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 10:30 AM ET, 12/ 9/2010

Forecasting turkeys: Bad predictions since 1970

By Don Lipman

Part II

1979-blizzard.jpg
Floral Street in Washington after the Presidents Day Snowstorm, February 19, 1979. Source: Weatherbook.com

In "Forecasting Turkeys" part I, I walked you through some of the worst forecast busts from the founding of the Weather Bureau (now the National Weather Service) in 1870 to 1969. Now, I highlight some of the most horrendous prediction failures in the modern age of weather forecasting.

THE PRESIDENT'S DAY STORM OF 18-20 FEBRUARY 1979

Now called the Presidents' Day Storm I, since a second Presidents' Day Storm has now occurred in recent years (February 2003), this storm developed, as might be expected, under weak El Niño conditions but only a slightly negative NAO (North Atlantic Oscillation), with deep Arctic air in place over much of the East. Temperatures in the single digits and teens were common in the DC area during the progress of the storm, which was considered to be the worst since the Knickerbocker Storm of 1922. (1)

Since numerical forecasting models significantly downplayed the developing storm system, the NWS forecast initially called for only 4-8 inches of snow in the DC area, far less than the 19 inches that actually fell (officially at Reagan National Airport).

Ultimately, the storm system developed a hurricane-like eye off the mid-Atlantic coast and, after hitting the New York region with about a foot of snow, skirted off toward the east-northeast, sparing most of New England.

THE VETERAN'S DAY SNOWSTORM OF 1987

In terms of unpredictability, this storm was one of the most unusual storms to hit the Washington area in recent decades, as it developed very suddenly, causing extremely heavy snowfall rates, partly from convective cells moving northeastward over the southeastern counties. It was the heaviest snowfall on record for so early in the season and was recently covered in a Veteran's Day post by Jason Samenow.

THE EAST COAST SNOWSTORM OF 24-25 JANUARY 2000

At the time, the winter of 1999-2000 was the warmest on record in the country as a whole, although during the last half of January, wintry conditions did make an appearance in the Northeast.

It was during this period that one of the more noticeable numerical prediction model failures of recent years occurred: up until six hours prior to the storm's onset in the DC area, little or no snow was predicted. A last-minute media release by the NWS on the night of 24 January provided for a substantial upgrade of the storm's impact on the DC to Boston urban corridor. The damage had been done, however, as most people never got the word and held the various media responsible.

Ultimately, this storm had the most serious effect on the entire area from western North Carolina to northern New England, with heaviest accumulations and near blizzard conditions on a track just east of DC, but somewhat west of NYC and Boston. (While Baltimore and Annapolis collected 15-17 inches, Raleigh, NC received 20 inches, about 5 times their seasonal average.)

See Steve Tracton's recap for more on this storm.

THE NORTHEAST SNOWSTORM OF 30-31 DECEMBER 2000

This was a storm that totally bypassed the Washington metropolitan area, although early advance predictions were not nearly so benign. By the time the pre-New Year's Eve storm was over, the entire corridor from central New Jersey, through western Long Island, including NYC, to northern New York and western New England was brought to a standstill by 1-2 feet of snow. Surely, New Year's Eve festivities were seriously affected. In Boston, where there was a changeover to rain, less than an inch of snow accumulated.

What was probably the most frustrating aspect of this storm (to forecasters) was the fact that the western edge of the main snow band had such a sharp cut-off, with the western suburbs of Philadelphia, for example, receiving almost no snow at all while the northeast suburbs got a foot or more.

THE 5-7 MARCH 2001 (NO-SHOW) SNOWSTORM

This storm was notable due to the public reaction--and even outrage--that resulted from the botched forecasts. Computer models on Friday, March 2, were suggesting that near blizzard conditions would develop from VA to Southern New England. NOAA's HPC (Hydrometeorological Prediction Center) said:

