The attenuation of solar radiation by atmospheric aerosols simultaneously decreases the amount of radiation reaching the surface and increases the fraction of radiation which is diffuse. Decreasing the total amount of photosynthetically active radiation (PAR, 400-700 nm) tends to decrease the amount of photosynthesis occurring in plant leaves. Increasing the fraction of PAR which is diffuse allows more PAR to reach shaded leaves and can thus increase the total amount of photosynthesis occurring in a plant canopy. In an attempt to quantify these two radiative influences of aerosols, the CERES crop model [Cereal growth, development and yield. In: Tsuji, G.Y., Hoogenboom, G., Thornton, P.K. (Eds.), Understanding Options for Agricultural Production. Kluwer Academic Publishers, Dordrecht, Netherlands.] was modified to estimate the impact on yield for maize, wheat, and rice under varying atmospheric conditions. The influence of aerosols on total and diffuse PAR was modeled using the National Center for Atmospheric Research (NCAR) Tropospheric Ultraviolet Visible (TUV) radiation model [UV radiation in the natural and perturbed atmosphere. In: Tevini, M. (Ed.), Environmental Effects of UV (Ultraviolet) Radiation. Lewis Publisher, Boca Raton, Florida.] under both clear skies and overcast skies and again when cloudiness conditions were allowed to vary to reflect actual meteorological conditions. The PAR intensity and diffuse fraction results from the radiation model were applied to existing meteorological data sets. These data sets were then used as input for the CERES model. The CERES model was modified so that plant radiation use efficiency (RUE) fluctuated as a function of the diffuse fraction calculated by the TUV model. Model simulations were performed using many years of meteorological data and different amounts of atmospheric aerosol loading for several different locations around the world. The change in crop yield due to the influence of aerosols was found to be extremely dependent on the magnitude of the increase in radiation use efficiency (RUE) incurred by increasing the diffuse fraction. The influence of aerosols on simulated crop yield was found to be more negative on overcast days since the diffuse fraction is already quite high even in a clean atmosphere. Consequently, the more overcast days there are in a growing season, the more negative is the influence of aerosols. For the most realistic set of model results based on location-specific aerosol loadings and crop-specific assumptions of RUE change, the influence of aerosols is estimated to be -10% on maize yield, ±5% on wheat yield, and ±10% on rice yield except for when grown under exceptionally sunny conditions as found in California's central valley, in which case, yields are predicted to increase by up to 30%. Aerosols also tend to decrease plant water stress by reducing soil evaporation and transpiration. When crops are grown under rainfed conditions, this reduction in water loss from evapotranspiration may offset the decrease in the photosynthetic rate and cause aerosols to have a positive influence on final grain yields. © 2005 Elsevier B.V. All rights reserved.