Counting the Capital's cats: Estimating drivers of abundance of free‐roaming cats with a novel hierarchical model

Abstract

Free-roaming cats are a conservation concern in many areas but identifying their impacts and developing mitigation strategies requires a robust understanding of their distribution and density patterns. Urban and residential areas may be especially relevant in this process because free-roaming cats are abundant in these anthropogenic landscapes. Here, we estimate the occupancy and density of free-roaming cats in Washington D.C. and relate these metrics to known landscape and social factors. We conducted an extended camera trap survey of public and private spaces across D.C. and analyzed data collected from 1483 camera deployments from 2018 to 2020. We estimated citywide cat distribution by fitting hierarchical occupancy models and further estimated cat abundance using a novel random thinning spatial capture-recapture model that allows for the use of photos that can and cannot be identified to individual. Within this model, we utilized individual covariates that provided identity exclusions between photos of unidentifiable cats with inconsistent coat patterns, thus increasing the precision of abundance estimates. This combined model also allowed for unbiased estimation of density when animals cannot be identified to individual at the same rate as for free-roaming cats whose identifiability depended on their coat characteristics. Cat occupancy and abundance declined with increasing distance from residential areas, an effect that was more pronounced in wealthier neighborhoods. There was noteworthy absence of cats detected in larger public spaces and forests. Realized densities ranged from 0.02 to 1.75 cats/ha in sampled areas, resulting in a district-wide estimate of ~7296 free-roaming cats. Ninety percent of cat detections lacked collars and nearly 35% of known individuals were ear-tipped, indicative of district Trap-Neuter-Return (TNR) programs. These results suggest that we mainly sampled and estimated the unowned cat subpopulation, such that indoor/outdoor housecats were not well represented. The precise estimation of cat population densities is difficult due to the varied behavior of subpopulations within free-roaming cat populations (housecats, stray and feral cats), but our methods provide a first step in establishing citywide

Estimating wolf (Canislupus) densities using video camera traps and spatial capture–recapture analysis

“…However, if the subsampling process that occurs when identifying individuals from the videos is not random (e.g., if some individuals are more identifiable than others), individual heterogeneity will be introduced in $$$ {\uplambda}_0 $$$. Where this problem is suspected, it can be addressed by removing individual heterogeneity in detection by combining the random thinning spatial capture–recapture model (Jiménez et al, 2021) with the categorical SCR model (Augustine et al, 2019) as described by Cove et al (2023). Another problem could arise if there were behavioral responses to the traps, because we cannot be sure whether some of the wolves detected (and identified) in one event might actually be present, but unidentified, in a previous video event.…”

“…Assuming that cohesiveness in wolf packs is related to their ability to kill larger prey and improve their competitive ability against scavenging animals (Schmidt & Mech, 1997; Vucetich et al, 2004), we would predict that pack cohesion in our study area would be medium/low, given that the main food source in the area consists of small prey and medium‐sized domestic livestock. The effects of overdispersion should therefore be minimal according to Bischof, Dupont, et al (2020) and could be addressed using random effects (Harrison, 2014; Kéry & Schaub, 2012, p. 82) or NB observation models (Cove et al, 2023).…”

“…We assumed that the event histories of the N observed individuals followed a Poisson (P) (2) or negative binomial (NB) (3) probability distribution of counts. The latter is considered adequate to describe aggregate event patterns (Anscombe, 1950) and has been previously used by Cove et al (2023) in SCR observation models to deal with overdispersion in detection.…”

“…Where there is no variation in K, we aggregate the observed number of encounters in each trap over the K sample occasions for more efficient computation (the sum of NB random variables is NB with size ¼ size 1 + size 2 + Á Á Á + size n ) simplifying the detection model (2 and 3) as in Cove et al (2023). Pr z i ¼ 1 ð Þ¼1 for the n observed individuals, and z i $ Bernoulli ψ ð Þ for the entire collection of M individuals.…”

Effect of species‐level trait variation on urban exploitation in mammals

“…However, we included domestic cats (Felis catus) in our analysis. Though domestic cats tend to associate with human-occupied landscapes (e.g., Cove et al, 2023), we included them in our analyses due to their unique ecological influence, position as free-roaming throughout much of the USA (Cove et al, 2018(Cove et al, , 2023Loss et al, 2013), and frequent presence in wildland areas, as well as across the rural-to-urban gradient (Crowley et al, 2020;Herrera et al, 2022;Lepczyk et al, 2003). Nonetheless, to confirm that our results were not significantly influenced by the inclusion of domestic cats, we conducted a secondary analysis of our dataset excluding domestic cats and found nearly identical results and unchanged model interpretations when cats were included versus excluded.…”