The National Hurricane Center routinely makes five-day forecasts of the track, intensity and size of existing tropical cyclones. The average track forecast error at five days is roughly 300 miles based on the previous five-year average. This large uncertainty is the reason NHC issues a Track Forecast Cone (that I call the Cone of Uncertainty) with each track forecast it makes.
Given that the track errors increase approximately linearly with time, one would expect the average ten-day forecast error – if one were made – to be somewhere in the neighborhood of 600 miles. With that understanding, I will share a couple of interesting graphics below.
The current ten day forecast (from 8 am EDT August 12, 2009) from the National Weather Service Global Forecast System shows a very strong tropical cyclone (the red bullseye in the above graphic) located in the Bahamas on August 22 (courtesy National Weather Service website).
The current ten day forecast (from 8 am EDT August 12, 2009) from the European Center for Medium-Range Weather Forecasts also shows a very strong tropical cyclone (the green bullseye in the above graphic) just east of the Bahamas on August 22 (courtesy ECMWF website).
In my opinion, there is a very good reason that the NHC only issues forecasts out to five days. At this time, the science doesn’t support an official forecast beyond that time period. However, the Internet has made it easy for anyone to view the longer range forecasts. No one can tell you with any certainty how accurate they will be. Sometimes the long range models show developments like this that never pan out. Still, these long range models should be a reminder to us that we are entering the peak of the Atlantic hurricane season and it would be wise to take this time to review your hurricane preparedness plan.