a Comparison of Polarimetric Radar Retrievals during Hurricane Harvey
Abstract
Hurricane Harvey hit the Texas Gulf Coast as a major hurricane on August 25, 2017 before exiting the state as a tropical storm on September 1, 2017. In its wake, it left a flood of historic proportions, with some areas measuring 60+ inches of rain over a five-day period. Although the storm center stayed west of the immediate Houston area, trailing bands of precipitation being fed by the Gulf of Mexico impacted the Houston area for five days. The National Weather Service (NWS) WSR-88D dual-polarimetric radar (KHGX), located southeast of Houston, maintained operations for the entirety of the event. The Harris County Flood Warning System (HCFWS) had 150 rain gauges deployed in its network. In this study, we used the full radar data set to retrieve daily and event-total precipitation estimates within 120 km of the KHGX radar for the period August 25-29, 2017. These estimates were then compared to the HCFWS and ASOS gauges. Three different polarimetric hybrid rainfall retrievals were used: Cifelli et al. 2011; Bringi et al. 2004; and, Chen et al. 2017 (known as Dual-pol estimators). Each of these hybrid retrievals have demonstrated robust performance in the past; however, most daily and the event-total comparisons from each of these retrievals compared to those of HCFWS and ASOS rain gauge networks resulted in significant underestimates. We then employed an attenuation-based polarimetric estimate following Ryzkhov et al. 2014 and found that it significantly outperformed the other hybrid approaches during periods when the deep moisture and heaviest rainfall occurred during this event. Following Tokay et al. 2008, we propose that the reason for this improvement was due to the presence of a large number of small drops, commonly seen in tropical cyclone rain bands, which limit the use of the differential reflectivity and specific differential phase in the retrievals. While the hybrid Dual-pol estimators mentioned above work quite well for continental convection, and did well on August 25, 2018 before the storm had made landfall, they do not perform well in deep tropical systems that dominated the precipitation during the period Aug. 26-28, 2017.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFM.H43F2463W
- Keywords:
-
- 3354 Precipitation;
- ATMOSPHERIC PROCESSESDE: 3360 Remote sensing;
- ATMOSPHERIC PROCESSESDE: 1817 Extreme events;
- HYDROLOGYDE: 1847 Modeling;
- HYDROLOGY