Our weather widget pulls your location based off of what your ISP returns for the city and state name. Sometimes this information is incorrect and in those cases we allow you to set a custom 5-digit ZIP code. Please enter your zip code below.

Showers remain from the coastal storm
on Tuesday, but abrupt changes coming
Join the Discussion!

EPAWA’s 2016-2017 Winter Outlook


It is that time of year when we look forward to the Winter months as we are now more than halfway through Autumn. On the heels of a strong El Niño winter last year, the transition to a weak La Niña occurred this Fall, and this outlook will cover what that and many other factors mean for the upcoming Winter of 2016-2017. We purposely waited to release at this juncture to incorporate all available data into the outlook, and not having to rely on speculation of the pattern we are headed into by releasing it early.

We will break down the technical of observational data that contributes to this outlook, and for those less versed, an easy to understand breakdown of temperatures, precipitation, and snowfall in map form follows the technical discussion. We even have a breakdown of expected snowfall totals in a chart at the end of this outlook for 28 select cities and towns within our tri-state coverage area. 

For coverage during snow events and daily short, medium and long-range outlooks our Premium Weather Forum has you covered. To learn more about that and all of our text message alert services, click here.

State of the Climate: La Niña

The current global Sea Surface Temperature (SST) anomalies are below… with colder waters being represented by the cooler colors, and the warmer waters represented by the warmer colors. SSTs drive the global climate, and that is no secret.

Note: All images hereinafter are clickable for better viewing.


Highlighted on the map above is the cooler anomalies in the equatorial eastern Pacific (west-based weak La Niña), a warm positive PDO that is trending in the direction of neutral, warm waters off of the US eastern seaboard in similar fashion to last year, and water surrounding Greenland – promoting an even more promising blocking signal than last year (see blocking/NAO discussion below).

The images below represent the current SST anomalies in the equatorial eastern Pacific, which continue to drop off with time, and the CFSv2 depiction of where the ONI index is headed, peaking low in January of 2017 from its current state.


The 850mb Zonal Wind Anomalies along the equator (5°S-5°N) continue to show a barrage of easterly winds, which promotes further cooling near the International Dateline especially in the near term. This supports the further drop in SSTs in this region, and the likelihood of a weak La Niña maintaining for at least the first half of Winter:


Taking a look a Philadelphia Winters in each climate state, the chart below highlights ENSO neutral and La Niña Winters and their local effects with snowfall on average. Since are likely to be in a weak La Niña for the first half of Winter, perhaps heading toward neutral in the 2nd half of Winter, both were considered, and both on average produce similar results. The stronger the La Niña, and the SE ridge begins to flex its muscles, and warmer temperatures occur in the SE and E US. The idea here that since we are projecting a weaker Niña this Winter, the SE ridge will have less of an influence.


Snow cover and the Arctic Oscillation


Siberian snow cover during October-November has a correlation with the Arctic Oscillation. Research done by Dr. Judah Cohen has shown this correlation very well. The more expansive the snowfall is in Siberia the lower the temperatures get. This surface cooling creates a stronger temperature gradient across Asia. The jet stream strengthens across Eastern Asia which creates a pattern that favors poleward heat flux/vertical wave propagation that disrupts the strong polar vortex. This leads to a weaker vortex that is more susceptible to being split or displaced out of the arctic circle. This then allows colder air to move into the middle latitudes.

The process of the vortex weakening and being displaced is what we call a negative Arctic Oscillation (-AO). The negative AO allows the jet stream to bend far enough to the south to allow the cold air locked up in the northern latitudes to be displaced southward into our region.

A positive Arctic Oscillation keeps the cold air locked up to the north.

Snow cover over North America also plays a key role in our weather. The colder airmasses moving in will begin to modify much quicker over snow free areas. If these airmasses cross over snow cover, they won’t moderate as much, delivering a more powerful and sustained chill.


The two images below show the current state of snow cover in the northern hemisphere in different ways. The first image (left) shows the current snow cover in Eurasia and North America, and the second image (right) shows that this level of snow cover is vastly above normal, especially in Siberia and Canada:

Note: All images are clickable for better viewing.


