Visual Representations of the Spatial Relationship Between Bermuda High Strengths and Hurricane Tracks

Jason T. Knowles, Michael Leitner


The 2004 and 2005 hurricane seasons dramatically demonstrated the magnitude of the societal significance of hurricanes, negatively impacting on all scales from the personal to the national. Although definitive identification of the forcing mechanisms controlling hurricane tracks and landfall patterns remains elusive, increasing evidence supports the hypothesis that the increase in hurricane activity along the Gulf Coast is due to a southwestward shift in the position of the Bermuda High. This research uses multiple visualization techniques to explore the spatial correlation between Bermuda High strengths - as interpreted from the North Atlantic Oscillation (NAO) index - and hurricane tracks. Using hurricane vector data from the National Oceanic and Atmospheric Administration(NOAA) Hurricane Data set (HURRDAT) and NAO index data since 1947, the hypothesized spatial relationships were investigated. Due to the vast number of storm track segments (more than 17,000), displaying all segments in the same map failed to reveal any coherent spatial pattern. For this reason, storm track segments were converted into a point coverage, each point representing the mid-point of an original storm segment. Other visualization methods were applied to this new point coverage, including choropleth mapping and continuous 2-D and enhanced 3-D surface displays. The latter two methods were novel approaches for the visualization of large numbers of hurricane tracks and can be applied to any large data sets consisting of linear features. Results visually support a spatial relationship between hurricane tracks and Bermuda High strengths.


geographic visualization; kernel density estimation; Bermuda High Hypothesis; hurricane tracks

Full Text: PDF


  • There are currently no refbacks.