Modelling seasonal habitat suitability for wide-ranging species: Invasive wild pigs in northern Australia

摘要: Invasive wildlife often causes serious damage to the economy and agriculture as well environmental, human animal health. Habitat models can fill knowledge gaps about species distributions assist planning mitigate impacts. Yet, model accuracy utility may be compromised by small study areas limited integration of ecology or temporal variability. Here we modelled seasonal habitat suitability for wild pigs, a widespread harmful invader, in northern Australia. We developed resource-based, spatially-explicit regional-scale approach using Bayesian networks spatial pattern analysis. integrated important ecological factors such variability environmental conditions, breeding requirements home range movements. The was parameterized during structured, iterative expert elicitation process applied wet season dry scenario. Model performance uncertainty evaluated against independent distributional data sets. Validation results showed that an expert-averaged accurately predicted empirical pig presences Australia both scenarios. largely associated with different assumptions pigs' resource-seeking varied considerably between seasons, retracting resource-abundant rainforest, wetland agricultural refuge expanding widely into surrounding grassland floodplains, savanna woodlands coastal shrubs season. Overall, our suggested suitable is less available than previously thought. Mapped used quantify impacts, assess risks, justify management investments target control activities. Our methods are applicable other wide-ranging species, especially data-poor situations.