Animals’ fear of people is widespread across taxa and can mitigate the risk of human-induced mortality, facilitating coexistence in human-dominated landscapes. However, humans can be unpredictable predators and anthropogenic cues that animals perceive may not be reliable indicators of the risk of being killed. In these cases, animal fear responses may be ineffective and may even exacerbate the risk of anthropogenic mortality. Here, we explore these questions using a 10-year dataset of movement and mortality events for the puma (Puma concolor) population in the fragmented Santa Cruz Mountains of California, for whom the leading cause of death was retaliatory killings by people following livestock loss. We modeled retaliatory killing risk and puma habitat selection relative to residential housing density to evaluate whether puma avoidance of human cues reflected their risk of being killed. We documented a mismatch between human cues, fear responses, and actual risk. Rather than scaling directly with housing density, retaliatory killings occurred at intermediate levels of human development and at night. Pumas avoided these areas during the day but selected for these high-risk areas at night, resulting in a mismatch between cue and risk impacting 17% of the study area. These results are unlikely to be driven by puma hunting behavior: livestock constitute a very small proportion of puma diets, and we found no evidence for the alternative hypothesis that state-dependent foraging drove depredation of livestock and subsequent retaliatory killings. Our findings indicate that puma responses to human cues are not sufficient to enable human-carnivore coexistence in this area and suggest that reducing risk from humans in places with few perceptible human cues would facilitate carnivore conservation in human-dominated landscapes. Furthermore, a mismatch between human cues and responses by carnivores can lead to selection rather than avoidance of risky areas, which could result in an ecological trap. Methods We captured adult and subadult pumas from 2009-2019 and fit pumas with GPS collars set to record a GPS location every 4 hours. For animals that died during the study, we recorded date, location, and cause of death. We 1) modeled overall and cause-specific mortality rates using the Kaplan-Meier procedure and non-parametric cumulative incidence functions, 2) modeled habitat selection using step selection functions, 3) modeled spatial predictors of the distribution of sites where retaliatory killings occurred, and 4) compared time since last predicted deer kill and body weights between animals killed after depredating livestock and the overall puma population. Included in this data product are the time-to-event data (tte_cause.csv) for analysis 1, covariate information for 4-hour puma locations (ssf_data.csv) for analysis 2, covariate information for locations of retaliatory killings (rk_rsf_data.csv) and other sources of mortality (other_mort_rsf_data.csv) for analysis 3, body weight data (body_weight_data.csv) and kill rate data (times_since_last_kill.csv and kill_rates.csv) for analysis 4. Please see Nisi AC, Benson JF, Wilmers CC. 2022. Puma responses to unreliable human cues suggest an ecological trap in a fragmented landscape. Oikos: 10.1111/oik.09051 for a full description of data collection methods, and the README file for detailed descriptions of each dataset.

This data breaks down estimated hunters as well as antlered, antlerless and total harvest numbers by: * wildlife management unit (WMU) * calendar year Harvest and active hunter numbers are estimates based on replies received from a sample of resident hunters and are therefore subject to statistical error. Additional technical and statistical notes can be found in the data dictionary.

Second-order habitat selection is influenced by a variety of factors, including individual- and species-specific traits and resource requirements, as well as landscape characteristics. By comparing home range characteristics across individuals, species, and landscapes, we can draw conclusions regarding whether and how different factors influence home range selection. Our objectives were to quantify home range characteristics of VHF- and GPS-collared coyotes and red foxes in urban and rural areas of southern Wisconsin, including home range size and shape, home range stability, and inter- and intraspecific overlap. On average, urban coyotes had smaller home ranges with apparently greater intraspecific overlap between neighboring individuals than rural coyotes. Similarly, urban red foxes had smaller home ranges with apparently greater intraspecific overlap between neighboring individuals than urban coyotes. We found no difference in home range boundary complexity or stability between urban coyotes and red foxes or between urban and rural coyotes. We did identify greater interspecific overlap between urban coyotes and red foxes than has been previously reported. Our results provide further evidence that intrinsic and extrinsic factors, such as species characteristics, resource predictability, and availability as well as the physical environment, influence home range selection of coyotes and red foxes. Methods

Data collection – Urban study area

We captured and placed Very High Frequency (VHF) or Global Positioning System (GPS) collars on urban coyotes and red foxes beginning with a pilot study in 2014 as part of the University of Wisconsin Urban Canid Project (UCP). We captured coyotes and red foxes in our urban study area annually between October and March using cable restraints while following trapping best management practices.

