Decision-makers in wildlife policy require reliable population size estimates to justify interventions, to build acceptance and support in their decisions and, ultimately, to build trust in managing authorities. Traditional capture-recapture approaches present two main shortcomings, namely, the uncertainty in defining the effective sampling area, and the spatially-induced heterogeneity in encounter probabilities. These limitations are overcome using spatially explicit capture-recapture approaches (SCR). Using wolves as case study, and non-invasive DNA monitoring (faeces), we implemented a SCR with a Poisson observation model in a single survey to estimate wolf density and population size, and identify the locations of individual activity centres, in NW Iberia over 4,378 km . During the breeding period, posterior mean wolf density was 2.55 wolves/100 km (95%BCI = 1.87-3.51), and the posterior mean population size was 111.6 ± 18.8 wolves (95%BCI = 81.8-153.6). From simulation studies, addressing different scenarios of non-independence and spatial aggregation of individuals, we only found a slight underestimation in population size estimates, supporting the reliability of SCR for social species. The strategy used here (DNA monitoring combined with SCR) may be a cost-effective way to generate reliable population estimates for large carnivores at regional scales, especially for endangered species or populations under game management.