Here is a look at my winter outlook for last year (along with links to previous outlooks):

https://ohwxramblings.wordpress.com/2014/10/12/jim-sullivans-2014-2015-winter-outlook-for-the-us-and-southern-canada/

Maps and Brief Summary:

This winter will be strongly affected by a very strong “El Nino” in the Tropical Pacific. This will likely result in a warm December, an up and down January, and a cold February in the Eastern US. The western US will become warmer as the winter moves on. Precipitation will be abundant across the Deep South in response to the El Nino. I expect the El Nino to contribute to continued rain/snow chances in California through December, before conditions gradually dry up in January and especially February.

There could be an active storm track from the southern Plains east towards the East Coast. Locations like the Great Lakes may stay north of the big storms, while locations like the Ohio Valley are near the northern edge of the active storm track later in winter.

The following are maps depicting departures from normal for the winter months across the country for temperatures, and general precipitation patterns. Remember a warm month overall can still have some cold days, and vice versa.

December:

January:

February:

Technical Discussion:

This will be my 6^th winter outlook; I’ve had two decent outlooks (2009-2010, 2014-2015), two outlooks that had varied success in different regions of the country but could’ve been much better (2012-2013, 2013-2014), and one that I’d consider subpar/not very helpful (2010-2011). I guess it’s good that I consider my most recent outlook to be one of my better ones.

I do arrive at a set of analogs at the end, which I do partially base my monthly maps off of, however extensive reasoning is provided herein on how I came to my conclusions.

Let’s Start With Sea Surface Temperatures:

Current sea surface temperatures as we head into October show several distinct features; a basin-wide and strong (bordering on “super”) El Nino in the Tropical Pacific, impressive warm waters well north of the ENSO regions from near Hawaii all the way to California, an abnormally warm Indian Ocean, very warm waters compared to normal, surrounding the Aleutian Islands, warm waters (that have cooled some recently) south of Alaska and off of the west coast of Canada, and an overall very warm Atlantic, with the exception of a large cold pool across the N. Atlantic.

Closer Look at El Nino:

The western Nino regions (3, 3.4 and 4) have all featured warmer than average waters since summer of 2014, although the eastern region (1+2) did cool during the winter of 2014-15. All Nino regions except for 4 (which was quite warm already) warmed strongly starting in the spring of 2015, and after a brief “leveling off” in the warming late in the summer, have warmed again in September. Nino 4 has the smallest warm SST anomaly, however this is quite typical, an anomaly of around +1C for Nino 4 is impressive. Regions 1+2, 3 and 3.4 all have SST anomalies of +2C or greater, which pushes them into the “super El Nino” range. These very warm waters haven’t persisted long enough for this El Nino to be officially classified as a super El Nino yet, although that may just be a matter of time.

A look at the 850mb winds over the equatorial Pacific shows that there have been persistent weaker than normal trade winds (denoted by the warm colors) since last Spring, with a few well defined westerly wind bursts (WWB). These weaker trades are important, as they help the warmer pool of water normally located closer to Australia and the Pacific Maritimes to spread farther east and limit upwelling of cooler waters. Westerly wind bursts are especially proficient at encouraging further warming, and usually help downwell warmer waters. There is a robust WWB occurring at present across much of the eastern equatorial Pacific, which is likely accounting for the recent warming discussed above. This suggests that the El Nino is not done strengthening as we head through October.

One distinct characteristic of an El Nino is a deepening of the thermocline east of the International Dateline. The thermocline has deepened considerably farther east over the last two months, all the way to the S. America coast. This suggests that El Nino’s effects are not confined to just the central and western Nino regions, this is a basin wide event (NOT a modoki or anything close). This is important as it has an impact on where the strongest tropical forcing may be located this winter. It’s also notable that the recent WWB has had a noticeable effect on deepening the thermocline in the central Pacific over the last two weeks.

