Forecasting Isaac

By: Stu Ostro , 1:28 AM GMT on September 10, 2012

Share this Blog
10
+

Jeff Masters, Bryan Norcross and I were among the meteorologists asked by TWC to participate in a Q&A (a blog roundtable of sorts) about forecasting and communication challenges with Hurricane Isaac.

The article is at this link.

In it my answers were interspersed amidst the others, and not all of my quotes were used. They are together and in full here:


1. What aspects of Isaac and its environment proved most challenging to the forecast models?

The subtle difference between just enough of a weakness in a ridge of high pressure to its north to allow it to move north into Florida, or that ridge being just strong enough to force Isaac farther west to the central Gulf Coast, as is what happened.


2. Why didn’t Isaac intensify as much as it could have?

It seemed to be fighting relatively dry, stable air throughout its whole life. There was a bit of wind shear at times, though generally Isaac had enough upper-level outflow (the storm’s “exhaust”) that is conducive to strengthening. Also, it can be easier for tiny tropical systems to be able to spin up in a shorter period of time than massive ones like Isaac, but such large size does not preclude rapid intensification of the core if conditions are right.


3. What turned out most surprising to you about the direction and development of Isaac?

There wasn’t much that was very surprising. Although the track had some wiggles, it was generally pretty straightforward from the Leewards to Haiti & Cuba and the Gulf of Mexico. As noted above, the meteorological difference between its ultimate destination being the central Gulf Coast vs. Florida was subtle but clear in why that was the case, and did not represent a huge geographical difference in those model track forecasts at the time frames involved. And Isaac’s pace of formation and strengthening was understandable given its size and the things in the atmosphere it was fighting.


4. What does Isaac teach us about the state of the science in predicting both path and intensity of tropical cyclones right now?

We already knew what the advancements and challenges are, and Isaac didn’t change that. Debby, earlier this season, was a much more interesting case forecast-wise, when two of our main models predicted it to go in completely opposite directions, and what is generally thought of as being the more reliable and accurate model had the storm going west to Texas, which was an epic FAIL.

One thing Isaac did was to reinforce that advancements have been made in models’ predictive ability in the long range – that they can sometimes be able to correctly identify two or more weeks in advance that a potent tropical cyclone is likely to develop – while also highlighting the limitations in track forecast accuracy and precision that can exist more than a couple days in advance, and expectations in that regard need to be realistic.


5. Is there a need for a scale that takes into account more of a storm’s variables than the Saffir-Smpson Hurricane Wind Scale?

Yes. We have the Saffir-Simpson scale for wind intensity. We have the “cone of uncertainty” for the track. There needs to be a way to better address the *size* of the storm. Size matters for geographical extent, duration, and in some cases magnitude of impacts. Compared to if Isaac had been a tiny hurricane of the same intensity, its power outages were far more widespread, and its storm surge affected more locations along the coast and in some places was much higher.

What is known as “IKE” – Integrated Kinetic Energy – is a measure that incorporates both intensity and size, and from which is computed a “destructive potential rating” on a scale of 0 to 6 for wind and surge, but it has its own share of technical limitations and communication complexities. The trick is to find a way to best make, for example, the message of "Isaac is immense and it shouldn’t be thought of as 'only' a Category 1" really resonate and stick.


6. Where is the breakdown between forecasters and the message they attempt to pass along regarding the impending threat of a hurricane and the public's uneven response to those warnings?

If there was one, it was the answer to #5 above.



Image credit: NASA Earth Observatory

Reader Comments

Comments will take a few seconds to appear.

Post Your Comments

Please sign in to post comments.

or Join

Not only will you be able to leave comments on this blog, but you'll also have the ability to upload and share your photos in our Wunder Photos section.

Display: 0, 50, 100, 200 Sort: Newest First - Order Posted

Viewing: 8 - 1

Page: 1 — Blog Index

7. WunderAlertBot (Admin)
8:36 PM GMT on October 10, 2012
stuostro has created a new entry.
6. Patrap
3:44 PM GMT on September 26, 2012
A cosy and most excellent entry.

