I am sure you have looked at the forecast track chart for Fay today, Tuesday 8/19 around noon.  Looks like the computers have no clue where this storm is going.  The track cone is both large and contorted.  The weather guys have the most/best data and largest/fastest computers available.  Why can’t they do better than this?

I actually saw much more disagreement in the computer models’ track for Fay at noon on Monday than at noon on Tuesday.  Models are now unanimous that the high pressure system over the Northeast U.S. will be strong enough, and in a good enough position, to slowly guide Fay back through Florida for the next several days.

In general, computer models do an EXCELLENT job processing weather data with sound meteorological calculations at light speeds.  I - a human forecaster - can’t imagine weather forecasting without help from computer technology (like my counterparts just 25 short years ago largely had to do). 

Even still, not all computer models are created the same - some emphasize different weather parameters more than others.  Also, weather data, though extensive, can never be totally complete (one cannot know for certain the temperature, air pressure, humidity, etc. of every single point of the atmosphere at all times).  These tiny “holes” in data and calculations can, unfortunately, lead to large spreads in forecast solutions. 

We human forecasters do the best we can with the weather data and computer models we have, and I am pleased that Fay seems to be “behaving” according to the solutions of a large camp of computer model runs from the past two days.

Thanks for writing!