In Situ Verfication of Polarimetric Radar-Based Hydrometeor Types

Hydrometeor typing algorithms, based on dual-polarization radar measurements, have been developed recently. The retrieval realized at the National Center for Atmospheric Research (NCAR) is based on a fuzzy logic approach and discriminates 17 different particle types. It is implemented to run in real time for S-Pol, the NCAR S-band dual-polarization Doppler radar. Even though the results look promising the algorithm has yet not been verified against in situ data. A systematic verification using airborne in situ measurements is the subject of this study.

Simultaneous radar and ground-based in situ measurements - necessary for a quantitative understanding of radar data - have been realized successfully over years. Comparing radar data and data recorded by a moving platform became reasonable due to the recent development of the Global Positioning System (GPS). The location of a radar volume and of an aircraft is now sufficiently accurate. Airborne in situ data are well suited for the verification of radar-based hydrometeor typing algorithms since the aircraft covers a wide horizontal range and provides information on different altitudes and therefore on different stages of the development of precipitation. In a first step of the present analysis radar and airborne data have been matched.

One specific focus of the Mesoscale Alpine Program (MAP) was on orographically induced heavy precipitation events including flash flooding ("Wet-MAP"). Within the scope of this international and interdisciplinary effort, special microphysical flight modules were performed by two research aircraft, the NOAA P-3 and the NCAR Electra. They flew in close vicinity to S-Pol for high-resolution comparisons. The precipitation systems during Wet-MAP ranged from frontal systems, stratiform events with embedded convection to pure stratiform precipitation. Rimed and unrimed ice crystals of different habit, aggregates and sporadically super cooled liquid water droplets were observed during flights. The collected data are so far the best data set in order to validate the ice particle classification of the algorithm.