Thanx from Se. Louisiana.
Member Since: July 3, 2005 Posts: 415 Comments: 125753
5. DocNDswamp
4:12 PM GMT on September 11, 2012
G'day Stu,
Thanks for your reply and link to your analysis - excellent!
Had much discussion here on WU at the time over classifying it, as a subtropical low, sheared TC, to non-tropical or semi-tropical hybrid... It was indeed an odd bird, appreciate your perspective on it. That was one screaming banshee of a low, a short-fuse situation that caught quite a few by surprise, made more difficult by the opposing model solutions leading right up to point of it's formation - from what I recall in my own observations, the first sign the GFS was on right track was viewing wind shifts / pressure falls / hints in vis sat imagery over SE / E Cuba the afternoon before, lifting NW-ward.
Thanks again!
Member Since: September 21, 2005 Posts: 94 Comments: 4783
4. Stu Ostro , Senior Meteorologist
11:02 PM GMT on September 10, 2012
Thanks for the comments, folks. DocNDswamp,you might be interested in this blog I posted on weather.com at the time about that October 2011 storm, which included this pressure trace from a C-MAN station near Cape Canaveral ...

Member Since: August 24, 2012 Posts: 30 Comments: 23
3. aislinnpaps
9:30 PM GMT on September 10, 2012
Very interesting, thank you.
Member Since: August 22, 2009 Posts: 0 Comments: 3113
2. DocNDswamp
3:46 PM GMT on September 10, 2012
Hello Stu,
Excellent responses...
Indeed, once again Isaac proved that with the subtle changes in steering ridge / trof evolution in our chaotic atmosphere, with many tropical cyclones the track forecast beyond 72 hours can not be stated in high-confidence other than within a 250-300 mile spread... That's roughly the difference from about Morgan City LA to Panama City FL, which means significantly different levels of actions either community must take. I can assure you once I saw the GFS westward trends on Sat Aug 25th, I began much more serious, intensive preparations here in SE LA, not waiting for official forecast outlooks that eventually followed later on Sunday and beyond... And in my case, with family to reign in and several properties to securely storm-harden / finish preps, it took every bit of those 3 days, completed right to the point when TS force winds began.

Despite all advancements, I don't see it getting any better than this 3 day limitation yet we're fortunate to have attained this degree of confidence and insight.

As for GFS verses ECMWF, despite questionable claims of model superiority (via NHC's comparative interpolation method) I haven't seen convincing evidence of such, quite the opposite in many cases over past few years are equally documented, not just this year as some want to believe due to recent GFS upgrades. I'll highlight this - TS Debby wasn't the 1st major blunder that NHC / NWS offices decided at crunch time to accept euro forecast and ignore the GFS as an outlier. Apparently an important lesson wasn't learned by forecasters from what happened almost a year ago with the Oct 7-9 2011 event, coincidentally involving Florida - acknowledging opposing model solutions, every NWS forecast office in FL bought into the ECMWF solution of an E Gulf subtropical low forming / hitting the W coast FL in their forecast, while the GFS correctly and consistently depicted the ULL-spawned sfc low developing more eastward over Bahamas, hitting E coast FL... NHC never declared the sfc low system as a tropical / subtropical entity, but if they had, their season's forecast model accuracy determinations might have concluded otherwise.

Member Since: September 21, 2005 Posts: 94 Comments: 4783
1. stormchaser19
2:21 AM GMT on September 10, 2012
With Debby NHC change the the forecast point about 180 grade....Incredible the science is far to be perfect
Member Since: June 20, 2012 Posts: 0 Comments: 2130

Viewing: 8 - 1

Page: 1 — Blog Index

Top of Page

About stuostro

Senior meteorologist at The Weather Channel. Proud to be a weather-obsessed weather geek. Would be a DJ if not a meteorologist.

Local Weather

Partly Cloudy
63 °F
Partly Cloudy