Retrieval of hydrometeor profiles in tropical cyclones and convection from combined radar and radiometer observations

A retrieval algorithm is described to estimate vertical profiles of precipitation ice water content and liquid water content in tropical cyclones and convection over ocean from combined spaceborne radar and radiometer measurements. In the algorithm, the intercept parameter N(0)s in the exponential particle size distribution for rain, snow, and graupel are adjusted iteratively to minimize the difference between observed brightness temperatures and simulated ones by using a simulated annealing optimization method. Sensitivity tests are performed to understand the effects of the input parameters. The retrieval technique is investigated using the Earth Resources (ER)-2 aircraft Doppler radar and Advanced Microwave Precipitation Radiometer data in tropical cyclones and convection. An indirect validation is performed by comparing the measured and retrieved 50-GHz (independent channel) brightness temperature. The global agreement shows not only the quality of the inversion procedure, but also the consistency of the retrieved parameters with observations. The direct validation of the ice water content retrieval by using the aircraft in situ microphysical measurements indicates that the algorithm can provide reliable ice water content estimates, especially in stratiform regions. In convective regions, the large variability of the microphysical characteristics causes a large uncertainty in the retrieval, although the mean difference between the retrieved ice water content and aircraft-derived ice water content is very small. The ice water content estimated by a radar-only empirical relationship is higher than those retrieved by the combined algorithm and derived by the aircraft in situ observations. The new combined algorithm contains information that should improve ice water content estimates from either radar-only or passive microwave-only measurements. An important caveat for this study is that it concerns precipitation estimates. In this paper, ice and liquid water content should be interpreted as precipitation ice and liquid water content.