Rainfall rate probability density evaluation and mapping for the estimation of rain attenuation in South Africa and surrounding islands

Abstract

The paper describes the modelling of the average rainfall rate distribution measured at different locations in South Africa. There are three major aspects this paper addresses: to develop a rainfall rate model based on the maximum likelihood method (ML); to develop contour maps based on rainfall rate at 0:01% percentage of exceedence; and re-classification of the ITU-R and Crane rain zones for the Southern Africa region. The work presented is based on five- minute rainfall data converted to one-minute equivalent using a newly proposed hybrid method. The results are mapped and compared with conventional models such as the ITU-R model, Rice-Holmberg, Moupfouma and Crane models. The proposed rainfall rate models are compared and evaluated using root mean square and chi-square (χ2) statistics. Then re-classification of the rain zone using ITU-R and Crane designations is suggested for easy integration with existing radio planning tools. The rainfall rate contour maps at 0:01% percentage of exceedence are then developed for South Africa and its surrounding islands.