Estimates of density, occupancy and activity patterns of wild-living cats in Foreste Casentinesi National Park (Italy) using data from a non-targeted camera array

Abstract

The European wildcat (Felis silvestris) is an elusive felid legally protected in Europe under the Appendix II of CITES, yet ecological knowledge for the Italian populations remains scant. Exceptions are a few studies that estimated population density in Southern Italy and the Alps. Here we estimated the population density of wild-living cats, i.e. wildcats, domestic cat (Felis catus) and their putative hybrids (Felis silvestris x catus; hereafter ‘hybrid’), while also investigating their occurrence and activity patterns in the Foreste Casentinesi National Park (Tuscany-Emilia Romagna), Italy. By using a camera-trapping grid with 59 cameras spread over 355.87 km2 from 31 August to 18 November 2020, we collected 180 wild-living cat detections, which we classified based on coat pattern as 65 wildcat, 23 domestic cats and 32 hybrid detections at 25, 4 and 10 stations respectively; we individually identified 15, 3 and 9 wildcats, domestic cats and hybrids, respectively. The most supported multi-session SECR model estimated the population density of wildcats, domestic cats and hybrids to be 3.09 (95% CI = 1.85 — 5.17), 0.61 (0.19 — 1.92) and 1.85 (0.96 — 3.59) individuals/100 km2, respectively. Basal detection rates g0 were affected by camera placement(road vs. trail), with higher values for the trails irrespective of the type of cats. The most supported multi-species occupancy model estimated the occupancy of cats to be 0.66 (95% CI = 0.40 — 0.84), 0.07 (0.02 — 0.28) and 0.29 (0.12 — 0.55) (95% CI = 0.40 — 0.84; 0.02 — 0.28; 0.12 — 0.55) for wildcats, domestic cats and hybrids, respectively. Distance to municipalities negatively affected not only domestic cats’ occupancy, but also the detection probability of all cat types whereas cameras located on trails resulted in lower detection probabilities for wildcat, but higher for domestic cats. The activity level of cats was relatively low, being estimated at 0.33, 0.48 and 0.35 (95% CI = 0.22 — 0.40: 0.24 — 0.59; 0.23 — 0.43) for wildcats, domestic cats and hybrids, respectively. Diel activity largely overlapped between cat types with an average estimate of 0.76. and nocturnal activity was the most supported diel niche model using the Maximizing Hypothesis set, irrespective of cat type. Although our camera array was not established specifically to monitor wildcats, it provided a novel baseline of ecological estimates for the wild-living cats within this protected area. Our results stress the importance of limiting the number of domestic cats that roam freely within this protected area of wildcat range.