Winter 2021-2022 Outlook

[40/70 Benchmark]

In like a lion, out like a lion.

https://easternmassweather.blogspot.com/2021/11/outlook-2021-2022-in-like-lion-out-like.html

 

[weathafella]

Thanks for your hard work on this Ray.   I think you’ll do well at least for this area.

[moneypitmike]

Thanks for this Ray--always great to see you presenting this, not just because of the thinking that goes into it but also for getting us excited that yes--winter is coming.

From that crystal ball, I should be making some plans to be up at Pit2.

[Ginx snewx]

Best of luck Ray. Great job as usual. 

[WxWatcher007]

Good stuff as always. 

[BuffaloWeather]

Great stuff Ray, always appreciate the work you put into these. 

[CoastalWx]

I need a few minutes to digest this, but looks well thought out. Good job as usual. 

[40/70 Benchmark]

I have tried to make it a bit less cumbersome over the past couple of years by just linking the basics for people who want to learn/refresh...its long enough as it is. A change I made this year was to really put more time into honing the monthly composites to better reflect my portrayal of the seasonal progression/evolution. In the past, they were too broadbrushed and didn't reflect my thoughts 100%, and I would have to communicate that in print. I was not nearly as rigid with respect to my analog composite seasons this year, especially with la nina being basin wide, mixed type. There are some SHITTY years in the composite at appropriate junctures, and likewise some epic ones at certain times. The weighting is applied to augment the strongest analogs.

The composites this year really reflect my thoughts well. That was a weakness before.

Enjoy guys.

[TugHillMatt]

@40/70 BenchmarkThanks for sharing! On your table above, do you mean "'21-22" for your column headings? 

[40/70 Benchmark]

27 minutes ago, TugHillMatt said:

@40/70 BenchmarkThanks for sharing! On your table above, do you mean "'21-22" for your column headings? 

Yes.

Good catch....fixed.

[MJO812]

Good luck and nice forecast 

[raindancewx]

It's amusing to see you come up with what I have, a month later, but with far more years used so the signal gets dulled out to almost nothing. You do have December different it looks like, but you've got Jan, Feb, almost identical to me, as well as the overall DJF look almost identical, just dulled out. Last year, you and Isotherm and Griteater all had similar looks to me, but with a very warm February swapped for the cold I had in the Plains....which was...eh.

[George001]

Great read, the fast start, mild period mid winter and big March you are forecasting is very similar to 2017-2018. That was a great winter.

[40/70 Benchmark]

59 minutes ago, raindancewx said:

It's amusing to see you come up with what I have, a month later, but with far more years used so the signal gets dulled out to almost nothing. You do have December different it looks like, but you've got Jan, Feb, almost identical to me, as well as the overall DJF look almost identical, just dulled out. Last year, you and Isotherm and Griteater all had similar looks to me, but with a very warm February swapped for the cold I had in the Plains....which was...eh.

Yea, missed the N plains last year, but did well otherwise.

The main reason I have that warm January and somewhat cooler Feb evolution this year is because alot of my analogs had a bit more PNA and less hostile arctic in February after the pattern really bottoms out in January. It's also due to the fact that I think the PV takes a pretty big hit early in Feb or maybe even the tail end of Jan. If it waits until March, or does not happen, then my second half will be off. As far as the mean DJF, I don't think there will be many voices of dissent with respect to that this fall...its essentially a la nina/E QBO look. As far as it being "dulled", it is like that for a reason. I don't see much support for exotically warm DJFM departures in the mid atl and NE....maybe the SE. I guess DJF maybe a bit easier to get a warmer composite since we both expect March to be cold. That maybe part of the reason why your map is +3 or greater and mine is +3 and under.

[40/70 Benchmark]

3 minutes ago, George001 said:

Great read, the fast start, mild period mid winter and big March you are forecasting is very similar to 2017-2018. That was a great winter.

That, along with 2000-2001 and 2005-2006 are my main analogs.

[OHweather]

Ray, that was a great read and forecast, thank you for taking the time to put all of that together! 

As you know I agree with most of your general ideas. I especially enjoyed how you looked a little closer at certain La Ninas and explored why they were milder or colder than other La Ninas...and based on how this one is leaning, it seemed to me that your research pointed to decent blocking potential? 

I also found it interesting that you found that eastern Pacific La Ninas, along with La Ninas in an easterly QBO (removing 11-12), tend to be colder later in the season than December, which kind of bucks the "typical" La Nina expectations. I am curious to see how this year plays out. We had a number of similarities in our analogs, and mine also bring back blocking for the end of the winter after relaxing for several weeks January into February. I do think December may be quite cold this year so it could be very hard to pull off a colder Feb than Dec this year, but that could at least indicate that we get a nice shot of cold and snow potential early, and that there's at least an argument for another shot late. 