http://www.esrl.noaa.gov/psd/enso/mei/

When comparing the 2015 El Nino to past strong El Nino events, using the Multivariate ENSO Index, which incorporates factors other than just sea surface temperatures when determining ENSO strength, it can be seen that for the August/September period, the only El Nino stronger than 2015 was the 1997-98 super El Nino. The only two El Ninos to peak higher than 2015 at any time of year are 1982-83 and 1997-98. Those two El Ninos are the only two “super” El Ninos since 1950 (MEI of +3 SD or greater), and both of those events had a secondary MEI peak in mid-winter.

http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ensoyears.shtml

Another method of determining Nino strength, called the Ocean Nino Index, relies strictly on sea surface temperature anomalies in the ENSO regions. The most recent tri-monthly value for JAS was +1.5, making this a strong El Nino. Weekly values have been generally +2 for quite some time, so it’s likely that the next two tri-monthly values continue to increase. The threshold for a “super” El Nino using the ONI is an anomaly of +2.0C or greater; that has only been done three times, in the 1972-73, 1982-83 and 1997-98 El Nino events. The JAS ONI value was the third highest since 1950, behind 1997 and 1987. 1987 did not get any stronger after JAS, while 1997 intensified into the early winter.

Either way, given the recent WWB near the Dateline and the continued deepening of the thermocline in the equatorial Pacific, further intensification of this El Nino is likely in the near future. A peak very close to or possibly stronger than 1982-83 is possible, although a peak matching or exceeding 1997-98 is a bit more of a stretch.

This very strong El Nino is a very good indicator of the general pattern that will likely be in place this winter across the CONUS; tropical forcing is usually focused in the central/eastern Pacific, east of the Dateline, which forces a persistent low pressure somewhere near or south of Alaska, which drives the polar jet northward across the west coast of the continent, causing warmer and somewhat drier conditions across much of the northern US in an El Nino winter. El Ninos also strengthen the sub-tropical jet, bringing increased moisture to the southern US and usually cooler weather to the Deep South.

Typically, stronger El Ninos match the above “perfect El Nino” pattern better than weaker ones, as the forcing with weaker El Ninos may not overwhelm other factors. However, not all El Ninos come in the exact same flavor (or else I’d be done with my outlook right now):

http://www.jamstec.go.jp/frcgc/research/d1/iod/enmodoki_home_s.html.en

This webpage from JAMSTEC explains the difference between a “Modoki” El Nino, one that has the warmest anomalies in the central Pacific, closer to the Dateline, with cooler waters off of the S. American Coast, and a normal El Nino, that has warmer waters farther east. The location of the warmest waters can have a big impact on where persistent tropical convection that ultimately drives the jet stream’s response to an El Nino occurs.

There are really three different types of El Nino: Modoki, basin wide, and east based:

All three different types, on average, yield a different positioning of the Gulf of Alaska low, and a different CONUS wintertime weather pattern:

Note how a central based/Modoki El Nino tends to have a weaker GOA low that is generally farther southwest, while the east based lows, in the means, are stronger and progressively farther northeast, flooding more of the CONUS with mild Pacific air, resulting in the negative anomaly moving farther southeast off the East Coast. These composites were made using all El Ninos since 1950 regardless of strength, however the Modoki and basin wide events have a reasonably large sample size and produced reasonably strong composite anomalies when blended together. Two of the three east-based El Ninos were strong-super, not a coincidence.

The difference in the GOA low position is due to differences in the intensity and location of the strongest tropical forcing in different El Nino winters. In an attempt to further diagnose this difference in forcing to use it as an aid in forecasting this winter, I looked at a few different methods to quantify the tropical forcing in the different El Ninos:

The first method was to look at 250mb height anomalies during DJF; the reasoning behind this is tropical convection gives off a tremendous amount of latent heat, which will thicken the atmosphere and raise the geopotential heights of the upper atmosphere. This is important because this tightens the pressure gradient immediately north of the higher heights and strengthens the jet stream, altering the downstream weather pattern. This first exercise was actually rather interesting, as it seemed as though the strongest anomalies in all three types of El Nino were located near 130W; it was the strength of the anomalies and eastward extent that differed. East based El Ninos had the strongest anomalies at 130W, with the anomalies extending farther east, likely suggesting that the convection along the equator in the Pacific in those El Ninos was more robust and also occurred farther east than the basin wide and especially modoki events. This may also be a by-product of the east based and basin wide El Ninos on average being stronger than Modoki events (with east based events being stronger than basin wide on average). With east based events, weak positive height anomalies extended well west towards the Indian Ocean; this was not the case with basin wide or Modoki events.

