MATYAS, TANG, ZICK – Using shape metrics to compare observed and simulated reflectivity during the landfall of Hurricane Isabel (2003)
Corene J. Matyas, Stephanie Zick, and Jingyin Tang
Conference paper presented: April 2016 at the American Meteorological Society‘s 32nd Conference on Hurricanes and Tropical Meteorology
ABSTRACT:
Measuring changes in the spatial attributes of precipitating regions within tropical cyclones (TCs) as they move over land facilitates the tracking of their location and how they change in size and shape. Observationally, the high spatial and temporal resolution data produced by ground-based radars permit the calculation of shape metrics of compactness (e.g., MacEachren 1985)to quantify changing storm structure over land(Matyas 2007, 2008; 2009, 2010).In this study, we use the method of Tang and Matyas (2016) to construct a 3D mosaic of Level II radar reflectivity values from the Weather Surveillance Radar 1988 Doppler (WSR-88D) networkand calculate a measure of dispersion to determine how the positions of rainbands evolve during landfall of Hurricane Isabel (2003)and its subsequent transition into an extra tropical cyclone. Much can also be learned about structural changes within TCs and TC interactions with the surrounding environment by conducting model simulations.The Weather Research and Forecasting model (WRF) is widely employed for TC research studies(e.g., Davis et al. 2008; Gentry and Lackmann 2010; Torn and Davis 2012). In this study, we use WRF to model the rainband structures of Hurricane Isabel and compare model results to those observed by the WSR-88D network. As the development of a TC is sensitive to model physics(Davis et al. 2008; Fierro et al. 2009), we take an ensemble approach in selecting two different cumulus parameterizations and three microphysics schemes.We expect the storm structure to evolve differently given these different setups.The goal of this study is to demonstrate how shape metrics can be used to compare observations and simulations of TC rainbands during landfall. To best match the observed rainband shapes to those in the simulations, we also identify and account for biases in radar reflectivity values. We emphasize that it is beyond the scope of this study to account for how differences in rainband structures evolve during the simulations, or determine which model setup performs the “best.”There are four reasons that we select Isabel for our analysis. First,it was large in size so that our innermost 3 km WRF grid should capture eyewall processes reasonably well(Gentry and Lackmann 2010). Second,operational forecast models had good predictions of track and extent of rainfall (NOAA 2003), so we expect that a research-grade simulation should initialize and perform well.Third, Isabel made landfall as a Category 2 storm at Drum Inlet, NC,on Sep.18 at 1700 UTC (Lawrence et al. 2005), and there is adequate availability of WSR-88D data while the storm is over land. Fourth, Isabel was declared post-tropical at 1200 UTC on Sep.19 while its center was still over the U.S., thus providing an opportunity to measure changes in the position of rainbands during the extratropical transition (ET) process.
Read the full publication at Using shape metrics to compare observed and simulated reflectivity during the landfall of Hurricane Isabel (2003)