I feel like a strat warming event, if one occurs, is a potential wrench. Like you, I'm skeptical of an early SSW (though there should be enough occasional disruptions to keep the PV from getting prohibitively strong). Typically when an SSW occurs  it's mild over the eastern CONUS when it happens, and gets cold a couple to a few weeks later, so an early SSW in December or early January could really change-up the December-January portion of the forecast (as we kind of saw last winter, with a mild period late Dec through mid-Jan as the SSW occurred, then a colder late Jan and Feb). However, I'm skeptical of an early SSW this year. But, if one occurs later in the season (such as 17-18) it could line up better with the expectation of a milder mid-winter period, then another crack at a colder / blockier pattern late. So, it will be interesting to see if we get a SSW that influences the pattern at some point. It's also possible the stratosphere is relatively more boring this year than in recent winters too. It seems like recent winters (since 17-18) have either had a true SSW event or have had a very strong PV, with nothing in between. 

[40/70 Benchmark]

@OHweatherYes, decent blocking potential in latter Novie and December, then again late in Feb into March. I feel while the December blocking is higher confidence, March has the greater ceiling. I don't expect anything epic in December and I am betting against an early seaso. SSW, though I agree the PV will take some shots.

[weatherwiz]

This is impeccable work, Ray! As I've said to you before, your ability to not only present your thoughts, but do so in a way that the reader can become educated is a true talent. I've always had a tremendous interest in long-range forecasting but over the past several-plus years I just haven't had time (or energy) to do much with it. I've read through the EP La Nina's portion of your outlook (so I have much more to go) and wanted to share some thoughts. 

I share the same view OHweather pointed out in which you took a closer look at why some Nina's were colder and why some were warmer. This is extremely critical (IMO) to the understanding of ENSO events and to the understanding of what may be in the driver seat each event. While a composite of all years will provide you insight of what may be "typical" or the "averaged outcome" each event has its own flavor and it's important to understand that. You explored this, analyzed this, and portrayed this beautifully. 

I also absolutely loved how you used the lines on the STA maps to illustrate where the core of the anomalies were located in relation to the ENSO regions to indicate where in the region the event was based...this is brilliant lol. I've been trying to find ways to illustrate this better for my own studies and never thought of this (I may have to steal that idea  ). 

There is certainly lots to digest here...tons to digest which is a beautiful thing because if there is anyone interested in not only long-range forecasting but understanding so many of the global drivers and teleconnections the best place to go if your outlook because you're going to learn tons...I just learned a tremendous amount myself. 

As always, great work!

[40/70 Benchmark]

https://easternmassweather.blogspot.com/2022/05/winter-2020-2021-seasonal-recap.html

Winter 2020-2021 Seasonal Recap & Forecast Verification; Mixed Results Abound

Pacific Timing Issues Plague the First Half and Misdiagnosed Arctic Nullifies Forecast Big Ending

Pacific Overruled the Arctic in December

 

While the overall tone of the 2021-2022 Eastern Mass Weather Outlook included a cold phase oriented Pacific, as evidenced by the primary extra tropical Pacific analogs of 2010-2011, 1970-1971, 1974-1975, 1955-1956, 1954-1955, 1964-1965, 2020-2021, 1988-1989, 2011-2012, 1998-1999, 1999-2000, 2007-2008, 2008-2009, 1975-1976, it was perceived that there was support amongst the analogs for a more variable PNA pattern during the early portion of the winter season. Thus the December portion of the forecast called for some variability within the PNA domain, which in conjunction with the forecast high latitude blocking would have ultimately led to an active month of winter weather culminating the week of Christmas. This proved to be a mistake. While a significant interlude of high latitude blocking did in fact verify, the Pacific volatility did not. In fact, December 2021 featured the most pronounced RNA pattern (-2.56) on record dating back to 1950. This was most similar to December 1955 (-2.07), December 1964 (-1.73), December 2008 (-1.41) and December 2010 (-1.78) from the extra pacific Pacific analog set. In hindsight, given the fact that the aforementioned composite contained the top 3 RNA months of December (1955, 1964, 2010) dating back to 1950, which included the top ENSO analog of 2010-2011, it was a glaring error to not forecast a very prevalent and static RNA in the mean for the month of December. 

There will be further exploration into the parallels to 1955-1956 later in this writing as it pertains to ENSO.

