The development of a satellite infrared technique for estimating convective and stratiform rainfall area and volume and its application in studying the diurnal variability of rainfall in Amazonia are presented. Cloud systems are defined in the technique by the 253 K infrared (IR) temperature isotherm. The convective and stratiform rain areas within these cloud systems are then related to morphologic characteristics of the IR temperature fields. Rainfall rates are assigned to the defined convective and stratiform areas using IR-microwave-derived rain rate probability matching relationships. The training data set consists of three months of collocated IR observations and microwave (MW) rain rate retrievals over a region in the Amazon Basin. Evaluation of instantaneous rain rate estimates over a second independent region in the Amazon showed 25% (-40%) systematic error and 55% (70%) residual random error standard deviation in morning (evening) MW overpasses. The method is used to derive the mean diurnal cycle of rainfall and investigate the relative contribution from its convective and stratiform components. Finally, the technique is applied to study the time evolution of rainfall and the transition from convective to stratiform over selected sites in the Amazon. ¿ 1999 American Geophysical Union