Comparison of Radar Rainfall Retrieval Algorithms in Convective Rain During TOGA COARE

We compare two deterministic and two stochastic rain retrieval algorithms by applying them to 14 GHz reflectivity profiles acquired during TOGA COARE. The first deterministic algorithm corrects the k – R relation, while the second corrects the Z – R relation and is equivalent to correcting the calibration constant. The stochastic algorithms are based on applying an Extended Kalman Filter to the reflectivity. One algorithm employs only radar data, while the other employs both radar and path attenuation. We find that the deterministic algorithm which corrects the Z– R relation and the two stochastic algorithms indicate a smaller mean diameter than would be expected for widespread, light or moderate rainfall. This finding seems in agreement with independent observations of the DSD in tropical convective rain. Only the algorithm which corrects the k-R relation suggests larger drops. This, combined with observation that the Z – R relation is much more variable than the k – R relation indicates that the Z – R relation should be corrected when using a deterministic algorithm.