Abstract
Wolves naturally recolonized France in the early 1990s from the Italian population, now covering all the alpine range. The highly human-dominated landscape of the Rhone valley, west of the French Alps, has slowed down the recovery of the species beyond the alpine chain. However, this barrier is still permeable to some dispersers as new packs have indeed appeared west of the river since 2019.About 3500 field experts are monitoring the wolf presence in France. This extensive network collects around 1800 biological samples per year which are then analyzed with microsatellite genotyping method. Using multi-event open capture-recapture models, population size is estimated at the end of each winter along with confidence intervals accounting for detection heterogeneity. Population size estimates are requested by authorities early spring to define the maximum number of wolves that can be culled to reduce depredation. Wolf population size in 2023 in France is expected to range between 1000 and 1210 individuals.A national action plan aims to reconcile outdoor livestock rearing and wolves, while maintaining the wolf population in a favorable conservation status. Governmental authority promotes and funds prevention measures against wolf depredation, including the use of culling. Wolf survival decreased from 0.76 to 0.68 as culling increased from 10% to 20% of the estimated population size. While Mortality remained stable for scent marking individuals (residents) the survival dropped down from 0.63 to 0.28 for transient individuals, thus being probably less impactful for the overall population dynamics. Transition probabilities from resident-to-transient showed significant higher turnover rates as culling increased, suggesting the destabilizing influence of wolf removals on pack dynamics, both producing more transients as well as a higher propensity to set up new packs.This high turnover associated with increasing pack density makes it very challenging to accurately monitor the population in space and time at the national scale, leading to greater uncertainty in population size estimates. Among the alternatives, patch occupancy models appear as a relatively cost-efficient metric, useful in adaptive management in order to detect changes in the population while also accounting for imperfect detection. We optimized these models to document changes in wolf pack dynamics across space and time in France. This metric appears reliable to identify pack occurrences and disappearance also avoiding false positives, but is less performant to identify pack disappearance in areas of high wolf density. Stakeholders and the general public often focus on the number of wolves in the country, the hardest number to document accurately at large scale. The change in wolf occupancy as materialized through dynamic maps across years combined with studies on the underlying mechanisms driving population trends and livestock damages may become robust and cost-effective indicators to monitor the conservation status of the wolf population at the national scale as well as useful for local management issues.