How to detect dormice: a comparison of footprint tunnels and camera traps as monitoring methods

Dormice (Gliridae) are hard to detect, as they are small, nocturnal, often arboreal when active and in or on the ground when hibernating. In Germany, the garden dormouse (Eliomys quercinus) and the edible dormouse (Glis glis) are afforded special protection by the German Federal Nature Conservation Act, while the hazel dormouse (Muscardinus avellanarius) is listed in Annex IV of the EU Habitats Directive and therefore strictly protected under the German Federal Nature Conservation Act. However, information on their exact geographic distribution is still incomplete. Reliable and efficient monitoring methods are therefore needed. The aim of this study was to compare the detection probability of footprint tunnels (FTs) and camera traps (CTs), taking potential effects of season into account. To achieve this, 25 FTs and two CTs were installed across 41 distinct transects and monitored from May until October between 2019 and 2022. CTs and FTs were installed to detect garden dormice, which might have influenced the performance of detection for the other two species. All three species could be detected with both methods. In the edible dormouse detection probabilities of both methods did not differ significantly and ranged around 50% and 70% for CTs and FTs, respectively. For garden dormice CTs had an almost 12 times higher detection probability than FTs, reaching detection probabilities above 90% in July and August. Hazel dormice showed an almost 34 times higher detection probability with FTs compared to CTs, with a detection probability of 95% observed in September. However, the effect of the month was not significant for either of the two methods in the three species. To survey all three dormouse species at a given site, we recommend using a combination of both monitoring methods between July and September. For the detection of garden dormice, we recommend to use CTs during July and August and for hazel dormice FTs in September. Both methods seem to work in edible dormice but showed only moderate detection probabilities.