Winter Outlook 2024-2025

[40/70 Benchmark]

I'll go through it all in great detail later this spring, but a bit of regression here in terms of my snowfall forecast....I have had some decent forecasts overall, namely 2022-2023, just not work out in terms of snowfall...but this year largely worked out, despite a pretty flawed forecast. I'm glad I was too warm, as opposed to cold because there was a definite bias in my forecasts.

I do not anticipate much more activity.

[40/70 Benchmark]

 

February Review & March 2025 Preview

Nature of Polar Vortex Split a Large Wrinkle for February Forecast

February Review

Here is the narrative forecast for February that was issued in November.

 

February Analogs: 2011, 2008, 2000, 1999,1972

The polar vortex should begin the month very strong, but watch for a potential SSW from about mid month onward.

The pace of moderate storms should increase for the first time all season, just as the warmer weather returns in February, however, a notable difference from many recent months of February is that it should not be prohibitively warm so as to entirely preclude some appreciable snowfall across the majority of the region. The month should finish 1 to 3F above average with near normal to perhaps just below normal snowfall. Northern New England should see above normal snowfall as the cold air source lurks close by in southeastern Canada, which will produce several front-end snowfall for much of southern New England and perhaps even into the northern mid Atlantic at times.

"SWFE"....AKA "Southwest Flow Events-

The primary source of deviation from the forecast this month was that a split of the polar vortex (PV) did indeed occur mid month, which represented the beginning of the window identified in the Eastern Mass Weather winter outlook.

 

 

However, this disruption of the PV initiated in the troposphere and then worked upwards. This obviously allowed the impact in terms of a severely negative AO to be realized immediately, whereas the forecast presumed the impetus for the disruption to be a sudden stratospheric warming (SSW), which would then take approximately 2-4 weeks to propagate down into the troposphere. 

 

 

Obviously the mean AO value ended up much lower for February than it would have had the PV disruption been triggered by stratospheric warming that would have taken until at least the end of the month to affect the lower levels.

It's clear in the 500mb plot how this altered the regime during the month of February across the continent relative to the forecast.

 

 

This faster than anticipated development of the AO blocking spilled over into the EPO domain in Alaska and combined with an unexpected +PNA to dump very cold air into the nation's mid section.

The strongly positive PNA this month will challenge the analog season of February 2000 as the most positive on record for a La Niña and is likely to end up the strongest La Niña seasonal +PNA on record.

 

 

This is likely to have longer term ramifications beyond the scope of this writing, but the implication for the month of February 2025 was a break in a longer term extratropical Pacific trend that combined with a polar vortex disruption that was triggered in the lower levels to produce a colder than forecast outcome.

 

 

 

The departure from the forecast is not as drastic locally as it was nationally, likely due in part to the well forecast +NAO allowing the cold to focus west of the region, as it often has over the course of the last several years. 

 

 

The +1-3 degree F regional forecast anomaly is likely to range instead from near normal to  -2 degrees F. Given that precipitation amounts are likely to finish near average, as expected, this has resulted in slightly above average snowfall instead of normal to slightly below average. 

 

 

 

 

However, the predominately positive NAO this month has indeed resulted in several mixed precipitation events, or southwest flow events, as expected. The positive nature of the NAO also likely played a role in the lobe of the PV post split initially drifting westward, which allowed one of the larger storms to phase of the west. The primary question left before entering is spring is whether or not a larger polar vortex disruption is triggered by a warming of the stratosphere, as implied by research conducted last fall pertaining to westerly QBO/solar max season.

March 2025 Preview

The focus of the March forecast last fall was whether or not there would be a Sudden Stratospheric Warming (SSW) that would induce a major disruption of the PV.

 

March Analogs: 2023x2, 2022, 2021, 2017 2014, 2001,1972

The polar vortex should weaken once again to an extent to close out the season, though confidence is not high in a major episode of blocking. The window for a SSW is through about the 20th of the month, so the impact on winter would largely depend on just how early it can take place. During February or at least the first week of March for maximum impact throughout southern New England. There is another window for a potential KU from approximately March 5 through the 19th, but confidence is low. The cold source should be relatively close by at the very least, not unlike February should major blocking not materialize. Near average to 2F above in the mean with near average precipitation inland and slightly below near the coast.  

 

The following is an excerpt from research cited pertaining to the increased likelihood for late season SSWs during westerly QBO seasons that occur near solar max.

