Comprehensive spatial risk assessment using community-based species distribution models stratified by ant damage types

Ants (Hymenoptera: Formicidae) include highly adaptable and aggressive species, some of which act as invasive species, posing serious threats to ecosystems and human society by causing biodiversity loss, agricultural damage, and adverse effects on human health. This study used an ensemble species distribution model (SDM) to predict the potential distribution of ant communities categorized by damage type, biodiversity impacts, direct or indirect agricultural damage, and threats to human health, including physical attacks or disease transmission, under various climate-change scenarios. The ensemble model was constructed using two machine-learning-based SDMs, MaxEnt and random forest, and applied under multiple general circulation models within shared socioeconomic pathway (SSP) climate-change scenarios, enabling robust identification of high-risk areas for ant-related damages while minimizing uncertainty. The ensemble model for ants based on damage type achieved an average true skill statistic (TSS) of 0.86. Temperature seasonality and the minimum temperature of the coldest month were identified as the key factors influencing ant distribution. Areas vulnerable to physical damage caused by ant attacks are expected to expand under future climatic conditions, representing the greatest high-risk zones. In contrast, the regions affected by disease transmission are projected to shrink, resulting in the least impacted areas. Additionally, spatial biodiversity loss patterns and both direct and indirect agricultural impacts are likely to vary among regions, underscoring the need for careful monitoring and management of the damage associated with the continued expansion of ant habitats. This study provides a foundation for developing tailored management strategies for areas at high risk of ant damage under various climate-change scenarios.