A reanalysis of twelve U.S. landfalling major hurricanes

The Atlantic hurricane database (HURDAT) contains track and intensity estimates for all tropical storms, subtropical storms, and hurricanes back to 1851. Wind (intensity) estimates from Atlantic basin tropical cyclones are recorded in HURDAT in 6-hourly intervals as the maximum 1-min surface (10 m) wind speed (in 5-kt increments in the 20th and 21st centuries) within the circulation of the tropical cyclone. Position estimates (also in 6-hourly intervals) are recorded as the location to the nearest 0.1 degree. HURDAT was first developed to provide objective track guidance tools during the 1960's and is utilized in a wide variety of ways, including climatic change studies, seasonal forecasting, risk assessment for emergency managers, analysis of potential losses for insurance and business interests, and the development and verification of both official National Hurricane Center and computer model predictions of track and intensity.

Using the original HURDAT file for twelve existing U.S. landfalling hurricanes, a wind swath analysis footprint was produced. The storm footprints are produced by the AEF RealTrack™ system, which guides a numerical weather prediction hurricane model along the prescribed track.

Using the storm footprint technology, it can be determined whether the maximum wind each state experienced for the prescribed track and intensity record can be validated. This is done by comparing the resultant maximum state winds from the footprints to the state-based Saffir-Simpson Hurricane Scale categories in the best track dataset. Through this analysis and comparison, it was revealed that large discrepancies often exist between the HURDAT record and the Saffir-Simpson state-based categorizations for the affected states.

A complete re-analysis was thus conducted to resolve the discrepancies between the original HURDAT and the state-based Saffir-Simpson categories of impact. The Atlantic basin hurricane database re-analysis project is an on-going effort to extend, revisit and revise the database, including official tracks and intensities of tropical storms. The re-analysis of the Atlantic Hurricane Database is both needed and timely as the records contain random and systematic errors that need to be corrected. In addition, the current knowledge of tropical cyclones has advanced beyond the point of those used in the past. An example arises from surface wind estimation techniques that have evolved over the years, leading to biases in the historical database that have not been addressed. Also efforts led by Jose Fernandez-Partagas to uncover previously undocumented hurricane extended the record from 1851 to 1885 and substantially revised the records from 1886 to 1910.

Preliminary results for the reanalysis of these twelve hurricanes – the 1915 Louisiana hurricane, the 1919 Florida Keys/Texas hurricane, the 1934 Louisiana hurricane, the 1938 New England hurricane, the 1944 Great Atlantic hurricane, the 1945 Texas hurricane, the 1947 Florida/Louisiana hurricane, 1954's Carol, 1954's Edna, 1954's Hazel, 1960's Donna, and 1985's Gloria – will be discussed. Each of these will be assessed by the National Hurricane Center's Best Track Change Committee before official changes to their HURDAT records will be made.