Matt Hewitt, Dana Karelus (TPWD), Amanda Veals Dutt, Louis Harveson, Russell Martin (TPWD), and Patricia Moody Harveson

Kit fox (Vulpes macrotis) and coyotes (Canis latrans) are known to both occur together in the grasslands of far west Texas. While it may appear that the two canids are happily coexisting, coyotes have been reported as a significant source of predation related mortality for kit fox elsewhere in the country. This, along with anecdotal evidence of a declining kit fox population in the area, spurred researchers and managers alike to tease out how these two carnivores are interacting on the West Texas landscapes. With this study, we aim to better understand the co-occurrence relationships between these species by identifying what effect, if any, coyotes are having on kit fox space use. To do this, we collected detection/non-detection data for the 2 species from camera traps at 732 baited survey sites, distributed across suitable kit fox habitat on public and private lands. Surveys resulted in 313 kit fox detections, and 1,430 coyote detections over a 2-year sampling period (March 2018–March 2020). We employed conditional two species occupancy models and determined coyote presence has no meaningful effect on kit fox probability of using a site, or our ability to detect them at that site if they were present. The two species also have a significant overlap in their temporal activity patterns, with kit fox being exclusively nocturnal and coyotes being primarily nocturnal with some crepuscular activity. These results tell us that coyotes are not displacing kit fox spatially and the two canids are not separating temporally. Co-existence is happening, and we suspect the drivers cannot be detected at the broad spatial and temporal scale of our surveys. Little is known about kit fox in Texas, and wide scale declines across their range underscores the need for further research to better understand the kit fox population in Texas. Our work is some of the first conducted on this population of kit fox and is a first look at important interspecific interactions with a co-occurring competitor and predator.

We conducted camera trap surveys at 772 baited sites from March 2018 to March 2020, with each survey lasting 14 days on average (11,205 trap-days). We captured 339 one-hour independent kit fox photos at 105 camera survey sites. We divided surveys into 3-day sampling occasions and fit single season occupancy models and tested for an effect of sampling occasion on detection and for additive effects of slope, elevation, shrub height, and shrub cover on occupancy.

Overall detection ± standard error was 0.46 ± 0.04 and was influenced by sampling occasion, whereby detection was highest on the first occasion and decreased afterwards. Average overall occupancy for surveys was 0.16 ± 0.01 (range: 0.00–0.82) and was influenced by all four environmental covariates, with greatest occupancy where there were shallower slopes, lower elevations, and shorter shrubs covering less of the area. We used the model to make a predictive map of occupancy across the Trans-Pecos region and overlaid kit fox sightings from other sources.