The second method was to examine outgoing longwave radiation (OLR) anomalies in the Pacific during the three types of El Nino during DJF. This helps show where tropical forcing is more prevalent because the colder cloud tops associated with thunderstorms radiates less outgoing longwave radiation; so a dearth of outgoing longwave radiation indicates more persistent thunderstorms. This method was similar to the 250mb height anomaly method in that the Modoki El Nino forcing was generally weaker than the basin wide and east based events, however the difference in this method is that the east based forcing shows up as being centered much farther east than the other two, unlike the height anomaly method where the max anomalies were near the same longitude (130W). It’s notable that the positive OLR anomalies (associated with less convection than normal) across the Maritimes into the Indian Ocean are much more distinct with east based events than basin wide or Modoki ones, indicated a lack of MJO activity overall with east based events.

A third proxy I used was 850mb zonal wind anomalies. This does two things; it determines where warmer/colder waters are likely to persist in the equatorial Pacific (westerly anomalies/weaker trades suggest warmer waters, vice versa with easterly anomalies/stronger trades), and also suggests where more persistent tropical convection is likely to be, as low level convergence along the eastern edge of the westerly anomalies promotes more active convection. Much as the case with the 250mb height anomalies and OLR anomalies, east based events feature stronger anomalies farther east, while basin wide are markedly farther west and a bit weaker, while modoki events on average feature the farthest west and weakest anomalies.

So, with all of this said, let’s start comparing to this year:

Height anomalies for September and first half of October:

September 1-October 18 OLR anomalies:

September 1-October 18 850mb zonal wind anomalies:

(note that I tweaked the scale some to show that while the anomalies near 150E are very strong, that strong anomalies extend well east of the Dateline)

Another method of looking at the position of the 200mb and 850mb anomalies:

A few things of note: The strongest negative OLR anomalies are located near the Dateline, much farther west than the DJF means for any type of Nino, however stronger anomalies along the equator do extend rather far east in a fashion most similar to basin wide events. The 250mb height anomalies are pushing 60m, close to the DJF mean for east based events, and stronger than the basin wide events mean. Also, the positive anomalies extend well east towards S. America, most like east based events. There’s also a stout positive height anomaly from NE Asia towards western Alaska. The 850mb zonal wind anomalies are centered much farther west than the east-based DJF composite; in fact, the max anomalies are centered near the DJF modoki composite, however the implied convergence at the eastern edge of the westerly anomalies is located farther east, closer to the basin wide DJF composite. The strongest 200mb velocity potential has been located along and a bit east of the Dateline, and has gradually nudged west over the last couple of months.

Before we go any farther, it’s important to note that these height/OLR anomalies for 2015 are for September-October, NOT DJF. First, I’m going to recreate the 250mb height anomaly, OLR anomaly and 850mb zonal wind anomaly for the three types of El Nino events for S-O:

250mb height anomalies:

OLR anomalies:

850mb zonal wind anomalies:

When comparing this September-October (so far) to the composites above, the 250mb height anomalies in the mid-latitudes are actually closest to the modoki composite; it’s very notable that really NO other El Nino since 1950 has had such high 250mb geopotential height anomalies over the tropics. The OLR and zonal wind anomalies are somewhat mixed in the comparisons; the strongest OLR anomalies in September-October are centered smack dab over the dateline, which is actually farther west than the strongest OLR in any of the September-October El Nino composite anomalies. However, the negative OLR anomalies extend farther east, so overall it has more of a basin wide event forcing look. The 850mb zonal wind anomalies are similar; the strongest westerly anomalies are well west of the dateline, centered somewhere between the modoki and basin wide strongest composite westerly anomalies, however per the time-longitude plot above, this is skewed a bit by a strong, but brief WWB in early October east of the Dateline. Overall, the strongest westerly 850mb anomalies have been west of the Dateline, most similar to basin wide events. In general, it seems like forcing is currently mirroring basin wide events the best this fall, however, as noted above it’s not a perfect match.

