Modeling sky luminance using satellite data to classify sky conditions

Abstract

A model for calculating sky luminance is presented. The earth–atmospheric reflectivity obtained from satellite data was used to classify sky conditions. The proposed sky luminance at a specific point in the sky hemisphere is a product of two functions, namely F 1 and F 2. For any point in a sky hemisphere, F 1 varies with the sky zenith of the considered point ( Z) and F 2 is a function of the angular distance between the considered point and the sun ( χ). The model uses simultaneous measurements of sky luminance and GMS-5 satellite observation, both taken at Nakhon Pathom, Thailand (13.70°N, 103.10°E), during November 2002 to May 2003. Both F 1 and F 2 are derived as polynomial functions of χ, Z and the solar zenith angle ( Z s) for each sky condition. The classification of the sky condition is based on the satellite-derived earth–atmospheric reflectivity. This sky luminance model was validated against independent measurements from the Asia Institute of Technology (AIT) (14.08°N, 100.62°E). The root mean square difference (RMSD) between the relative sky luminance calculated from the model and that obtained from the measurement is 0.133.