Population modelling to assess supplementation strategies for the European pond terrapin Emys orbicularis in Liguria

We present the results of modelling the supplementation program for Emys orbicularis in Liguria. We evaluated three ossible alternative strategies for the reintroduction, releasing 3-, 4- or 5-year-old turtles. In particular, we wanted to assess the expected population sizes that could be achieved by these release strategies, given environmental and demographic stochasticity, and the possibility that captive-bred individuals, when released, suffer greater mortality than those born in the wild. We built a stage-structured model for a reintroduced population and parameterised it using published and unpublished information. We compared the outcomes of population viability analyses for releases of 3-, 4- or 5-yr old turtles, explicitly accounting for uncertainty in the estimated parameters and investigating the effect of an increased mortality in the year after release. Assuming post-release effects would affect all life stages equally, releasing 5-yr old turtles was always the most effective option, with the highest predicted number of mature individuals in the wild population after 20 and 50 years. However, releasing 3- and 4-yr-old turtles was also predicted to provide positive results, and may prove a cheaper strategy since it requires a smaller captive population. In the event that post-release survival has a greater impact on older individuals, their release may become less advantageous and even sub-optimal. Therefore, future monitoring and analysis should concentrate on resolving the uncertainty for this parameter, since it is the most likely to affect management decisions and outcomes.