Mihai Pop, Ruben Iosif, Viorel Popescu

Noninvasive genetic monitoring is a powerful tool for conservation and wildlife management and has been growing in popularity as a relatively cost-effective and accurate method of estimating population sizes. Particularly, noninvasive genetic monitoring using scat and hair samples has become one of the primary methods of studying bear populations. This study conducts a meta-analysis of noninvasive genetic monitoring studies in brown bear (Ursus arctos) and black bear (Ursus americanus) populations over the past 10 years from across their ranges in North America, Europe, and Asia. Data were extracted from 108 studies on sampling methods, study area, number of samples analyzed and successfully genotyped to an individual bear, number of individuals identified, sex ratios, and population estimates. Preliminary results indicate that genotyping success rates tended to be higher for hair than for scat (medians = 62.50% and 52.83% genotyped, respectively), with scat sampling increasing in genotyping success over time, but hair sampling remaining relatively constant. Studies were inconsistent in how or if recapture rates were reported, but comparison of numbers of genotyped samples and individuals identified showed that studies using scat samples tended to identify more unique individuals relative to the number of genotyped samples. Of the studies estimating population size or density, 30 used spatially explicit capture-recapture models, while 27 used non-spatial models. Population density estimates varied widely between 23-840 bears per 1000 km^2 for black bears and 0.7-186.6 bears per 1000 km^2 for brown bears. This analysis can serve as a baseline to compare with future studies to evaluate the plausibility of density estimates and improvements in the efficacy of genetic monitoring methods for brown and black bears.