 

 

 

 

 

These relationships between solar behavior and the NAO are evident in the graph above, with reds, denoting +NAO, very evident in the declining phase of the last several solar cycles. And Blue, indicative of -NAO, prevalent in the ascending portion of the cycles. What can also be deduced from the graphic above is that while solar max seasons are not as favorable as the ascending phase of the cycle for incidences of high latitude blocking, nor are they as hostile as the descending seasons. Thus winter seasons such as 2024-2025, which are near solar max, are not entirely devoid of blocking, however, nor is this season as prone to an intense round of late season blocking the magnitude of March 2023 given that the solar cycle was still ascending at that point. Be that as it may, solar max seasons are not entirely hostile to incidences of a disturbed polar vortex and this is evident when considering the best solar analogs of 1970 and 1999.

 

There was a split of the polar vortex on January 17, 1971, a displacement on March 20, 1971 and a displacement on March 20, 2000. While there was an easterly QBO evident during the 1970-1971 winter, the polar vortex displacement that took place in March of 2000 occurred during a westerly QBO, as will be the case this season. Thus the latter displacement seems worthy of more consideration for the coming season. When considering the three primary QBO analogs of 2016, 2020 and 2022 within a solar context, although none took place during solar max, 2022 was the closest, followed by 2016 and 2020, the latter of which is a poor solar match having taken place near solar minimum.

 

Given that the QBO analog of 2020-2021 took place near solar minimum, the early January 2021 SSW  is of least relevance as a viable analog occurrence this season. However, the February 2023 SSW, which is a better solar analog, lends more support to a later season potential polar vortex disruption along the lines of March 1971 and 2000. This notion is supported by research on high solar, westerly QBO seasons, which lends credence to the late winter/early spring displacement scenario.

 

 

 

 

The research clearly implies that a displacement of the PV triggered by a SSW beyond mid month is a distinct possibility based on the history of westerly QBO seasons that have occurred near solar max. There is some recent guidance that has aggressively indicated that a SSW would occur earlier than this, and yet other guidance suggests that 

 

 

 

 

While other guidance is more subdued and implies marginal conditions late in the month, which would like be of little consequence.

 

 

 

 

 

Regardless, any late SSW is unlikely to have a major impact on the nation in terms of notable winter weather during March, rather it would likely have implications for the weather pattern during the spring season. 

 

Here is a comparison of the March forecast composite from this past fall (left) with the latest CFS guidance composite (right).

 

 

And a similar temperature anomaly juxtaposition.

 

 

 

The risk to the forecast is likely a slighly warmer outcome and reduced likeliehood of a KU caliber winter storm owed to the expedited manifestation of last month's PV split impacting the February patternas , opposed to early March.

What is likely to be the final winter storm threat of the season for the region looks to take place in the vicinity of March 9. This signal is supported by teleconnector convergence in both the Pacific.

 

 

And the Atlantic side.

 

 

Bolstering confidence in the threat period is cross ensemble support.

 

 

The primary mitigating factor at this time would appear to be the positioning of the PNA ridge slightly too far to the east, which would perhaps favor an out to sea track and/or later development.

 

Thereafter the early indications are that the MJO may approach phase 3 and then the Maritime continent, which would effectively end winter 2024-2025 with a deep RNA derived warm burst.

 

 

 

 

 

[40/70 Benchmark]

 

Winter Coming to a Close Despite Expected Late Stratospheric Warming

Final Stratospheric Warming Too Late for Much Fanfare as Per Analogs

Analog Data of Great Utility

The EasternMassWeather March preview cited last fall's research of solar max/westerly QBO seasons as support for an early spring Sudden Stratospheric Warming (SSW) event that would occur too late to have major ramifications in terms of winter weather for the forecast area. The March 20, 2000 SSW and resultant displacement of the polar vortex was listed as a primary analog last fall and guidance has continued hone in on a very similar stratospheric evolution throughout the first half of the month of March. 

 

 

 

Accordingly, the resultant weather pattern over the arctic if forecast to respond in a similarly like manner to latter March 2000.

Late March Blocking Likely to Interrupt Pleasant Spring Onset

The weather pattern next week looks seasonably mild and relatively pleasant, as the early seeds of change are planted in the arctic stratosphere.

 

 

The evolution over the arctic stratosphere will begin to manifest itself into the 500mb pattern during the final week of the month, as the development of a negative AO and NAO will be signify the development of high latitude blocking in response to the ongoing stratospheric warming.

