Weather Blog

Let the debate begin: Will it be a cold winter?

Let the debate begin: Will it be a cold winter?
Photo submitted by users "B & K LaPointe"
Far and away, *the* most popular e-mail we get into the KOMO weather bin is some variance on: "Do you think we'll get any big snows this winter?"

Long range forecasting is an inexact science. At best, we can usually just give odds on whether it'll end up being warmer or colder, or wetter or drier than normal.  But even then, it's marginal (like, "a 55% chance of a warmer than normal year") and most of the time, the long range predictions bust, such as the really wet spring this year, or the mega rains of the 2006/7 winter.

But there's always speculation, and while that mutual fund saying of "past performance is no guarantee of future results" rings generally true for climate forecasting, it's sure fun to ponder, "what if?" To that end, I've been exchanging a few e-mails with some of my resident climate junkies over what the winter could behold. Most revolves around circumstantial evidence of what a neutral year (i.e., not La Nina or El Nino) typically means, and correlate that with what a warm October or November has meant for January to March in the past.

One tidbit discovered is that November 1949 is the warmest November on record -- an average high of 56.17 degrees and just a half degree cooler than that October was. (Typically, November is 9 degrees cooler than October.)

But then if one were to look at the daily record lows for January and February, you'd find 13 of them between Jan. 1 and Feb. 2 were set in just two months later in 1950, including the all time record low of 0 degrees set Jan. 31 and 1 degree set on Feb. 1. The records are spread out over the entire period showing that Seattle was generally in a frozen arctic pattern for the better part of five weeks.  Seattle also set the still-standing daily snowfall record of 21.5" on Jan. 13, 1950.

Fast forward to this year. November 2008's average high temperature is now 55.6 degrees, not too far from 1949. And Sea-Tac just missed tying a record for most consecutive days to begin November over 50 degrees. (The record was 22; we got to 21 this year before a 48 degree highs on November 22nd.)

Now, while the forecast is generally for continued above-normal temperatures through the end of the month, they'll still expected to be in the low 50s, so I gather this average high temperature will drop in the coming days and we won't break the record. But it's still looking like a pretty warm November.

Does that mean a frigid Jan-March? No. At least, it's not a slam dunk. As we said, weather doesn't necessarily repeat itself, or else it would be really easy to do long range forecasting. Plus, there have been plenty other warm Novembers that haven't lead to squat in the rest of the winter. But I throw this out there just for discussion sake.

You can see how the months break down for yourself here.

Jason Phelps, one of the aforementioned climate junkies, has this take on the November '49/January '50 winter:


That would be interesting, wouldn't it? :)
Our past snowstorms of 16-18"+ include...
1971-72 La Nina
1968-69 El Nino (most interestingly enough we had a big snow event during an El Nino year)
1949-50 La Nina
1922-23 La Nina
1915-16 Neutral
1898-99 Neutral
1896-97 El Nino (most interestingly enough we had a big snow event during an El Nino year)
1892-93 La Nina
1880-81 El Nino
There have been some close-call winters like 1996-97 (a neutral winter) while Seattle southward only got 10-12", Bellingham received 38" of snow!!!  1979/80 there was a big snowstorm but it went heavier south.
Looking at past major snowstorms, it has been a while since we've had a snowstorm on a magnitude like that of 1949/50, in fact 37 years!!
Now, this winter is a neutral winter.  Out of the past nine major snow events two were neutral, though there were also two close calls that were neutral.  1915/16 was quite an interesting winter when it snowed 26" on Ground Hog's Day :)
Neutral winters tending to be more volatile can lead to significant snows most of the time in the 8-10" or less range but sometimes there will be a bigger snow :)  Other factors include no sun spots on the sun which can sometimes lead to a temporary global cooling, though I'm not sure if the sunspots are starting to come back.
Over the last couple years the Pacific Decadal Oscillation has switched to the colder one which could be why the last couple winters have featured more snow, in addition to last winter being La Nina.
With this all factored in I would say a very interesting February is possible, but I wouldn't jump to conclusions about snowstorms on the magnitude of 1949/50.  I believe the probabilities of an arctic outbreak in February are high because it hasn't happened in eight years (a snow of significance and widespread nature in February).  It could be though that it's more like 2-10" or less than a foot.  However, the possibility and liklihood of another mega Arctic Outbreak like 1949/50 remains a possibility, and with probability one would think it would be likely to happen again eventually.  It has been 37 years, so I'm not ruling out the possibility of a mega-arctic outbreak in February like 1949/50, but I suspect that there may be a snowstorm that's more like maybe getting as much as 2-10" (obviously a wide range due to varied terrain in Washington) widespread, and maybe a foot in isolated foothill locals :).


And here is some more evidence from another climate guru Jim Hinkle:


The first piece of evidence is years that met the following criteria in November.

1. A very warm and wet November in our region

2. zero or below ENSO [This is the El Nino/La Nina period. "Zero or below" means neutral or La Nina conditions]
3. zero or below PDO [This is the Pacific Decadal Oscillation, or a still-being-studied phenomenon where the Pacific Ocean can undergo periods of warming and cooling on 30-40 year time frames. Right now, most seem to think we're in a neutral phase, just coming out of a warm phase from 1970-2000]

4. cold in the eastern half of the country

The years are 1901, 1932, 1949, 1967, and 1995

1901, 1932, and 1949 had legendary cold spells during the winter; 1967 had some decently cold and snowy weather in January; and 1995 had a very nice mid Jan to early Feb cold snap with widespread snowfall.  Attached is a composite map of those winters.

(Second E-mail)

Years with a combination of negative PDO and a positive AMO (Atlantic Multidecadal oscillation).  The years are: 1948, 1951, 1952, 1955, 1961, 1962, 1999, and 2001.  Most of those winters had very cold weather.  There are a couple of exceptions, but that is inevitable in this game.


NOAA also seems to think at least this winter will be colder than normal. Here is their Jan-March forecast:

So, what's your take? Use the commenting area below to sound off on your thoughts or research for this winter and get in some predictions. Then next March, we'll revisit this story and see who was right!