AI-Enhanced CLIMEX Modelling for Predicting Invasive Pest Risks in Climate-Smart Agricultural Systems

Abstract

Invasive pest species create progressive threats in the global agricultural productivity, food security, and ecosystems stability especially in changing climatic conditions. This paper will describe an unified framework which combines the CLIMEX ecological niche model with the advanced artificial intelligence (AI) algorithms to enhance the forecasting of climate-sensitive pest distribution and estimating future invasion. The hybrid CLIMEXRandom Forest models have been constructed through the combination of historical temperature data, precipitation data, soil data, and crop distribution data to represent the non-linear interactions that are affecting the suitability of the habitat of pests. The CLIMEX interpretation and predictive performance can be optimised by introducing AI, which enables it to be calibrated adaptively to empirical occurrence data. Assessment with international pest occurrence datasets showed that the hybrid model had a better predictive power, and much better area under the curve (AUC) and true skill statistic (TSS) than conventional modelling methods. Scenario-based projections under CMIP6 RCP 4.5 and RCP 8.5 climate pathways revealed a potential expansion of pest suitability zones by 15–25% in tropical and subtropical regions, driven by rising temperatures and altered rainfall patterns. The findings underscore the growing vulnerability of climate-smart agricultural systems to invasive pests and highlight the importance of integrating ecological modelling with AI-driven analytics for proactive risk assessment. The designed framework is a dynamic decision support tool to boost early warning frameworks, refine precision pest management plans and resilient agricultural, rural innovation that issustainable and dynamic in the climate change era.