Development of a hybrid method for estimating land surface shortwave net radiation from MODIS data

Accurate estimation of shortwave net radiation (Sn) at a high spatial resolution is critical for regional and global land surface models. Current surface radiation budget products have fine temporal resolutions, but only coarse spatial resolutions that are not suitable for land applications. A hybrid algorithm was developed in this study to estimate Sn from MODIS data under both clear and cloudy sky conditions without requiring coarser resolution ancillary data. This algorithm was validated using ground measurements at seven sites of the SURFace RADiation budget observing network (SURFRAD) in the United States. Instantaneous Sn estimated by this method was also compared with GEWEX/SRB and ISCCP data, and other methods. The results indicate that our algorithm can produce Sn at 1-km resolution with improved accuracy and is easily implemented to generate operational global products. Daily integrated Sn is estimated at 1-km resolution using instantaneous Sn. These finer spatial resolution datasets capture the specific sequence of the redistribution of the available energy at the Earth's surface. Therefore, they support recent high resolution land surface models.