Non-invasive genetic population density estimation of mountain hares (Lepus timidus) in the Alps: systematic or opportunistic sampling?

Abstract

The development and evaluation of a reliable non-invasive genetic sampling (NIGS) is a crucial step towards accurately and reliably estimating population size for the long-term monitoring of wildlife species. We used NIGS data to obtain population density estimates of a mountain hare (Lepus timidus) population in the Swiss Alps. We evaluated and compared the effectiveness of systematic and opportunistic NIGS and their combination in spring 2014. Extraction success rate of DNA from faeces, hair and urine samples, their age-dependent variation as well as the completeness of microsatellite genotyping data were used as measures of effectiveness. We applied a spatially explicit capture-recapture (SECR) approach to estimate the minimum population size. We found that the extraction success of faecal samples decreased with the time since excretion and that urine and hair samples mostly yielded insufficient DNA for the successful genotyping of individuals. Mountain hare faeces up to 5 days old are most appropriate for NIGS because the risk of unsuccessful DNA extraction or genotyping errors/failure is considerably lower in these samples. Systematic sampling revealed more genotypes than opportunistic sampling, but the latter resulted in higher numbers of recapture and thus, increased the spatial resolution of the data. Depending on the sampling design, the population density estimates ranged from 3.2 to 3.6 mountain hares per 100 ha. This study informs ecologists and wildlife managers about suitable survey techniques for the monitoring of free-ranging lagomorph populations and addresses important principles for the development of accurate survey methods for other elusive wildlife species that inhabit difficult, mountainous terrain.