Here is the original narrative and forecast composite for the period:

December Analogs: 2007, 2000 (x2), 1974, 1970, 1995, 2005 (x2), 2017 (x2), 2008, 1983, 2020

"The polar vortex should take some assaults throughout the latter portion of November and much of December, however, it should remain intact and ultimately intensify and consolidate beyond the new year. This is consistent with basin-wide, mixed-type la nina events. While there may be a mild spell early in the month, it should ultimately turn active with a multitude of moderate snowfalls for the northeast, with the most significant event likely to occur the week of Christmas between the 19th and 26th. A white Christmas is highly likely for the majority of the region and a storm on the holiday itself is a distinct plausibility. Plan accordingly. Although the focus will be on the northeast via a mix of Miller B systems and southwest flow events whenever blocking relaxes, the mid atlantic should receive some snowfall due to a subtle subtropical influx coupled with a variable PNA.

The month should not be exceedingly frigid, as the very mild fall has ensured a late developing cold reservoir. Anywhere from near normal close to the coast to as much -2 departures well inland should cover the monthly departures".

Although the most prolific winter storm of the month throughout the northeast did in fact occur near the onset of the identified December 19 through December 26 period, it primarily impacted the northern half of the region, and was mainly rainfall throughout southern New England. More importantly, this particular event was the only noteworthy winter storm of note for the month in large part due to the development of the poorly forecast record RNA pattern. While the forecast episode of high latitude blocking in the NAO domain did in fact materialize as forecast by the holiday period, its development was concurrent with the formation of the strong ridge of high pressure to the south in response to the historically low heights along the west coast in conjunction with the extreme RNA pattern.

 

 This acted to create a very compressed flow between the southern ridge, and high latitude ridge in the vicinity of Greenland, which sheared off storm activity approaching from the active Pacific pattern.

 

Note in the annotation above that the two respective ridges of high pressure were situated at just the right distance from once another, at the appropriate latitude to attenuate the storm systems on approach to the Great Lakes. Although this evolution is similar to the year 1970 from the list of analog composite, which featured near record snows for the  month of December, in this instance the heights out west were just so low that heights to the east were elevated enough to be create said compression. Storm systems ejecting out of the PNW were sheared by the compressed flow as they neared the great lakes, so that what we were left with at the longitude of New England was a multitude of minor disturbances, instead of any larger, more discrete, congealed or phased system. This pattern acted to negate what was envisioned last fall as being a very active period with a multiple moderate snowfall events from the period leading up to the holidays and beyond. In turn, the higher heights downstream from the deep trough in the west shifted the core of the warm anomalies from the forecast position in the southwest, into the Texarkana region. It also caused temperatures in the east to be slightly warmer than anticipated.

 

Forecast:

 

Actual:

 

 

 

In summary, the most important aspect of monthly forecasting is correctly identifying periods of volatility from the seasonal base state. While this does not show up in the overall DM seasonal composite, it is key to correctly diagnosing the sequential evolution of a season. And while the Pacific pattern throughout the month of December was essentially a more amplified version of the overall DM forecast base state of the Pacific, it certainly was not the forecast period of variability and perhaps slight variation that was forecast. This essentially rendered the well forecast high latitude blocking a moot point, and the forecast sensible appeal did not materialize due to the severe and static RNA throughout the month that was flagged by the extratropical Pacific Composite.

 

January 2022 Colder & Snowier Than Forecast 

 

The most notable issue with respect to the first half of the seasonal forecast was that the fact that the timing of the forecast evolution was rushed by approximately two weeks due to the misdiagnosis of the PNA period as occurring during the month of December,  rather than January. Thus while the first half of winter was fairly well forecast in the aggregate, the monthlies left something to be desired in that December was milder with much less snowfall than forecast and January more wintery than anticipated.

Timing is everything, indeed. And given that the primary ENSO analog was in fact in 2010-2011, the severe December RNA and PNA driven severe winter stretch in January should have been more accurately forecast with respect to the monthly progression.

 

Given the fact that the PNA pattern did evolve during the month of January as it did in the analog year of 2011, the forecast harsh period ultimately evolved into more of a "delayed but not denied" element of the winter outlook. As evidenced by the excerpt pertaining to the January portion of the outlook released on November 6, 2021, the month was expected to begin cod and snowy, before transitioning to a mild, Pacific jet dominated look around mid month.

 

January 2022 Outlook

 

January Analogs: 1955, 1956, 1971, 1996, 2006 (x3), 2001, 2018, 2012, 1989, 1965, 1976

 

There is a slight chance of a KU magnitude, Archambault type east coast snowfall between approximately December 28th and January 11th, as any residual high latitude blocking erodes and heights lower in the vicinity of Alaska and western Canada. Heights will also lower in the western US as the aleutian ridge retrogrades into the pacific and a sustained RNA accompanies the development of an active, zonal pacific storm track. The core of the warmth should center in the northern plains as the pacific jet eradicates the supply of cold from much of the continent, however, the degree of warmth in the northeast should not be prohibitive of some wintery weather during this peak climo period. Although mixed precipitation and southwest flow events should be the rule, and major pure snow events will be difficult to achieve outside of the higher terrain of upstate New York and central/northern New England. The forecast is for monthly departures to range anywhere from +1 to +3 degrees, but the ceiling is quite high and the potential is there for several degree positive departures. This has the potential to bean exceptionally mild month akin to February 2018.

