Consistent bear population DNA-based estimates regardless molecular markers type

Abstract

Reliability of population size and density estimates is one of the most contentious issues when evaluating the conservation status of species. Non-invasive DNA monitoring, combined with spatially explicit capture-recapture approaches (SCR), is recurrently presented as a reliable procedure to achieve accurate, precise and feasible estimates. However, conservation practitioners frequently face the dilemma of which molecular marker type would perform better for their objectives. Here, using as case study the eastern subpopulation of the endangered Cantabrian brown bear, and the same dataset of non-invasive samples, we evaluated which molecular marker perform better (microsatellites vs. SNPs), and whether the selection of the marker could impact on estimates. Our results show similar posterior estimates for bear density and for the scale parameter, resulting in consistent population estimates, around 50 bears, regardless the molecular marker type selected, microsatellites: 48.4 (95%BCI = 33.8–67.3) or SNPs: 52.7 (95%BCI = 36.0–74.8). Using our dataset, we provide evidence for the importance of running several replicates for both microsatellites and SNPs when genotyping non-invasive samples, and how for the same number of molecular markers, microsatellites were statistically more powerful. The positive trend observed in small large carnivore populations requires the adaptation of monitoring approaches. The quality of indexes commonly used, such as the minimum annual counts of females with cubs of the year in our study case, may be more difficult to maintain over time as populations increase in size and range. We therefore recommend the implementation of a regular monitoring based on non-invasive DNA monitoring and SCR approaches.