September 09, 2017

Hurricane Irma Takes a Western Route Northward

It is looking increasingly likely that Hurricane Irma will take a more western route than initially predicted, with negative implications for the west coast of Florida and more benign conditions for Miami and the populous eastern side of the peninsula.

The NWS Key West weather radar at 9:34 PM EDT Saturday was clearly picking up the storm's eye and spiral rain bands.

Wind gusts so far have gotten to 55-65 mph at some coastal locations, with the worst yet to come.

The European Center Model has been the most skillful weather prediction model for this storm, and its latest forecasts (initialized Saturday morning) has most forecast tracks (from its ensemble system) moving up the Florida West coast (see graphic courtesy of WeatherBell Analytics).

As I noted in my last blog, this is a VERY hard forecast, with the storm predicted to make a sharp right turn near a relatively narrow peninsula.   Small errors in position and time of the turn have HUGE implications for the forecasts.  The storm has also weakened substantially (now a category 3), with the latest model runs suggesting some intensification (perhaps to a cat 4, but NOT a cat 5).

Let me show you the pressure and surface (sustained) wind predictions from the European model, courtesy of WeatherBell.  At 8 AM EDT Sunday, Irma is nearly over the Florida Keys,

 By 8 PM Sunday (EDT), the eye is near the SW Florida coast,
 And near Tampa at 8 AM Monday.  The strongest winds stay on the west side of the Peninsula, where substantial damage would occur.
Similarly, the largest precipitation totals (5-15 inches) remain over the west side the peninsula (see graphic).  A huge storm surge is probable along the western shores.

In a later blog, I will analyze the predictions of the various modeling systems for this event.


  1. How confident are predictions of a northward turn vs a more northwesterly route into the gulf proper then on to one of the gulf states?

  2. This is turning into an obvious "bust" by the NHC. They caused the evacuation of the wrong side of the FL coast. Many Miamians went to Tampa which was outside the predicted path.! The national news media is also to blame for their over exuberant scare stories. There was way to much reliance on the HWRF models vs the European ensembles. I suspect in the future the "track cone" will be wider for forecasts beyond 72 hours.

  3. I had a good laugh watching The Weather Channel yesterday morning (ok I was indulging in a bit of disaster porn). The commentators were commenting as the storm moved west along the coast of Cuba and one of them said "it must turn north, all the models predict it!". What a mindset- the climate models are the arbiters of reality.

  4. Some reasonable predictions, if things play out as now expected:

    1) Some of will note that the inherent error in any predictive model is reason to dismiss any modelling as unreliable.

    2) Some ( perhaps the same as above) will note that scoring less damage on Miami than the predicted potential is every reason to assume that Miami risk management is once again proven superior and reliable ( I can hear the Rush Limbaugh radio show now)

    3) Some will assume (etc) that now the storm has passed, Miami is off the hook for the next hundred years.

    4) Some will assume correctly that this single storm is no conclusive result of global warming.

    5) Those precise same people will ignore the obvious: While not directly implicated, it none the less reveals the growing vulnerability of Miami and the entire state just wonderfully, if we are ever to manage risk as humans have successfully done to date by forecasting skillfully under conditions of uncertainty, then acting in advance.

    Other than that, I sincerely hope all modelling is off and it spins out its energy offshore somewhere. Good luck Florida!

  5. This is the biggest storm from a "hype" perspective. The media pumped and primed all week and the spigot yields a few rusty drops at the end

  6. Dr. Jeff Masters, co-founder of the Weather Underground, has a good discussion of the models used to predict the track of hurricanes.

  7. Cliff
    It's so easy for the critics ( I can think of a few "renown" "Florida" ones) to say how far off this track was. For those of us who love science, we ought to be passing out kudos for the incredible forecast as of 10 days ago that brought a track from the Atlantic to within miles of the actual track. The emphasis on storm surge is equally noteworthy.
    But, I would say that we need to figure out a way to convey a sense of uncertainty of a forecast, and to convey, equally, that living with that uncertainty ( and telling millions to move) is still a better option than doing nothing.
    Maybe a summary of the devastation of Typhoon Haiyan which left about 7300 dead and missing might add something to this perspective.

  8. Won't it die out once it turns northward without the warm water to power it?

  9. @Unknown, but it's the Global Warming Apocalypse, because Seattle "progressives" have never heard of hurricanes until this summer! Get with the program!

    Oh Category 2? No way! Category 12, damn it!

  10. Cliff, this storm is undoubtedly powerful, but I'm a child of the Columbus Day 1962 storm because I was in 8th grade that year. I've seen it characterized as category 3 equivalent. As bad as the damage is in Florida, setting the media speculation aside, the damage I experienced in 1962 leads me to wonder if the Columbus Day storm wasn't more powerful than Irma has proven to be in the U.S. I have to believe that the Caribbean suffered the most being hit in Irma's category 4 and 5 phases, but other islands seem to have been on the eyewall's left, less powerful side, and the Keys and Florida to be hit while Irma was, overall, about comparable in power to the Columbus Day storm. After all, Columbus produced measured winds in excess of 145 and one gust measured at 160. Any thoughts on this matter or is it extraneous? Thanks.

  11. To Penquin who said: "What a mindset- the climate models are the arbiters of reality."

    Arbiter is the wrong word. Predictor is more accurate. And the models were right. It did turn north and it stayed within the earlier predicted cone of uncertainty.

    As for those who thought the forecast was blown, remember that the NHC claims that a track anywhere inside the forecast cone is possible and to prepare accordingly. The entire state of Florida has been inside that cone for most of the week. West and east coast. Anyone who traded one coast for the other simply didn't understand how hurricanes are forecast (which unfortunately, the media always distorts and exaggerates.)

    What people are focusing too much on is the consensus track, which is the average of many tracks, and running from that line as if it was magical. It is an interesting line as it reflects the changes in guidance, and moves with updated observations and model runs, but I think NHC would do some people a service by not showing that line and just stick to the cone of uncertainty.

    Bottom line, if you are anywhere in or near the projected cone, you should prepare as if it's coming over your house. Any other action is the result of magical thinking, IMHO.

    And yes, the media went nuts as always. Hard to believe that anyone could watch CNN for more than few minutes without suffering a little nausea. But I'm sure they sold a lot of soap with all their ads. The reality show junkies probably stayed awake for days glued to their screens watching reporters struggling to make it sound scary.

    Note: I used to live in FL and have seen a few hurricanes. The storms never bothered me all that much (a good excuse for a party), but rather it was the cleanup afterward from the "veggie shredders" that I hated.

  12. I gave up on The Weather Channel years ago, when they turned away from forecasting the weather towards ghoulish wishcasting of imminent doom and death for everything from thunderstorms to N'oreasters. Watching short clips of their jackass "reporters" straining mightily over the weekend to showcase all of the alleged disasters just waiting in the wings was pathetic.


Please make sure your comments are civil. Name calling and personal attacks are not appropriate.

The Other Type of Mountain Wave Cloud

 Folks love to talk about lenticular clouds , which are generally produced by air moving up (and down) downstream of a mountain barrier (see...