 

Note that the powerful bomb cyclone that became known as the Blizzard of '22, similar to the one that occurred during the analog season of 2017-2018 on 1/4/2018, took place roughly two weeks beyond the 12/16 to 1/11 favored window of time, during which the January 11, 2011 blizzard also took place. Additionally, Boston, MA recorded its 8th snowiest month and 4th snowiest month of January on record with 36.2",  just 2.1" behind the month of January 2011, which occurred during the primary ENSO analog season. Thus while the first half of winter 2021-2022 largely paralleled the overall tenor of the primary analog seasons, the timing in relation to the forecast sequence of events left much to be desired due to the failure to forecast the PNA period correctly. This discrepancy between the forecast timing of the pattern evolution and what actually transpired dictated that the second half of January did not turn mild, and instead featured the climax of the harsh period with several bouts of arctic air masses, including sub zero temperatures, as well as the blizzard. The month of January clearly finished colder than forecast, just as the month December was milder than forecast because the Pacific patterns were reversed.

 

 

The month finished approximately 1-2 degrees below average across the region, which was about two degrees colder than the forecast range. In summary, while the DJ first half of the boreal winter season was fairly accurately forecast accurately forecast in the aggregate, the monthly evolution with respect to the timing sequence was relatively poor due to the +PNA interlude occurring during the month of January, instead of December, as forecast. 

The month of February, on the other hand, was by far the best forecast from a sensible weather stand point.

February 2022 Well Forecast at the Surface; Poorly Forecast Stratosphere

 

Here is a review of the February 2022 portion of the winter outlook that was originally issued early last November.

Forecast Review:

 

 

February 2022 Outlook

February Analogs: 1955, 1956, 1971, 1996, 2006 (x2), 2001 (x3), 2018, 2012, 1965, 1975, 1972, 2008, 2011,1985

 

"The month of February is the most difficult portion of the forecast because there will be a great deal of flux. Not only between the potential SSW that may parallel that which was observed on February 13, 2018:

But also a PNA that is more likely to fluctuate from the more stagnant RNA of January, as la nina begins to decay and loosen its grip on the pacific. Here is an extended range forecast for the polar vortex from the European model courtesy of meteorologist Giacomo Masato:

Note that guidance implies a recovery to a strong vortex in January, as per the Eastern Mass Weather forecast, however, the analogs suggest that the disruption could even begin to take place in later January. This is crucial to the forecast because if it does in fact delay until March, then blocking will not redevelop until spring and the second half of the season could be mild.

Nonetheless, the month of February should begin very mild, but winter is likely to gradually return as the Pacific also becomes a bit less hostile and the Aleutian ridge that had retrograded last month begins to edge back closer to the west coast and a bit more poleward towards Alaska. This will allow more proficient cold delivery through the plains to bleed into the northeast, leading to an active pattern of overrunning, SWFEs and Miller B cyclogenesis.

The month should average anywhere from near normal to 2 degrees above average. Warmest near the east coast and the mid atlantic, coldest north and well inland. But again, the risk is for a much warmer outcome.

The primary risk during February is that if a major disruption of the PV does NOT occur, then the month may remain fairly mild".

 

 

The Results:

 

In hindsight, this was a very accurate forecast overall in terms of sensible weather, as the region did indeed finish slightly above average for the month, and the primary modes of cyclogenesis were in fact overrunning, SWFEs and Miller B events. 

 

Forecast Precipitation:

 

Versus Actual:

 

 

 

Actual Temps: 

 

 

Versus Forecast Temp Departures
 

 

Forecast H5 anomalies:

The poleward ridging in the Pacific that is so reminiscent of the eastern biased la nina composite was very well forecast. 

 

This did indeed allow for the delivery of arctic air throughout the east at times, however, note the absence of high latitude blocking

 

 Versus the forecast analog composite:

 

The primary reason for this discrepancy with respect to the polar domain is the failure of the major SSW predicted to occur between approximately January 23 and February 11. 

 

While there is no way to know for certain why the negative  Quasi Biennial Oscillation (QBO)  and eastern biased cool ENSO couplet failed to result in a SSW as forecast, one potential reason is the sharper than anticipated increase in solar activity via a  solar flare in the fall of 2021. This acted to effectively end the solar minimum and initiate solar cycle 25.  

