Microwave radiometric technique to retrieve vapor, liquid and ice: part ii-joint studies of Radiometer and Radar in Winter Clouds

A neural network-based retrieval technique is developed to infer vapor, liquid, and ice columns using two- and threechannel microwave radiometers. Neural network-based inverse scattering methods are capable of merging various data streams in order to retrieve microphysical properties of clouds and precipitation. The method is calibrated using National Oceanic and Atmospheric Administration (NOAA) results in a cloud-free condition. The performance of two- and three-channel neural network-based techniques is verified by independent NOAA estimates. The estimates of vapor and liquid agree with NOAA values. In the presence of ice, the liquid estimates deviated from NOAA's estimates. One of the major contributions of the threechannel radiometer is the estimation of ice in a winter cloud. The three-channel radiometer not only improves estimates of vapor and liquid, but also retrieves the ice column. Passive remote sensing can be ameliorated with the help of active remote sensing methods. The three-channel radiometer is used for estimating columnar contents of vapor, liquid, and ice in a cloud. It is shown that vertical profiles of median size diameter, number concentration, liquid water content, and ice water content can be inferred by combining radar reflectivity and radiometer observations. The combined remote sensor method is applied to Winter Icing and Storms Project (WISP) data to obtain detailed microphysical properties of clouds and precipitation. We also derived Z- Ice Water Content (IWC) and Z- Liquid Water Content (LWC) relationships and they are consistent with the earlier results.