In Autumn, many eyes drawn to the North Pacific

With Autumn very obviously and officially underway (have you been outside the past few days?), the questions have begun to surface regarding the upcoming winter — and if the Autumn pattern will foreshadow it. Unfortunately, it isn’t quite as simple as “A cold October means a cold winter” or ” A warm November means a warm winter”. If that were the case, we’d have long range forecasting figured out by now, and there would be much less urgency to winter outlooks. The best we have for now are correlations, analogs, and pattern drivers. When used together, the summation of these factors provide a higher probability of success in medium to long range forecasting. But these methods are far from perfect. The evolution of these different atmospheric oscillations during the autumn season can often give us substantial clues into the ensuing winter’s dominant index modalities.

One major piece of the Winter’s puzzle can come in the form of “clues” from the North Pacific pattern in Autumn. The pattern in the Pacific Ocean nearer to our area is often represented as the East Pacific Oscillation (EPO). Like the North Atlantic Oscillation (NAO) the EPO has certain generally predictable outcomes when it oscillates from positive to negative phases. The positive phase of the EPO is usually associated with lower than normal atmospheric heights in the Northeast Pacific, Alaska, and Northwestern Canada. Conversely, the negative phase of the EPO generally produces higher than normal heights over the same region. The positive (+EPO) phase tends to flood the United States with maritime pacific air due to the lower pressure around Alaska, while the negative (-EPO) phase often sets up a cross-polar flow in the winter months with higher pressure centered over Alaska, and the atmospheric flow oriented across the North Pole from Siberia into Canada. -EPO patterns have yielded some of the most impressive, severe arctic outbreaks on record.

Data since 1990 presents us with some interesting clues when analyzing the state of the North Pacific and EPO in October, rolled forward to the following winter. Breaking the data down to be simply understood produces some compelling graphics. The Octobers preceding our colder than normal/snowier than normal winters since 1990 produce the following 500mb atmospheric height composite:

October pattern preceding colder than normal/snowier than normal winters since 1990.

October pattern preceding colder than normal/snowier than normal winters since 1990.

The most notable features are a large area of below normal 500mb heights near Alaska, the Gulf of Alaska, and the North/Eastern Pacific.

Interestingly enough, when you roll the composite forward to the month of November, the pattern becomes more anomalous — and changes quite dramatically. Ridging develops from the Northeast Pacific along the West Coast of the United States. This ridging is very important in allowing colder than normal air to filter southward into Canada and eventually the United States during the month of November:

November 500mb pattern preceding the snowier/colder winters since 1990.

November 500mb pattern preceding the snowier/colder winters since 1990.

One thing becomes clear in the analogs — the ridging that develops in the Northeast Pacific is critical for the winter ahead.

The analog(s) continue to become apparent when analyzing the November composite for our less snowy/warmer winters since 1990. The Gulf of Alaska and Northeast Pacific is generally dominated by below normal heights — and the EPO modality is positive. This has major implications on the downstream pattern. Notice how higher than normal heights generally are present across the Continental United States:

November pattern preceding the warmer/less snowy winters since 1990.

November pattern preceding the warmer/less snowy winters since 1990.

While these conclusions seem fairly startling, there are a few things to consider. First, remember that the sample size is quite small relative to time. These analog composites only consider roughly 23-24 years of data. So additional variability has to be considered within a longer time span. Second, the impacts of the global pattern can’t be understated. Each of these years, despite the sample size, had seaprate surrounding conditions. Consider the pattern in the North Atlantic, the ENSO state (this year features a strong El Nino), etc.

The North Pacific Pattern is just a piece of a very large puzzle which we will try to piece together over the next few weeks as Winter approaches. That being said, time has proven that it is a very important piece to the puzzle — one which may very well give us some important clues into the Winter ahead.