Predicting the impact of climate change on desert locust (Schistocerca gregaria) breeding areas using an ensemble model: A case study of the Wabe Shebelle River Basin, Ethiopia

Desert locusts (Schistocerca gregaria) pose a significant threat to arid and semi-arid regions due to their rapid reproduction, long-distance migration, and potential to devastate environmental resources. Climate change plays a crucial role in shaping desert locust breeding habitats. This study aims to investigate the impact of climate change on desert locust breeding areas in the Wabe Shebelle River Basin, Ethiopia. Using presence and pseudo-absence data along with environmental variables, an ensemble modeling approach in R (version 4.4.1) was employed to assess breeding areas. The ensemble model incorporated the Generalized Linear Model (GLM), Multivariate Adaptive Regression Splines (MARS), Bioclimatic Modeling (BIOCLIM), Maximum Entropy (MAXENT), Random Forest (RF), and Support Vector Machines (SVM). Pearson correlation tests and Variance Inflation Factors (VIF) were applied to mitigate multicollinearity. Model performance is evaluated using 10-fold cross-validation, spatial partitioning, and AUC analysis. The ensemble model demonstrated strong predictive performance, with AUC values exceeding 0.7 for most models. Under a high-emission scenario (SSP5-8.5), breeding areas are projected to expand significantly, reaching 44,072.46 km² by the 2050s and 45,739.60 km² by the 2070s. Conversely, under a moderate-emission scenario (SSP2-4.5), the breeding area is expected to decrease to 35,865.08 km² by the 2050s but expand again to 44,913.33 km² by the 2070s. These findings highlight the critical influence of climate change on desert locust breeding habitats. Therefore, policymakers should implement effective and sustainable locust management strategies that incorporate climate change projections to safeguard agricultural productivity and livelihoods in vulnerable regions.