Now, I’m going to look at all moderate-super El Ninos (per ONI) since 1950 to determine which El Ninos had similar September-October patterns to this one, and determine what kind of El Nino they were and how their upcoming winters turned out. It was actually tough to find similar analogs to this year, with very few years that looked anywhere close to the N. American pattern we have seen thus far this fall during a moderate-super El Nino. Another complicating factor was this year’s odd convection placement along the equator; there’s a clear max in convection centered at the Dateline, however a lesser but still abnormal amount of convection has occurred rather far east.

After looking over all moderate-super El Nino Septembers and Octobers, I determined that 1957, 1986 and 1987 were the “closest” matches to our current tropical forcing. With this said, the overall pattern in the mid-latitudes doesn’t really match what we’ve seen this fall thus far. Because 1986 was the first year of a two year El Nino and was strengthening through DJF (very unlikely in this second year El Nino), I tossed it out. The overall OLR composite for September and October for 1957 and 1987 is “OK”:

The following DJF mean OLR anomaly looks close to what you’d expect for a basin wide El Nino event, with the GOA low positioning also close to what you’d expect for a basin wide event:

After the review above, my best guess at this time is that the Gulf of Alaska low placement will be somewhat close to what is “typical” for basin-wide El Ninos. This makes sense; this is clearly NOT anything near a Modoki based on the exhaustive review done thus far, but warm SST anomalies along with the location of the most persistent tropical forcing are most certainly farther west than east based events, despite very warm SST anomalies in the eastern Nino regions.

It is worth noting that per the MEI, this is the strongest basin-wide El Nino since at least 1950, and if the ONI continues rising as it should per the recent weekly ENSO region weekly anomalies, there’s a good chance this will beat 1972-73 for the strongest basin wide El Nino per the ONI as well. So, we’re already in potentially uncharted territory when looking at just the El Nino strength and type.

Other factors heading into this winter are also rather unusual, and will be investigated now. The very warm waters off of the western N. American coast, in the sub-tropics and in the northern Pacific are very impressive. These waters aren’t paramount to tropical forcing that is a significant driver of the wintertime jet stream strength/position, but seem somewhat important.

The first area examined specifically will be the area of very warm waters extending from south of Hawaii east-northeast towards California. This year’s waters are off the charts warm in this area, with widespread anomalies of +2-4C right now. There are a few possible things I can think of this effecting: It could possibly weaken trade winds farther east (weaker temperature and pressure gradient between the EQ and the sub-tropics in this area), possibly forcing convection farther east. It may also allow additional moisture into the Southwest US. To examine this area, I looked at all moderate or stronger basin wide El Ninos since 1950 (per ONI), and looked for years that also had warmer waters in this region. No year has waters as anomalously warm as this year, however a few did have generally warmer waters in this same area. Her is the September-October SST difference map in the four Ninos I have with warmer waters from Hawaii up to near California compared to the rest of the basin wide, moderate plus El Nino events since 1950:

When looking at the difference of several potentially key DJF variables between years that featured warm waters from near Hawaii to California, a few interesting features were evident:

It is worth noting that these years ignore all other variables. This is just the difference in moderate or strong El Nino winters between ones that had the warmer HI-CA waters compared to years that didn’t. It’s possible that other variables are affecting some of these differences.

One interesting difference is an apparent tendency to pull tropical forcing a bit farther west in years where the warm water is present; the Pacific jet is stronger overall, but strongest over the central Pacific, with stronger trade winds and less OLR (indicative of more clouds/convection) along and west of the Dateline, potentially indicative of more MJO activity, which is normally limited during stronger El Nino winters. This results in the heights over the Aleutians being lower (GOA low farther west) and heights along the west coast being a bit higher (overall a slightly more favorable +PNA look). There is also a distinct tendency for a strongly west based –NAO. The look also looks potentially wetter for CA (stronger westerly 850mb winds, a stronger jet, lower OLR).

