There is a good reason we don’t
post our winter forecast in October or even earlier. It’s not because we want to see what the
other guys forecasted before we make ours.
It’s just we’ve learned from past experience that Fall can give us a lot
of clues as to what the winter will be like.
As we see it, this Fall is no exception.
The two keys are the recent warmth and also the developing La Nina.
The whole beauty of the blog is
that it can be a stream of consciousness to what we are currently researching. You may recall in my first blog we found this
nugget: when there is 90 degrees or warmer in the last half of September (it
was 91 on September 25th) we
tend to have above normal snow during the following winter. It was an eye
opener but it was a small sample; only 3 pervious years since 1950.
Now fast forward to October. It
ended up as the warmest October on record.
This is the kind of anomaly that we like to use when we make a seasonal
forecast. We took a look at past warm
Octobers and that opened things to a much larger sample of 18 years. A majority
of those years DID NOT have big snow years in the following winters. In addition, when there are an unusual amount
of 80 degree days in October (3 or more days) there has never been a winter
with more than 130” of snow. That’s a 0
for 7 slump. These are some of the
reasons we are not going with a big snow year.
Now on to the La Nina. That’s the cooler than normal water
temperatures in the Equatorial Pacific and that has been unfolding over the
last couple of months. Digging through
the numbers of past La Ninas we found the signals point away from big snow
years. Only 3 out of 16 La Nina winters
had snowfall above 130” .The other thing about La Nina is that it does have an
impact on the jet stream here in the United
States.
Unlike during an El Nino (when there is a strong southern branch of the
jet stream) the northern branch is the main player during a La Nina. With this
stronger northern branch we still do see lots variability in our winter weather
and we think that will still lead to near normal snowfall.
Keep in mind when you are looking at large
samples of past winters there is no black and white: 100% of the years do not show
up as all above or all below normal. There are usually no ‘slam dunks.’ That’s
the forecasting game; you have to play the percentages. So based on these
numbers we are forecasting a winter with near
normal snowfall and above normal temperatures. That’s it.
From this point on there is no more tinkering. Seasonal forecasts are not our main focus in
the office but we’ve done pretty well in the past. Let’s see what happens over the next few
months. Meanwhile, we now turn to the short term and forecasting the daily
weather that will make up the winter.
