Atmospheric Radar Research Seminar Series presents...
Simulations and Processing of Polarimetric Radar Signals Based on Numerical Weather Prediction Model Output
Lei Lei
School of Electrical and Computer Engineering
The University of Oklahoma, Norman, OK
12 March 2009, 1:00 PM
National Weather Center, Room 5600
120 David L. Boren Blvd.
University of Oklahoma
Norman, OK
Directions to the NWC (.pdf, 60 kb)
Simulations of radar signals are important in understanding the performance of a radar system and its signal processing. Recently, a weather radar simulator has been developed to generate time series radar signals from numerical weather prediction (NWP) model output. The radar signal is a coherent sum of the echoes from all the individual scatterers while their size and motion are determined by the Advanced Region Prediction System (ARPS). Currently, the simulator generates single polarization radar signals.
The operational WSR-88D (Weather Surveillance Radar 88-Doppler) radar network is being upgraded to dual polarization and will be completed within the next few years. Therefore, it is worthwhile to extend the radar simulator with polarimetric capability. In this study, we present the added feature of generating radar signals for dual-polarization and with the propagation effect included. Further, polarimetric radar parameters such as differential reflectivity, differential phase and copolar cross correlation coefficient are estimated and are shown to be consistent with polarimetric radar measurements.
In order to avoid bias by omnipresent but varying white noise, least square fitting on multi-lag autocorrelation function is introduced. Reflectivity, differential reflectivity and correlation coefficient derived from the fitting are compared with those derived from the conventional estimators.