The overall tendency appears to be for winters that have a more favorable N. Pac orientation and NAO for troughing over the eastern US…however, the sample size is small, and it’s quite possible other factors influenced/caused these differences. It IS worth noting that my two tropical forcing “analogs” (1957-58, 1987-88) and 1986-87 (not one of my forcing analogs, tossed due to being a first year Nino that strengthened during winter) are in the list of moderate or stronger basin wide El Nino winters that had warm water from HI to CA.

The next area/factor examined specifically will be the very warm N. Pacific waters and strongly positive Pacific Decadal Oscillation (PDO). The PDO is a pattern of north Pacific SSTs that tend to persist for around 30 years. A warm phase of the PDO features colder waters of the north-central and northwest north Pacific, with warm waters off of the west coast of N. America, while a cold phase of the PDO features warmer waters of the northwest and north-central north Pacific, with colder waters off of the west coast of N. America.

http://research.jisao.washington.edu/pdo/

We had been in a 10-15 year old cold PDO phase, but have seen a strong rise over the last couple of years, which isn’t uncommon during El Nino events that occur during an otherwise cold PDO. Sea surface temperatures in the PDO region are not nearly warm enough to force convection that completely re-shapes the jet stream, however, PDO trends may be more indicative of persistent weather patterns, and although the sea surface temperatures here aren’t nearly as significant as over the tropics, they may cause small differences in the track/intensity of mid-latitude cyclones in this region that does have some impact on the overall pattern.

September’s PDO value came in a +1.94, per the University of Washington, which is the highest it has been since March. There are only two moderate or stronger, basin wide El Nino events since 1950 that had a September PDO of greater than +1.50 per the University of Washington dataset (or greater than +1 for that matter); 1957 and 1987. These are ironically the two best tropical forcing “analogs,” and are also two of the four moderate or stronger basin wide El Nino events since 1950 that featured warm water from HI to CA. When looking at difference maps as done above between the moderate or stronger basin wide El Nino events with strong warm September PDOs and the rest of the moderate or stronger basin wide El Nino events, some similar trends are noted:

One initial warning here is that we are starting to deal with small sample sizes (two years), and that factors other than PDO may influence some of these differences. However, some trends that are again apparent is for stronger trade winds east of the Dateline, perhaps suggesting tropical forcing will be located somewhat farther west than most basin wide El Nino events. The 250mb zonal wind difference also suggests a stronger jet over the Pacific, although in this instance the jet is even stronger in the west coast of North America. There is also a tendency for a more strongly west based –NAO in 1957-58 and 1987-88 compared to the rest of the basin wide moderate or strong El Nino winters…again however, this is a two year sample, and should be used with caution.

The next factor examined is the very warm Indian Ocean; much of the basin is experiencing SST anomalies of near or above +1C; since 1950, only 3 moderate or stronger basin wide El Nino events have featured a warm overall Indian Ocean (and most not as warm as this year), when looking at September-October Indian Ocean SSTs. The years are: 1987-88, 2002-03 and 2006-07. Here are the difference maps compared to basin wide moderate or stronger El Nino winters that had normal or cooler than normal Indian Oceans:

A few interesting trends; again, much like warm HI to CA SSTs and strong +PDO years, the jet is stronger across the northern Pacific; however, in this instance, the OLR is lower west of the Dateline, indicating more convection than the rest of the basin wide Nino years in this areas, perhaps due to a more active MJO than what normally occurs in El Nino years (which is potentially a direct result of the warm Indian Ocean supporting lower pressures and more convection than what would occur farther west in a typical El Nino). There’s also a strong high latitude blocking signal across N. America, although it’s possible other factors such as the stratosphere or solar influences are at work here. There is an overall trend for the GOA low to be centered a good bit farther west in these winters than in typical basin wide El Nino winters.

