Statistical modeling of space‐time rainfall using radar and rain gage observations

Abstract

A statistical framework for modeling space‐time rainfall using radar and rain gage observations is developed. Three principal tasks are involved in implementing our statistical model. These tasks are referred to as sampling (that is, characterization of the error structure of radar and rain gage measurements of rainfall, modeling), (that is, specification of a stochastic model for space‐time rainfall), and parameter estimation. It is emphasized that sampling, modeling, and parameter estimation are interrelated and equally important tasks. Our statistical model is applied to daily rainfall fields in the tropical Atlantic region covered by the GATE experiment. Copyright 1987 by the American Geophysical Union.