Although bulk microphysics parameterizations are used extensively in mesoscale models to simulate precipitation, verification of the simulated microphysical processes against observations has been difficult and with relatively poor results when possible. The recent launch of new passive and active microwave space based sensors, such as in TRMM (Tropical Rainfall Measuring Mission), provide a new data source not only for the observation of precipitation as intended in their mission, but also for indirectly diagnosing the microphysical processes ongoing. These observations are useful to the development of microphysical parameterization. In this study, we consider a numerical simulation of hurricane Bonnie employing an explicit bulk microphysics. Six categories of hydrometeors were simulated including 2 categories of droplets and 4 categories of ice crystals. During the period of the simulation, a TRMM satellite passed over the storm. Direct in situ observations of microphysics were also made as part of the CAMEX-3 field program. Combined, the satellite and in situ observations provided a powerful data set that can be used to verify and improve simulated hydrometeor size distributions and growth processes. The method involves the iterative use of the modeled and observed data to improve the microphysics parameterization and verification against in situ observations. Comparisons with observed data suggested important inconsistencies of the simulated microphysical processes. In each case, the problem was with conversion processes for which observational justification was not previously available. When the simulated processes were modified in accordance with these findings, resulting comparisons with satellite observations improved. The results were checked against the in situ observations available in limited regions throughout the storm.