 

 

The similarity to the latter March 2000 blocking pattern to the long range modeled evolution is apparent.

 

 

However, the north Pacific pattern looks slightly more conductive to potentially wintry weather over the northern New England ski resorts, and some nuisance or trace amounts of snowfall over southern New England cannot be ruled out. Be that as it may, as has been the case throughout the winter season, the cold appears merely seasonable in severity, which is unlikely to be supportive of significant early spring snowfall across the forecast area by the end of the month and into early April.

 

[40/70 Benchmark]

 

 

March Slightly Warmer Than Forecast

 West Pac Changes Lead to Warmer Outcome Despite PV Disruption

The major assault on the polar vortex that begin just prior to the midway point of the month was very well forecast by Eastern Mass Weather.

The following is an excerpt from research cited pertaining to the increased likelihood for late season SSWs during westerly QBO seasons that occur near solar max.

 

 

 

 

 

These relationships between solar behavior and the NAO are evident in the graph above, with reds, denoting +NAO, very evident in the declining phase of the last several solar cycles. And Blue, indicative of -NAO, prevalent in the ascending portion of the cycles. What can also be deduced from the graphic above is that while solar max seasons are not as favorable as the ascending phase of the cycle for incidences of high latitude blocking, nor are they as hostile as the descending seasons. Thus winter seasons such as 2024-2025, which are near solar max, are not entirely devoid of blocking, however, nor is this season as prone to an intense round of late season blocking the magnitude of March 2023 given that the solar cycle was still ascending at that point. Be that as it may, solar max seasons are not entirely hostile to incidences of a disturbed polar vortex and this is evident when considering the best solar analogs of 1970 and 1999.

 

There was a split of the polar vortex on January 17, 1971, a displacement on March 20, 1971 and a displacement on March 20, 2000. While there was an easterly QBO evident during the 1970-1971 winter, the polar vortex displacement (PV) that took place in March of 2000 occurred during a westerly QBO, as will be the case this season. Thus the latter displacement seems worthy of more consideration for the coming season. When considering the three primary QBO analogs of 2016, 2020 and 2022 within a solar context, although none took place during solar max, 2022 was the closest, followed by 2016 and 2020, the latter of which is a poor solar match having taken place near solar minimum.

 

Given that the QBO analog of 2020-2021 took place near solar minimum, the early January 2021 SSW  is of least relevance as a viable analog occurrence this season. However, the February 2023 SSW, which is a better solar analog, lends more support to a later season potential polar vortex disruption along the lines of March 1971 and 2000. This notion is supported by research on high solar, westerly QBO seasons, which lends credence to the late winter/early spring displacement scenario.

 

 

 

 

The research clearly implies that a displacement of the PV triggered by a SSW beyond mid month is a distinct possibility based on the history of westerly QBO seasons that have occurred near solar max, and this is precisely what took place. Note the displacement of the PV in concert with the warming polar stratosphere that had already begun prior to mid month.

 

The event continued to evolve until a reversal of the zonal winds occurred around March 20th, as suggested by the analogs.

 

 

The research also suggested that any cold resulting from the PV disruption would be insufficient and/or too late to bias the monthly mean negative and that has also proven correct.

 

Warmer & Drier Than Average March With Very Little Snowfall

Eastern Mass Weather forecast anywhere from near normal top +2F positive mean monthly departures for the region, but this is likely to verify anywhere within a range of approximately +3 to +5F due in large part to the reversal of the prominent -WPO pattern this season.

 

 

 

The lower heights in the vicinity of Northeastern Siberia in the +WPO composite are very apparent.
 

 

This is consistent with the 500mb pattern throughout most of the month.

 

 

As well as the aforementioned analog season of 2000, which perhaps should have been weighted more heavily throughout the pattern half of the season.

 

 

 

It is fair to consider whether or not the behavior of the polar stratosphere had any impact on the evolution of the Pacific given the uncanny similarities. 

The drier than average pattern, especially near the coast, was well forecast and undoubtedly played a role in with southern New England saw very little snowfall throughout the month.

 

 

March Conclusions

The forecast for the month of March was not bad overall, however, its success was limited by an incorrect assessment of the WPO in conjunction with an underestimation of just how unfavorable the polar domain would be despite the major polar vortex disruption.

