October 26, 2012

A More Perfect Storm Will Strike the Midlatlantic Coast

There is now a high probability that an historic storm will strike the mid-Atlantic states next week.  

A storm far more intense than the Perfect Storm, of book and movie fame.

One of the most intense landfalling cyclones in several years and perhaps the equal of the great Columbus Day Storm of 1962 that brought 100 mph plus winds to large swaths of the Pacific Northwest.

The storm that will strike the East Coast early next week is eerily like the Columbus Day Storm and the famous Perfect Storm in several ways.  Each of these started as a typhoon or hurricane and  transitioned into a powerful midlatitude cyclone, undergoing what we call extratropical transition.

As I will describe below, although there is still some uncertainty, the major modeling systems, American and European, have now locked on to a very similar solution.

Hurricane Sandy is now a Category 1 storm with sustained winds about 75 mph and a central pressure of 971 hPa.  Now located just north of the Bahamas and still over warm water.

At the same time, an upper level trough is approaching from the west and the interactions between the two will cause Sandy to morph into a hybrid storm, expand,  and intensify rapidly.  The National Weather Service is doing something very rare:  adding more radionsonde (weather balloon) launches (four times a day instead of twice day)...a very wise investment.

This figure shows the situation on Saturday, showing the upper trough over land (red line) and an upper low over Sandy (purple circle) (the figure is for 500 hPa--about 18,000 ft above sea level).  These will join up during the next few days.

Sandy will not only grow larger, but the pressure will fall, and the winds will increase.  Perhaps the most dramatic description of this process (but not technically quite accurate) was given by a supposed meteorologists in the Perfect Storm movie, click on the picture to view it:

Click to see video
Until recently, two of our most important models (the European Center model and the U.S. GFS) were substantially at odds, with the European model taking the storm into Maryland/Delaware and the U.S. model taking the storm out to sea.   In more recent runs, the U.S. high resolution model began recurving the storm back to the East Coast, first over New England and now across Long Island, with the European Center model not changing very much.

But now European Center model has moved northward, the GFS model swings inland earlier, so the differences have diminished.

Lets talk a look.  First the U.S. GFS.  Here is a series of surface pressure analyses from this morning's run (solid line are sea level pressure and shading is relative humidity.  A very deep low moves in off the ocean at 8 PM PDT Monday with a pressure of around 950 hPa (the Columbus Day Storm was about 960 hPA), then swings in over Long Island on Tuesday morning, then moves in a circle just inland, and eventually moves northward on Wednesday.  This is extraordinary behavior--to have a deep low just circle about for a day, before heading out of the region.  And deadly.




The winds would be very strong and damaging...sustained winds of 50-70 knots over the water with higher gusts.  Here is the wind forecast at 2 PM on Tuesday morning.  Winds are in knots (see legend).  Lots of sustained winds above 50 knots over the Atlantic and some coastal locations.

 What about the European Center forecasts?  A sub-940 hPa low makes landfall on central New Jersey, circles around, and then moves northward.  The European Center and GFS models tend to overdo the deepening in this situations, so I don't believe we will see anything below 950-960 hPa.


We are talking about huge events, if either of these forecasts are correct--or if the truth is somewhere in between.  Heavy rain, powerful winds over a long period, storm surges north of the low center.   This will make the Perfect Storm look like a zephyr.  Take a look at the cumulative rainfall through October 31st at 11 PM PDT.  5-10 inches over New Jersey and adjacent areas.  And this is a relatively coarse model....could well be more at some locations.  Cascade Mountains do almost as well!



Is there still uncertainty now that the major models are converging?  You bet.  There is some probability that the real storm will moves out to sea.  To get a handle on this, lets take a look at the last European ensemble prediction (left panel below) and the their high-resolution forecast (right panel) for 5 AM PDT on Tuesday.   The shading is a measure of uncertainty...how much the ensemble members disagree.

