Challenges in the use of genetic mark-recapture to estimate the population size of Bwindi mountain gorillas (Gorilla beringei beringei)

Monitoring the population dynamics of endangered species is a critical component of conservation management strategies, but attaining accurate and precise estimates of population sizes using cost and time effective methods can be challenging. Routine censuses of the two populations of critically endangered mountain gorillas (Gorilla beringei beringei) have been conducted over the last decades to monitor populations and evaluate the effectiveness of conservation strategies. A census in 2006 of the mountain gorillas in Bwindi Impenetrable National Park, Uganda, showed the value of genetic analysis of fecal samples collected at nest sites by revealing discrepancies between the numbers of nests and uniquely identified gorillas. In this study, we censused the Bwindi gorilla population using a 'mark-recapture' method which involved genetic analysis of fecal samples collected in 2011 during two 'sweep' surveys of the entire park. We found that a notable proportion of gorillas were missed in either of the two sweeps (minimum 35% and 31%, respectively). Based on the number of genotyped gorillas and correction factors, we estimated the population to contain a minimum of 400 individuals. Using the mark-recapture approach, we infer possibly as many as 430 gorillas (95% confidence interval: 398-487). As the 2010 census of the Virunga Massif population found 480 gorillas, the total number of mountain gorillas worldwide is at least 880 individuals. Simulations using different mark-recapture models suggest that a future census of Bwindi mountain gorillas would benefit by increasing the number of sweeps in order to achieve accurate and precise results. Finally, based on our results, we recommend a sequential approach incorporating a pilot study and simulations for optimizing time and resources in large mammal genetic census studies.