 

The PNA was not a prevalent factor for the majority of March, but its decided descent into strongly negative territory to close the month may assist in mitigating any potential impacts of the ongoing stratospheric warming for the region.

 

 

 

 

[40/70 Benchmark]

On 11/15/2024 at 11:22 AM, 40/70 Benchmark said:

Just realized I never threaded this in the NE subforum, which I usually do....I prefer to update in here, rather than the main forum.

 

https://easternmassweather.blogspot.com/2024/11/using-past-forecasting-difficulty-to.html

 

 

 

 

 

 

I'll go over the post-mortem in depth about a month from now, but looks like the snowfall worked out for the most part. The overall forecast def. had issues, though.

[40/70 Benchmark]

Looks like Worcester may have been my largest miss...followed by Concord, NH.

[40/70 Benchmark]

Just eye-balling, I think I landed in my range on 8/15 locales, which would be a new high-water mark....and desperately needed after the past couple of horror-shows. There is obviously luck involved with snowfall because its subject to so much variance...I don't think some of the "poor" forecasts were as bad as the snowfall made them appear, and this seasons's forecast wasn't as good as the snowfall made it appear...kind of like 2014-2015 in that a flawed forecast got "lucky" in that the snowfall was right for the wrong reason. It evens out in the end.

[40/70 Benchmark]

Winter 2024-2025 Recap: Slightly Warmer Than Forecast

Irregularities of Polar Vortex Made for Challenging Forecast

 December 2024 Marked By Hint of El Niño Pattern

Snowfall was not expected to be abundant during the month of December, so it comes as no surprise that it was in fact below normal across the vast majority of the region. However, that is about where the similarities between the December forecast and the actual sensible weather end. 

Here is a juxtaposition of the forecast temperature departures across the CONUS in degrees Fahrenheit for the month of December and the actual anomalies.

 

 

PNA Correlation to DM 500mb Heights

Cleary the forecast for the month of December went awry and any accuracy with respect to regional snowfall was largely a coincidence.  Temperatures ranged near normal, anywhere from about +1 to -1F versus the forecast of +3 to +5F. The expectation was for a predominately -PNA and a variable to perhaps modestly negative NAO in the mean, but what actually verified was an extremely positive PNA along with an strongly positive NAO.

 

 

AO Averaged Slightly Negative

The robustly positive monthly mean NAO of 1.21 for the month of December was indicative of a seasonal trend in that it represented a notable departure from the more volatile AO, which averaged modestly negative at -.34.

This regime represents a more protracted early to mid winter deviation from the prominent MC forcing of the modern era along the lines of the 2017 mismatch period considered in the Winter Outlook.

It was intimated by Eastern Mass Water in November that this was likely to occur, once it became apparent that there would be a delay in the development of La Niña that would ultimately lead the previously issued December portion of the seasonal forecast astray.

 

When La Niña is Away (Delayed) The Aleutian Low Will Play

The seeds of the December forecast demise were planted during the month of November, when a very pronounced westerly wind burst (WWB) throughout the tropical Pacific halted the development of La Niña, and allowed for an episode of Positive Global Absolute Angular Momentum that was more protracted than expected. 

 

 

Since positive GLAAM is indicative of an atmosphere that is generally redolent of El Niño, and negative GLAAM La Niña, this likely contributed to the development of an Aleutian low in the December monthly mean as opposed to the Aleutian ridge evident in the forecast composite. This would become a theme throughout the majority of the season and is likely at least partially responsible for the persistent +PNA

 

 

This Aleutian low enabled the potent PNA ridge to sustain throughout the majority of the month, which of course promotes lower heights and thus colder temperatures downstream over the eastern US, despite the passage of the MJO through the Maritime Continent. However, the journey of the MJO through phases 5 and 6 for the vast majority of the month of December may have played a role in the maintenance of a predominately positive monthly NAO, which likely limited regional snowfall. 

 

The evolution of both the tropical and extra tropical Pacific during the month of December led many forecasters to abandon the notion La Niña. And while this may have proven to be a wise course of action per the official ONI criteria used for its designation, the atmospheric La Niña imprint remained very discernible throughout the balance of the season.

 

 Reemergence of La Niña Well Predicted Using Primary ENSO Analog of 2008

Clearly the reports of the failure of La Niña to materialize that ran rampant throughout the forecasting industry were premature, as was asserted by Eastern Mass Weather throughout the fall and into the early winter season. 