The chart shows the effects of snow cover in Siberia on the Stratosphere and Polar Vortex (PV). The PV is markedly weaker than it was last year, which allows it to become displaced or split more easily. Last year we dealt with a strong PV, which was stubborn and difficult or displace or split. As a result, we saw a +AO for much of the Winter aside from a few times, and warmer Winter temperatures as a whole. One perfectly-timed PV displacement occurred in the middle of January last year, which led to essentially one big storm a week later. Aside from that, however, Winter was basically non-existent, with most locations receiving less than 10″ if it wasn’t for the fluke blizzard. 

Quasi-biennial Oscillation

A factor that also determines the strength of the vortex over the Polar regions is the Quasi-Biennial Oscillation (QBO). The stratosphere temperatures are very cold over the Arctic, and warmer of the tropical stratosphere. During a negative QBO, it is easier for the polar vortex to split or be displaced. In a positive QBO, the gradient is strong and results in a stronger polar vortex. However, values of the QBO that fall between +10 and -10 are weak, and don’t have the same effects as a moderate or strongly positive QBO. Latest indications are that the QBO continues to rise at both 30mb and 50mb, but not too high to be alarmed just yet. With early recent splits and displacements of the PV, and the PV rendered much weaker certainly than last year, this may be offset somewhat by other factors, such as low solar activity, which we’ll cover more later in this outlook.

The two QBO indices below are taken at 30mb and 50mb, respectively. The data shows a westerly (+QBO) at 30mb, and a weak westerly at 50mb:





Latest values are shown to be positive and still increasing at a slow rate. However, in 1995 we saw a big drop towards neutral which supported a cold winter with blocking, which as you’ll see later in the outlook, is our top analog for this Winter. The combination of increased snow cover and weak westerly +QBO, and low solar activity supports the PV becoming displaced or split more easily. This supports the argument of a colder than average Winter, and certainly much colder than the Winter of 2015-2016.

The Madden-Julian Oscillation


The Madden-Julian Oscillation is an atmospheric oscillation, in which intense convective activity begins over the warm waters of the Indian Ocean, travels eastward, and dissipates over the colder waters of the eastern Pacific Ocean. Every 30 to 60 days, on average, a new burst of convection forms, and the eastward march begins again. The position of these areas of greatest convection are represented on phases, or octants, on a 1-8 scale as shown above. MJOs occur regardless of the season or year, but the stronger easterlies in a La Niña typically halt the eastward advance of these MJO waves at or near the international dateline. Since they remain strong up to that point, it can still remain favorable for an active pattern in the eastern US.

The phases of 8, 1, and 2 near the International Dateline is what typically supports cold and an active pattern the eastern US as shown below. This is a shorter range product that we will use frequently during the Winter to determine cold vs. warm periods, and active precipitation periods.

Screen Shot 2015-11-06 at 12.50.13 PM

Phase 7 in January with the tropical forcing around the dateline

Screen Shot 2015-11-06 at 12.50.20 PM

Phase 8 with the tropical forcing just east of the dateline


Other influences

The Pacific Decadal Oscillation (PDO)


The Pacific Decadal Oscillation (PDO) is a recurring pattern of ocean-atmosphere climate variability centered over the mid-latitude Pacific basin. The PDO is detected as warm or cool surface waters in the Pacific Ocean, north of 20° N. Currently this value is +0.45, making it a warm PDO. But it should also be noted that the trend of the PDO has been toward neutral recently. Winter will come down to is how long the warm +PDO stays intact, which will dictate a -EPO pattern and cold/trough in the NE US down the road. Right now it remains positive, but continues to head toward neutral territory.

The -EPO is what drove the Winter of 2013-2014 with frequent nickel and dime events, adding up to a lot by Winter’s end.


Solar influences


Based on the above chart, a solar minimum is on the way.  There is debate on what effects less sunspots has on our weather, but newer studies show that less sunspots (solar minimum) results in less solar radiation impacting the Earth’s atmosphere. As a result, more ozone can build in the stratosphere, and more warming occurs in the stratosphere and upper levels of the atmosphere. By warming the stratosphere over the Arctic regions, it helps to render the polar vortex (PV) weaker, where it can become easily displaced or split. As we talked about above, a disruption of the polar vortex from the polar regions helps to funnel colder air southward into the United States and other mid-latitude locations around the world, such as Europe and Asia.

