The able and competent forecasters at the National Hurricane Center are proud of their accomplishments but humbled and honest about their failures. While their record on predicting the track of the storm has shown slow but steady improvement the last decade their skill in predicting the intensity of the storm over the same period has been abysmal.
The science and forecasting tools (satellites, planes, dropsondes, computer models) have improved dramatically, just not in the area of intensity forcasts. The average forcast error at 24 hous before impact remains at 10 knots. The rule of thumb I learned 15 years ago, plan for an impact one category higher than forecast, still applies. The NHC documented that they are wrong by 2 categories 5-10% of the time.
Here are some notable comments that I wrote down over the last two days of listening to NHC and National Weather Service forecasters here the the Governor's Hurricane Conference in Ft. Lauderdale:
- The Saffir Simpson Scale (SSS) is now used only to categorize wind speed. A cyclone has other impacts, inland flooding, surge or tornadoes, that must be considered separately.
- Freshwater flooding from rainfall is the biggest killer in cyclones.
- Cyclone induced tornadoes most often come in an extended rain ban in the right front quadrant of the storm.
- The size and strength of a hurricane not a reliable indicator on inland flooding. Get the six hour flash flood guidance values from your river forecast center.
- The Tropical Storm Surge Probabilities Product (created in 2008) is produced only when there is an active storm threatening land. This looked like a good product to evaluate the dangers of surge during an event, but I wasn't very familiar with it. That could be because Florida hasn't been hit by a storm since 2005.
- The SSS for each storm is derived from the highest measured estimate of one minute intensity wind speed. This intense wind speed is normally only in a very small part of the right front quadrant of the storm. Thus, category 4 intensity winds, for example, may only make up 5% of the total hurricane force winds generated by the storm.