 

 

Indeed, the parallels drawn between the late-blooming cool ENSO event of 2008 and last year's modest La Niña in terms of both the subsurface, as well as the December easterly trade surge proved valid. Note the parallels in the late recovery of each event during the month of December.

                            Region

Date                   1.2          3               3.4             4

04DEC2024     0.1     -0.4     -0.3     -0.2
11DEC2024     -0.2    -0.4     -0.6     -0.3
18DEC2024     -0.2    -0.5     -0.8     -0.4

03DEC2008     -0.6    -0.4     -0.6     -0.8
10DEC2008     -0.3    -0.6     -0.8     -0.7
17DEC2008     -0.2    -0.7     -0.8     -0.8
24DEC2008     -0.3    -0.8     -1.0     -0.9

While the hedge towards an official La Niña designation per the Climate Prediction Center criterion of five consecutive ONI readings of -0.5 or lower ultimately proved to be incorrect, the atmosphere was indeed reflective of a cool ENSO event of very comparable intensity (and structure) to the primary ENSO analog of 2008. This is apparent when considering the peak DJ bi-monthly MEI value of -1.0 versus -1.1 in SO of 2008, which is likely attributable to why the monthly precipitation distribution was similar to both the Eastern Mass Weather forecast and weak La Niña  composites.

 

 

In fact, an argument can be made that this most recent La Niña event was actually slightly stronger than the 2008 analog when accounting for the degree to which the expression of the 2025 event was masked by climate change. This is precisely what the Relative Ocean Nino Index (RONI) attempts to account for. It is calculated in trimonthly periods like the ONI, and the DJF peak of -1.12 this past season as compared to the -.91 peak in DJF 2008-2009 implies that La Niña was actually slightly stronger when considering the gradient between the SSTs within the ENSO region and the adjacent Pacific waters, which have warmed considerably over the past 16 years.

January 2025 Review

The main premise for the January 2025 portion of the winter outlook was that a potent -WPO/+PNA couplet would fuel a January 2022 like mismatch from the Maritime Content forcing that has been so prevalent over the course of the last several years. 

 

 

This portion of the forecast was a partial success, as the MJO did indeed make it into the west-central Pacific to re-enforce the anticipated +PNA Pacific mismatch response during the month of January. 

 

However, the behavior of the MJO during the month of January more closely resembled the mismatch period of January 2018 in that it featured a strong MJO peak in phase 2, before ultimately traveling into the MC continent. In the 2022 analog, the MJO wave collapsed following a strong peak in phase 8, which represents a notable departure from the strong phase 2 peaks noted in both January 2025 and January 2018.

 

This deviation from the 2022 analog is presumably why the January 2025 mismatch period acted as a essentially a compromise between the preferred extreme -WPO/+PNA driven January 2022 analog, and the modestly +WPO/+PNA driven January 2018 analog.

 

Amplified MJO Phase 2 Peak in January 2025 Triggered Strong +WPO Response Instead of Expected Strong -WPO/+PNA Couplet

 

MJO Wave Amplifying in Phase 8 Before Dying Allowed Strong -WPO to Persist with +PNA In January 2022

It is apparent that there are notable similarities between the temperature anomaly distribution in January 2025 and January 2018 that are consistent with a +PNA pattern, thus it is the +PNA that can be identified as the primary driver in both instances.

 

 

 

 

 

This is also very apparent in the contrast between the January forecast composite and the actual January 2025 temperature departures across the southeastern US.

 

The very pronounced discrepancy across the southeastern third of the nation is as expected given how prominent of a driver the +PNA proved to be even in the absence of the forecast -WPO. The forecast was more accurate further to the north, across the forecast area of New England given the lower correlation to the PNA. The forecast for near normal to +2F was slightly too warm, with readings averaging anywhere from +1 to -1F. The only two areas across the country that did not verify at least somewhat colder than forecast were portions of the west coast, obviously attributable to the aforementioned error regarding the PNA, and northern New England. The latter is due to some forecasting error with  respect to the polar domain.

Nuances of the Polar Vortex that Contributed to January Forecasting Error

January 2025 was forecast to feature a very strong polar vortex (PV) and consequently, a +AO and NAO in the mean, much like January 2022 (.85, 1.08 mean monthly AO/NAO). However, as was the case with respect to the PNA, reality strayed from this concept somewhat. Here is the polar vortex composite from January 2022, with the stratosphere pictured on top, and the tropospheric PV on the bottom.

