Patterns of exposure of coyotes to anticoagulant rodenticides in California, USA

Abstract Secondary exposure to anticoagulant rodenticides (ARs) causes the death of mammalian predators and scavengers directly and indirectly through sublethal effects that reduce fitness. Poisoning by ARs has been proposed to be a significant source of mortality for coyotes ( Canis latrans ), a medium‐sized canid that thrives at the urban–wildland interface and may prey upon species targeted by pest control efforts. However, only 1 study, with a relatively small sample size, documented the prevalence of AR exposure in a free‐roaming coyote population. We quantified AR exposure in carcasses of 365 urban and suburban coyotes in southern California, USA, and compared AR prevalence and hepatic residue concentrations to those of 120 rural coyotes collected elsewhere in the state. For urban coyotes, we also examined demographic (sex, age, body mass, cause of death) and environmental factors (season, degree of urbanization, diet) that could influence the number of AR compounds and residue concentrations. Nearly all urban coyotes (98.1%) were exposed to at least 1 AR, compared to 41.7% of rural coyotes, and most individuals had residues of both first‐generation (FGAR) and the more potent second‐generation (SGAR) compounds, often at concentrations exceeding thresholds considered lethal in other mammals. Anticoagulant rodenticide exposure of urban coyotes did not vary by sex or season, but the number of compounds detected increased with mass, and adults tended to have residues of more compounds and at higher concentrations than juveniles, suggesting repeated and chronic exposure. Livers of road‐killed coyotes had higher SGAR concentrations than those euthanized as nuisance animals, which had lower SGAR concentrations in intensively urbanized areas. Concentrations of SGAR and FGAR residues were highest in suburban areas with natural open space and lower intensity development, and stable isotope values suggested that these coyotes were exposed to ARs by consuming commensal rodents and possibly mesocarnivores. In contrast, coyotes from urbanized areas had lower AR concentrations possibly because less AR is applied in these settings or because coyotes consumed foods with less AR, such as domestic cats and anthropogenic resources. Although some coyotes showed evidence of internal bleeding consistent with AR toxicosis and were in poorer body condition, there was no clear relationship between the extent of hemorrhaging and AR exposure. Despite statewide legislation to restrict their use and mitigate non‐target impacts, AR exposure remains ubiquitous in southern California and represents another stressor of urban life to which coyotes have successfully adjusted, making them a potential sentinel of environmental contamination.