 

 

 

 

Many papers (such as this one: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2005JD006283) have attempted to establish a link between the solar cycle and stratospheric blocking patterns during the northern hemisphere winter season. In general, years near and shortly following a solar minimum tend to have a weaker stratospheric polar vortex and are more prone to blocking, with the opposite true in solar maxima winters. Although an increase in activity was noted last fall, the forecast rationale was that it was unlikely have a major impact due to a potential atmospheric lag in conjunction with the depth of the previous solar minimum. However, this was cited as a potential forecast risk with respect to the second half of the winter that would appear to have come to fruition with greater than 55 sunspots likely to be observed between June 2021 and July 2022.
 

Ultimately this did not impact the February monthly temperature departure as theorized in the winter outlook last fall because the more hostile polar fields were offset by a slightly more favorable Pacific than advertised, thus the forecast was still successful. And the warmer risks mentioned as a result of SSW failure did not materialize. However, the failure of the SSW was of a great detriment to the March portion of the forecast.

 

Overall Poor March Verification Due to Misdiagnosis of Polar Stratosphere

 

The March portion of the winter outlook released this past fall was predicated upon a major disruption of the the polar vortex via a February Sudden Stratospheric Warming (SSW) event, which obvious did not happen. While the forecast was for monthly temperature departures across the region anywhere from 1 to as much as 3 degrees below average with above average snowfall, it was explicitly stipulated that if the aforementioned SSW did not occur, than the risk was significantly warmer with less snowfall.

This is precisely what transpired.

 

Note thE pronounced and distinct lack of blocking for the vast majority of the month, as evidenced the the NAO/AO charts depicting largely positive values of each below.

 

 

There was a slight disruption of the polar vortex during the latter third of the month, which led to a fleeting period of warming throughout the polar stratosphere. 

 

It is this period that corresponded with the window of slightly greater polar heights and colder air across the forecast area due to a bifurcation of the vortex in the lower levels of the stratosphere, near the tropopause.

 

 

However, this did not represent a major SSW, nor did it result in any significant snowfall to speak of across the area. despite the colder temperatures.

This is again theorized to be at least in part due to the robust onset to solar cycle 25.

 

March 2022

 

Here is how the month actually evolved:

Actual temperature anomalies were anywhere from 1-3 degrees above average, as compared to the forecast, which called for monthly departures of  1-3 degrees below average. 

 

The result was not only warmer weather than forecast for the month of March, but also less precipitation, as the primary storm track remained inland due to the prevalent southeast ridge as a result of the dearth of high latitude blocking.

Thus a predominate storm track through the Great Lakes, as opposed to the coastal NE or offshore meant both less precipitation and a larger percentage of said precipitation falling as rain. This equates to below average snowfall and was obviously in contrast to the March forecast from this past November.

 

March Analogs: 1956 (x2), 1984, 1996, 2006 (x2), 2001 (x3), 2018 (x2), 1975, 2011

 

The first half of March should be blocky with an improving Pacific. The month should feature both Miller A & B cyclogenesis, as there are signs that the subtropical jet may enter the scene in addition to the active northern stream. This could be a very volatile set up with immense storm potential should the two streams phase. There is a relatively strong chance of a KU magnitude event between March 1st and 15th. The failure risk is that the SSW does not materialize, which could alter the forecast dramatically.

Monthly departures anywhere from 1-3 degrees below average with the coldest across the interior and above average snowfall from the mid atlantic into New England. This should be a very active east coast pattern rivaling that of March 2018 and 2001, though with perhaps more southern stream involvement.

The pattern may rapidly break down later in the month with an early spring to reward beleaguered residents of the northeast if the vortex does indeed weaken a bit earlier than the Euro guidance suggests, as forecast by Eastern Mass Weather.

This March forecast, again, is contingent upon the development of high latitude blocking, and the risk is for a warmer outcome should that not occur. or be delayed, as the European guidance hints at.

 

 

 

 

 

While ultimately long term seasonal guidance such as the European was correct in forecasting a dearth of high latitude blocking during the 2021-2022 winter season, conceivably due to the faster than anticipated onset of solar cycle 25, this cited forecast risk was primarily only a factor with respect to sensible weather during the month of March. This is likely why winter ended much milder and benign than anticipated last fall.

 

December-March 2021-2022  Composite Verification & Summarization

Here is the H5 forecast composite for the December through March 2021:

 

 

 

 

Versus reality:

 

 

Its clear why December was so much milder than expected, as the immense RNA materialized then, instead of January, as forecast. And the major NAO block that developed was rather ephemeral and frankly overwhelmed by the magnitude of the record RNA block to the west.

There were two primary weaknesses of the outlook issued from last fall, which served to negate one another to a degree and drove the pattern for the final 3/4 of the forecast period post New Year.

