Aerosols affect the Earth's radiation balance directly and cloud microphysical processes indirectly via the activation of cloud condensation and ice nuclei. These two effects have often been considered separately and independently, hence the need to assess their combined impact given the differing nature of their effects on convective clouds. To study both effects, an aerosol–microphysics–radiation coupling, including Goddard microphysics and radiation schemes, was implemented into the NASA Unified Weather Research and Forecasting model (NU-WRF). Fully coupled NU-WRF simulations were conducted for a mesoscale convective system (MCS) that passed through the Niamey, Niger area on 6–7 August 2006 during an AMMA special observing period. The results suggest that rainfall is reduced when aerosol indirect effects are included, regardless of the aerosol direct effect. Daily mean radiation heating profiles in the area traversed by the MCS showed the aerosol (mainly mineral dust) direct effect had the largest impact near cloud tops just above 200 hPa where short-wave heating increased by about 0.8 K day−1; the weakest long-wave cooling was at around 250 hPa. It was also found that more condensation and ice nuclei as a result of higher aerosol/dust concentrations led to increased amounts of all cloud hydrometeors because of the microphysical indirect effect, and the radiation direct effect acts to reduce precipitating cloud particles (rain, snow and graupel) in the middle and lower cloud layers while increasing the non-precipitating particles (ice) in the cirrus anvil. However, when the aerosol direct effect was activated, regardless of the indirect effect, the onset of MCS precipitation was delayed about 2 h, in conjunction with the delay in the activation of cloud condensation and ice nuclei. Overall, for this particular environment, model set-up and physics configuration, the effect of aerosol radiative heating due to mineral dust overwhelmed the effect of the aerosols on microphysics.