As a note, 1987 is the only September-October moderate or stronger basin wide El Nino since 1950 that featured a similar N. Atlantic SST configuration to 2015. That seems to be a top analog based on global SST patterns.

Moving on and looking more specifically at high latitude blocking now, there is a clear trend in all moderate-super El Nino winters since 1950 (regardless of east based, basin wide or modoki) for a progressively more negative NAO and to an extent AO as the winter progresses (along with a trend for the GOA to retrograde through the winter). This signifies that there is a strong chance that tropical forcing associated with stronger El Nino events tends to favor a –NAO. The question is, why, and will a –NAO be favored later this winter as well?

When taking the basin wide events (which we will be dealing with this winter) and subtracting all other moderate to strong events from it, the –NAO signal becomes weaker! Granted, a great –NAO winter like 2009-10 is removed in this process. Also, there’s a tendency for higher heights in the PNA/EPO domains, potentially due to removal of east based super El Nino events in 1982-83 and 1997-98. There is still a –NAO signal in the means for basin wide events for January and February, although the above implication is that removing the modoki events likely weakens the overall –NAO signal for moderate to strong El Nino winters:

QBO/Stratosphere Stuff:

The Quasi Biennial Oscillation refers to phases of more westerly or easterly winds in the stratosphere over the Tropics. Easterly phases are typically associated with a weaker winter time stratospheric polar vortex (and a better likelihood of sudden stratospheric warming events, leading to potentially periods of deep –AO blocking), while westerly phases tend to be associated with a stronger winter time stratospheric polar vortex (and a lower likelihood of the vortex getting disrupted in winter, potentially leading to periods of +AO conditions). There are exceptions to the rule, but the correlation is fairly well understood.

As seen on the above graph, there are westerly winds descending through the stratosphere right now, suggesting that the winter QBO will be positive or westerly. There have been 5 moderate to strong El Nino winters since 1950 that featured a +QBO (per 30mb wind anomalies), per these data sources:

http://www.cpc.ncep.noaa.gov/data/indices/qbo.u30.index

https://climatedataguide.ucar.edu/climate-data/qbo-quasi-biennial-oscillation

There was a pretty strong signal for retrogression of the GOA low through the winter and a developing –NAO by February. The February pattern looks quite favorable, even with the super east based El Nino of 1982-83 in that composite.

When removing the modoki and east based El Nino winters from the composites, the overall message doesn’t change a ton; a fairly warm December with a GOA low flooding the CONUS with Pacific air, with no AO or NAO help, followed by a retrogression of the GOA low for January and February allowing for West Coast ridging, with a declining AO and NAO for February. The AO decline is much more noticeable in this composite than the one including all moderate-strong +QBO winters.

Another factor that influences the potential for high latitude blocking during winter is solar activity; I’m not extremely well versed on how this process works so I won’t try to explain it, but in general, the best combination for a deep, -AO winter is a solar minimum and –QBO. Monthly solar radio flux data from the sun can be found here back into the 1940’s:

http://www.spaceweather.gc.ca/solarflux/sx-6-mavg-en.php

We are currently coming off of a weak solar maximum; slightly weaker than the maximum in the early 1970’s; radio flux output will likely be in the neighborhood of 85-105 if the current rate of decline continues into this winter. Out of the +QBO basin wide ENSO years above, 1988 is an OK but not great solar match, as the radio flux was very similar to what’s expected this winter, however 1987-88 was near the beginning of a significant ramp up towards the next solar maximum near 1990. 2002-03 is also not the best match, as the radio flux values are higher, however, 2002-03 was a winter that also occurred in the declining phase of the solar cycle. 2006-07 occurred before the most recent solar minimum, with numbers slightly lower than what will likely occur this winter. If we see a ramp up in solar activity, 2002-03 could become a better analog; if we see a decrease, 2006-07 could become a better analog from a +QBO/ENSO/solar perspective.

Overall, the current solar activity doesn’t appear extremely helpful or unhelpful in attempting to get high latitude blocking this winter.

