Abstract / Summary
Background In Sudano-Sahelian regions of sub-Saharan Africa, malaria transmission is strictly seasonal and governed by climate variability. Climate change is altering these predictable patterns, increasing rainfall unpredictability and shifting vector ecological niches, which threatens ongoing control efforts. This study analyzes the relationship between environmental variables, vector dynamics, and malaria morbidity in Keur Socé, Senegal. Methods Monthly meteorological data (rainfall, temperature, and maximum relative humidity), entomological data, and epidemiological records (confirmed malaria cases) were collected from 2021 to 2025. Spearman correlations and Cross-Correlation Functions (CCF) were used to identify optimal time lags. A SARIMAX predictive model was developed and calibrated using data from 2021–2023, then tested and validated against observations from 2024 and 2025. Results A systematic time lag of 4 to 8 weeks was observed between peak rainfall and peak morbidity. Maximum relative humidity (UMAX) emerged as the primary predictor (ρ = 0.91) at Lag 2. High vector parity rates (94.5% in October) confirmed that UMAX levels above 80% ensure vector longevity and sustain transmission beyond the rainy season. While the SARIMAX model accurately predicted 2024 epidemiological trends (R 2 = 0.69), a significant divergence in 2025 (R 2 = 0.03) highlighted the impact of Dual-active ingredient long-lasting insecticidal nets (Dual LLINs) in successfully decoupling malaria morbidity from climatic factors. Conclusion This study demonstrates that SARIMAX models incorporating maximum relative humidity as an exogenous factor serve as powerful early warning tools in central Senegal. However, the scalar deployment of targeted public health interventions can significantly alter predicted environmental trends, shifting transmission dynamics from climate-driven to intervention-controlled realities.