Comments on: Ike back over water; still moving due west

By: Nathan

Nathan — Mon, 08 Sep 2008 21:49:51 +0000

This probably isn’t news to anyone here, but the last few hours of visible satellite imagery show a rightward bend. This vindicates the models somewhat, but more importantly it means Cuba will interfere with the circulation more over the next day or so. Bad news for the poor folks who live in Cuba though–this has been a ridiculous hurricane season for them.

By: Greg

Greg — Mon, 08 Sep 2008 21:28:49 +0000

Unlike the New Orleans area where everyone has to run for the hills, Houston has zonal planning for evacuation or not.

If you are near Galveston or even some areas of Houston proper, you should go if a Cat 3 storm is bearing down on you.

http://www.houstontx.gov/oem/evac.html

By: Bo McIlvain

Bo McIlvain — Mon, 08 Sep 2008 20:50:59 +0000

I haven’t been saving copies of the model run forecasts, so I can’t dispute you with any certainty, but 3DayLoopLooksSmoothUntilToday ModelsWereAccurate in programmerspeak. The first two points in the cone wiggle around like crazy. I have a daughter living in Key West and I’ve been watching the models carefully with lots of worry about her. It’s my recollection that nearly all the models since late last week have been consistently to the north of the actual track in the 24 hour period, FWIW. You’re surely right about the southward turn Thursday and Friday, but it seems to me the storm went further southward than was predicted even then. Anyway, without firm data in my hand I’ll not press the point further.

By: Hucbald

Hucbald — Mon, 08 Sep 2008 20:22:29 +0000

This storm reminds me a little of Mitch when I was withh FEMA. It was a colossal monster and everyone said it would turn north. It never did, plowing into Mexico instead.

By: The Monster

The Monster — Mon, 08 Sep 2008 20:21:16 +0000

The key rule is “Run from the water; hide from the wind.” Low-lying coastal areas that take a lot of time to evacuate, like the Keys, NOLA, and Galveston, simply have a lower risk threshhold than higher elevations further inland.

The Keys are probably the most extreme case, because there is a single two-lane highway connecting them to the mainland. If you wait until a hit is “likely”, you’re out of time to do anything about it, and there are no structures on those islands where it’s safe to ride out the storm surge a major hurricane can produce.

If you’re in Houston proper, where a storm surge isn’t much of a problem, you can wait until the storm is practically on top of you and head for the room where you’re best equipped to survive a tornado (which you’d better have anyway, because tornadoes hit Houston even without tropical storms). Those who don’t have adequate provisions in their homes can wait nearly as long before going to stronger buildings nearby that do. There is no reason to evacuate the entire city of Houston like there is in NOLA (where a storm surge could legitimately flood the whole city).

By: Brendan Loy

Brendan Loy — Mon, 08 Sep 2008 20:08:32 +0000

P.S. Also, “all of them (except simpleminded XTRP) have predicted a northward turn in less than 24 hrs for the last WEEK” is certainly wrong. Back on Thursday and Friday, the models were — correctly — predicting a southward turn within hours.

By: Brendan Loy

Brendan Loy — Mon, 08 Sep 2008 20:06:17 +0000

Bo, if you look at the 3-day forecast loop, I think you'll find that what you've said isn't quite right. The models overstated the predicted northward motion back on Monday and Tuesday, but the Wednesday-Friday short-term forecasts were pretty good. The early Saturday forecasts failed to predict a slight southward jog, but the late Saturday and early Sunday forecasts were quite accurate out to 24 hours. Only since midday Sunday do we see serious inaccuracy in the 24-hour forecasts.

By: Bo McIlvain

Bo McIlvain — Mon, 08 Sep 2008 19:50:08 +0000

There is an amazing story here that many seem to have missed – Ike has confounded the models for days in the SHORT TERM. This is not a routine failure of long term forecasts, but a persistent failure of all the models to predict Ike’s track over the near term 24 hour period. All of them (except simpleminded XTRP) have predicted a northward turn in less than 24 hrs for the last WEEK. If I was in Cancun right now, I’d be getting worried…

By: Brendan Loy

Brendan Loy — Mon, 08 Sep 2008 18:54:45 +0000

cbrown, I did not mean to imply that the entire Houston/Galveston region would, or should, be evacuated. My wording “evacuation orders for Houston/Galveston will begin” was intended to imply that we’ll start seeing some orders for that area at that time. But I could have been clearer. I’ve re-worded it to “Berger thinks any evacuation orders for parts of the Houston/Galveston area.”

By: Greg

Greg — Mon, 08 Sep 2008 18:39:33 +0000

Brendan,

You were quoted in the New York Times. I will have to think if that is congratulatory or not.

Well, congratulations.