Peaking at “newer” research done by others:

https://www.aer.com/science-research/climate-weather/arctic-oscillation

Dr. Judah Cohen has done extensive research that suggests that the late winter AO state is strongly correlated to October snowcover advance in Eurasia south of 60N (especially in Siberia); the reasoning is found in the graph above. The “Snow Advance Index” did very well in predicting AO state until the last two winters (with the 2014-15 winter being a fairly large failure); there is valid meteorology behind the idea, so although the index failed to properly predict last winter’s AO state, it is still worth looking at for this winter.

The above plots, found Dr. Cohen’s blog linked above, it can be seen that so far, through the first half or so of October, Eurasian snowcover is advancing quicker than normal, but a bit behind the last two years.

The GFS has been persistently showing a significant amount of snow south of 60N across Eurasia over the next 10 days. Considering the model has been showing this fairly consistently, and considering the current Eurasian pattern features a few upper lows/troughs well south of 60N, likely supporting snow, and considering the thus far decent (but not as good as last year) snow advance this month, it seems more likely than not that the SAI will finish higher than normal, suggesting a more –AO winter is potentially more likely. It remains to be seen whether it’s well above normal or just a bit above, but currently this factor doesn’t look to hurt things.

Research done by Al Marinaro (wxmidwest) suggests that the winter NAO state in El Nino winters is correlated to N. Pacific sea level pressure in October. This correlation makes sense when thought through, as El Nino winters in general appear to favor a –NAO, with some exceptions, generally based on where/how strong the tropical forcing is. This may be an early way of diagnosing the tropical forcing, and seeing if it’s in a favorable location for a winter –NAO.

SO FAR, through October 18^th, the N. Pacific SLP appears to be averaging lower than 1013mb, which would favor a –NAO. The catch is it’s a bit early. The “good” news is, the European ensembles insist on a very active weather pattern in the N. Pac, with a persistent Aleutian low through the remainder of October with a recurving typhoon perhaps deepening the feature quite a bit:

It appears possible that the N. Pac SLP in Al Marinaro’s box will be quite a bit below 1013mb on the mean for the month of October, which tends to correlate to a –NAO in an El Nino winter.

In Conclusion:

We are essentially in uncharted territory since at least 1950 in terms of a basin wide El Nino event of this strength, when combined with a warm Indian Ocean, very warm waters compared to Hawaii to California and a very warm N. Pacific. However, there is evidence that the warm Indian Ocean may act to cause somewhat more MJO activity that what is common during stronger El Nino winters, and in general, the warmer waters elsewhere in the East Pacific tended to pull the Aleutian low farther west when compared with other moderate to strong basin wide El Nino events. This suggests that a +PNA may become more likely as the winter goes on, after Pacific air likely floods a good portion of the CONUS in December.

In general, basin wide El Nino winters favor a negative NAO as you head farther into winter, although not as strongly as modoki events. The three moderate to strong basin wide El Nino events since 1950 that had a +QBO as we’ll have this winter featured decent AO/NAO blocking by February, after a warm December and up and down January. Research done on October snow advance in Eurasia by Dr. Judah Cohen and on October N. Pacific sea level pressure by Al Marinaro both suggest that trends this month also support a –AO/NAO this winter.

In general, a rising +PNA could limit California/west coast rainfall by later in winter. Hopefully they’re very active before then, as they desperately need the rain. The eastern US looks cold/active for the second half of winter, after a potentially very slow start.

The potential bust factor here revolves around how the Nino behaves. Other basin wide El Nino events with such a warm Indian Ocean (1987-88, 2002-03) saw more MJO activity than normal for stronger EL Nino winters, which likely contributed to a favorable pattern for eastern US snow. If this event becomes more strongly east based, this may not be able to occur. Another overall question mark is if all of the correlations discussed, small sample size for some of them notwithstanding, hold up with such a strong El Nino. We may learn quite a bit this winter…hopefully not at the expense of this forecast.

Analogs, based on good SST matches globally and QBO:

1957-58, 1987-88 (2), 2002-03, 2006-07