Improved prospects for estimating insolation for calculating regional evapotranspiration from remotely sensed data

Abstract

Insolation is a key parameter in most formulations for estimating regional evapotranspiration and/or primary productivity. Many attempts have been made to replace several parameters in these formulations with remotely sensed data. Insolation, however, has been obtained from conventional observations. The objective of this paper is to demonstrate that satellite methods for deriving insolation can be improved on the regional scale beyond what is achievable on large scales. We have developed a realistic surface albedo model, which was used as an improved boundary condition in an insolation model, and also, to specify clear/cloudy thresholds. It was demonstrated that as a result, the mean bias error (MBE) was reduced by as much as 50%, to about 6 W m −2 for a monthly mean. Such improvements and accuracies have implications for local evapotranspiration estimation methods.