Modeling topographic solar radiation using GOES data

In this paper the authors present an algorithm that combines solar radiation fields derived from Geostationary Operational Environmental Satellite (GOES) observations with digital elevation data to produce topographically varying insolation fields at fine grid spacing. Cloud-modulated irradiances are obtained using hourly 8-km resolution GOES observations. These irradiances are then spatially integrated to the grid spacing of the digital elevation data. The integration accounts for uncertainties in satellite navigation, the limited sensor resolution relative to the hemispheric field of view of a terrain element, and the mismatch between the instantaneous fluxes estimated by GOES observations and the time-integrated quantities typically used in distributed modeling, such as hourly fluxes. The integrated fields are partitioned into direct and diffuse components and then adjusted for the effects of elevation. Lastly, other topographic effects, such as slope orientation, shadowing, sky obstruction, and terrain reflectance are modeled using fields derived from the digital elevation data. The final product is a map of solar radiation that marries coarse-scale variability in insolation caused by clouds with the finescale variability caused by topography. The authors demonstrate the technique for a portion of the Rocky Mountains, using a 90-m digital terrain model covering over 1 degrees x 1 degrees of latitude and longitude. Lastly, assumptions, limitations, and sources of error in data and algorithms are discussed.