Polarimetric Radar Analysis of Hurricane Harvey’s (2017) Record-Setting Rainfall

As Hurricane Harvey (2017) stalled over southeast Texas, excessive rainfall flooded the region, despite Harvey weakening rapidly after landfall. 1,538 mm of rainfall was observed in Nederland, TX and set a new record for precipitation in the continental United States by a tropical cyclone and its remnants. Additionally, a large region between Houston and Beaumont underwent substantial flooding due to rainfall in excess of 800 mm. A better understanding of the precipitation characteristics is critical to understanding how extreme rainfall events associated with tropical cyclones evolve, both spatially and temporally. To achieve this goal, this study capitalizes upon the excellent coverage of the polarimetric WSR-88D radars and extensive rain gauge network in Texas.

First, we analyze rain gauge data to determine the relative contributions of varying rainfall intensities to the total precipitation. These results are placed in the context of Harvey’s evolution and location relative to storm features. Next, the rain gauge data are compared with collocated polarimetric observations from the nearby radars to analyze the hydrometeor types identified by the NCAR particle identification algorithm and precipitation types determined by a convective-stratiform separation algorithm. These comparisons enable an examination of the nature of the precipitation that fell over southeast Texas. Finally, output from a quantitative precipitation estimation (QPE) algorithm is compared with the rain gauge data to evaluate its accuracy and to identify potential areas of improvement for QPE in tropical cyclone environments.