Behaviour 2019
Determining recapture probability of the free-ranging wood mouse using infrared thermography
Yiannis Zevgolis, Stylianos Zannetos, Triantaphyllos Akriotis. Biodiversity Conservation Laboratory, Department of Environment, University of the Aegean, Mytilene, Lesvos, Greece

In this study, we estimated, if changes in the surface temperature of Apodemus sylvaticus, during a stressful challenge in first capture occasions can sufficiently explain its recapture probability. For this purpose, A. sylvaticus individuals were handled and temporally released into a box, in order to reduce their stress. For each individual two thermal images were taken; the first during the handling challenge and the second before their release. Temperature data from six thermoregulatory regions of each animal were extracted and the temperature difference (ΔT) between handling and releasing, was calculated. To examine if individuals’ ΔT explains their recapture probability, a logistic regression (LR), was computed. Results showed that the LR model identified the ΔTeye as the most important factor which best separates recaptured from non-recaptured individuals, followed by the ΔTbody with a predicted classification accuracy of 89.5% and 73.7 respectively. This study provides a primary assessment of the importance of small mammals’ thermal window measurements in the field, in order to assess the effect of handling stress in the capture occasions.