Lessons from the use of non-invasive genetic sampling as a way to estimate Eurasian otter population size and sex ratio

Given the difficulties in establishing population parameters of elusive animals in the wild by traditional methods, such as trapping, much attention has been given in recent years to non-invasive genetic sampling. Our work compared estimates of population size and sex ratio derived from genetic sampling with the known number and sex of animals released during an otter reintroduction and reports on the pitfalls and opportunities that may be encountered in studies of this kind. This study makes use of 121 samples of otter spraints (faeces) collected over 7 months during a reintroduction in the Upper Thames (UK) where a total of 17 otters was released in two consecutive phases. Spraints were processed with a multiple tubes approach and seven microsatellites were used. Of all collected samples, 19 % were complete for at least five loci, the minimum required for discrimination between individuals. Six out of nine of the otters that were released in the first phase were detected, four males and two females, while none of the otters released in the second phase was detected probably due to a combination of sampling pitfalls and otter behaviour. In particular, the specific sex (mostly females) and dominance composition (lower) of the individuals in the second release group may explain our failure to detect individuals in this group. Taken together, our results add further evidence that genetic sampling approaches represent a potentially accurate and non-invasive route to census populations of otters but that the sampling design should take into account factors like the sex ratio and dominance composition of the population in order to maximise detection and minimise error.