Comparison of spotlighting monitoring data of European brown hare (Lepus europaeus) relative population densities with infrared thermography in agricultural landscapes in Northern Germany

A successful wildlife management requires monitoring. Including non-scientific volunteers into monitoring actions is a common way for obtaining long-term and comprehensive data. Hunters present a valuable target group as they are spread out nationwide in Germany and additionally, they provide a know-how regarding game species. Since 1990s, various German hunting associations established monitoring programs and motivated hunters to join, in order to record population sizes of huntable game species under standardized census methods. The aim of this study was to compare instructed hunters performed spotlight counts of European brown hares with thermography in three federal states (Lower-Saxony, Saxony-Anhalt, North Rhine-Westphalia) in 2015–2018 in Northern Germany. Therefore, we modelled the number of hares counted by both methods with the associated observed area. Moreover, we performed repeated thermographic counts in selected areas and performed distance sampling to test the deviations of estimated population densities within a short time period. Repeated infrared thermographic counts on three consecutive nights show a coefficient of variation from 6.6% to 15.5% with deviations of 2.2–2.7 hares per 100 ha, while the method of distance sampling reveals minor deviations of 0.9–1.7 hares per 100 ha and a coefficient of variation from 3.1–7.4%. The coefficient of variation value between spotlight and infrared thermographic count lies between 0 to 21.4%. Our model confirmed no significant differences between the European brown hare density estimations based on a spotlight count and an infrared thermographic count on the following night. The results provide insight into the dimension of the error margin of density estimations performed by spotlight counts. Therefore, we recommend to take possible counting errors into account and to ideally perform repeated counts to assess the error margin for each counting site. This would help for example to quantify the uncertainty in the calculation of mortality rates. Additionally, our results show that monitoring data generated by instructed hunters can provide reliable and valid data, if implemented and conducted in a standardized scientific way.