Efficacy of Proximity Loggers for Detection of Contacts Between Maternal Pairs of White-Tailed Deer

Contact frequency and duration estimates between individuals are important to understanding the behavioral ecology of wildlife species and the epidemiology of infectious diseases. A new technology uses proximity data loggers to record time and duration of contacts. We conducted an experiment at Sand...

Full description

Saved in:
Bibliographic Details
Published inWildlife Society bulletin (2011) Vol. 35; no. 4; pp. 452 - 460
Main Authors Ryan Walrath, Timothy R. Van Deelen, Kurt C. Ver Cauteren
Format Journal Article
LanguageEnglish
Published Wildlife Society 01.12.2011
Subjects
Chronic wasting disease
Deer
Disease transmission
Global positioning systems
Loggers
Observational research
Original Article
Tuberculosis
Ungulates
Wildlife ecology
Wildlife management
Online AccessGet full text

Cover

More Information
Summary:Contact frequency and duration estimates between individuals are important to understanding the behavioral ecology of wildlife species and the epidemiology of infectious diseases. A new technology uses proximity data loggers to record time and duration of contacts. We conducted an experiment at Sandhill Wildlife Management Area, located near Babcock, Wisconsin (USA) to compare probabilities of detecting intraspecific contacts among white-tailed deer (Odocoileus virginianus) maternal pairs (dams/fawns) based on detections from proximity loggers deployed on collars versus those obtained from direct observation. We defined 5 discrete probabilities of detection of a contact in terms ofP(probability of detection by a single proximity logger) andV(probability of detection by visual observer) and estimatedPandVby minimizing the Kullback–Liebler distance between distributions of theoretical probabilities and observed distributions in experimental data. We used parametric jackknifing to estimate means and variances forPandV. Mean estimates ofPandVwere 0.64 (95% CI = 0.62–0.67) and 0.34 (0.32–0.35), respectively. Estimates ofPandVenabled the calculation of the probability that an encounter was undetected by both proximity loggers and the visual observer, which was 0.09 (95% CI = 0.073–0.094). Estimates ofPandVprovide estimates of nondetection bias for future studies that use proximity loggers to estimate frequencies of encounters and help quantify the usefulness of this technology relative to visual observation. Management concerns such as chronic wasting disease and bovine tuberculosis could be better understood and addressed by using proximity loggers because they are better able to quantify close contact than conventional methods such as radiotelemetry or Global Positioning System telemetry.
ISSN:2328-5540