Turbulence intensity retrieval in precipitation via optimal estimation using polarimetric radar

In this presentation we present a generalized methodology to retrieve the turbulence intensity from polarimetric radar data within the framework of optimal estimation. A forward model is developed, where as a function of turbulence intensity, a parametric solution or synthetic 3D random wind field (e.g. Mann, 1998) can be chosen. In this work we use the rain droplet radar cross section calculations from Mishchenko (2000). The simulated radar observables consist of equivalent radar reflectivity, differential reflectivity, linear depolarization ratio, correlation coefficient, mean Doppler velocity and Doppler spectral width (HH, HV, VV). Next to that the Doppler spectrum and polarimetric spectra (specific differential reflectivity, specific linear depolarization ratio, and specific correlation coefficient) are modeled. One of the challenges that is addressed in this presentation is the analysis of the influence of the inertial effect on the rain droplet movements. We estimate the droplet inertial effect by comparing the following cases: a) not including it in the forward model (droplet velocity = air velocity + terminal fall velocity) and b) numerically solving the Stokes equation for each individual droplet for a small part of its trajectory. The forward model with the parameterized options is then used to retrieve the turbulence intensity from polarimetric radar measurements. The proposed technique has been tested using data from the S-band profiling polarimetric radar TARA at the Cabauw (the Netherlands) research site. The estimated turbulence intensities were validated via comparison with in-situ turbulence intensity measurements from sonic anemometer.