A technique for mapping global illuminance from satellite data was developed. A five-year (1998 - 2002) climatology of global solar illuminance obtained from this technique is presented for Thailand using hourly GMS-5 satellite data. The technique is based on a radiation budget model which traces solar radiation as it is scattered, absorbed and reflected back to space. The model produces an earth-atmospheric albedo in the satellite spectral window as well as global illuminance at the earth's surface. The model is tuned using surface illuminance measurements at four stations in Thailand: Chiang Mai (18.78 degrees N, 98.98 degrees E), Ubon Ratchatani (15.25 degrees N, 104.87 degrees E), Songkhla (7.20 degrees N, 100.60 degrees E) and Nakhon Pathom (13.82 degrees N, 100.04 degrees E). In the mapping process, a satellite earth-atmospheric albedo at any locations yields a cloud-atmospheric albedo in the satellite band, which is then transformed into a cloud-atmospheric albedo in the photopic band. Having obtained the photopic cloud-atmospheric albedo, the model calculates surface illuminance. The model gives a root mean square difference of 8.1% and a mean bias difference of - 2.6% when tested against an independent data set. Monthly average maps are presented covering Thailand for local times of 10:30, 12:30 and 14:30. (c) 2007 Elsevier Ltd. All rights reserved.