Comparing Teleconnection Patterns to Historic Snowstorms

By Mark Sannutti

February 2019

Overview: Several cities along the East Coast of the United States are shown to compare their biggest snowstorms with the various patterns of Climate Variability. The following patterns were used:

The table below shows the top snowstorms for Philadelphia, PA against the various index values and phases for each pattern. Several other cities are also included and listed below the table. The data range for the snowstorm events are the months of Decemeber, January, February and March from 1950 to 2018. The colors represent the various phases of the patterns, with red being warm, blue being cold and green being neutral. The light and dark shades of color represent the strength of phase such as strong or weak. ENSO values are based using the Nino 3.4 region for weekly, monthly and 3 month running mean. Daily MJO phases prior to 1974 are not avilable. The data sources for each of the index values are provided at the bottom of this page.

Top 15 Snowstorms(Dec. to March 1950-2018) Philadelpha, P.A.
Philadlephia International Airport(KPHL)
Storm Event Snowfall(inches) NAO(7-day Average) PNA(7-day Average) AO(7-day Average) ENSO PDO(monthly) MJO(daily)
January 7-8, 1996 30.7 1/2 to 1/8 = -0.925 1/2 to 1/8 = 0.245 1/2 to 1/8 = -1.644 Weak La Nina Weak Positive Phase 1
February 5-6, 2010 28.5 1/31 to 2/6 = -0.463 1/31 to 2/6 = 0.523 1/31 to 2/6 = -3.499 Moderate El Nino Weak Positive Phase 7
December 19-20, 2009 23.2 12/14 to 12/20 = -1.843 12/14 to 12/20 = 0.212 12/14 to 12/20 = -4.185 Moderate El Nino Neutral Phase 7 and 8
January 22-23, 2016 22.4 1/17 to 1/23 = -0.329 1/17 to 1/23 = 1.161 1/17 to 1/23 = -2.079 Very Strong El Nino Strong Positive Phase 2
February 11-12, 1983 21.3 2/6 to 2/12 = -0.527 2/6 to 2/12 = 0.717 2/6 to 2/12 = -2.416 Very Strong El Nino Positive Phase 2 and 3
February 16-17, 2003 18.7 2/11 to 2/17 = 0.789 2/11 to 2/17 = 0.487 2/11 to 2/17 = 0.202 Weak El Nino Strong Positive Phase 7 and 8
February 9-10, 2010 15.8 2/4 to 2/10 = -0.936 2/4 to 2/10 = 0.786 2/4 to 2/10 = -4.597 Moderate El Nino Weak Positive Phase 8
January 26-27, 2011 15.1 1/21 to 1/27 = -0.684 1/21 to 1/27 = 1.124 1/21 to 1/27 = -1.187 Moderate La Nina Weak Negative Phase 8
December 11-12, 1960 14.6 12/6 to 12/12 = -0.168 12/6 to 12/12 = 1.482 12/6 to 12/12 = -0.885 Neutral Neutral N/A
February 18-19, 1979 14.3 2/13 to 2/19 = -0.280 2/13 to 2/19 = -0.515 2/13 to 2/19 = -0.263 Neutral Negative Phase 3 and 4
February 6-7, 1978 14.1 2/1 to 2/7 = 0.116 2/1 to 2/7 = 1.011 2/1 to 2/7 = -4.494 Weak El Nino Positive Phase 7 and 8
January 21, 2014 13.5 2/1 to 2/7 = -0.007 2/1 to 2/7 = 1.089 2/1 to 2/7 = -1.722 Neutral Neutral Phase 6
January 19-20, 1978 13.2 1/14 to 1/20 = 0.529 1/14 to 1/20 = 0.240 1/14 to 1/20 = -0.032 Weak El Nino Neutral Phase 3 and 4
January 19-20, 1961 13.2 1/14 to 1/20 = 0.227 1/14 to 1/20 = 1.296 1/14 to 1/20 = -1.116 Neutral Positive N/A
February 15-16, 1958 13.0 2/10 to 2/16 = -0.513 2/10 to 2/16 = 0.794 2/10 to 2/16 = -1.725 Moderate El Nino Weak Positive N/A



Results for the rest of the cities.

Observations: The negative phase of the Arctic Oscillation showed the most occurances of historic snowstorms among all the patterns. Two conditions are especially important:

This is highly significant for Philadelphia, with the top five biggest snowstorms and 12 out of 15 during this period all occuring when these conditions are met. This city has also shown the highest signficance with the AO for monthly and seasonal snowfall through my own research. In addition, a college study I worked on back in 2004 also showed the AO to be the most significant against other climate patterns.

Several other features are also observed:

As mentioned already, there are several occasions where many or all of the patterns were not in the ideal phase for historic US East coast snowstorms. Some of these events were also poorly forecasted and busted big time. Some examples include the January 25-26, 2000 event that was a memoriable surprise, which shifted track at the last minute. Another is the January 26-27, 2015 event where many forecasters were predicting a double digit snow for Philadelphia with a moderately +AO(+1 standard deviation or greater), even though there had only been one time since 1950 when that occured(Superstorm 1993). This also was a Miller B storm, which in some cases has a sharp cutoff of the precipitation on the western side and can destroy a forecast. One great example of this was the December 30, 2000 event, where Wilmington, DE saw very little, Philladelpha had 7" and Northern NJ had 20"+. In addition, these storms more often affect New York City to New England. Lastly, there is the December 9, 2018 event over Richmond with a moderately +NAO and neutral AO and PNA. Some forecasts were very underdone until the last minute when totals were greatly increased.

Lastly, these results don't guarantee a historic snowstorm for any one city. It is a guide and comparision to show which patterns separately or together increase the chances of seeing such an event.

Data Sources

NAO, AO, PNA
Daily --> NAO - Climate Prediction Center
          PNA - Climate Prediction Center
          AO - Climate Prediction Center

PDO
Monthly --> JISAO

ENSO 
Weekly and Monthly --> Climate Prediction Center
3-Month Running Mean--> Oceanic-Nino Index - CPC

MJO
Daily and three month running mean--> Australian Bureau of Meteorology

Snowfall
Various National Weather Services Offices--> NWS