On the role of drop size distribution in TRMM rain profiling algorithm

This paper presents the variations of the drop size distribution at different climate regimes in conjunction with its importance on TRMM precipitation radar (PR) rainfall retrievals. The precipitation radar is an active sensor on aboard the NASA Tropical Rainfall Measuring Mission (TRMM) satellite. It operates at attenuating13.8 GHz frequency. The rain retrieval algorithm that utilizes the PR measurements employs relationships between rain attenuation and reflectivity (K-Z) and rain rate and reflectivity (R-Z). Both relations require the knowledge of the drop size distribution. The current version of the rain retrieval algorithm combines the surface based drop size distribution measurements at different tropical sites and offers an initial set of relations of K-Z and R-Z for convective, stratiform, and other precipitation. Here the other precipitation represents neither convective nor stratiform precipitation. The precipitation type is determined based on the characteristics of the drop size distribution at the surface. The final sets up of relations is determined though surface reference technique.

In this study, we develop K-Z and R-Z relations for seven different sites around the tropics. The objective of this study is to evaluate the accuracy of operational K-Z and R-Z relations at different tropical regimes. In that regard, we compare the rain accumulations that are derived from operational and site specific R-Z relations with in situ observations. Overall, the characteristics of drop size distribution and subsequent relations of bulk descriptors of rainfall may be grouped for a oceanic and continental regime. However, at some sites like in Darwin, seasonal shift in wind direction results in both continental and oceanic (monsoon) rainfall. Similar trend is also evident in India and in Amazon basin. A specific attention is therefore given to the characteristics of drop size distribution in dual wind regimes.

Although the precipitation type is determined from vertical and horizontal features of radar echoes, the TRMM PR algorithm employs the R-Z and K-Z relations of convective, stratiform precipitation where the classification is based on surface drop size distribution characteristics. In this study, we examine the sensitivity of the K-Z and R-Z relations to the precipitation type algorithms. Due to the differences in sampling volumes, it is difficult to simulate the PR precipitation algorithm utilizing scanning and/or vertically pointing ground-based radar observations. Nevertheless, we will determine the variations in the drop size distribution observed by a disdrometer due to the differences in precipitation type algorithms. Here, the texture algorithm utilizing constant altitude radar measurements and bright band algorithm utilizing vertically pointing radar are primarily considered.