Extended Abstract (260K)P2.4
An inter-comparison of raindrop size distributions retrieved from polarimetric radar parameters
Michael P. Morris, University of Oklahoma Atmospheric Radar Research Center, Norman, OK; and P. B. Chilson, A. V. Ryzhkov, T. J. Schuur, M. Teshiba, G. Zhang, Q. Cao, R. D. Palmer, and L. M. Kanofsky
In precipitating systems, the drop-size distribution (DSD) is critical for characterization of the microphysical processes responsible for the development of the precipitation. In addition, the DSD and its moments form the basis of measurable parameters such as radar reflectivity factor and rainfall rate, and as such a great deal of effort has gone into methods for measuring, modeling, and understanding the properties and evolution of DSDs. One of the main methods for exploring the DSD is to analyze the fallspeed spectrum recorded by Doppler profiling radar, but the relative paucity of profiler systems as well as the planned dual-polarization upgrade to the existing WSR-88D network indicates the need to develop methods of extracting the DSD from polarimetric measurands. Central Oklahoma provides an ideal test bed for such research due to the presence of a prototype polarimetric weather radar near the University of Oklahoma campus (hereafter referred to as KOUN), an operational NEXRAD WSR-88D installation at Twin Lakes (KTLX), as well as a 915-MHz wind profiler (OU profiler) and 2-D video disdrometer (2DVD) located at the University of Oklahoma's Kessler Farm Field Laboratory (KFFL), a remote instrumentation site in McClain county. The Oklahoma Mesonet, a network of 117 remote hydrometeorological observing stations spread across the state, also allows for high spatial and temporal resolution surface observations.
From the KOUN values of radar reflectivity factor at horizontal polarization (Zh) and differential reflectivity (Zdr), a DSD is extracted using the constrained-gamma approach detailed by Cao et al., JAMC, 2007. The spectra of Doppler fallspeeds measured by the OU profiler are time averaged over a one-minute interval and then converted to a DSD, and the two sets of retrieved DSDs are compared against the corresponding data collected by the 2DVD. To account for the intrinsic variability of raindrop size distributions, several microphysical parameters are calculated from each set of DSDs and intercompared to quantify the accuracy of both retrieval methods relative to the 2DVD measurement. Additionally, since a key function of scanning radar is to remotely estimate the rate of rainfall accumulation at the surface, the rainfall rates estimated from each of the three instruments are compared against the Oklahoma Mesonet observations.
Poster Session 2, Recent Developments in Atmospheric Applications of Radar and lidar
Wednesday, 23 January 2008, 2:30 PM-4:00 PM, Exhibit Hall B
Previous paper Next paper