Estimating invasive species elimination probability and abundance trends using multiple operational data streams

Models of species distribution and abundance are important for evaluating and planning regional management of species with heterogeneous densities. Feral swine (Sus scrofa) are a widespread and destructive invasive species in many countries, causing damage to ecosystems, agricultural resources, and property, as well as being a vector of numerous diseases that pose risk to human, livestock, and wildlife health. In the United States, the state of Missouri has a spatially heterogeneous distribution of feral swine, with some regions of moderate to high density and some regions at or near elimination. Thus evaluating management progress requires methods that can track changes in species abundance and distribution concurrently across the landscape. To address this need, we developed a joint occupancy-removal model to estimate elimination probability at a watershed scale and abundance trends at an aggregated, elimination area scale. We fit operational data collected from southern Missouri's feral swine monitoring program in conjunction with removal data from October of 2020 to December 2024. The spatial extent of feral swine elimination remained relatively constant over the study period (24%-29% of the study area having an elimination probability >0.95), and elimination areas with high amounts of feral swine removed exhibited trends in feral swine density ranging from -19.2% to 40.9% population changes within watersheds per year. Posterior mean population growth rate, excluding removals, was negatively influenced by oak tree cover and northing. Running systematic baiting stations at densities of one trap per mi² for 45 days could determine a 95% probability of feral swine elimination at the watershed level within a season, but densities of one station per two or four mi² for 45 days could not. We created a web application that allows interactive exploration of our model inference for understanding trends in occupancy and abundance and planning ongoing management. Our approach leverages multiple data types to evaluate efficacy of invasive species management across the continuum from abundance to elimination and provides a tool for prioritizing monitoring and removal efforts.