 

 

Now here is the PV composite from January 2025 for comparison.

 

 

The vortex is of considerable intensity and reasonably well centered over the pole, but note that it is neither perfectly aligned or coupled down through the troposphere, where the vortex is slightly weaker and more displaced away from the NAO domain. This allowed for high latitude blocking to prevail in this region, as evidenced by the fact that January was the only month of the DM period to register a -NAO in monthly mean ( -.52):

 

Here is the 500mb composite as compared to the forecast.

 

 

This clearly mitigated the forecasting error across especially far northern New England, where -NAO is not correlated with cold as it is further south.

 

 

The only extensive period of -NAO thoughout the season actually contributed to less snowfall than anticipated, as it teamed with the persistent -EPO/+PNA tandem to produce a very suppressed pattern that resulted in an even drier month than forecast.

 

 

The error was most pronounced within the NAO domain due to the poorly coupled PV, as illustrated in the previous annotation, which is why the volatile AO averaged essentially neutral in the monthly mean (-0.07), as opposed to the solidly negative NAO. 

 

 

Conflict between the AO and NAO due to irregularities regarding a stout PV would go on to become a seasonal theme; however, other Idiosyncrasies within the polar domain would dictate that the latter half of boreal winter would feature a return to predominantly positive NAO conditions despite a rather significant PV disruption.

Nature of Polar Vortex Split a Large Wrinkle for February Forecast

February Review

 

Here is the narrative forecast for February that was issued in November.

 

February Analogs: 2011, 2008, 2000, 1999,1972

The polar vortex should begin the month very strong, but watch for a potential SSW from about mid month onward.

The pace of moderate storms should increase for the first time all season, just as the warmer weather returns in February, however, a notable difference from many recent months of February is that it should not be prohibitively warm so as to entirely preclude some appreciable snowfall across the majority of the region. The month should finish 1 to 3F above average with near normal to perhaps just below normal snowfall. Northern New England should see above normal snowfall as the cold air source lurks close by in southeastern Canada, which will produce several front-end snowfalls for much of southern New England and perhaps even into the northern mid Atlantic at times.

"SWFE"....AKA "Southwest Flow Events-

The primary source of deviation from the forecast during the month of February was with regard to the origin of the mid month PV split, which represented the beginning of the window identified in the Eastern Mass Weather winter outlook.

 

 

This disruption of the PV initiated in the troposphere and then worked upwards. This obviously allowed the impact in terms of a severely negative AO to be realized immediately, whereas the forecast presumed the impetus for the disruption to be a sudden stratospheric warming (SSW), which would take approximately 2-4 weeks to propagate down into the troposphere, as was the case in 2018. This represented a significant point of inflection because it is the point at which winter 2024-2025 parted ways with the 2017-2018 analog.

 

 

Obviously the mean AO value ended up much lower for February (-.86) than it would have had the PV disruption been triggered by stratospheric warming that would have taken until at least the end of the month to affect the lower levels, as was the case in February 2018.

It's clear in the 500mb plot how this altered the regime during the month of February across the continent relative to the forecast.

 

 

This faster than anticipated development of the AO blocking spilled over into the EPO domain in Alaska and combined with an unexpected record monthly +PNA to dump very cold air into the nation's mid section.

 

The February 2025 mean PNA value of 1.41 eclipsed the previous record of 1.12 set in the year 2000, making it the most extreme mean +PNA value on record for the month of February during a La Niña Season. This may potentially have longer term ramifications beyond the scope of this writing, but the implication for the month of February 2025 was to protract the break from the longer term extratropical Pacific trend towards -PNA that combined with a polar vortex disruption from the bottom up to produce a colder than forecast outcome.

 

 

 

This bottom-up polar vortex disruption in conjunction with the strongly +PNA that is so uncharacteristic of cool ENSO months of February represented a hand-off of sorts between the heretofore exemplary 2018 analog and the 2000 analog that would provide the script for the final chapter of winter 2024-2025.

 

 

The departure from the forecast is not as drastic locally as it was nationally, likely due in part to the well forecast +NAO that is consistent witjh cool ENSO February climatology allowing the cold to focus west of the region, as has often been often been case over the course of last several years. 

 

In fact, the degree of warmth over the southeast is indicative of the fact that the +1.60 mean NAO value for the month of February was even higher than anticipated, potentially at least in part due to the translation of La Nina westward into more of a Modoki configuration.

