The calendar just flipped over to fall on Sunday, but it’s never too early to look forward to winter. Meteorologists – or at least those in the seasonal prediction business – have already been and will be checking various atmospheric/oceanic signals to guide their outlook for the coldest part of the year. Will the winter be cold and snowy, mild and dry, or some other combination?

The atmosphere is not offering many hints thus far, so it is difficult to tell if this one will be a 2009-10 Snowmaggedon repeat or something more like 2011-12, with wall-to-wall warmth and just a few sympathetic snowflakes.

One of the most elemental forecast factors is the projected late fall/winter state of El Niño-Southern Oscillation (ENSO). In terms most relevant to this post, ENSO reflects the relative temperature of the sea surface in the eastern tropical Pacific Ocean, where abnormally warm SSTs signify the El Niño phase of ENSO and abnormally cold SSTs indicate the La Niña phase. The term “La Nada” or “neutral” describes the phase when there is a neither an El Niño or La Niña.

These ENSO phases reach peak strength in the winter and early spring, but fade as summer approaches. Often times, the transition from the warm phase (El Niño) to the cold phase (La Niña) and vice versa will be obvious by mid-late summer. Constant measurements of the ENSO regions help meteorologists denote the transition and make informed predictions on future trends.

Sometimes, though, the atmosphere doesn’t behave so predictably. This summer-into-fall period offers a prime example. For the first time since the winters of 1992-93 and 1993-94, two consecutive winters (2012-13 and the upcoming 2013-14 season) will have likely featured neither the El Niño nor La Niña phases. Instead, the ENSO-neutral phase is almost certainly poised to hold serve into and through the winter.

Most if not all of the objective forecasts for ENSO call for the neutral state (average SSTs between 0.5 degree below normal and 0.5 degree above normal) to persist long term. Two of these outlooks are shown in graphical form below.

The top forecast provided by the Australian Bureau of Meteorology shows the ENSO prediction from seven different models. Consensus favors a SST reading very close to neutral, with a slight lean to abnormal warmth (El Niño-like). Note that the UKMO forecast is a major outlier as it nears weak La Niña territory. The bottom forecast, courtesy of CPC/IRI, is probability-based and highly favors ENSO-neutral conditions through the wintertime period (encompassing NDJ, DJF, and JFM.)

Why is it important to know which phase – El Niño, La Niña or neutral – will take hold for the winter? Because atmospheric patterns supporting defined areas of warmth or cold air and determining the chances for stormier or drier weather are driven by each specific ENSO phase.

Other signals can drive a winter pattern, of course – think of high-latitude blocking in the form of the negative NAO, such as we saw late last winter – but typically ENSO is a primary driver.

El Niño and winter weather patterns

Take, for instance, El Niño. Under this warm phase of ENSO, relatively mild temperatures affect the Northern U.S. and relatively cold temperatures impact the Southern states. At the jet stream level (250 mb, or roughly 34,000 feet in the air), two separate streams of air – the Polar jet and the subtropical jet – move through North America. Outbreaks of very cold Arctic air are much less common under the El Niño regime given the fact that low pressure draws warmth into Canada and the Northern U.S. Areas from the Southwest to the Southeast are often stormy and chilly thanks to the highly active subtropical jet, along which moisture channels into Mexico and the southern tier of the U.S. from near the Hawaiian Islands.

Top image shows temperatures compared to normal for the composite of all El Niño winters since 1950 (note the warm North and cold South). Bottom image shows the jet stream at 250 mb, with low pressure south of Alaska directing a highly active subtropical branch of the jet from Hawaii into Mexico and points downstream. (Both composite maps provided by NOAA/ESRL Physical Sciences Division)

El Niño winters and D.C. snow

El Niño events can be feast or famine for Washington, D.C. in terms of snowfall. If warmth from the North overwhelms, most of the storms riding along the subtropical jet will cause rain and/or mixed precipitation. On the other hand, a well-timed push of cold air (or sustained period of frosty weather aided by a blocking pattern), can meet one or a few of the Southern storms and produce heavy snow here. Interestingly, the median snowfall for the 14 El Niño events recorded since 1950 is 21 inches, or about 6 inches more than the climatological season average of 15.4 inches at Reagan National.

La Niña and winter weather patterns

In the case of La Niña, the temperature pattern is nearly reversed. Cold temperatures overspread western Canada, the Pacific Northwest and North-Central U.S. Above-normal warmth dominates over the Southern states and, at times, over the Lower Midwest and Mid-Atlantic (including D.C.). The typical jet stream is also nearly reversed. Instead of a controlling low pressure system in the northeast Pacific Ocean, high pressure governs the flow into North America. The Polar branch of the jet stream rides around the northern periphery of the high into Alaska and then dives south into Canada and the Northern U.S. Frigid air from the Arctic and, sometimes, Siberia can follow the jet if the transport mechanism is just right. Some of our most historic cold air outbreaks have occurred during La Niña winters (i.e. January 1985).

Top image shows temperatures compared to normal for the composite of all La Niña winters since 1950 (note the generally cold North and warm South). Bottom image shows the jet stream at 250 mb, with high pressure south of Alaska dominating the flow pattern. Note the easterly winds from the Gulf of Mexico to Hawaii; these were westerly under the El Niño regime. As a result, the subtropical branch of the jet stream is much less efficient at steering moisture needed for winter storm development into the Gulf of Mexico. (Both composite maps provided by NOAA/ESRL Physical Sciences Division)

La Niña winters and D.C. snow

The record including all 11 La Niña events since 1950 underscores a paucity of heavy snow seasons. Nine of the 11 winters have produced fewer than 12 inches of snowfall, and the highest seasonal total was 16.7 inches, recorded for the winter of 1973-74. Not surprisingly, the median snowfall is just over 10 inches, less than half of the median amount recorded for all El Niño events.

ENSO-neutral and winter weather patterns

Then there’s all of the ENSO-neutral events – for which we have a considerably larger sample size – to consider. What sticks out on the temperature map is the large expanse and impressive intensity of cold anomalies over Canada. A “finger of chill” points at the Great Lakes and Northeast. Little warmth is apparent except for near Greenland (hinting at a propensity for blocking).

Temperatures compared to normal for all ENSO-neutral winters. (NOAA/ESRL Physical Sciences Division)

The jet stream is extremely interesting. As with La Niña, the steering flow rides south across Alaska and western Canada, but unlike the La Niña pattern, it does not weaken so much over the Northern U.S. Cold air from the Arctic source region typically carries farther south and east into the U.S. compared to the Niña setup. Influence from the subtropical jet appears to be limited, but any surge of moisture from the Gulf of Mexico would tend to aid secondary storm development over or off the Southeast Coast given the cold-to-warm temperature boundary to the north and northwest.

Jet stream at 250 mb for all ENSO-neutral winters. Note the clockwise circulation over western Greenland and the dominant easterly flow across the North Atlantic. This blocking pattern is key to driving more of the available cold air from Canada southeast into the U.S. (NOAA/ESRL Physical Sciences Division)

ENSO-neutral winters and D.C. snow

Ironically – or perhaps not – D.C.’s median seasonal snowfall calculated for all ENSO-neutral events since 1950 is 15 inches, which equals the normal amount. There have been some real duds, including last winter’s 3.1 inches, but there have also been some bonanzas, like in 1960-61 (40.3 inches) and in 1978-79 (37.7 inches).

ENSO’s impact will be difficult to judge for this winter, and it won’t be the only driving force on our local temperature and snowfall pattern. Despite those arguments, ENSO correlations with winter temperatures and snowfall show us certain patterns that might be helpful in creating a forecast.