Warm Atlantic SSTs


The Sea Surface Temperatures along the US East Coast have been anomalously warm, and remain warm heading into the Winter. These warm waters off of the coast can lead to rapidly deepening storms (cyclogenesis) and if tracked right, larger snow events. This has always been a wildcard in past Winters where SSTs were warm in this region, and that is no different this Winter.

SST are currently ranging from 15°-20°C (59°-68°F) from the immediate Mid-Atlantic US coast out to the area just beyond the Hudson Canyon. Beyond that, SST are ~26°C (79°F) in the Gulf Stream. With that said, all SST under the traditional coastal snow storm tracks are running 1°-4°C above average. This creates a double-edged sword for any low that tracks near the benchmark or just SE of such. The warmer water can help amplify the intensity of the surface low which ultimately can lead to greater adiabatic cooling, and a larger precipitation shield in the comma head of the mid-latitude cyclone. This, however, comes at the cost of warmer onshore flow during any easterly wind components. So for those lows that track between the benchmark and just SE, this would traditionally result in more rain/mixing between the onset of overrunning precipitation, and a back-end deformation band (most pertinent for the I-95 and SE area). We saw this kind of “snow-mix-snow” sandwich during the recent January 2016 snow storm.

The North Atlantic Oscillation (NAO)


The North Atlantic Oscillation (NAO) consists of two pressure centers in the North Atlantic: one is an area of low pressure typically located near Iceland, and the other an area of high pressure over the Azores (an island chain located in the eastern Atlantic Ocean). It is important to note that these two locations are most commonly used to measure the NAO, but studies have found that the pressure centers move around on a seasonal basis, and other locations have also been used for measuring this index. Fluctuations in the strength of these features significantly alters the alignment of the jet stream, especially over the eastern U.S., and ultimately affects temperature and precipitation distributions in this area. It is also important to note that the AO and NAO are two separate indices that are ultimately describing the same phenomenon of varying pressure gradients in the northern latitudes and the resultant effects on temperature and storm tracks across the continent.

Positive NAO

During a positive NAO there is a strengthening of the Icelandic low and Azores high. This strengthening results in an increased pressure gradient over the North Atlantic, which cause the westerlies to increase in strength. The increased westerlies allow cold air to drain off the North American continent rather than letting it build up and move south.


A negative NAO indicates weakening of both the Icelandic low and Azores high, which decreases the pressure gradient across the North Atlantic. This decreased pressure gradient results in a slackening of the westerlies. The decrease in the westerlies allows cold air to build up over Canada, and this combined with below average heights (troughing) over the eastern U.S. gives the cold air a greater chance to move south and affect the eastern United States.

There are also differences between east and west based negative NAO, as shown below. A west-based negative NAO is more favorable for cold/snow in our region:


Indications are that blocking will occur more frequently than the last few years. Those indications are supported by a dominant -AO, anomalously increased snow cover over Eurasia, and low solar activity. In addition, sea surface temperatures in the North Atlantic (AMO) support blocking. The cold pool in the North Atlantic is in a promising spot heading into this winter, and as long as the warm anomalies across Greenland maintain, combined with a weaker polar vortex than last winter, the indications are more favorable for a –NAO blocking pattern given the warm anomalies around the cold pool as shown below:


Using past analogs


The current SSTA map above points out a few major trends that could lead to an active winter:

  1. La Niña becoming focused in the central and western basins.
  2. Warm +PDO enhancing ridging in Western Canada – expecting a +PDO to start, but will trend toward neutral toward the end of Winter most likely.
  3. Cold pool in North Atlantic. This SST configuration supports a -NAO set up.

Analog matching involves where Sea Surface Temperatures are a process of comparing to years of the past that are similar. Since they are the biggest driver of climate, this is the biggest weight factor. There are other factors such as solar activity that we consider into our analog choices, however. We narrowed the list to 7 years, and weighted each year/season according to closest match. On the chart below, we outline the years that are the closest matches. You’ll take note that there is no 100% match to this year, but these seasons are ranked according to the closest match (1995-1996) to the least closest match on the list (1966-1967), and are weighted as such in rank order.


Using those same analog years, these were results at those climo stations. Keep in mind, we can only use climo stations because they keep official records. So if you live in an area in between these areas, pick out the closest location to where you live and go off of that. This is to be used for trends only. For example, if you live in Reading, PA, use Allentown for a point of reference, perhaps subtracting a few inches.


