Share this article

print logo

Don Paul: Irma's threat to Caribbean and Florida continues to grow

East of the Caribbean, Irma has already been the strongest hurricane on record in the Atlantic basin.

An anemometer on Barbuda recorded a gust of 155 mph before breaking, very early on Wednesday. At the time, top winds of Irma, which moved directly over Barbuda, were estimated at 185 mph with occasional gusts breaking 200 mph. Shortly afterward, as the eye passed overhead, the wind dropped to 6 mph briefly at another Barbuda site, before roaring upward again as the other side of the eyewall hit.

Click here to see what it looked like, but allow loading time.

Irma will have fluctuations in strength over the next five days. There will be what are called eyewall replacement cycles, in which a new eyewall takes shape as an older eyewall dissipates. During this cycle, a hurricane will temporarily weaken before regaining its intensity. During other timespans in its path toward the west and northwest, stronger winds aloft — wind shear — may increase from time to time, disrupting Irma’s circulation and causing some weakening.

The NOAA forecast, which is updated every six hours.

However, virtually all models keep Irma as an extremely dangerous and powerful Category 4-5 hurricane at least until it draws near Florida. At left is the National Hurricane Center track and intensity forecast for Irma as of midday Wednesday.

There are two primary caveats on Irma’s intensity: One, if the center tracks over Hispaniola or Cuba, mountains there would disrupt and weaken the circulation, though the devastation on those islands would be catastrophic. Two, if more wind shear enters Irma’s circulation, that could cause weakening. For all the winds hurricanes produce, they don’t thrive when stronger winds aloft are blowing IN to them. You might say they can dish it out, but they can’t take it.

As of this writing, the majority of models and ensembles keep the center tracking just north of Hispaniola and Cuba, which would allow the full potential of heated ocean waters to fuel Irma’s intensity with less disruption from land terrain.

(NOAA graphic)

At left is a current depiction of sea surface temperature anomalies — departures from normal readings this time of the year. You’ll note waters are warmest over the northwest Carribean, but they are above average all around Florida and Cuba. See that patch of blue in the northwest gulf? That’s the cooling that occurred when Harvey extracted much of the heat from those waters prior to and during landfall.

Most of you are at least vaguely aware meteorologists have a huge arsenal of models and model ensembles.

(Ensembles are multiple runs of models, each with slightly varied “initial conditions” because we cannot know the precise state of the atmosphere at any given moment, due to sparsity of observations. So, modelers create a reasonable range of differing initial conditions. For example, the vaunted European model has 51 different runs, due to superior computer crunch power. The American GFS model has 21 different runs in its ensemble. The GFS has improved, but it has a way to go to consistently match the European for accuracy. None of this means the European should always be favored over the GFS. All models have good days and bad days. The European, however, has more good days than the GFS.)

The track forecast above has shifted the most likely range of tracks farther east, near Florida, than Tuesday’s forecasts. In fact, many models now support such a shift, and some shift the track farther to the east of Florida. Why wouldn’t the NHC jump all over this trend? Because meteorologists want to see more consistency from model run to run. Things can shift back to the west again in the next few runs. The NHC appropriately is more conservative with track forecast changes because that’s usually good science. Jumping on a sudden change is often bad science.

Here’s an example of just one model, the GFS, showing an eastward shift. (Due to heavy internet traffic, it may take a bit longer to load.) This particular track would be catastrophic for at least eastern Florida if the track verifies. It is somewhat east of yesterday’s run.

Here is a Canadian model that shows a little less intensity due to prior interaction with Cuban terrain. It begins to recurve Irma toward the northeast as it nears Jacksonville, and turns it toward the Carolinas.

The European model has also shifted east from yesterday’s runs, which had actually focused on Irma being near or just west of southwest Florida before turning more north. The overnight run also focuses now more closely on the east coast of Florida.

None of these more easterly track solutions means the west coast of Florida gets off scot-free, but the worst destruction with such an easterly track would be in southern and eastern Florida.

Here’s what’s clear as of this writing: People in Florida should be rushing preparations to protect their property by Friday. They should be prepared to obey evacuation orders, the sooner the better. Highway infrastructure will make near-gridlock a certainty, so I suspect the governor and local authorities will not waste time in ordering early evacuations. Yes, there is still a chance model solutions will point to the storm staying farther off the coast in later runs. But if Floridians wait to take action and that shift does not occur, they will be trapped where they are.

The first impacts will reach the Keys, probably building during Saturday, and then reach at least South Florida by early Sunday.

As always, beware of fake weather forecasts on Facebook and Twitter. Know your source! There are so many rank amateurs out there, just waiting to add to the alarm and panic because they get some sociopathic thrill out of stirring the fright pot. This is no time for what I call “Toy Weather.”

The National Hurricane Center issues all watches and warnings for hurricanes, though web traffic may slow response time on their sites a little. Here is a valuable forecast product from NHC, Hurricane Force Wind Speeds, that will be updated regularly.

Story topics:

There are no comments - be the first to comment