Assessing the reliability of biologger techniques to measure activity in a free-ranging primate

Abstract

Assessmentof animal activity patterns can shed lighton a rangeof behavioural and ecological processes. The timing of activity canhelp improve understanding of a species’ spatial ecology, chrono-biology, energetic demands, reproductive strategy and nutritionalrequirements (Halle & Stenseth 2000; Erkert 2003). Activity dataare also useful for the assessment of animal welfare and conser-vation planning (Kitchen & Martin 1996; Broom & Fraser 2007;Cooke 2008); Activity patterns allow us to measure the impact ofhuman disturbance on habitat viability and behaviour (Williamset al. 2006; Yang et al. 2007), and to predict the influence of localor global climate change on the survival and distribution of a spe-cies or population (Walther et al. 2002; Hetem et al. 2012). Reliablemeasures of activity are thus fundamental to furthering our un-derstanding of animal behaviour and ecology.Two main methodological approaches have been used to recordthe activity patterns of free-ranging animals: behavioural obser-vation and biologging (reviewed in Cooke et al. 2004; Nathan et al.2012; Ropert-Coudert et al. 2012). Instantaneous scan sampling(Altmann 1974) is the most commonly used method to record ac-tivity and behaviour data in traditional studies of animal behav-ioural ecology. Scan sampling involves human observers recordingthe activity state (i.e. resting, foraging, travelling or social) of studyanimals at predetermined time intervals, to provide a descriptivemeasure of an individual’s or group’s activity. Scan sampling allowsarangeofbehaviouralinformation(e.g.diet, posture andproximityto conspecifics), in addition to states of activity, to be recordedsimultaneously. However, scan sampling often requires studysubjects to be individually identifiable and habituated to thepresence of human observers, and data can, by definition, only becollected during those periods when human observers are present.These prerequisites for successful scan sampling mayconstrain thecollectionofactivitydata.Insomespeciesorpopulationsitmaynotbe feasible to follow animals regularly, for example, if they arenocturnal or live in an inaccessible habitat, or if the researcher’spresence disrupts normal activity patterns.Biologgers,eitherattachedexternallyorimplantedintheanimal’sbody, have become increasingly popular for recording activityremotely(Cookeetal.2004;Ropert-Coudertetal.2012).Thesedevices