Merging integrated population models and individual-based models to project population dynamics of recolonizing species

Abstract

Recolonizing species exhibit unique population dynamics, namely dispersal to and colonization of new areas, that have important implications for management. A resulting challenge is how to simultaneously model demographic and movement processes so that recolonizing species can be accurately projected over time and space. We introduce a framework for spatially explicit projection modeling that harnesses the rigorous parameter estimation made possible by an integrated population model (IPM) and the flexible movement modeling made possible by an individual-based model (IBM). Our framework has two components: [1] a Bayesian IPM-driven age- and state-structured population model that governs the population state process and estimation of demographic rates, and [2] an IBM-driven spatial model that allows for the projection of dispersal and habitat colonization. We applied this model framework to estimate current and project future dynamics of gray wolves (Canis lupus) in Washington State, USA. We used data from 74 telemetered wolves and yearly pup and pack counts to parameterize the model, and then projected statewide dynamics over 50 years. Mean population growth was 1.29 (95 % Bayesian Credible Interval = 1.26–1.33) during initial recolonization from 2009 to 2020 and decreased to 1.02 (95 % Prediction Interval = 0.98–1.04) in the projection period (2021–2070). Our results suggest that gray wolves have an ~100 % probability of colonizing the last of Washington State's three specified recovery regions by 2030, regardless of alternative assumptions about how dispersing wolves select new territories. Our spatially explicit projection model can be used to project the dynamics of any species for which spatial spread is an important driver of population dynamics.