Discussion
We provide the first assessment of bobcat density for the recovering population in Ohio using eDNA from scats and SECR analysis. Bobcat density differed between the AEP and Vinton-Zaleski study areas (17.9 ± 4.3 SE and 11.3 ± 2.9 SE bobcats/100 km2respectively). Our results support prior telemetry data which indicated that bobcats in eastern Ohio (AEP area) had smaller home-ranges than bobcats in southern Ohio (Vinton-Zaleski area) and thus could support a higher density of individuals (Prange and Rose 2020) and camera trap studies that indicated lower habitat occupancy by bobcats in the Vinton-Zaleski area (Bencin 2018; Rich et al. 2018). However, the mechanism/s driving these differences are not fully understood. Prange and Rose (2020) hypothesized that differences in home-range sizes between the two areas were a result of differences in food availability, habitat quality, and body size as other studies have shown (Litvaitis et al. 1986; Anderson 1987; Knick 1990). While Rich et al. (2018) found landscape variables had little effect on occupancy, but coyote presence had a strong negative influence on bobcat occupancy. Bobcats and coyotes are sympatric throughout most of North America and interference competition between the two predators is rare (Dyck et al. 2022). However, bobcats and coyotes have only recently begun to co-occur in the US Midwest (Deems and Pursley 1978; Woolf and Hubert 1998). The concomitant recolonization by bobcats (Reding et al. 2012) and range expansion by coyotes (Hody and Kays 2018) represents a unique situation as species are co-occurring that have shared evolutionary histories but have not coexisted in recent time. While our study provides additional support for regional differences in bobcat density and abundance, it does not address the mechanisms driving these differences, and additional research into this topic is needed and would provide useful insights for bobcat management.
When comparing density estimates across the bobcat range from SECR analyses, our estimates for AEP are at the higher end of those reported elsewhere, while estimates from Vinton-Zaleski are within the average (Clare et al. 2015; Thornton and Pekins 2015; Rounsville Jr 2018; Morin et al. 2018; Jacques et al. 2019; Greenspan et al. 2020). However, when comparing SECR density estimates for another recovering Midwest bobcat population, the estimates from both study areas are much higher than those reported for Illinois (1.4 individuals/100 km2; Jacques et al. 2019). There are several factors that may explain differences in bobcat density between our study and Illinois.
First, there are distinct differences in the type and distribution of habitat between the two studies. Our study areas are characterized primarily by forested habitat (60.1% - 92.7%) with minor areas of developed land (2.8% - 3.8%; Figure 1), while the study site in Illinois was dominated by agriculture (53.2%) and pasture-hay (12.6%) with a lesser-degree constituting forest (27.3%; Jacques et al. 2019). Prior investigation into habitat suitability for the bobcat population in Ohio indicated that bobcats select for forest habitat, natural herbaceous vegetation habitat, and areas of low road density (Popescu et al. 2021). Bobcat habitat suitability is also highest in southern and southeast Ohio (Popescu et al. 2021), given our estimates are from high suitability habitat it is likely that they represent the higher end of density for the state.
Second, field methods between the two studies differed. Jacques et al. (2019) used infrared-triggered cameras which are a common non-invasive tool that can be used with capture-mark-recapture methods to estimate density and abundance of bobcats and other felid species due to their unique recognizable pelage patterns (Karanth and Nichols 1998; Silver et al. 2004; Greenspan et al. 2020; Iosif et al. 2022). However, this method requires quality images of both flanks of an animal and sufficient variation in pelage/individual markings for observers to accurately identify individuals. Jacques et al. (2019) deployed cameras for 77 days and had 139 unique bobcat events but had to discard 18.7% due to low image quality and a small subset (6.2%) of the remaining images were classified as tentative identifications. Bobcat pelage varies across their range (Young 1978; Croteau et al. 2012) and some areas are known to have individuals with less distinct markings (Morin et al. 2018).
Third, harvest of bobcats (hunting or trapping) is currently not permitted in Ohio, while Illinois has had regulated hunting or trapping season for bobcats since 2016 (Illinois DNR). The protected status of bobcats in Ohio has played a role in their recovery and may act as a mechanism for ‘mesopredator release’ allowing population densities to exceed those in states without protection. Legal harvest accounts for a large portion of bobcat mortality in exploited populations (Rolley 1985; Knick 1990; Fuller et al. 1995; Chamberlain et al. 1999; Blankenship et al. 2006) and therefore if harvest mortality is additive with other sources of mortality it is expected that population size would be higher for unexploited populations such as Ohio.