A better understanding of aerosol optical, microphysical and radiative properties is a crucial challenge for climate change studies. In the present study, column-integrated aerosol optical and radiative properties observed at a rural site, Mbita (0.42°S, 34.20 °E, and 1125 m above sea level) located in Kenya, East Africa (EA) are investigated using ground-based Aerosol Robotic Network (AERONET) data retrieved during January, 2007 to December, 2015. The annual mean aerosol optical depth (AOD500 nm), Ångström exponent (AE440–870 nm), fine mode fraction of AOD500 nm (FMF500 nm), and columnar water vapor (CWV, cm) were found to be 0.23 ± 0.08, 1.01 ± 0.16, 0.60 ± 0.07, and 2.72 ± 0.20, respectively. The aerosol optical properties exhibited a unimodal distribution with substantial seasonal heterogeneity in their peak values being low (high) during the local wet (dry) seasons. The observed data showed that Mbita and its environs are significantly influenced by various types of aerosols, with biomass burning and/or urban-industrial (BUI), mixed (MXD), and desert dust (DDT) aerosol types contributing to 37.72%, 32.81%, and 1.40%, respectively during the local dry season (JJA). The aerosol volume size distribution (VSD) exhibited bimodal lognormal structure with a geometric mean radius of 0.15 μm and 3.86–5.06 μm for fine- and coarse-mode aerosols, respectively. Further, analysis of single scattering albedo (SSA), asymmetry parameter (ASY) and refractive index (RI) revealed dominance of fine-mode absorbing aerosols during JJA. The averaged aerosol direct radiative forcing (ARF) retrieved from the AERONET showed a strong cooling effect at the bottom of the atmosphere (BOA) and significant warming within the atmosphere (ATM), representing the important role of aerosols played in this rural site of Kenya. Finally, the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model revealed that aerosols from distinct sources resulted in enhanced loading during JJA.