Predicting human-wildlife interaction in urban environments through agent-based models

Abstract

Synurbic species adapt to global urbanization by increasingly inhabiting urban environments, where social and ecological factors, such as anthropogenic food resources and habitat alterations, promote close human-wildlife interactions. Ineffective management of these interactions can result in conflicts, altered animal population dynamics, and increased public and private expenditures. This study presents the Barcelona wild boar (BCNWB)-prototype model, a spatially explicit, incremental agent-based simulation that captures interactions between citizens and wild boar (Sus scrofa) agents in fine-scale GIS-based scenarios in Barcelona. Developed using GAMA software, the model's results were analyzed with QGIS and R software. The model aims to simulate the dynamics of the social-ecological system underlying the urban ecosystem use by synurbic wild boars and their interactions with humans in the (peri)urban area of Barcelona, Spain.The BCNWB-prototype model demonstrated high accuracy in predicting the magnitude and location of wild boar movements (multiple-resolution-goodness-of-fit = 0.73) compared to reported wild boar presences in Barcelona. The model also forecasted 115 attack events and 1,442 direct feeding events during a one-year simulation period, as compared to the actual 150 attacks and 1,858 feeding events reported annually. The model’s strong performance highlights its potential as a predictive tool for identifying priority areas for human-wild boar interactions and conflicts. Additionally, the model could be employed to assess the cost-effectiveness of management strategies and evaluate the spread, transmission risks, and public health implications of pathogens carried by wild boars.