Results from a top-ranked negative binomial regression model of high or moderate severity (Class 1–2) bear conflicts, New York, USA, July–September 2006–2019.

Covariates included mild severity (Class 3) conflicts early in the year (April–June; PRIOR), bear abundance index as a metric of relative bear population size (POP), cumulative precipitation difference from mean early in the year (PRECIP), and whether the Wildlife Management Unit (WMU) was within a park (PARK). Continuous covariates were standardized before analysis.