A method for delineating and estimating rainfall fields

Abstract

A geostatistical model has been defined to deal with intermittent rainfall fields. The intermittency is represented by a binary random function, while, inside the rainy areas, the rainfall variability is rendered by an intrinsic random function. A rain gage data set from the Cevennes region (France) is used to infer the model. The inner variability is independent of the intermittency, thereby allowing a simple estimation relying on the separate kriging of these two components. Comparison of the rainfall assessments made by a classic global kriging and the proposed method shows its clear superiority for delineating and estimating rainfall fields.