There are some high snow totals in there for 1995-1996 and 2013-2014. But don’t get too hung up on the amounts. 1996 featured the blizzard in early January, and you can’t rely on a blizzard coming to fruition to reach projected totals and fit an outlook agenda. Still, 130-150% of normal snowfall would still have occurred in the Winter of 1995-1996 if we had NO blizzard, which is the key takeaway. Blocking is realized, and it increases your chances for that to occur.

2013-2014 did not see blocking occur, but that looks to be due to a high +QBO and high solar year, which was still overcome with a dominant -EPO pattern, thanks in part to the warm +PDO.

Blending the top 3 Winters according to rank, we come up with this temperature and precipitation blend for December, January, February, and March individually:

Note: All images are clickable for better viewing.

December temp


January temp


February temp


March temp


December precip


January precip


February precip


March precip


Most of the top analogs we used had more than just a few things in common, but none more common than the notion that Winter began early… which would be a change from what we saw over the past couple of years. The last Winter where we saw a flip in the pattern to a consistently cold and snowy one as early as late November was the Winter of 2010-2011. Climate models and analogs are in close agreement that this flip occurs somewhere around or just after Thanksgiving (November 24th) and continues into the Winter months. Models were quick to jump the gun on this transition, which is typical with seasonal change. The Pacific is still not cooperating in the overall pattern, and until it does, anomalous warmth will continue in western Canada, preventing a snow cover buildup in that area for now.

We are projecting the pattern to flip during the last week of November, not mid-month as some models had previously shown, likely near or after Thanksgiving.

My Pocket Meteorologist Alerts

Stay on top of winter weather this season with EPAWA’s Winter Weather Alerts and Premium Forum. Now in its fourth year, everyone from commuters to businesses that depend on the weather are benefitting from dependable forecasts direct to their cell phones. Tailored for those who want early warnings for any significant weather system moving into their specific county during winter only, November 1 through March 31st, or last snowfall.

To sign up for the Winter-only text alerts, for one monitoring county or two if you commute across county lines, and/or to sign up for the EPAWA premium forum, visit:  http://epawaweather.com/winter-weather-alerts/ or click on the products image below:


For those that had the Winter alerts last season:  

If you had Winter-only text alerts last year, now is the time to get signed up for them again. Log into your account in the billing system at: https://mpm.epawaweather.com/ under “secure client login”. If you don’t remember your password, that’s ok, there is a password reset link on the next page. Once logged in, select either the MPM – Winter One County Weather Alerts Service (3rd down on the list, for one monitoring county) or MPM – Winter Two County Weather Alerts Service (4th down on the list, for those that commute across county lines) by pressing the “order now” button next to it. Verify the cell number you want to receive the texts to is correct – and the one or two counties that you want monitored in the drop down menu below it. It will take you to an order summary page, make sure the info so far is correct, and then click on “checkout”. Verify the billing information on the next page, with either credit card or PayPal, and then “complete order” when finished. You’re signed up through March 31st, 2017, or last snow. If you are already signed up for monthly, quarterly, semi-annually, or annual text alerts, you need not purchase this option and can ignore the steps above. Winter alerts are for those that want a one-time payment option for Winter months only.

EPAWA Winter projections by month

MonthTemperatures (month)Precipitation (month)Snowfall
December 2016Below normalSlightly above normalAbove normal
January 2017Well below normalSlightly above normalAbove normal
February 2017Near normalNear normalNear to slightly below normal
March 2017Near to slightly aboveNear normalNear to slightly below normal

Unlike recent Winters, this year Winter starts early. We went against analogs for the 2nd half of Winter, as the PDO continues toward neutral territory, perhaps even becoming negative in late February and March. This will, in theory, allow for a +EPO and faster Pacific pattern, less dominated by the -AO in latter months. Should the PDO remain positive for most if not all of the Winter, and/or the -AO becomes more dominant than thought in part to expansive snow cover and frequent PV displacements/splits, the 2nd half of Winter could be just as busy as the first half. Given the trends recently, it is a tough call, but siding with the PDO continuing toward neutral/negative, which will introduce warmer air and an early Spring by late February and/or March.  

