This paper grew out of two ad hoc conferences held last year to discuss the problem of estimating precipitation from satellite data. Although developments in an area as new as this are rapid, there was sufficient interest in the problem, progress towards a solution, and ignorance of the work of others to warrant drawing the conference summaries together in a review paper of wider circulation and general interest.
Our purpose is to examine, from the perspective of GATE rainfall requirements, methodologies in use and under development for the estimation of rainfall from satellite visible and infrared images. As an introduction to critical descriptions of each method, we touch upon the uses of rainfall observations in GATE, the advantages of satellite monitoring, and the purposes of the two conferences held to date. A discussion of calibration and accuracy requirements, including ground truth and accuracies for GATE, is followed by descriptions of two field programs—one completed, the other planned—which have provided and should provide excellent “ground truth” data. We conclude with a discussion of remaining problems and recommendations for observations in GATE.