1) As previously discussed, the sun became more active faster than originally anticipated, which ultimately served to augment the polar vortex more than forecast. This is why the latter portion of the season, namely March, ended up so much milder than anticipated.  But the question of why January and February did not remains.

2) Note that in the juxtaposition of the DM forecast composite and the verification composite above, the Aleutian ridge was more pronounced in reality than forecast. Additionally, the polar vortex is also stretched and elongated onto our side of the globe, which also aided in the delivery of cold air.

This is what produced some rather harsh periods of winter weather throughout the eastern US in especially January, but even into March, despite a fairly robust polar vortex.

 

However, as the season aged and the month of March ensued, the cross polar flow abated and the active end of winter that was forecast across the east failed to materialize in the absence of high latitude blocking that was presumably attributed to a sharp increase in solar activity. 

 

 

Thus winter truncated in the east, despite the easterly QBO working in concert with what ultimately verified as a very significant and solidly eastern biased la nina. In fact, this was the most intense and eastern centered la nina since 1955-1956.

 

Poleward Aleutian Ridging a Function of East-Based La Nina

La Nina was forecast by Eastern Mass Weather to peak near the threshold of weak-moderate within an anomaly range of -.9C to -1.1C during the NDJ tri-monthly period. This portion of the forecast was exactly correct, as it achieved a peak ONI value of -1.0 during the NDJ period. 

However, it was measured at approximately -.8 by JAMSTEC on the el nino modoki index (EMI), which with the term "modoki" referring to west based, means that this particular event was very solidly east-based.

 

The forecast was for it to register between -.2 and -.4 on the EMI scale, or slighly east of neutral, making it a hybrid event similar to this past 2020-2021 and the 2010-2011 seasons, respectively. 

 

Here are these two seasons as compared how this past event materialized:

 

2010-2011:

 

 

2020-2021:

 

2021-2022:

 

Note that how in contrast to the two aforementioned hybrid seasons, just about all of the significant cool anomalies are relegated to east of 150 degrees longitude and hug the coast of Peru.

This is much more reminiscent of how our east-based composite looks.

 

Here is the synoptic forcing scheme associated with the above east-based la nina composite:

 

 

Considering the close similarity to the east-based la nina composite at the surface in terms of placement of max SST anomalies, it would be a reasonable inference to assume that the forcing from this past la nina was redolent of the above east-based composite:

 

Indeed was the case. Note that the warmer anomalies near the dateline ensure that forcing remains over the central and western Pacific, similar to the weaker la nina composite. Sinking air that discourages convective forcing is focused well east of the dateline. Also evident is that the Aleutian ridge focuses more to the northwest relative to the modoki, cp event. This often entails a protrusion of said ridge into the polar region at times, which likely contributes to the lower heights over the mid latitudes:

 

 

This explains the significant expansion of the Aleutian ridge into the polar domain this past season, which countermanded the stronger than forecast polar vortex to a degree, thus salvaging some semblance of a boreal winter across the eastern US.

 

 

Note that of the eastern-based la nina data set, the most intense cool-ENSO event of the composite, 1955-1956, was comparable to this past event in terms of both intensity and structure, thus it is prudent to conclude that this season should have been at least the primary ENSO analog and quite possibly a primary general analog. This harkens back to the major oversight alluded to at the onset of this writing. That being not having weighed heavily enough into the forecast the fact that the December forecast composite contained the most severely RNA Decembers on record, which included the aforementioned 1955-1956. This season was not only the superior ENSO analog in hindsight, but also featured the most prominent December RNA on record until this past December. Additionally, it also would have allowed more insight to be gleaned into the fact that this past la nina would very biased fairly heavily to the east.

 

Here is how the national forecast temperature composite verified:

 

 

Vs reality:

 

 

 

DM mean temperature anomalies were generally near normal to slightly above average across the northeast, as forecast.

 

Finally, the precipitation pattern verified as forecast, although the anomalies were slightly more extreme and the mean storm track slightly further inland than forecast due to the absence of polar blocking.

 

 In summary, Eastern Mass Weather correctly identified that winter 2021-2022 would have interludes of rather severe winter weather due to the spasmodic, poleward expansion of the Aleutian ridge that induces cross polar flow and is inherent of an east-based la nina. However, Eastern Mass Weather poorly forecast the state of the polar stratosphere and thus the strength of the polar vortex, presumably attributable to a failure to appreciate of how quickly and fervently solar cycle 25 began this past fall and into the boreal winter season. The end result was a fairly well forecast winter from a sensible weather appeal that ended sooner than anticipated across the eastern US.

Here is a closer examination of how the teleconnection and snowfall prognostications fared.

