer ST, Schroepfer RL (2005) Long-term residence ofred snapper on artificial reefs in the northeastern Gulf ofMexico. Trans Am Fish Soc 134:315–325
Teo SLH, Blackwell S, Boustany A, Walli A, Weng K, Block BA(2004) Validation of geolocation estimates based on lightlevel and sea surface temperature from electronic tags.Mar Ecol Prog Ser 283:81–98
Tuck GN, Polacheck T, Croxall JP, Weimerskirch H, PrincePA, Wotherspoon S (1999) The potential of archival tagsto provide long-term movement and behaviour datafor seabirds: First results from wandering albatrossDiomedea exulans of South Georgia and the CrozetIslands. Emu 99:60–68
Tyus HM (1990) Potamodromy and reproduction of Coloradosquawfish in the Green River basin, Colorado and Utah.Trans Am Fish Soc 119:1035–1047
Vilella FJ, Hengstenberg DW (2006) Broad-winged hawk(Buteo platypterus brunnescens) movements and habitatuse in a moist limestone forest of Puerto Rico. OrnitholNeotrop 17:563–579
Voegeli FA, Smale MJ, Webber DM, Andrade Y, O’Dor RK(2001) Ultrasonic telemetry, tracking and automated mon-itoring technology for sharks. Environ Biol Fishes 60:267–281
Wade P, Heide-Jørgensen MP, Shelden K, Barlow J and oth-ers (2006) Acoustic detection and satellite-tracking leadsto discovery of rare concentration of endangered NorthPacific right whales. Biol Lett 2:417–419
Webster MS, Marra PP, Haig SM, Bensch S, Holmes RT (2002)Links between worlds: unraveling migratory connectivity.Trends Ecol Evol 17:76–83
Weimerskirch H, Capdeville D, Duhamel G (2000) Factorsaffecting the number and mortality of seabirds attendingtrawlers and long-liners in the Kerguelen area. Polar Biol23:236–249
Welch DW, Boehlert GW, Ward BR (2002) POST — ThePacific Ocean Salmon Tracking Project. Oceanol Acta25:243–253
Welch DW, Turo S, Batten SD (2006) Large-scale marine andfreshwater movements of white sturgeon (Acipensertransmontanus). Trans Am Fish Soc 135:386–389
Weng KC, Castilho PC, Morrissette JM, Landeira-FernandezAM and others (2005) Satellite tagging and cardiac physiol-ogy reveal niche expansion in salmon sharks. Science 310:104–106
White GC, Garrott RA (1990) Analysis of wildlife radiotrackingdata. Academic Press, San Diego, CA
Wikelski M, Cooke SJ (2006) Conservation physiology.Trends Ecol Evol 21:38–46
Wikelski M, Kays R, Kasdin J, Thorup K, Smith JA, CochranWW, Swenson GW Jr (2007) Going wild: what a globalsmall-animal tracking system could do for experimentalbiologists. J Exp Biol 210:181–186
Williams BK, Nichols JD, Conroy MJ (2002) Analysis andmanagement of animal populations: modeling, estimation,and decision making. Academic Press, San Diego, CA
Wilson RP, McMahon CR (2006) Measuring devices on wildanimals: what constitutes acceptable practice? Front EcolEnviron 4:147–154
Wittmer HU, McLellan BN, Seip DR, Young JA, Kinley TA,Watts GS, Hamilton D (2005) Population dynamics of theendangered mountain ecotype of woodland caribou(Rangifer tarandus caribou) in British Columbia, Canada.Can J Zool 83:407–418
Zeller DC (1997) Home range and activity patterns of thecoral trout Plectropomus leopardus (Serranidae). Mar EcolProg Ser 154:65–77
Zurstadt CF, Stephan K (2004) Seasonal migration of wests-lope cutthroat trout in the middle fork salmon riverdrainage, Idaho. Northwest Sci 78:278–285
185
Editorial responsibility: Rory Wilson, Swansea, UK
Submitted: August 6, 2007; Accepted: October 26, 2007Proofs received from author(s): December 23, 2007