Total snowfall, Nov. 15th - Mar 31st as a percentage of normal


As you can see from the image above, the farther in the interior you are located, the higher above normal snowfall you can expect. The coastal regions still receive snow, but a lot of mix/slop storms will hold accumulations down. Typically I-95 is a good waypoint marker that separates coastal regions from the interior, but areas north and west of I-95 will also see their fair share of mix/slop storms, just not as frequent the farther NW one travels.

Select cities snowfall expectations within the EPAWA coverage area

LocationNormal annual snowfall this locationEPAWA projected snowfall in 2016-2017
State College, PA46"64" or greater
Mount Pocono, PA48"57" to 67"
Sussex, NJ41"49" to 57"
Williamsport, PA39"54" or greater
Towanda, PA44"62" or greater
Hazleton, PA44"53" to 62"
Pottsville, PA43"52" to 60"
Stroudsburg, PA39"47" to 55"
Harrisburg, PA33"46" or greater
Hackettstown, NJ44"53" to 62"
Scranton/Wilkes-Barre, PA46"55" to 64"
York, PA25"30" to 35"
Allentown/Lehigh Valley, PA33"40" to 46"
Reading, PA32"38" to 45"
Lancaster, PA26"31" to 37"
Quakertown, PA31"37" to 43"
Pottstown, PA30"36" to 42"
Somerville, NJ30"36" to 42"
Newark, NJ25"30" to 35"
Wilmington, DE19"19" to 23"
Trenton, NJ23"23" to 28"
Philadelphia, PA22"22" to 26"
Freehold, NJ25"25" to 30"
Mount Holly, NJ20"20" to 24"
Dover, DE14"11" to 14"
Vineland, NJ18"14" to 18"
Atlantic City, NJ17"14" to 17"
Cape May, NJ15"12" to 15"

Winter Outlook Summary

Plain English Summary:

  • In comparison to last year, this Winter season will be much colder than last year, which isn’t really difficult to do, but overall late November through the end of March averages slightly below normal when averaging the entire period.
  • Winter will start early this year, which is a big change from recent previous years. Wintry precip could start in late November for far interior and far northern areas
  • Although a milder period is expected late February or March, colder shots will still occur during that period, and snow is still possible, especially for the interior, through Winter’s end
  • Higher frequency of storms expected this year, with many Lake Effect snow episodes and fast-moving clipper systems at the start of Winter for the interior
  • The middle of Winter will feature a higher than normal amount of SWFE (southwest flow events) with overrunning snow and ice along thermal gradients/boundaries, and coastal redevelopment possible near the coast
  • Storm tracks along the coast will have to be watched for rapid deepening with warmer SSTs along the US Eastern Seaboard
  • Greater potential for blocking this year, highest potential in several years
  • Overall, most areas see near to above normal snowfall, especially the farther NW in the coverage area, even without a major storm or blizzard – a blizzard is considered a wildcard that can skew numbers higher than what is listed in this outlook, and the projected numbers do not reflect that potential

What can go wrong?

  • QBO further strengthens and becomes more strongly positive. This would increase the chances of the AO remaining predominantly positive, but this is unlikely based on current observations and modeling down the road
  • The PDO drops toward neutral or negative faster than expected. This would allow for a warmer 2nd half of Winter starting earlier in February instead of late February or March, as warm Pacific air will flood the country in a +EPO configuration
  • Blocking is not realized. Last year’s cold pool in the northern Atlantic looked promising at this time, but the warmth over Greenland eroded, keeping low pressure at higher latitudes instead of farther south where we needed it for blocking. Blocking potential looks promising thus far, but this is never a guarantee that it maintains. The QBO continuing to rise could be a negative factor with regard to blocking.
  • The AO becomes too strongly negative. This is also possible, where too much cold air overwhelms the pattern, and suppresses storms south of the region. This limits precipitation overall, and limits bigger storm potential if this happens
  • The Winter takes on a 1995-1996 look, but like January 1996, a blizzard hits that skews totals much higher. It is always a wildcard in any outlook, as these cannot be projected this far in advance
  • The PDO maintains positive/warm longer than expected, keeping the cold and snow around longer than expected. This isn’t that far fetched, as our leading analogs suggested a colder/snowier March, but that will be predicated on how long the PDO stays positive

Outlook prepared by:  Martrich/Carmella, with contributions to the outlook by Jonathan Carr, Weather NJ

Outlook prepared and finalized:  Friday, November 4th, 2016

Click the image below to get alerts and forum!