Evaluating population resilience to anticipated stressors using integrated population modeling: a case study of Peregrine Falcons

Reliable estimates of demographic parameters are fundamental to understanding population dynamics and guiding conservation efforts. Integrated population models (IPMs) provide a powerful framework for jointly analyzing diverse data sources to estimate demographic rates and population trajectories, evaluate resilience to environmental stressors, and project population dynamics info the future. We applied a Bayesian IPM to assess the demographic performance and long-term viability of a Peregrine Falcon population that recently recolonized Hungary after regional extinction in the 1960s. Using 27 years (1997–2023) of count, productivity, and mark–recapture–recovery data, we estimated age-specific survival, productivity, and recruitment rates, and used the fitted IPM to conduct stochastic population viability analysis (PVA) to explore how plausible demographic perturbations (e.g., arising from predation, habitat alteration, or climatic variability) might influence persistence. Annual adult survival averaged 0.79 (95% CRI: 0.73–0.85), juvenile survival 0.50 (0.35–0.69), and productivity 1.9 fledglings per breeding attempt (1.44–2.53), yielding a mean national, annual population growth rate of λ = 1.21 (0.81–2.07). The recruitment probability of two-year-old females was estimated at (0.76; 0.39–0.98). Scenario-based PVAs showed that the population growth rate responded more strongly to changes in adult survival than to equivalent changes in juvenile survival and productivity. They also revealed that moderate declines in adult survival (<0.70) or productivity (<1.3 fledglings per pair) had the potential to push λ below unity, underscoring the need to maintain both high survival and productivity. By linking hierarchical modeling with stochastic forecasting, we demonstrate how an IPM can be extended into a PVA framework to evaluate demographic resilience and identify quantitative thresholds associated with population vulnerability. This approach provides a transparent, generalizable workflow for translating long-term monitoring data into predictive tools for assessing population vulnerability.