We selected trap sites based on landscape characteristics such as greenspace and sightings reported to the project’s iNaturalist page (“UW Urban Canid Project iNaturalist Project”), and we used bait including carcasses of road-killed deer for coyotes and nuisance-trapped beavers for red foxes to attract canids to a site, especially when there was minimal snow cover. We chemically immobilized captured canids with an intramuscular injection of 4-10 mg/kg ketamine and 2-4 mg/kg xylazine based on estimated weight of the individual. We then determined weight and physical condition, collected biological samples, and fitted the animal with either a VHF radio collar (2014 and 2015 capture seasons; Advanced Telemetry Systems, Isanti, MN; Model M1950 for red fox and M2220B for coyote) or a Lotek LiteTrack Iridium GPS collar (2016 through 2022 capture seasons; Model #360 for coyotes and #150 for red foxes). After handling, we reversed the immobilization with an intramuscular injection of either yohimbine (0.1 mg/kg) or antisedan (0.6-0.7 mg/kg). We then released the animal at the capture site.

Location schedules varied with collar type (VHF or GPS). For individuals with VHF collars, we located each radio-collared individual weekly using a five-hour bout, with a triangulated location recorded once every hour, during each bout, for as long as the VHF collar was active and the individual was alive. We rotated tracking bouts across the entire diel cycle to capture variation in temporal activity. For individuals with GPS collars, we programmed each collar to collect GPS fixes every hour only between 9pm and 4am for coyotes and 1am and 4am for red foxes. Species-specific time periods were selected to maximize locations during periods of high activity in urban areas and optimize battery life of the telemetry collars. Individuals were tracked until death or the end of the collar’s battery life.

Data collection – Rural study area

The Wisconsin Department of Natural Resources (WDNR) collected GPS data from coyotes captured in Iowa County as part of an independent research project on deer and their predators. The WDNR started this project in fall of 2016. Coyotes were captured using cable restraints or foothold traps either through collaboration with trappers and landowners who would voluntarily report a captured coyote or through traps set by WDNR staff. Captured coyotes were anesthetized via an injection of ketamine-dexmedetomidine-butorphanol (4 mg/kg ketamine, 0.2 mg/kg dexmedetomidine, and 0.4 mg/kg butorphanol) based on estimated weight. Staff then weighed and fitted each animal with a Lotek LiteTrack GPS collar (Model #360) which they programmed to collect a GPS fix every three hours throughout the 24-hour diel cycle. Individuals were tracked until death or the end of the collar’s battery life.
Home range delineation Using radio-location data from both study sites, we calculated home range (95%) size for individual coyotes and red foxes using Minimum convex polygons (MCP) and Fixed kernel density estimators (KDE) with the amt R package. We chose to retain all locations for each individual rather than thinning data to standardize the tracking schedules between urban and rural individuals and VHF-collared and GPS-collared individuals. There were several individuals in the dataset with relatively few location points (n ≤ 50), so we used area-observation curves to determine whether each individual had sufficient data to reach an asymptote, and excluded individuals with too few locations. Similarly, to ensure that dispersing individuals and transients did not artificially inflate average home range sizes, we excluded individuals with large data sets whose home range area also failed to reach a stable asymptote as determined by area-observation curves.

Number and rate (per 100,000 population) of homicide victims, Canada and Census Metropolitan Areas, 1981 to 2023.