# 2262
During the months of August, September, and sometimes into October my attentions are divided between pandemic issues and storms forming in the Caribbean and tropical Atlantic.
These three months are the peak of the six-month Atlantic Hurricane season, and this year, like many years in the recent past, looks to be an active one.
It wasn't always so. The decades of the 1970's and 1980's, and the first half of the 1990's, were relatively quiescent as far as 'canes went. Sure, we had them, but more infrequently than we've seen over the past 10 years.
When I was a kid, growing up in the 1950's and 1960's, Hurricanes were plentiful. Much like we see today.
These cyclical patterns aren't well understood, but they are well documented. We are in the midst of an upswing in tropical activity, and that is forecast to last for another decade . . . maybe longer.
On the eve of the third anniversary of Hurricane Katrina's landfall along the Louisiana/Mississippi coast we have a new threat lurking south of Cuba, with the potential to visit upon the Northern Gulf Coast a tremendous blow sometime in the next 120 hours or so.
I say `potential', because we do not know yet where Gustav is going, or even when it will arrive. We also don't know how strong it will be when it hits . . . someplace.
What we have are models, complex mathematical expressions of how this tropical system may interact with the atmosphere around it, giving us some ideas where Gustav may be going.
You may have seen these models, often called `spaghetti models' for obvious reasons, before. As you can see from this latest run, the models put Gustav approaching landfall somewhere between Texas and the Florida Panhandle on Monday or Tuesday.
Believe it or not, these models are showing a rare degree of unanimity. We don't often see all of the models pointing in the same general direction. Even so, that's a lot of territory to warn, particularly if your job is to decide whether or when to order an evacuation.
The NHC's (National Hurricane Centers) `consensus' plot takes all of these models and tries to formulate a `best guess' of where this storm is going.
Here is this afternoon's latest track. Don't just look at the center-line track . . . five days out the average margin for error is 300 miles either side of the track.
You have to look at the `cone'.
And in this case, it is a very large cone. And this `cone' may very well shift left or right several times over the next 72 hours, as new data gets input into these computer models.
Now, how do you decide who to evacuate . . . and when?
Adding to the confusion, we really don't have a very good handle on intensity forecasts. Right now, the smart money is on Gustav becoming a CAT 3 storm by Sunday . . . but nobody really knows.
And what will he be by Monday or Tuesday . . .?
The bottom line, right now, is we probably won't have a good idea where Gustav is going to land until Saturday morning, or perhaps Saturday night. And even then, the 48 hour forecast could be a couple of hundred miles off base.
Evacuating a city like New Orleans isn't something you do on a whim. It is a horrendously expensive undertaking, and it puts lives at risk as well. If you `pull the trigger' early enough to give people time to get out . . .you risk ordering an unnecessary evacuation.
If you wait . . . well, we saw what happened in 2005.
And it's not as if New Orleans is the only populous area that may need to evacuate. Depending on the strength of storm, many low lying areas would be highly vulnerable to a storm surge or hurricane force winds.
What should authorities be doing about people in Pensacola, or Mobile, or Beaumont, Texas?
They are all in the cone.
These are not easy times for emergency planners, and many tough decisions will have to be made in the coming hours.
Millions of gulf coast residents right now are watching, and waiting, essentially prisoners of these model runs. These are our best tools for forecasting the movement of a storm, even though they are subject to increasing margins for error the further out in time they project.
These `model prisoners' will have to decide, probably on Saturday, whether or not to evacuate. And where they will go. No easy task when hundreds of thousands of other people are all trying to do the same thing.
Yesterday, I was in the `cone'. Today I am not. Tomorrow, well . . . we'll know tomorrow.
That's life in Hurricane Alley.
Hopefully Gustav won't come my way. There's always Hanna . . . and her sisters, for me to worry about later next week - but that's next week.
If you've ever wondered why I am such a proponent of being prepared, the little graphic above should give you a pretty good idea.
