Ask Weather Dan: Pulling out the crystal ball (part 1)

We had a crazy winter and a crazy summer, and Richmonders definitely noticed. Here Weather Dan addresses questions from three readers concerning the extreme conditions we’ve been facing over the last several months.

September 9, 2010; 12:00 PM • by Weather Dan

I’ve received several questions of the same variety, so I’m going to try to answer them all at once.

Nicole asks:

Is the weather more bipolar than normal lately? It seems it went from one extreme (snow! snow! snow! more snow!) to another (hot! hot! hot! more sunshine!) in a short time. What’s up with that? Is it always like this and I just don’t remember it. What happened to easing into it?

Valerie also asks:

What with all the “holy hell, my face is melting” Spring/Summer we’re having, and the “SNOMG MORE SNOMG” Winter we had, I got to thinking: does a more extreme Winter increase the likelihood for a more extreme Summer (and vice versa)?

And finally, Shannon asks:

Will extremely hot summers like these become the norm, or is this just a fluke?

On the flip side, can we expect more snowy winters like last winter? (please say yes!)

It does seem that this year has been a bit extreme, doesn’t it? Between this winter’s snowfall and this summer’s heat, it’s been a bit of an active year so far. 2009 was was the year of the mild winter and cool summer, and now we’ve got 2010 with it’s extra-snowy winter and extra-sticky summer. So what does this mean down the road?

The short answer is: we don’t know.

Seasonal predictions are far from an exact science at this point. We’ve come a long, long way in just the last thirty to forty years, but there’s still a long way to go. The longer answer is pretty long, and I think it’s worth talking about in two parts. In this first portion, we’ll talking about the arm of the National Weather Service dedicated to long-range seasonal forecasting.

The Climate Prediction Center is an office within the National Centers for Environmental Prediction, which is itself a division of the National Weather Service. The CPC provides long-range and seasonal forecast predictions, but it’s not like forecasts most people are used to. Rather than being exact forecasts (high of 35 on Thursday, overnight low of 57 on Saturday, mostly cloudy skies, etc.), they’re probabilistic forecasts. The tools simply don’t exist to give an exact forecast three months out just yet.

By probabilistic forecasts, I mean that the CPC gives their forecast as the percentage chance that the given conditions will actually occur. Let’s take a look at their most recent 3-month forecast, for the months of September, October, and November.

We’ve got two maps, one for temperature and one for precipitation. Both maps are created with the same forecast in mind — the probability that a forecast will deviate from the average conditions for the period. Rather than being absolute, the contours differentiate between different probabilities for conditions to vary in a given direction from the mean.

For example, the desert southwest – Arizona and New Mexico primarily – show a greater than 50% chance of above-normal temperatures over the next three months, but also show a greater than 40% chance of below-normal precipitation amounts during the same time period.

Looking at Virginia, we’re in the white on both the temperature and precipitation maps. What does this mean? Basically, the CPC meteorologists believe there are equal chances of either above- or below-normal conditions over the next 90 days. I realize it’s not exactly helpful, but it’s giving us more information than you think.

Let’s take a look at some of the averages for Richmond.

Average high/low temperature, September 1: 84/64
Average high/low, September 30: 75/54
Average September precipitation: 3.98”

Average high/low October 1: 75/54
Average high/low October 31: 65/42
Average October precipitation: 3.60”

Average high/low November 1: 64/42
Average high/low November 30: 55/35
Average November precipitation: 3.06”

The Climate Prediction Center actually creates each of these plots for a rolling 3-month period, going out 13 months. So we can get a picture of what the CPC is thinking for the next year:

This is hard to read without enlarging (click to do so), but it gives you a good idea of what the CPC is thinking for the next year.

As it is with local short-term forecasts, forecasting confidence decreases with time, and certainly nothing presented in this image is set in stone. As time passes and seasonal fluctuations appear, these will be refined and updated, and give us a clearer picture of what to expect.

BUT — this isn’t all! There are several atmospheric patterns — dubbed oscillations — that influence these observations as time passes. We’ll explore these more in part 2.

September 9, 2010 12:00 PM 1