Tropical Cyclone Regions

Tropical Cyclone Info

INTRODUCTION

        These forecasts have always been targeted more to the landfall than the actual numbers of tropical cyclones.  The reasoning being that the numbers do not always reflect the probability of a tropical cyclone impacting the coastal areas surrounding the Atlantic basin.  The best example of a "slow year" in numbers but a "large year" in terms of human impact is 1992, when there were only 7 named storms, but 2 landfalls, one of those being Andrew, one of the most devastating ever in U.S. history.   And, those years with a high number of storms do not always produce a large number of land-falling cyclones, 2000 for example with 15 storms and 4 landfalls.  It is for this reason that we have focused on forecasting those areas of the populated coastline that are most likely to experience a land-falling tropical cyclone.  The "numbers" forecast is merely a by-product.  It is our belief that there are several atmospheric and oceanic patterns in the 3 months prior to the beginning of the "Atlantic Hurricane Season", June 1 through November 30, that are statistically viable precursors to those atmospheric conditions that will be the drivers behind the eventual track patterns for those tropical cyclones which develop in the following season.

        These atmospheric and oceanic predictors are not useful in the day to day forecasting required during the actual events themselves, but do demonstrate a viable linkage in the long term.  The following factors are to be considered in the use of these predictors .....

1. the predictors by themselves do not always have a "one to one" relationship with the numbers of tropical cyclones that form
2. there is no single predictor that is statistically much greater than all of the others
3. there are innumerable outside influences that can affect the outcome in a single season which can make the forecast wrong
4. based upon hind-casting, or the comparison of previous years forecasts made after the fact, these predictors can explain approximately 80% of seasonal variation
5. we can never perceive of a time when there will be enough knowledge of all of the individual determining factors to say unequivocally that any forecast is 100% accurate. the probability that the predictors will vary enough to be statistically significant and thus alter the forecast significantly during the forecasted season is very small, but it does happen, 2006 being a prime example
7. there are predictors that we have as yet not determined that could have a significant impact on the future forecasts

FORECAST AND METHODOLOGY EXPLANATION

Previous forecasts verifications graphics.