THERE MAY BE BEACH EROSION AND COASTAL FLOODING ... ALONG WITH WINDS EXCEEDING GALE FORCE ALONG THE IMMEDIATE COAST. FARTHER INLAND ... MODERATE TO HEAVY SNOW AND BLOWING SNOW FROM VA THRU THE SOUTHERN HALF OF NEW ENGLAND. LOW WIND CHILLS WITH TEMPS GENERALLY IN THE HIGH TEENS TO MID 20S. THERE COULD EVEN BE THUNDERSNOW. SNOW ACCUMULATIONS MAY EXCEED A FOOT FROM NORTHERN AND CENTRAL VIRGINIA NORTHEASTWARD THRU THE MAJOR METROPOLITAN AREAS

Although the storm ultimately did hit parts of New York and New England with heavy snowfall (12-18 inches), it was essentially a no-show in most of the mid-Atlantic, where many schools and businesses announced advance closings due to the predicted blizzard.

In the Philadelphia area, this storm is still known as the John Bolaris Storm. Bolaris, who was the face of the weather for WCAU, (then) NBC4 in Philadelphia, supposedly suggested (during the rating period) on the Wednesday before the expected Sunday-Monday storm that it could be "historic" and that all interests should be prepared.

In a 2008 interview, however, Bolaris says that he only mentioned that the "potential" existed for such a storm. By the time the weekend arrived and he believed, contrary to the NWS, that the storm should be greatly downgraded, the weekend "weather lady" was doing the honors.

Bolaris says he was asked early that Sunday night to support her and explain the genesis of the storm but refused because by then he believed it wouldn't happen. (He was allowed, however, to break in during prime-time to announce that the storm would be a no-show.) I don't know what the other media were announcing at that time. Do you?

The media never seemed to forgive Bolaris, even though, as he tells it, he did nothing wrong. In any event, he left that station subsequently. Some say he was discharged because of all the negative publicity. But on the other hand, as he puts it, "I am in a Hurricane Hunter going on a mission and a captain comes up to me and goes, 'Yeah, there was this weatherman in Philadelphia that predicted like forty-eight inches of snow and they ran him out of town.' And I said, "That would be me and I did not predict forty-eight inches of snow."

Do you remember other forecasting turkeys since 1970? Comment below...

*Unless otherwise noted, most material from this articles was drawn from Kocin and Ucellini's monograph, Northeast Snowstorms, Volume II
(1)National Weather Service monograph, Mid Atlantic Winters, SNOW, WIND, ICE AND COLD
(2) National Weather Service monograph, Mid Atlantic Winters, SNOW, WIND, ICE AND COLD

By Don Lipman  | December 9, 2010; 10:30 AM ET
Categories:  History, Humor, Latest, Lipman, Media  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   Del.icio.us   StumbleUpon   Technorati   Google Buzz   Previous: Forecast: Wintry cold but winds "brake"
Next: Will Sunday's storm produce any snow?

Comments

The "TWA 514" storm, Dec. 1, 1974:

This storm was rather well-forecasted, so it may not qualify as a "forecasting turkey", but it certainly qualifies as a "weather turkey"!

I believe the cause for the plane crash on Mt. Weather was officially determined as "pilot error". It's possible the crew of TWA 514 had a weather briefing which forecast conditions less intense than what eventually developed. As I recall, Flight 514 originated in Columbus, OH and was headed to Reagan rather than to Dulles.

Note on the Veterans' Day 1987 snowstorm:

We had thundersnow at Baileys Crossroads between noon and 2 pm. After the thunderstorms moved out, another four hours of heavy snow dropped about six inches for a total accumulation of a foot to 14 inches.

Posted by: Bombo47jea | December 9, 2010 11:10 AM | Report abuse

don,
perhaps you could devote part III to over-hyped "joey fearless" predictions that didn't materialize...

http://voices.washingtonpost.com/capitalweathergang/2010/12/a_white_christmas_for_santa.html

Posted by: walter-in-fallschurch | December 9, 2010 11:11 AM | Report abuse

Wasn't there a bad freezing rain event in early January 2001 or 2002 for which there was no forecast at all?

Posted by: gilmoredaniel | December 9, 2010 2:32 PM | Report abuse

I recall a clipper in March 1999 that produced more than 8" in the DC area when a dusting was forecast. Tony Kornheiser wrote a scathing a column "Eat My Dusting" which you can read here: http://www.windspring.net/tony_stuff/march211999.html

Posted by: Jason-CapitalWeatherGang | December 9, 2010 2:43 PM | Report abuse

I seem to remember that the Blizzard of 1996 was basically un-predicted... at least in the TV media; not sure about the NWS.