The high resolution forecast shows a huge, powerful storm making landfall.  Importantly, the ensemble (the mean of many different forecasts) show a low center at a similar position, perhaps a bit farther north.  This provides some confidence in the high-res forecast.  But the shading shows there is considerable uncertainty still, with the shading offshore suggesting a few of the forecasts were for a low farther offshore.  So, this is not 100% now...but a very good bet at this point.

 I think you can appreciate the technology...high resolution forecasts show us some possibilities, with ensemble prediction giving us a measure of the uncertainty in the forecasts.  No forecast is complete without a quantification of the uncertainty.  Nothing is 100%.   And certainly not a complex, sensitive situation like this one.

13 comments:

  1. The Andrea Gail- Unavailable for comment

    ReplyDelete
  2. Thanks for the discussion. I have relatives SE of Boston who will be affected by this storm. Hold on to your hats!

    ReplyDelete
  3. Below are interactive map links that should always display the most recent data files that NOAA has posted for Hurricane Sandy. The maps are displayed by the browser app Gmap4. The data files are hosted by http://www.hurricane.com/.

    Hurricane SANDY Advisory GIS Forecast Track
    http://www.mappingsupport.com/p/gmap4.php?t=h&q=http://www.nhc.noaa.gov/storm_graphics/AT18/AL1812_TRACK_latest.kmz

    Hurricane SANDY Advisory GIS Forecast Cone of Uncertainty
    http://www.mappingsupport.com/p/gmap4.php?t=h&q=http://www.nhc.noaa.gov/storm_graphics/AT18/AL1812_CONE_latest.kmz

    Hurricane SANDY Advisory GIS Watches/Warnings
    http://www.mappingsupport.com/p/gmap4.php?t=h&q=http://www.nhc.noaa.gov/storm_graphics/AT18/AL1812_WW_latest.kmz

    Hurricane SANDY Best Track Information
    http://www.mappingsupport.com/p/gmap4.php?t=h&q=http://www.nhc.noaa.gov/gis/best_track/al182012_best_track.kmz

    Hurricane SANDY Tropical Storm (34-kt) Wind Speed Probabilities - 0.1 Degree Multiple Basins
    http://www.mappingsupport.com/p/gmap4.php?q=http://www.nhc.noaa.gov/gis/forecast/archive/latest_wsp34knt120hr_tenthDeg.kmz&ll=35.925443,-72.742187&t=h&z=4

    Hurricane SANDY Gale (50-kt) Wind Speed Probabilities - 0.1 Degree Multiple Basins
    http://www.mappingsupport.com/p/gmap4.php?q=http://www.nhc.noaa.gov/gis/forecast/archive/latest_wsp50knt120hr_tenthDeg.kmz&ll=34.275166,-73.939453&t=h&z=4

    Hurricane SANDY Hurricane (64-kt) Wind Speed Probabilities - 0.1 Degree Multiple Basins
    http://www.mappingsupport.com/p/gmap4.php?q=http://www.nhc.noaa.gov/gis/forecast/archive/latest_wsp64knt120hr_tenthDeg.kmz&ll=32.543476,-74.480469&t=h&z=4

    Gmap4 default map: http://www.mappingsupport.com/p/gmap4.php

    Gmap4 homepage: http://www.mappingsupport.com/p/gmap4.html

    Joseph, the Gmap4 guy
    Redmond, WA

    ReplyDelete
  4. Both models seem to bring it over or close to my hometown (NE of Baltimore). The weather nut in me wishes I was there to watch the fury. Thanks to technology I will watch from the comfort of vineyard country in E WA. Thanks for the insights.

    ReplyDelete
  5. The storm surge and rain will probably be profoundly devastating but forget Columbus Day Storm-esque winds...no coastal locations will match Mt. Hebo or Cape Blanco, even with elevation (Blanco is only a few hundred feet) and as Dr. Mass has said, most instruments were destroyed before the storm's peak wind. "At Oregon's Cape Blanco, an anemometer that lost one of its cups registered wind gusts in excess of 145 miles per hour (233 km/h); some reports put the peak velocity at 179 miles per hour (288 km/h)." At least Mt. Washington will put on a show ;)

    ReplyDelete
  6. And with it becoming an extra-tropical storm, no weakening once it moves onshore right?

