Statistical Methods of Estimating Average Rainfall over Large Space–Timescales Using Data from the TRMM Precipitation Radar

Abstract

Data from the Tropical Rainfall Measuring Mission (TRMM) precipitation radar represent the first global rain-rate dataset acquired by a spaceborne weather radar. Because the radar operates at an attenuating wavelength, one of the principal issues concerns the accuracy of the attenuation correction algorithms. One way to test these algorithms is by means of a statistical method in which the probability distribution of rain rates at the high end is inferred by measurements at the low to intermediate range and by the assumption that the rain rates are lognormally distributed. Investigation of this method and the area–time integral methods using a global dataset provides an indication of how well methods of this kind can be expected to perform over different space–timescales and climatological regions using the sparsely sampled TRMM radar data. Identification of statistical relationships among the rain parameters and an understanding of the rain-rate distribution as a function of time and space may help to test the validity of the high-resolution rain-rate estimates.