Now on ScienceBlogs: Attack of the pregnant cannibal fathers

Seed Media Group

Collective Imagination

Dynamics of Cats

Speculations on astronomy, astrophysics, news I find interesting, theoretical issues, science and science policy. I will digress into computational physics, science fiction and general issues and basically whatever I feel like whenever. And, of course, cats.

Search

Profile

MyPicture0609c.jpg Still working on an analytic exact approximate solution to the herding problem.

Recent Posts

Recent Comments

Archives

Blogroll

Other Information

« word peeps? | Main | Leave John Alone! »

Computing Gustav's Path

Category: science
Posted on: August 27, 2008 2:16 PM, by Steinn Sigurðsson

Every now and then I get a message which informs.

"If your jobs are unfortunately killed by the hurricane..."

"...simulation runs, please let us know and we will refund you the used SUs.
Sorry for the inconvenience."

LONI - the Lousiana Optical Network Initiative - is flushing its computing queue for pre-emptive hurricane modeling, this afternoon, and, I presume, through the weekend.

Current National Hurricane Center path predictions have NOLA right in the center of the predicted path - with landfall late sunday/early monday.
Cone of uncertainty stretches from NW Florida to NE Texas.



Click to enlarge

Looks like it will be category 3 or higher, depending on whether it grazes Cuba.

PS: Interesting supplemental data - summary of forecast for oil/gas interests - and predicted production list


PPS: Aug 28 06:00 forecast - that is Hanna coming in from the Atlantic

There's another major low coming off Africa for later.
Worst case scenario is Gustav mimicking Katrina or Rita, and Hanna copying Andrew.
Best case scenario - Gustav get shredded by Cuban mountains (unlikely) and coming ashore away from a city as a small low category hurricane; and Hanna curving back out to sea and dropping rain on Iceland or UK.

Share this: Stumbleupon Reddit Email + More

TrackBacks

TrackBack URL for this entry: http://scienceblogs.com/mt/pings/79812

Comments

1

Just what New Orleans needs, another hurricaine....

Posted by: Brad Holden | August 27, 2008 3:17 PM

2

Please remember that the cone of projection covers the actual path only 2/3 of the time. That means there is a good 33% chance that the hurricane will fall outside the cone you see on the map. New Orleans needs to be prepared, but so do Houma, Beaumont, Galveston, Houston, etc.

Posted by: speedwell | August 27, 2008 4:18 PM

3

What he says.
Cone is 1-sigma error. Non-gaussian truncated.

Posted by: Steinn Sigurdsson | August 27, 2008 4:21 PM

4

Also, the cone is a prediction of the path of the center of the hurricane. Strong winds and heavy rain can easily occur outside of the cone even if the center stays within it.

And while the "cone of projection" has a track record of being right roughly 2/3 of the time, the models for Gustav are actually in pretty good agreement.

(Shouldn't "NE Texas" be E or SE Texas? Can you add a link to the prediction map to embiggen?)

Posted by: Tophe | August 27, 2008 5:09 PM

5

National hurricane center:
http://www.nhc.noaa.gov/

Posted by: Lab Lemming | August 27, 2008 7:20 PM

6

Click through embiggen enabled
these are from NHC as Lab Lemming said

added oil field impact link

Posted by: Steinn Sigurdsson | August 28, 2008 12:29 AM

Post a Comment

(Email is required for authentication purposes only. On some blogs, comments are moderated for spam, so your comment may not appear immediately.)





ScienceBlogs

Search ScienceBlogs:

Go to:

Advertisement
Enter to win a free copy of The Monty Hall Problem
Visit the Collective Imagination blog
Advertisement
Collective Imagination

© 2006-2009 Seed Media Group LLC. ScienceBlogs is a registered trademark of Seed Media Group. All rights reserved.

Sites by Seed Media Group: Seed Media Group | ScienceBlogs | SEEDMAGAZINE.COM