Multisensor Synergy to Improve Ground-Based Temperature and Humidity Profiling

Continuous profiling of the atmospheric thermodynamic state is highly valuable for understanding atmospheric processes and weather forcasting, but atmospheric sounding by radiosondes is usually available 2-4 times. While multi-channel microwave radiometers convey certain information on the vertical variation of temperature and water vapor, their vertical resolution is highly limited. To overcome this limitation, we propose a hybrid algorithm to retrieve the profiles using merged data from ASSIST and Raman Lidar (called A-R retrieval below). Comparison between sounding data and retrievals for a 2-month observation is present. In cloud-free situations, the mean bias errors with the radiosonde are about 0.2K and 0.25g/kg below 3km for temperature and water vapor mixing ratio, respectively. The maximum root-mean-square errors are less than 1.2K and 0.8g/kg. Compared with cloud-free situations, it has larger retrieve errors due to contributions in scattering for downwelling radiance. Using both of two passive/active instruments provides more information content than that for ASSIST-only retrieval. The degrees of freedom for signal which obtained from A-R retrieval is 1.5 times than ASSIST-only retrieval. It is worth noting that using A-R retrieval can provide better temperature retrieval and capturing elevated inversion. The combination of these two instruments can considerably improve retrieval accuracies for both.