Recent updates of a GPM Radar Simulator using CRM data with Bulk Microphysics

The goal of Global Precipitation Measurement (GPM) mission is to provide more accurate global precipitation information within ±65° latitude every 2~4 hour. GPM core satellite which carries the Dual-frequency Precipitation Radar (DPR) operating at Ku- and Ka-band will be launched in February, 2014. A number of radar precipitation retrieval algorithms are being developed for Level-2 radar processing. To support radar algorithm developers who need test data to evaluate the performance of their algorithm, a GPM radar simulator has been developed to generate synthetic Level-1 products which include the most important component, the received echo power, as well as radar parameters required for Level-2 algorithm processing. The GPM radar simulator consists of two modules. One is the DPR scanning geometry module that specifies the characteristics of the DPR sensor and emulates the scan and antenna geometry of the DPR. The other module is the forward calculation module that takes as input the prognostic variables (mixing ratio of hydrometeors) of high resolution cloud-resolving model data, using a one moment bulk microphysics scheme, and computes radar scattering parameters consistent with the microphysical assumptions of the hydrometeors. The forward calculation module also includes a surface scattering model to compute the radar return power from the illuminated surface area by means of a model for the normalized radar cross section (σ0 or NRCS), given as a function of incidence angle and surface type (land/ocean). The simulation of the surface return power plays an important role for one of primary DPR retrieval algorithms, the Surface Reference Technique (SRT), that estimates path-integrated attenuation (PIA) from the surface returns under rain and no-rain condition. Recently, we have completed an update of the surface scattering model for land. Like the NRCS over ocean, the NRCS over land is modeled by taking realizations from a jointly Gaussian random variable with means, variances and correlations obtained from measurements of the NRCS from the APR2 (2nd generation Airborne Precipitation Radar), which operates at approximately the same frequencies as the DPR. The APR2 data that was analyzed was taken during the GPM Cold-season Precipitation Experiment (GCPEx) and provided courtesy of Simone Tanelli of JPL. To classify the surface type into land and ocean, the MODIS-derived high resolution land/ocean mask (0.05°) was used in this study. The upgrade of the simulator to both ocean and land surfaces will enable tests of the SRT and other retrieval algorithms over both surface types. We have been running the simulator with a number of different CRM inputs so that the algorithms can be tested under a variety of meteorological conditions. For light, wide-spread rain data, we have used WRF model data tuned to the Light Precipitation Evaluation Experiment (LPVEx) that took place in the Gulf of Finland in September and October, 2010. For studying heavy precipitation event, we are using WRF model simulation data for the Mid-latitude Continental Convective Clouds Experiment (MC3E) that took place from April 22 - June 6, 2011, in Oklahoma. To evaluate the performance of the algorithms, truth data derived directly from the model data serve as a standard against which the Level 2 algorithm outputs can be compared.