A GIS Analysis of Radar and Surface Wind Data from a Landfalling Hurricane
Corene Matyas email@example.com
Department of Geography, University of Florida
Due to the popularity of employing a geographic information system (GIS) to analyze climatologic and meteorologic data, surface winds from hurricanes and WSR88-D radar data are now available in GIS-friendly formats. This study examines several methods used within a GIS to analyze these data for the Florida landfall of Hurricane Charley (2004). GIS-ready surface wind data are acquired from the H*Winds website. Radar reflectivity returns and hourly precipitation totals are obtained through the National Climatic Data Center website, and their Java NEXRAD tools are used to georeference the data prior to importation into ArcGIS. Within the GIS, the wind and radar-derived data are converted into polygons, whose centroid, area, perimeter, compactness, elongation, and orientation are calculated. The perimeters of the rain- filled polygons are defined by the outer edge of the 20, 25, 30, 35, and 40 dBZ reflectivity values. The polygons representing the wind fields are created using 64, 50, and 34 kt values as the perimeter. Merging the wind field and rain shield polygons allows a calculation of the percentage of the convection located within the radius of a given wind speed, or a specified distance from the storm’s circulation center. Spatial overlay tools allow the determination of the portion of the wind field or rain shield that intersects the land surface to be calculated. Converting the polygon data into raster format allows cell values to be added to determine the storm total rainfall in a particular location.
Results of analyses from Hurricane Charley are presented for a 24 hour period that encompasses its rapid intensification, landfall in southwestern Florida, and emergence over the Atlantic Ocean off the coast of Daytona Beach, Florida during the early stages of an extra tropical transition. Examination of the elongation ratios show that the main region of strong convection is twice as elongated as the rest of the rain shield as it is located outside the radius of hurricane-force winds, but becomes more circular during landfall while the remaining rain-filled polygons elongate. Prior to landfall, the main area of stratiform precipitation and strong convection are similarly compact while regions bounded by 30 and 35 dBZ reflectivity returns have a longer perimeter. After landfall, the perimeter length increases for all regions of the rain shield until the extra tropical transition commences. Examination of orientation values confirms these observations as all polygons are oriented in the same direction as the storm heading at the end of the study period, indicating strong atmospheric influence on the storm structure. During rapid intensification, the wind field of Charley elongates, and then becomes more circular and increases in area during the early stages of the extra tropical transition. These results are in good agreement with the expected changes to the storm’s structure, indicating the potential for these methods to facilitate the comparison of the structures of additional landfalling hurricanes.