Changes in the Radial and Tangential Distribution of Radar Reflectivity During Tropical Cyclone Landfalls Over the United States

The primary and secondary wind circulations in a tropical cyclone (TC) help to organize the spatial configuration of its rainbands. When TCs make landfall, they weaken dynamically. The somewhat drier continental air mass over the U.S. can begin to interact with the storm hours before it makes landfall, causing rainfall to decrease on its left side. For a TC that dissipates, we expect rainfall to be distributed asymmetrically about the storm center due to differences in friction and moisture availability in the environmental air mass. As TCs interact with the middle latitude westerlies, rainfall should dissipate behind the storm and increase ahead of the circulation center. This study employs Level II data from the Weather Surveillance Radar 1988 Doppler network to examine the changing spatial configurations of rainbands in TCs while they are within detection range of the radar network. We employ at new technique that allows for rapid and accurate data processing using parallel computing to perform quality control, project data into a Cartesian coordinate system, mosaic data from multiple radars onto a single grid, and perform interpolation to determine values in grid cells within which a radar gate does not fall. For this study, we create a mosaic of reflectivity values every 30 minutes on a grid with horizontal resolution of 1 km and vertical resolution of 0.5 km using radars within 600 km of the storm center. We then take a constant altitude slice through the grid at 3.5 km to remain below the freezing level. Polygons are created by contouring reflectivity values of 20 and 40 dBZ to represent regions of light and heavy rain rates. We calculate two spatial metrics to represent the tangential and radial distribution of reflectivity values in accordance with the primary and secondary wind circulations within the storm. We calculate dispersion as our radial spatial metric. The area of the polygons and the distance of the polygon centroids relative to the storm center are used to determine whether reflectivity values are located closer to or farther from the storm center with the largest weights being given to the largest polygons. We employ a 600 km search radius and metric values range 0-1 with lower values signifying that reflectivity is clustered near the storm center while values near one indicate reflectivity is scattered towards the outer edge of the search radius. Tangentially, we measure closure by determining whether reflectivity values are present each one degree around the 360 degree circle. Dividing this count by 360 yields a metric whereby a completely enclosed circulation center yields a value of one and completely exposed center yields a value of zero. As a TC moves over land, the storm’s center should become increasingly exposed to the continental air mass, thereby decreasing the amount of reflectivity values around the circle and decreasing the closure metric. Once the core becomes exposed, reflectivity values should become more dispersed as rainfall decreases near the storm center and occurs mainly in the outer rainbands. The rates of exposure and dispersion differ according to whether the TC dissipates after landfall, becomes absorbed by or merges with a trough, or completes an extratropical transition.