Winter is approaching, time to start talking about what might drive the winter pattern to enhance or decrease this winter’s snowfall.
For snow lovers, the last several winters have been boom or bust. The winter of 2009-2010 was the snowiest on record while last winter Reagan National Airport (DCA) recorded the third lowest seasonal snowfall on record. The three lowest snowfall winters have all occurred since 1970.
El Nino and La Nina (ENSO)
In short, El Nino winters are generally good for snow in D.C., La Ninas are bad, and neutral years (neither El Nino or La Nina) can go either way. This year, neutral conditions will probably win out.
To try to understand why we get more snow in some winters than others, I’ve listed the 10 snowiest winters since 1950 at DCA (above left) and the 10 least snowy (above right). El Nino winters are listed in red while La Nina winters recorded in blue. Neither El Nino nor La Nina conditions were present during the years in black.
(Remember that El Nino is associated with warmer than normal water temperatures in the tropical Pacific Ocean from South America to near the date line while La Nina conditions are present when those same water temperatures become colder than normal. The temperatures of the waters in the Pacific Ocean play a role in determining the general circulation pattern over the U.S. during winter.)
Six of the 10 snowiest winters since 1950 have occurred during El Ninos and 7 of the top 11 (around 64%). The prevalence of El Nino conditions during our snowiest winters is not too surprising since El Nino winters typically have a stronger and more active subtropical jet.
An active subtropical jet injects heat and moisture into the southern U.S. which increases the chances of low pressure systems having a storm track across the Gulf States to a position just off the coast of North Carolina. That track is our most favorable for producing a major snowstorm. Such a track is far enough south to pick up moisture from the Gulf of Mexico while also often helping to hold off the really warm air from the Gulf or Southeast Atlantic from streaming northward into our area.
While having an El Nino increases the likelihood of getting a really snowy winter, four of the ten snowiest winters did not occur during an El Nino. One happened during a weak La Nina and three coincided with neutral winters (la Nada)..
Three El Nino winters resulted in abysmally low seasonal snowfall (see above right), two of them recording the smallest snowfall totals in the entire Washington D.C. climate record. La Nina conditions resulted in 50% of the relatively snowless winters.
While El Nino winters are on average snowier than normal and La Nina’s generally result in less snow than normal, there are marked exceptions. The El Nino Southern Oscillation (ENSO, i.e. the alternating of El Nino, La Nina and neutral conditions) is not the only thing that influences whether D.C. will experience a mega-snow year. There are obviously other important players that help determine whether a winter ends up being an exceptionally snowy one or one that leaves snow lovers in the damp rainy blues.
The Arctic Oscillation (AO)
To show the patterns associated with either a lot or a little snow, I composited (averaged) the difference from average (anomaly) in the pressure patterns at around 18,000 feet (or 500 mb) for the 10 snowiest winters (below left) and for the 10 years that produced the least snow (below right). Areas with cold colors (blues and violets) represent where the pressures averaged below normal while the yellow and orange colors denote where pressures averaged above normal.
The difference between the two groups is striking and is indicative of different phases of the Arctic Oscillation (AO). During the snowy winters (bottom left) when the AO is in its negative phase, pressures across much of Canada and Greenland are above normal (warm shades) while lower than average pressures (cold shades) are found across the southern U.S. extending eastward across the Atlantic into southern Europe. Such a configuration keeps the main storm track to D.C.’s south while the high pressure to the north helps feed cold air into the U.S. from Canada.
The non-snowy winters were characterized by having a positive AO with below normal pressures (cold colors) across Canada and Greenland with a ring of above normal pressures to its south. This second configuration keeps cold air bottled up in Canada as air moves from regions of higher pressure to lower pressure. Also, the storm track is typically to our north rather than to our south. No matter how El Nino is shaping up, the AO and its cousin the North Atlantic Oscillation (NAO) will be major players in determiner whether we will end up colder and snowier than normal.
To get a feel for the importance of the AO in determining which years we see anomalously heavy snow, I plotted the 17 seasons since 1950 that have received at least 20 inches of snow against the ENSO and AO indices during the winter season. Only one season (1982-1983) received more than 20 inches when the AO index averaged positive (located above the red line) during the winter (Dec-Feb). However, that season benefited from having a strongly negative AO during February. That month more than 20 inches of snow fell, the bulk coming from one monster storm. The remaining 16 seasons coincided with a negative AO (diamonds located below the red line).
A similar plot of the NAO index garners almost identical results. Snow lovers should root for the AO and NAO to become and stay negative during the winter. That’s your main hope for a snowy winter. Snow haters, you should be lusting after a positive AO and NAO. That would probably lead to a warmer than normal winter while also keeping snowfall on the lighter side.
However, one of the two least snowy winters occurred during an El Nino winter that had a negative AO and a favorable southern storm track but still did not produce snow. At our latitude, serendipity or luck also plays a role in which years will be snowy or non-snowy.
Can the AO (or NAO) be predicted in advance?
Unfortunately, most climatologists believe that the NAO and AO phases can only be skillfully forecast a week or two in advance.
October Siberian snow cover?
The one exception is Judah Cohen who has developed an index of how fast snow cover expands across Eurasia during October and found a fairly robust correlation with the winter AO for a 20 year period. His research would suggest that the AO might end up being negative this winter. However, when he used a longer time period, his correlations were not as robust making any forecast of the AO based on his research iffy.
October AO/NAO phase predicts the winter?
A colleague, Allan Huffman, in the Raleigh Examiner notes that of the 22 most negative AO Octobers, 18 of the 22 averaged having a negative AO during the following Dec-Feb period. Of those 22, 13 also had a negative AO in November. Of those 13 only one year recorded a positive AO. This October the AO was strongly negative. Allan’s research, like Cohen’s, offers snow lovers some hope that the AO may not end up ragingly positive this winter.
Other AO/NAO indicator are not as favorable for snow lovers. The latest European Monthly forecasts have been advertising a positive NAO. For that reason, I’m not sure what the AO and NAO will do this winter. If I were to make a forecast, I’d guess that the AO would be negative more than positive but not with much confidence.