New parametrization of the rain drop size distribution

This paper revisits the problem of finding a parametric form for the rain drop size distribution (DSD) which 1) is an appropriate model for tropical rainfall, and 2) involves statistically independent parameters. Using TOGA/COARE data, we derive a parametrization which meets these criteria. This new parametrization is an improvement on the one that was derived in [3], using TRMM ground truth data from Darwin, Australia. The new COARE data allows us to verify that the spatial variability of the two `shape' parameters is relatively small, thus confirming that this parametrization should be particularly useful for remote sensing applications. We also derive new DSD-based radar-reflectivity-rain-rate power laws, whose coefficients are directly related to the shape parameters of the DSD. Perhaps most important, since the coefficients are independent of the rain rate itself, and vary little spatially, the relations are ideally suited for rain retrieval algorithms. It should also prove straightforward to extend this method to the problems of estimating cloud hydrometeors from remote-sensing measurements.