Conserving viable large carnivore populations requires managing their interactions with humans in increasingly anthropogenic landscapes. Faced with declining budgets and escalating wildlife conflicts, agencies in North America continue to grapple with uncertainty surrounding the efficacy of socially divisive management actions such as harvest to reduce conflict. We used multistate capture–reencounter methods to estimate cause-specific mortality for a large sample (>3500) of American black bears Ursus americanus in north-western New Jersey, USA over a 33-year period. Specifically, we focused on factors that might influence the probability of bears being harvested, lethally managed, or dying from other causes. We further analysed temporal correlations between >26 000 human–black bear incidents reported between 2001 and 2013 and estimates of total mortality rates, and specifically, rates of harvest from newly implemented public hunts and lethal management. Adult females were twice as likely (0·163 ± 0·014) as adult males (0·087 ± 0·012) to be harvested during the study period. Cubs (0·444 ± 0·025) and yearlings (0·372 ± 0·022) had a higher probability of dying from other causes, primarily vehicle strikes, than adults (0·199 ± 0·008). Reports of nuisance behaviours in year t + 1 declined with increasing mortality resulting from harvest plus lethal management in year t (P = 0·028, R2 = 0·338). Adult bears previously designated as a nuisance and/or threat were more likely to be harvested (0·176 ± 0·025) than those never identified as a problem (0·109 ± 0·010). Across age classes, individuals assigned problem status, were significantly more likely to be lethally controlled. Synthesis and applications. Given continuing failures in conserving exploited carnivores, their recreational harvest and lethal management remain polarizing. Within this social-ecological system, the well-regulated harvest of carefully monitored black bear populations represents a pragmatic approach to achieve population objectives. Furthermore, the integration of harvest and incident-response management (both lethal and non-lethal practices) with educational programmes aimed at reducing anthropogenic attractants can result in subsequent reductions in problem behaviours reported.

NJ black bear encounter histories for multistate modelsIndividual black bear capture histories from northwestern NJ between 1981-2014. States are: A=Alive; H=Harvested; M=Management Mortality; D=Dead All Other. Class at first capture: c=cub (0-1); y=yearling (1-2); a=adult (2+). Sex: f=female; m=male. Behavior: b=normal bear; p=nuisance or threatening bear.Raithel-JAPPL-2016-00639_bear_encounter_histories.txt