Evaluation of Synoptic Patterns for Atlantic Hurricanes Categorized by Ensemble Track Error and Variance

Since the 1970s, Numerical Weather Prediction Models have made significant improvements in forecasting the large scale forcings on tropical cyclone motion. As a result, there have been significant improvements in tropical cyclone track forecasts during this time span. However, there are still instances where several independent forecasts are in agreement on the track of a storm, yet the actual track is different (e.g. Hurricane Alberto in 2000). This is an example of a forecast with low variance among the forecast members, yet high error. Lives and property are most at risk when this situation occurs near land. This study uses the Global Ensemble Forecast System (GEFS) data to analyze Atlantic tropical storms from the years 2004-2008. The ensemble variance and mean error in the GEFS track forecast of all tropical storms existing during those five years is calculated. The ensemble variance of the track forecast as a function of forecast hour is used to classify the track forecast as having high, medium, or low confidence. The ensemble mean track compared to the best-track is used to classify the forecast as having high, medium, or low error. Each hurricane is placed into a single confidence/error bin, which is determined by the ensemble spread and error of the track forecast. The more frequently studied cases where there was low confidence and high error in the track forecast will be identified, but not further examined. The more disturbing cases where a highly confident track forecast was accompanied by high error are focused on in this study. Synoptic environments common to these cases will be examined. Track forecasts with low ensemble spread and high error have a high potential to hinder hurricane preparedness actions. They could lead to excessive evacuation orders, or worse, failure to order necessary evacuations. Identifying commonalities in the synoptic set up among tropical cyclones that fall into this category could be useful in future hurricane track forecasting. It could serve as a signal to alert forecasters when a highly confident track forecast may be misleading.