    ReplyDelete
  7. Cliff I've been following you since my grad school days at UW. Now that I'm in DC I still find your insight and analysis above all else. We're gathering supplies and getting prepped for a major impact from Sandy. Thanks for helping us separate the hype from the fact...lives may depend upon it!

    ReplyDelete
  8. Some confusion here...possibly based on typos. Have relatives who are scheduled to leave NYC on a plane Monday Afternoon. Have other relatives outside of Raleigh NC and several relatives throughout New Jersey and New York (upstate).

    Confusion: Direct quote from blog:
    "Here is the wind forecast at 2 PM on Tuesday morning. "

    Should that read: 2 A.M.? or 2 P.M Tuesday AFTERNOON?

    OR... is this a reference to a forecast that was made on Tuesday October 23rd?

    Since there are several typos in the blog...I'm wondering what part of the above quote is a typo... The Tuesday Afternoon or the 2 PM.

    I certainly want to assume because of relatives leaving at 4 p.m. from New York City on Monday the 29th... that it's 2 p.m. Tuesday Afternoon.

    Anyway... perhaps you can clarify.

    ReplyDelete
  9. I've been watching the weather channel a lot these days.
    They keep saying Sandy will turn left into the coast. Then they say this is very different than anything they have ever seen. They say there is no modern comparison.

    My question is, HOW DO THEY KNOW it will turn left into the coast? Yes the American and European computer models all suggest the left turn, but WHAT IS IT about this particular storm that makes the computer models think it will turn left? WHY don't they think it will just veer to the right- or even continue to parallel the coast????
    Is it something to do with barometric pressure?
    The jet stream?
    The air temperature?
    The water temperature?

    I just want to know the science behind it. Thanks.

    ReplyDelete
  10. Heya Cliff,

    I've been following your blog for years for Seattle weather. My husband's from Southern New Jersey and I'm from Baltimore originally, so we tend to watch these kind of big storms with great interest, as our families are still there. Outside your blog, I've seen nothing that isn't full of sensationalism or hyperbole, except reading the NWS discussions. Can you recommend any blogs or other weather news sources who focuses on East Coast weather who have a similar realistic science-based take to yours?

    ReplyDelete
  11. Unknown, I'd bet $10 I can't afford to lose that your relatives will NOT fly from NYC on Tuesday afternoon.

    Bethday, I was born in Northern Jersey and have lots of friends and family from Raleigh, NC, up through DC and Philly, to NYC and even Boston. I might be in Bellevue outside Seattle, but I am WATCHING this thing. I like this blog:
    Capitol Weather Gang They were great during the DC snows a few years ago, and while they get excited, they don't lose their minds. ;-)

    ReplyDelete
  12. Snapdragon, this from the Capitol Weather Gang blog might be helpful to you: 11:15 p.m. update: Hurricane Sandy continues its trek northward off the Carolina coast. The ”turn” back to the coast is about to occur as the trough to its west begins merge with Sandy and slingshot it back toward the coast as it transitions into an extratropical storm. Sandy remains a 75 mph hurricane, with a pressure now of 950 mb (28.05”). That’s the lowest the pressure has been in Sandy’s life and it appears the storm is either slowly strengthening now or may do so shortly.

    ReplyDelete
  13. Dreaming in Waves has a quote: "At Oregon's Cape Blanco, an anemometer that lost one of its cups registered wind gusts in excess of 145 miles per hour (233 km/h); some reports put the peak velocity at 179 miles per hour (288 km/h)." Does he have a source other than Wikipedia? Wikipedia wants a source for that quote.

    ReplyDelete

Please make sure your comments are civil. Name calling and personal attacks are not appropriate.

Should I Let AI Write My Blogs?

I spend a lot of time thinking about and writing my blogs.   Could AI apps do the work for me?   Could any of you tell the difference? Well,...