Genetic structure and assignment tests demonstrate illegal translocation of red deer (<i>Cervus elaphus</i>) into a continuous population

Molecular forensic methods are being increasingly used to help enforce wildlife conservation laws. Using multilocus genotyping, illegal translocation of an animal can be demonstrated by excluding all potential source populations as an individual's population of origin. Here, we illustrate how this approach can be applied to a large continuous population by defining the population genetic structure and excluding suspect animals from each identified cluster. We aimed to test the hypothesis that recreational hunters had illegally introduced a group of red deer into a hunting area in Luxembourg. Reference samples were collected over a large area in order to test the possibility that the suspect individuals might be recent immigrants. Due to isolation-by-distance relationships in the data set, inferring the number of genetic clusters using Bayesian methods was not straightforward. Biologically meaningful clusters were only obtained by simultaneously analysing spatial and genetic information using the program baps 4.1. We inferred the presence of three genetic clusters in the study region. Using partial Mantel tests, we detected barriers to gene flow other than distance, probably created by a combination of urban areas, motorways and a river valley used for viticulture. The four focal animals could be excluded with a high certainty from the three genetic subpopulations and it was therefore likely that they had been released illegally.