Network News

X My Profile
View More Activity
The new Washington
Post Weather website
Jump to CWG's
Latest Full Forecast
Outside now? Radar, temps
and more: Weather Wall
Follow us on Twitter (@capitalweather) and become a fan on Facebook
Posted at 11:00 AM ET, 09/24/2010

Tropical Storm Matthew may threaten U.S.

By Greg Postel

* Near-record heat today: Full Forecast | Hurricane Tracking Center *

Tropical Storm Matthew formed in the Caribbean Sea yesterday afternoon, roughly halfway between Jamaica and South America. There's already been much speculation on the potential for Matthew to impact the U.S., and how an active tropics and overall change in the weather pattern may bring substantial rains to what's been a dry Eastern U.S.

Location of Tropical Storm Matthew as of this morning. Credit: National Hurricane Center.

At this point, there are too many unknowns in the forecast to say with any certainty how (or if) Matthew might impact the Eastern Seaboard. The global weather models are still are not overly confident in the position and intensity of the steering currents and other features that will guide Matthew along, let alone the complex interaction between these features and the storm itself. But there is enough evidence to suggest Matthew, or the remnants of Matthew if it weakens, may at least get pulled northward into the Gulf of Mexico at some point next week.

Keep reading for more on the outlook for Tropical Storm Matthew...

As of 11 a.m. Friday, Matthew's maximum sustained winds were estimated at 50 mph. The storm was moving westward at 20 mph and showing signs of significant organization. Unfortunately, as highlighted in my last post, steering currents in the upper levels of the atmosphere have recently shifted in a way that will eventually try to tug Matthew northward toward the Gulf of Mexico.

The satellite presentation of Matthew is impressive at this early stage, with a spiral-shaped appearance in both the lower clouds and the high-altitude cirrus outflow.

With very warm ocean waters and not much wind shear (winds that change speed and/or direction with height and can weaken storms) near Matthew, the only apparent hinderance to its intensification over the next day or two is its proximity to land. The official forecast from the Nation Hurricane Center (NHC) actually has Matthew crossing over Central America tonight and then over Belize and the Yucatan Peninsula early next week.

As I detailed in a post earlier this week, what's most concerning for the U.S. is that the upper-level circulation over North America and the Atlantic is beginning to outline a path for landfall along the Gulf Coast and/or in the Caribbean. Computer model forecasts for Matthew are consistent with this idea, with some models turning Matthew northward sooner than others.

Over the next several days, an upper-level area of of low pressure (with a counter-clockwise circulation) will migrate southeastward from the Northern Plains. Forecasts from global weather models suggest it will arrive over the Southeast U.S. early next week. The figure below shows a forecast of upper-level winds for Tuesday of next week. The southward extension of counter-clockwise flow all the way to the Gulf of Mexico is something we haven't seen for several months, and is consistent with the expected pattern shift.

Computer model forecast of upper-level winds for Tuesday Sept. 28. Image credit: University of Wyoming.

It's quite possible that the flow from the south, on the south and east side of this vortex, will tug Matthew northward toward the Gulf of Mexico early next week. Indeed, the NHC's forecast for Matthew shows a northward turn associated with the storm's interaction with this feature early next week.

However, it's also important to note that there is by no means a consensus in the various forecast tracks for this fledgling system. Matthew could ignore the pull of the upper-level low and continue westward across Central America/Mexico, never visiting the Gulf of Mexico. For the East Coast in general and mid-Atlantic in particular, small differences in the upper-level pattern and storm track and intensity could be the difference between wet/humid or cool/dry weather later next week.

But the continued suggestion by global weather models that a U.S. landfall somewhere along the Gulf of Mexico is possible in the next week must be watched closely.

Fortunately, there are obstacles that may prevent Matthew from intensifying significantly, even if it takes a turn toward the Gulf Coast. For example, its proximity to Central America during the next day or two will likely disrupt the circulation. The official intensity forecast from NHC keeps Matthew below hurricane intensity prior to reaching Belize (just south of the Yucatan Peninsula). In addition, a curve northward may, as suggested by NHC, occur over the Yucatan Peninsula. This could further delay intensification by several days, or even weaken the circulation irreversibly.

It's also possible that if Matthew latches on to the southern edge of the upper-level low shown above, it may encounter a wind and humidity regime hostile to its development.

These are details we should be able to better predict in the coming days as Matthew matures and as the upper-level pattern over North America evolves.

By Greg Postel  | September 24, 2010; 11:00 AM ET
Categories:  Tropical Weather  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Video: 90+ degree day contest drawing
Next: DCA shatters record of 94, soars to 99


It's clear from the wide divergence of model solutions that Mathew's track is currently unpredictable after moving inland over Central America. That is, for this case at this time the limits of predictabilty have been exceeded

Posted by: SteveT-CapitalWeatherGang | September 24, 2010 12:15 PM | Report abuse

hi Steve

That assumes each forecast is equally skillful. Which is not the case. Folks at NHC might be able to tell you which models are performing better, and which ones to weigh lightly. But, you are right to point out that there is a very high degree of uncertatiny in Matt's future.


Posted by: gregpostel | September 24, 2010 12:24 PM | Report abuse

It seems more to becoming more likely that whatever tries to impact the U.S. won't actually be Matthew but a new system forming in its wake.

Posted by: Ian-CapitalWeatherGang | September 24, 2010 12:57 PM | Report abuse

Predictability uncertain...but the Weather Channel has us rainy all next week from Sunday on. Channel 4 seems to think that only Monday and Tuesday will be rainy.

WEAU-13's seven day outlook for NW Wisconsin has no widespread frost in their forecast--so a real cool air push down here seems unlikely in the near future. We're more likely to see a rather cooler but humid period with poor drying conditions.

Posted by: Bombo47jea | September 24, 2010 1:11 PM | Report abuse

Well my weather guesser friends, here's the NWS precipitation forecast 4-5 days:

That forecast suggests 3.5" which so long as it's over a period of days, probably is about as good as it gets....

Posted by: blackbear336 | September 24, 2010 1:28 PM | Report abuse

Sorry related, but wrong map:

Posted by: blackbear336 | September 24, 2010 1:31 PM | Report abuse

Hey- I'm headed to Orlando next week. What's the likelihood of Matthew's impact there Mon-Wed?


Posted by: mdem929 | September 24, 2010 2:56 PM | Report abuse

Greg, you are correct in that this ensemble of models (single run of different models) the ensemble members are not equally likely. But, I have it on the best of authority that forecasters do not have the ability to identify which model(s) is (are) most likely to perform better in THIS case, especally beyond 4-5 days.

On the other hand the ensembles built independently around the GFS (NWS) model and ECMWF (European) model point to the most likely track north/northeast along the eastern U. S.(but with each having uncertainty on details).

Posted by: SteveT-CapitalWeatherGang | September 24, 2010 3:52 PM | Report abuse


Odds are pretty low Orlando would be impacted, but I'd keep an eye on things this weekend.

Posted by: Jason-CapitalWeatherGang | September 24, 2010 4:07 PM | Report abuse

Hi Steve: Unless there is no autocorrelation in skill, one should be able to identify which models are likely (not certainly) to perform better in the near term (not sure what the e-folding time scale is for the correlations)

Posted by: gregpostel | September 24, 2010 5:40 PM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2012 The Washington Post Company