 

 

The +1-3 degree F regional forecast anomaly was again slightly too warm, as temperatures across the forecast area were essentially near normal, ranging anywhere from -1 to +1 F.  This is the product of two opposing strong drivers in the potent -AO/+NAO essentially negating one another. The strong positive NAO allowed for a return to a more active pattern with seasonable cold in place, which triggered a sequence of wintry events. This consequently led to slightly above average snowfall relative to the expectation for normal to slightly below average snowfall.

 

 

 

The predominantly positive NAO during the month resulted in a southwest flow event character to the vast majority of storm systems, which was consistent with the forecast. As a consequence of the +NAO initially forcing the lobe of the PV post split to drift westward, not only was the bulk of the cold relegated to the high plains, but the signature storm of the month tracked to the west of the area. This limited the extent to which snowfall could exceed normal relative to if the NAO had been negative. The rest of the season behaved as expected once the impact from the polar vortex disruption abated, and the 2000 analog led the way in instructing how this westerly QBO/solar max season would conclude.

 

 West Pacific Changes Lead to Warmer Outcome Despite PV Disruption

March Review

There was potential for March to begin on a more wintry note prior to turning warmer had the February PV disruption been triggered by a more traditional SSW, which propagated down from the stratosphere. However, once it was clear that was not the case, there was little doubt that the final disruption would come with little fanfare. The major assault on the polar vortex that began just prior to the midway point of the month was very well forecast by Eastern Mass Weather.

The following is an excerpt from research cited pertaining to the increased likelihood for late season SSWs during westerly QBO seasons that occur near solar max.

 

 

 

 

These relationships between solar behavior and the NAO are evident in the graph above, with reds, denoting +NAO, very evident in the declining phase of the last several solar cycles. And Blue, indicative of -NAO, prevalent in the ascending portion of the cycles. What can also be deduced from the graphic above is that while solar max seasons are not as favorable as the ascending phase of the cycle for incidences of high latitude blocking, nor are they as hostile as the descending seasons. Thus winter seasons such as 2024-2025, which are near solar max, are not entirely devoid of blocking, however, nor is this season as prone to an intense round of late season blocking the magnitude of March 2023 given that the solar cycle was still ascending at that point. Be that as it may, solar max seasons are not entirely hostile to incidences of a disturbed polar vortex and this is evident when considering the best solar analogs of 1970-1971 and 1999-2000.

 

There was a split of the polar vortex on January 17, 1971, a displacement on March 20, 1971 and a displacement on March 20, 2000. While there was an easterly phase of the Quasi Biennial Oscillation (QBO) evident during the 1970-1971 winter, the polar vortex displacement (PV) that took place in March of 2000 occurred during a westerly QBO, as will be the case this season. Thus the latter displacement seems worthy of more consideration for the coming season. When considering the three primary QBO analogs of 2016, 2020 and 2022 within a solar context, although none took place during solar max, 2022 was the closest, followed by 2016 and 2020, the latter of which is a poor solar match having taken place near solar minimum.

 

Given that the QBO analog of 2020-2021 took place near solar minimum, the early January 2021 SSW  is of least relevance as a viable analog occurrence this season. However, the February 2023 SSW, which is a better solar analog, lends more support to a latter season potential polar vortex disruption along the lines of March 1971 and 2000. This notion is supported by research on high solar, westerly QBO seasons, which lends credence to the late winter/early spring displacement scenario.

 

 

 

 

The research clearly implied that a displacement of the PV triggered by a SSW beyond mid month was a distinct possibility based on the history of westerly QBO seasons that have occurred near solar max, and this is precisely what took place. Note the displacement of the PV in concert with the warming polar stratosphere that had already begun prior to mid month.

 

The event continued to evolve until a reversal of the zonal winds occurred around March 20th, as suggested by the 2000 analog.

 

 

The research also suggested that any cold resulting from the PV disruption would be insufficient and/or too late to bias the monthly mean negative, again much like March 2000 and other La Nina seasons , and that also proved correct.

 

Warmer & Drier Than Average March With Very Little Snowfall

Eastern Mass Weather forecast anywhere from near normal to +2F positive mean monthly departures for the region, but this verified within a range of approximately +3 to +5F, due in large part to the restrengthening of the predominant +WPO pattern this season.

 

 

 

 

 

The lower heights in the vicinity of Northeastern Siberia in the +WPO composite are very apparent.
 

