Analysis of Upstream Synoptic Conditions for Tropical Cyclones That Pass near or over Florida

During the course of the Atlantic hurricane season, numerous tropical cyclones (TCs) recurve around the Bermuda high. Where this recurvature occurs can have a large influence on societal impacts. Even with the reduction in official forecast track errors observed over time, it is at times difficult to forecast exactly when a storm will recurve (e.g., Irma 2017). This study analyzes the upstream synoptic scale patterns for both landfalling and recurving TCs near the Florida coastline. The goal is to determine if there are specific synoptic conditions that can help predict whether a TC is more likely to make landfall in Florida or recurve to its east 1-7 days before the landfall or recurvature occurs. The National Hurricane Center Best-Track and the ERA-Interim (ERA-I) reanalysis datasets were used during the period 1979 to 2017. The TCs were sorted into three categories, with the following times noted in parentheses: landfall (landfall in Florida); recurving set 1 (first motion vector with a greater northward component than a westward component); and recurving set 2 (first motion vector with an eastward component). The 500 hPa geopotential height fields were recorded every 24 h from 24 to 168 h prior to the times of landfall or recurvature using ERA-I data. Composites were created for each category and each time prior to landfall or recurvature. Differences in the composite means were then calculated and tested for statistical significance. Preliminary results indicate that the landfalling storms occurred when a weak ridge was over the southwestern United States during the time frame of 24 to 168 h prior to landfall. Similarly, both recurving categories occurred in the presence of a much stronger ridge over the southwestern United States during the time frame of 24 to 168 h prior to recurvature. These results suggest that there is a correlation between whether or not a TC will recurve depending on the strength of upper level synoptic patterns.