Uncertainties in the rain profiling algorithm for the TRMM Precipitation Radar

The rain profiling algorithm for the TRMM Precipitation Radar (PR) has been producing reasonable rain estimates. However, their statistics do not completely agree with the statistics of the simultaneously measured TMI rain estimates. There are some differences. To resolve such discrepancies, critical reviews of uncertainties in both algorithms are wanted. In this paper, uncertain factors in the rain model used in the PR algorithm and their effects on the rain estimates are summarized.

The uncertainties include: Absolute bias in measured radar reflectivity factor, quantization error in measurement, vertical profiles of parameters to specify Z-R and k-Z relationships, particle state (phase, shape, orientation, temperature), drop size distribution (DSD), non-uniform distribution of hydrometeor particles, vertical wind, errors in the attenuation estimates from the surface reference, rain echo cluttered by surface echo, and attenuation by cloud water. The effects of these uncertainties are examined quantitatively. In particular, effects of DSD, storm model and non-uniform beam filling are investigated.