Whilst climate change is expected to tremendously influence the regional transmission of malaria, the available data reveal conflicting results. This study provides contextual evidence. We adopted multi-scale geographically weighted regression (MGWR) modelling approach. AICc and local r(2) were used to evaluate performance of the MGWR.. The MGWR analysis showed that LST (beta = -0.667), maximum temperature (beta = -0.507), mean temperature (beta = -0.480), and distance from streams (beta = -0.487) were negatively associated with malaria prevalence. However, enhanced vegetation index correlated positively with malaria prevalence (beta = 0.663). Our results may be important for public health interventions.