Saturday, October 24, 2015

Hurricane Patricia: Extreme, Poorly Predicted, and Soon to Flood Texas

Hurricane Patricia, which made landfall on the western Mexico coast, has been a jaw-dropping hurricane, with a long-list of superlatives:

The lowest sea level pressure ever observed for a hurricane in the Western hemisphere:  879 hPa (millibars).  The absolute lowest pressure anywhere was 870 hPa in Typhoon Tip in 1979 over the western Pacific.  We are talking about bringing the normal pressure at 5000 ft down to sea level----amazing.

The sustained winds reach 200 mph, with gusts substantially higher  This greatly  exceeds the criteria for a Category 5 storm (sustained winds equal or greater than 156 mph).

The storm developed faster than any other Western Hemisphere storm on record, going from a a tropical disturbance to a category 5 storm in less than 24 h.   We are talking about a 100 hPa pressure decline in less than a day.  Unbelievable.

The eye of Patricia was quite small (about 8 miles) at landfall yesterday, something obvious in the satellite picture shown above.

Why did Patricia rev up so quickly?   I don't think we understand the origins of the extreme behavior, but the conditions were ideal:  very warm ocean temperatures, deep warm water, low wind shear in the vertical, unstable conditions.  In fact, the waters off Mexico are now some of the warmest on the planet (see image) and a big reason for that is the major El Nino that is occurring currently.
Forecast Problems

Patricia is a poster child, perhaps the worst case in a while, of a major problem for meteorologists:  we have gained substantial skill in forecasting hurricane tracks, but we often fail in predicting hurricane intensification.

Let's be honest here: NONE of our numerical prediction models, or our statistical aids, predicted the extreme intensification of Patricia.

Consider the National Weather Service's primo modeling system for hurricane forecasting (HWRF), something they have invested tens of millions in.  Here is the HWRF forecast available late Wednesday. HWRF  forecast (purple lines) lowest pressure only declines to 950 hPa and sustained winds reach 95 knots (109 mph).  Not good--they are predicting a storm that is half as intense as reality.

The forecasts the day before improved substantially--but that left little time for preparation.  Here is the forecast available about 12h prior to landfall.

In contrast to intensification forecasts, the track forecasts were skillful and stable for days before.  Why the difference?   To forecast track one does not have to predict the complex inner dynamics of the hurricane;  if you can skillfully predict the general environment (the steering flow), you can get the track correct.   And our large scale models (like the NOAA GFS and the European model) can generally do that.

But predicting intensity is a very different thing.  You need to know the inner structure of the storm and predict its evolution.   The size and evolution of the changing eyewall clouds, the development and contraction of new eyewalls, and much more.   This is very hard in many ways.   Some research suggests that it may not be even possible to do this more than a day or two out.  An area of active research.

Texas flooding

Patricia, although radically weakened by its passage over land (robbing it of moisture) and terrain (which tears the lower structure apart), is not done causing problem.  The residual circulation and moisture of the storm will meet up with an approaching upper level disturbance over the southwest U.S, and moisture from off the Gulf of Mexico. to produce extreme precipitation over southern Texas.  Here is the predicted precipitation total for the next 48h from the National Weather Service GFS model:  10-15 inches over coastal Texas.   Serious trouble and the potential for flash flooding.

The National Weather Service forecast for Houston says it all:


Tim L said...

Yep, deploying the pontoons to our vehicles here in NE Houston. We've been very dry the last six weeks or so, now we get our reprieve over a period of about 36 hours. Crazy. The lines at the grocery stores yesterday and this morning have been ridiculous. You'd think for people that live in a tropical storm prone area, that they'd be a bit more sensible about this. Good thing I did my long run today.

clive boulton said...

Doesn't NOAA somehow fly into or above the eye of hurricanes to measure intensity?

Glen Hiemstra said...

"They call it “the end of stationarity,” a term with a powerful meaning: We can no longer rely on past events to predict future probabilities. 1 The ground is shifting beneath our feet." From the book Carbon Shock. I know that Cliff argues that climate change will only impact weather on very very long time scales, and that such impacts are not yet present or measureable. But the book above is worth checking out.

Richard said...

Don't I recall H. Katrina and H. Rita crossing an unusually warm Gulf on their run-ups? I think a degree or two. Maybe one built faster, and one built higher, than predicted. Hazy on that...

JewelyaZ said...

I'm a total amateur, but have been fascinated by hurricanes all my life, having grown up on the East coast and spending many weeks at the coast. My dad and I started tracking hurricanes together on paper maps when I was 5, 43 years ago. Right, so... I watched Patricia, and thinking about all the El Nino warm water and the lack of shear, I mused aloud on Tuesday, "I wonder if this is going to be a 200-mph hurricane". If I can call it, based on an "educated guess," why can't they even come close 48 or 72 hours out? Sure, you wouldn't muse like that on TV, but the sheer level of professional surprise is depressing.

When will we see a Category 6, and perhaps even a Category 7, added to the hurricane scale? Because I think we need it. This is not the last super-intense hurricane, given the changes that we are seeing in climate (who cares if it's natural variation or human-caused climate change? I don't, I just think it's useful to describe events unambiguously). Even if these monster storms remain uncommon or even very rare, it would be useful to distinguish a "weak" Cat 7 from a strong Cat 5.

Also, given that the winds at landfall were EF-4+ tornado strength, what keeps us from CALLING this a tornado? It is solely the tropical formation of the storm? Is it that it wasn't built by a thunderstorm cell? I realize that with hurricane winds spanning 35-miles across, with an eye 6-miles wide, it's hard to call this a tornado, but I really can't see the difference... rotation, extreme winds, damaging forces. Is it the lack of vacuum sucking debris up into the storm and depositing it miles away? That's a pretty flimsy criteria. I think there's probably something that I'm missing that makes the hurricane break the tornado "rules," but I wonder what it is.

I'm really glad that the storm came ashore in a less-populated area. Nothing in Mexico is built to code that would guarantee surviving a direct hit from a 200-mph storm with 20-foot storm surge.

lhsouthern said...

Yes they do: hurricane hunters have a special plane that is aquipped to measure the hurricane while they fly in and through the eye wall.

John McBride said...

8 miles across? Has there been another storm that compact to which to compare this hurricane? One that intense, and small, almost sounds like some hybrid of a hurricane and tornado, although obviously the development factors of the two are different. Or, you being the meterologist, were they that different in this case?

I'm curious to know, and there is likely data that defines this, how much "water" gets lifted into the atmosphere in storms like this? Is it 10s of thousands of acre feet? 100s of thousands? Millions? It has to be billions of gallons, doesn't it?

Thanks Cliff. Great, informative post.

Brendan Mccormack said...

Hurricane Wilmas eye contracted to just over 2 miles wide at peak intensity. Look at satellite images of wma at peak and its just a dot at the center.