Solar radiant energy incident at the earth's surface is a quantity of increasing importance in agricultural monitoring and solar power development. An experiment was undertaken in the summer of 1977 to determine if incoming visible radiation at the surface could be estimated from geostationary meteorological satellite data. The experiment entailed the collection of coincident satellite, conventional meteorological and pyranometer data over the Great Plains of the United States. Regression techniques were used to estimate hourly insolation from the satellite data. Hourly estimates were summed to give daily total insolation. The standard error of the satellite-derived daily insolation when compared against pyranometers was 10% of the mean, an accuracy more than sufficient for most agricultural uses. Problems of producing insolation operationally from both geostationary and polar-orbiting satellites are discussed. Abstract Solar radiant energy incident at the earth's surface is a quantity of increasing importance in agricultural monitoring and solar power development. An experiment was undertaken in the summer of 1977 to determine if incoming visible radiation at the surface could be estimated from geostationary meteorological satellite data. The experiment entailed the collection of coincident satellite, conventional meteorological and pyranometer data over the Great Plains of the United States. Regression techniques were used to estimate hourly insolation from the satellite data. Hourly estimates were summed to give daily total insolation. The standard error of the satellite-derived daily insolation when compared against pyranometers was 10% of the mean, an accuracy more than sufficient for most agricultural uses. Problems of producing insolation operationally from both geostationary and polar-orbiting satellites are discussed.