I lived just east of Dulles Airport and distinctly remember that the TV weather consensus seemed to be a 'few inches of snow.' Then all of a sudden they upped the forecast dramatically right around the same time it actually started snowing, not giving people very much time to prepare.

I remember as soon as they heard the increased forecasts, my parents drove my sister and me to the grocery store for a few necessities and then to Blockbuster to load up on rental movies & games while the snow was already coming down.

By the time we got back there was already a good inch on the ground, and by the time all was said and done we had almost two feet of snow (and much worse in the drifts -- at least 4 or 5' against the front of our garage).

My memory may be a bit fuzzy; I was a kid, and pretty much drunk on the prospect of many days off school ;-).

Posted by: scott_bradford | December 9, 2010 3:10 PM | Report abuse

walter-in-fallschurch:
Good idea for a future post. Right now, people may be tired of "turkey" for a while.

Don-Capital Weather Gang

Posted by: Weatherguy | December 9, 2010 3:47 PM | Report abuse

The 2003 President's Day storm was a forecasting turkey for NYC. As of Sunday of that weekend, NYC was forecast to get hit with only a couple of inches of snow at best. My husband was due up in NYC on Monday, and based on that forecast, wanted to get out of DC Sunday so he could get back to NYC for the week. The trip was treacherous.

He did get to NYC and dry pavement, but ultimately even more snow fell in NYC than in DC. Everything was cancelled. He would have been better off staying in DC.

Posted by: kperl | December 9, 2010 4:22 PM | Report abuse

In regard to the THE NORTHEAST SNOWSTORM OF 30-31 DECEMBER 2000, the official forecasts for Washington as late as Firday night at 10PM read :

WINTER STORM WARNING FOR LATE TONIGHT THROUGH SAT EVENING, 5-10” EXPECTED FROM THIS STORM .... strong northerly winds wil result in considerable blowing and drifitng of snow..,

CNN pronounced that "more than 300 snowplows and salt trucks were ready to go into action Saturday, according to city officials

Well, Washington woke up to bright sunny skies with no threat of snow. a surprise No Snowstorm.

Posted by: SteveT-CapitalWeatherGang | December 9, 2010 4:57 PM | Report abuse

Re: 1979 "President's Day Storm." The NWS called for 4 inches. I drove to Rockville from my home in Fairfax to meet my girlfriend's parents. We didn't get the 19" that fell at National Airport--we had 40" inches. I was snowed in at their house for three days. Been married 30 years now--best snowstorm ever!

Posted by: h0db | December 9, 2010 9:35 PM | Report abuse

There was another no-show-snow within a few years of that one, but right around Christmas. I had to make a decision about which emergency vet to use for a client's cat, and chose the closer, rather than the better, one because of the impending blizzard. Next day, dry road, cloudless sky. Didn't end well.

Posted by: Rhea1 | December 9, 2010 10:38 PM | Report abuse

Sorry-- I probably meant Steve T's December 2000 no-show.

Posted by: Rhea1 | December 9, 2010 10:40 PM | Report abuse

The THE 5-7 MARCH 2001 (NO-SHOW) SNOWSTORM was one of the biggest busts of my career, that discussion probably had my name on it. The models were evenly split on the storm with the newest, highest resolution models loving the storm and the GFS disliking it.

I don't remember the 1996 blizzard as being poorly forecast in this area. We knew it was going to be a major storm before it even started and I think Young might have mentioned up to two feet in the area in her discussion prior to it starting.

Posted by: wjunker | December 10, 2010 2:49 PM | Report abuse

If we are still adding to this list, I would like to add the "12/10/10 dusting of doom" storm (probably more like 1/2 inch in the N. Shenandoah Valley). Result: accidents and traffic jams. Not predicted by forecasters AFAIK.

Posted by: eric654 | December 10, 2010 2:50 PM | Report abuse

The comments to this entry are closed.

 
 
RSS Feed
Subscribe to The Post

© 2012 The Washington Post Company