Aerosol feedbacks are becoming more accepted as physical mechanisms that should be included in numerical weather prediction models in order to improve the accuracy of the weather forecasts. The default set-up in the Aire Limitee Adaptation dynamique Developpement INternational (ALADIN) — High Resolution Limited Area Model (HIRLAM) numerical weather prediction system includes monthly aerosol climatologies to account for the average direct radiative effect of aerosols. This effect was studied using the default aerosol climatology in the system and compared to experiments run using the more up-to-date Max-Planck-Institute Aerosol Climatology version 1 (MACv1), and time-varying aerosol data from the Monitoring Atmospheric Composition and Climate (MACC) reanalysis aerosol dataset. Accounting for the direct radiative effect using monthly aerosol climatologies or near real-time aerosol distributions improved the accuracy of the simulated radiative fluxes and temperature and humidity forecasts in the lower troposphere. However, the dependency of forecast meteorological conditions on the aerosol dataset itself was found to be weak.