December-March Teleconnection Forecast

 

 

The verification of the mean teleconnection values for the DM were congruent with the overall synoptic verification laid out above. Clearly Eastern Mass Weather underestimated the mean DM negative value of the PDO by a great deal (biased .82 positive) this past season, however, the seasonal PNA value was forecast relatively accurately (actually biased -06 negative) despite the early season timing issue.

 

 

 

 This validates the position that the first half (DJ) of the winter season was forecast fairly well in the aggregate, despite mistakenly having forecast the extreme period of RNA in January, as opposed to December. This timing evolution was on par with the 2010-2011 and 1955-1956 analogs. 

 

Index Value	Predicted '21-'22 DM   Value Range	Actual  '21-'22 DM Value	Forecast Error
Pacific Decadal Oscillation (PDO)	-1.00 to -1.30	-2.12	Biased .82 Positive
Perennial North American Pattern (PNA)	-.25 to -.55	-.19	Biased .06 Negative
ENSO	NDJ -0.9 to -1.1 EMI: -.2 to -.4 (slightly east-hybrid)	NDJ -1.0 EMI: -.8  ( moderately east-based)	Timing & Intensity of ENSO Peak Verified. However, forecast was biased .4 West-based.
(J-M) East Pacific Oscillation (EPO)	+.45 to +.75	+.01	Biased +.44 Positive
Arctic Oscillation (AO)	-.45 to -.75	+.72	Biased Negative 1.17
North Atlantic Oscillation (NAO)	-.10 to -.40	+.96	Biased Negative 1.06

ENSO was forecast with absolute precision as it pertains to the peak ONI value of -1.0 during the NDJ tri-monthly period. But it was forecast to be a hybrid season with an eastern tilt, registering an EMI value of between -.2 and -.4 during its peak. The fact that la nina ended up being firmly east-based (-.8 EMI) led to the forecast poleward protruding Aleutian ridging being even more prominent than forecast, especially during the month of January, which is evidenced by the fact that the DM mean EPO value (+.01) verified as essentially neutral instead of the solidly positive forecast range of +.45 to +.75. This helped to counteract the fact that a more rapid start to solar cycle 25 prevented much in the way of blocking in the polar domain, as forecast. Thus despite the DM mean AO/NAO values registering as solidly positive (+1.17 & +1.06, respectively), in contrast to the forecast -.45 to -.75 & -.10 to -.40, respectively), the overall forecast synoptic appeal was fairly well forecast with the notable exception of the month of March.

 

 

 

 That being said, the snowfall forecast, which is inherently subject to a high degree of variance, anyway, was certainly flawed across parts of the region.

December-March Snowfall Outlook Review

 

 

City	Predicted Snowfall	Actual	Forecast Error
Boston, MA	45-55"	54"	Verified
NewYork, NY(Central Park)	25-35"	17.9"	Snowfall Overforecast by 28.4%
Philadelphia, PA	17-27"	12.9"	Snowfall Overforecast by 24.1%
Baltimore, MD	10-20"	14.4"	Verified
Washington, DC	7-17"	12.3"	Verified
Albany, NY	65-75"	36.4"	Snowfall Overforecast by 44%
Hartford, CT	45-55"	31.5"	Snowfall Overforecast by 30%
Providence, RI	30-40"	43.5"	Snowfall Underforecast by 8.75%
Worcester, MA	80-90"	53.6"	Snowfall Overforecast by 33%
Tolland, CT	55-65"	50.6"	Snowfall Overforecast by 8%
Methuen, MA	70-80"	42.5"	Snowfall Overforecast by 47%
Hyannis, MA	15-25"	10"	Snowfall Overforecast by 33%
Burlington, VT	85-95"	70.7"	Snowfall Overforecast by 14.3%
Portland, ME	80-90"	44.1"	Snowfall Overforecast by 44.9%
Concord, NH	75-85"	44.6"	Snowfall Overforecast by 40.5%

 

 

The mean snowfall forecasting error for the 2021-2022 winter season throughout the 15 select mid atlantic and northeast locations was 25.4%. This ranked 4th amongst the eight seasons forecast. The seasonal forecast ranges verified in the cities of Boston, MA, Baltimore, MD & Washington, DC.

