Making The Best Possible Choice: Using Agent-Based Modelling To Inform Wildlife Management In Small Communities

Human-wildlife interactions occur where human societies interface with the natural landscape. Some of these interactions can be problematic and lead to negative outcomes, such as the destruction of property or injury to humans or wildlife, especially when they involve large carnivores such as bears. Developing management strategies to mitigate these negative outcomes can be costly, time-consuming, and ethically problematic. We developed an agent-based model of black bears foraging in the forest-urban interface of a small resort community to simulate and test a variety of management strategies. The model allows to quickly and cost-effectively identify those strategies, such as attractant management and deterrence measures, that are most likely to reduce the number of bears that become habituated to human areas and conditioned to anthropogenic food sources. Although increasing the likelihood that a bear is deterred from human areas did have some effect, model outcomes indicated that the elimination of sources of anthropogenic food was the more effective management strategy. However, the model also revealed that interactions between the spatial configurations of these two approaches impacted the overall effectiveness of each. Notably, the random distribution of an increase in deterrence across the area of interest in conjunction with a focus on attractant removal around the perimeter of said area was the most effective spatial implementation of the two strategies together. This result provides a quantitative confirmation that reducing anthropogenic food availability should be the highest priority for managers seeking to reduce the number of conflict bears in urban areas, but also that the spatial configuration of any applied strategies is important to consider. The agent-based modelling approach proved a fast and costeffective way to compare a variety of possible management strategies and the different ways in which they can be arranged in a specific landscape.