Profiling atmospheric graupel, snow and ice using TRMM´s PR and TMI

Of TRMM´s three instantaneous algorithms, the radar-only and combined radar-radiometer algorithms currently produce profiles of liquid rain only. These two algorithms do not attempt to estimate the graupel, snow or other ice distributions. This severely limits their usefulness in estimating the underlying latent heating due to the precipitation. While the radiometer algorithm does produce estimates of liquid rain and undifferentiated solid ice, it does so at a much coarser resolution than the other two and accounts for only a small fraction of the available data. In order to improve the accuracy of TRMM´s current estimates of the vertical distribution of the various hydrometeor species, we have refined the combined algorithm to expand its scope and estimate graupel, snow and ice profiles along with the current rain-only estimates, at the resolution of the TRMM radar. The associated uncertainties are automatically quantified and should allow TRMM to produce robust latent heating estimates that have unprecedented accuracy