Snowfall was generally over forecast between 24.1% (Philadelphia, PA) and 47% (Methuen, MA) across the majority of the forecast area due in large part to the failure of the late season high latitude blocking to materialize. However, there was a narrow corridor in the vicinity of I-95 and I-84 in the mid atlantic and southern New England, where snowfall was fairly accurately forecast and even under forecast by 8.75% in Providence, RI. This is due to the fact that the aforementioned area proficiently accumulated snowfall during the favorable period of poleward Aleutian ridging during January and February, where as points slightly further inland experienced some close misses and lighter snowfalls. This was most notable with respect to the late January blizzard in southern New England, in which more high  latitude blocking likely would have led to faster and more proficient phasing and a wider expanse of blizzard conditions. Southeast of the narrow region of well forecast seasonal snowfall, across the cape and islands, snowfall fell short of the  forecast by 30% in Hyannis, MA due to mixed precipitation. Snowfall outside of the narrow I-84 and I-95 corridor was never able to reach the forecast totals because later in the season the high latitude blocking never materialized a a result of the more active sun. Snowfall totals across far northwestern New England, in Burlington, VT, ended up a mere 14.3% (70.7") under the forecast range of 85-95" due to the fact that blocking is less of a factor for snowfall in this region due to its distant interior location. 

 

The mean forecasting error with respect to the eight seasonal snowfall outlooks are as follows:

2021-2022: 25.4%

2020-2021: 11.7%

2019-2020: 678.3%

2018-2019: 90.4%

2017-2018: 15.6%

2016-2017: 36%

2015-2016: 71.4% 

2014-2015: 13.7%

Eight season running mean error: 117.8%

29/117 (24.8%) of snowfall outlooks have been under forecast.

61/117 (52.1%) of snowfall outlooks have been over forecast.

26/117 (22.2%) of snowfall outlooks have verified within the forecast range.

Below is the verification for ENSO, as well as the various teleconnections over the past several years:

 

Index Value	Seasons Verified	Mean Forecast Error	Forecast Bias
Pacific Decadal Oscillation (PDO)	0/6 Seasons	.66	Biased Negative 3/6 Seasons
Perennial North American Pattern (PNA)	1/5 Seasons	.15	Biased Negative 4/6 Seasons
ENSO	7/8	.01 too cold	Biased cold 1/7 Seasons
(J-M) East Pacific Oscillation (EPO)	3/6 Seasons	.39	Biased Positive 2/3 Seasons
Arctic Oscillation (AO)	0/6 Seasons	1.20	Biased Negative 5/6 Seasons
North Atlantic Oscillation (NAO)	0/6 Seasons	.95	Biased Negative 5/6 Seasons

 

   

Final Grade for 2021-2022 Season: C-

 

[StormchaserChuck!]

Nice job.. I thought early that -NAO was establishing itself, veering off later in the Winter. The SW/West coast High pressure is really becoming a feature like Global warming, a little underestimated in your forecast. That High pressure is climbing right up the gravity of the Rocky Mountains all the time now.. something to consider going forward (if the drought busts, I'd expect a stronger High pressure after in duality- that sort of thing). Really looked like up to Jan 1st that it was going to be a -NAO Winter, if so you would have nailed it at 85-90%. I give you a B+. No A- because ENSO is easy  . -PNA in Dec (630dm vs 370 #2 or something) is always followed by +NAO Jan and Feb at a +70dm-100dm signal as an average of 20/75 total analogs.. ridiculous strong. So there is no real contesting that after the record -PNA December, and it could be something to remember forward if it happens again. Be careful about places like Portland and Seattle going forward.. it will be hard for them to go below average believe it or not. 

[40/70 Benchmark]

On 6/12/2022 at 2:14 AM, StormchaserChuck! said:

Nice job.. I thought early that -NAO was establishing itself, veering off later in the Winter. The SW/West coast High pressure is really becoming a feature like Global warming, a little underestimated in your forecast. That High pressure is climbing right up the gravity of the Rocky Mountains all the time now.. something to consider going forward (if the drought busts, I'd expect a stronger High pressure after in duality- that sort of thing). Really looked like up to Jan 1st that it was going to be a -NAO Winter, if so you would have nailed it at 85-90%. I give you a B+. No A- because ENSO is easy  . -PNA in Dec (630dm vs 370 #2 or something) is always followed by +NAO Jan and Feb at a +70dm-100dm signal as an average of 20/75 total analogs.. ridiculous strong. So there is no real contesting that after the record -PNA December, and it could be something to remember forward if it happens again. Be careful about places like Portland and Seattle going forward.. it will be hard for them to go below average believe it or not. 

Yea, I thought the Pacific would end up a hair less favorable, and the Atlantic/polar domain a bit better...in the end, not bad. I def. had the idea of poleward Aleutian ridging owed to east biased la nina. Interesting regarding the correlation....that is one area that I am weak on. I don't really incorporate those, aside from ENSO forecasting. Those correlations are something that I see you and raindance use alot.

[StormchaserChuck!]

Yeah it was cool to see that ridiculous -PNA December and -NAO, then the analogs showed a -70dm 500mb anomaly over +NAO for Jan and Feb both combined 20/75 analogs! It happened. Blew the trend at the time out. Would like to see more happenings to really span it out, it's probably planetary or something.