More Thoughts on the D.C. Forecast Miss

Addition: Dr. Cliff Mass has some thoughts here

Original Posting...
The above is a screen capture from a Washington, D.C. television station this morning where a meteorologist was made to "sit in the corner" and take a "timeout" because of yesterday's missed snow accumulation (not missed snow; it snowed but was too warm in the District to accumulate).

And, you should see some of the comments on various Facebook pages, "Idiots all." "Pay them only when they were accurate!", etc., etc. Below, I talked about the unique way this storm seemed to wrap around the District and how that is simply not forecastable.

Now, look at the actual (as sensed by NOAA weather satellite) snow accumulation from this storm.
NOAA image. White = 1. Blue-gray = 2", medium blue is 4" and heavier
accumulations associated with darker blue colors. 
Note there is snow cover in all directions from the District except straight east. There is even snow cover northeast and southeast! While I certainly concede the forecast was missed in the District, that forecast hardly seems to deserve the contempt being heaped upon it.

Tornado warnings by themselves save 1,000 to 2,000 lives a year. The warnings for Hurricane Sandy saved thousands of lives. Planes no longer crash in wind shear. I fully get that yesterday's miss caused problems and money. But, while weather forecasting and storm warnings are rapidly improving, they remain and inexact science.


  1. I have been pleased with the forecasts in Kansas this winter. Since I do snow removal, I pay very close attention to the weather, and everyone has been at least 99% accurate.

    But I remember a few years ago here in Hutchinson that we started in the north end of town with a 4" snow along and north of 30th Avenue, only had about 2" in the area of Main and 13th, and downtown it was melting and not building up.

    By the time we crossed the river into South Hutchinson we were in rain. A very crazy storm, and I don't see how anyone could be that exact.


  2. I think AccuWeather, NWS, and Weather Channel have to tone down the hype machine a bit when there is so much uncertainty about where the rain/snow line would set up - as the local weather discussions CLEARLY kept saying, but were overwhelmed by the need to hype.

    There's been a lot of good forecasts this year - but, as they say, one "oh shoot" cancels out a lot of "attaboys."

  3. Hi Brian, can you be specific as to where you think AccuWeather has "hyped" a winter storm? I would genuinely like to know. Mike

  4. Certainly Mike.

    Just look at the links from this blog. "major blizzard in the Mid Atlantic region." Personally, I don't know the difference between a 'blizzard' and a 'major blizzard.' Probability of heavy snow is listed as "High" right over DC.

    Here they started backing off:

    But then they consistently overforecasted the snow amounts, even as NWS models (I was paying particularly close attention to this storm as it directly affected my travel) were backing off - or certainly not converging to a solution.

    We're about ready to have a snow storm here in Denver. It's supposed to be strong. There may be conditions on the plains that meet blizzard criteria. Is it a "major blizzard"? A "regular" blizzard? A "small" blizzard? A "strong" winter storm? A "historical" storm? (No to this's moving too fast.) See what I mean?

    Again, there's been a lot of good forecasts this year, with deserved use of hyperbole. This storm never seemed to meet that criteria (exception: coastal flooding along the NJ coast which as we all know has no defenses). But AccuWeather took the opportunity in their site blog to hype this storm. Unfortunately, if they consistently hype every storm, then people tune them out when the hype is necessary.

    I must emphasize, I appreciate all you do on this site. It is a welcome change to the incessant global warming stuff that Dr. Masters spews on wunderground.

  5. Brian, thank you for your comment and allowing me to clarify the situation.

    First of all the term "major blizzard" was mine, not AccuWeather's. I'm sufficiently "experienced" to remember the time the NWS had a criterion for "severe blizzard" over and above "blizzard." That is what I was thinking of when I wrote "major" but I used the incorrect wording. I apologize.

    Second, the forecasts on this blog are mine (so blame me) but, in this case, I was linking to my colleagues at AW State College HQ because I thought our graphics and forecasts captured the threat well. We were wrong that far south (although, farther north, the forecast worked out well).

    Now, let me try to clarify a point: "Hyping" a forecast implies intent to make it seem worse than it actually is. I can assure you that neither AccuWeather nor I ever do that. You wouldn't believe the number of forecast discussions I sit through where we worry about a forecast even SEEMING like "hyping." That is not to say we don't overforecast on occasion but it is unintentional. We always strive to get it right.

    In this case, the southern end of the storm produced less than we thought. But, it was our honest opinion with zero intent to portray it worse than we thought it would be.

    Thanks again for the comment and the chance to clarify, Brian.

  6. Just another opinion on weather/storm "hype":

    I believe the "naming" of storms adds unneeeded "drama" to forecasts. Recent examples: "The Blizzard of Oz" and "Blizzard of Oz II". At least for me, this seems to be an attempt to add an aura of "seldom have we seen the likes of this before," when in fact the storms were pretty much run-of-the-mill winter storms.

    It is my understanding that The Weather Channel began naming winter storms this year, but I wouldn't know for sure as I don't watch because of the consistent "hype" there.

    Truly great storms rightfully gain names, but mostly after the fact, to wit, "The Blizzard of '88" (1888, that is) or "The New England Hurricane of 1938" (or Great New England Hurricane, Yankee Clipper, Long Island Express, or simply the Great Hurricane, per Wikipedia)

    I believe the same point applies to "naming" hurricanes. Simply designating storms by year and sequence number would remove the emotional element of anthropomorphizing nature.

  7. Interesting comments, Richard. Thank you.


Post a Comment

Popular posts from this blog

[1:10am Update] Tornado Forecast for Rest of the Night

First Tornado Watch of the Day Issued

Hilary's Forecast Path Shifts West; Updated 9:20am PDT