 

This is consistent with the 500mb pattern throughout most of the month.

 

 

 

 

As well as the aforementioned analog season of 2000, which represented perhaps about as strong an analog as possible for the month of March.

 

 

 

It is plausible to consider whether or not the behavior of the polar stratosphere, or the westerly phase of the QBO had any impact on the evolution of the extra tropical Pacific pattern given the uncanny similarities and the fact that each season occured during a westerly phase of the QBO. One significant difference between the two seasons is that the predominate +PNA pattern did neutalize during March 2025, whereas it persisted in thoughout March 2000. This likely played a role in the drier than average pattern that was especially prevalent near the coast, which when paired with warmer than average temperatures is undoubtedly in why southern New England saw very little snowfall throughout the month.

 

 

March Conclusions

The forecast for the month of March was not bad overall, however, its success was limited by an incorrect assessment of the WPO in conjunction with an underestimation of just how unfavorable the polar domain would be despite the major polar vortex disruption that was already anticipated to have little sensible impact during the month.

 

The strong +PNA pattern finally relented during the month of March, as it made a decided descent into strongly negative territory to close the month (-.10 monthly mean). This conceivably played a role in mitigating any potential latter month impacts of the ongoing stratospheric warming for the region, which was always going to be in large part reserved for April, regardless .

 

It is fair to conclude that the month of March would have been much more wintry and in more line with the 2018 analog had the season occurred at a different point in the solar cycle and/or during the easterly phase of the QBO, which are both historically more highly correlated to major and more protracted incidences of high latitude blocking.

December-March 2024-2025 Composite Verification and Summarization

There were two significant issues with the DM 2024-2025 Eastern Mass Weather forecast composite issued last November. The first issue is that the anticipated catalyst of the deviation from the Maritime forcing regime was not a -WPO pattern, as forecast last fall, but instead a -EPO/+PNA tandem.

 

 

This rendered the December 2017 to January 2018 mismatch period (-EPO/+PNA) a better fit than the preferred January 2022 analog.

 

 

Negative NAO style mismatch analogs were discounted due to the expectation of a significant +NAO in the seasonal mean due to being near solar maximum.

The December to January 2017-2018 period was also more comparable to this past period in length, as it was longer in duration than the January 2022 period.

 

 

 

The second issue, referenced in the annotation above, is with respect to the WPO. Whereas the primary January 2022 analog featured a strongly -WPO (-1.44), January 2025 registered a mean value of 1.02, which ensured that the DM season would average positive (.45). This runs contrary to the premise of the Outlook, which identified the primary driver of the season as a -WPO that would focus on the month of January, akin to the 2022 analog.  This +NAO/+WPO couplet not only provided resistance to the core of the cold reaching the east coast, but also an avenue for storms to track inland. This is at least partially attributable to the absence of a KU event this season despite seasonable cold, which is why snowfall finished well below normal across southern New England, and normal to perhaps even slightly above normal across far northern New England.

 

+WPO & +NAO Correlate to Warmth for Most of the NE and Inland Storm Tracks

Below is the complete table of forecast mean DM index values versus verification. The DM mean PDO value registering significantly more positive than expected is likely at least somewhat connected to the aforementioned positive PNA. The previously discussed disconnect between the NAO (+) and AO (-) is a product of the stout stratospheric PV being poorly coupled with the troposphere, as discussed in the January review. While this did preclude a very mild winter, it did not necessarily represent a favorable pattern for NE US coastal storms. 

 

 

Index Value	Predicted '24-'25 DM   Value Range	Actual  '24-'25 DM Value	Forecast Error
Pacific Decadal Oscillation (PDO)	-1.85 to -2.15	-1.46	Biased .39 Negative
Perennial North American Pattern (PNA)	-.40 to +.10	+1.02	Biased .92 Negative
ENSO	DJF -0.6 to -0.8 ONI EMI: -.5 to -.7 (Mixed-Type)	DJF -0.6 ONI DJF EMI: -.7	Verified
(J-M) East Pacific Oscillation (EPO)/DM West Pacific Oscillation (WPO)	+.30 to +.60 JM EPO -.40 to -.70 DM WPO	-.30 JM EPO +.45 DM WPO	Biased .60 Positive Biased -.85 Negative
Arctic Oscillation (AO)	+.26 to +.56	-.18	Biased .44 Negative
North Atlantic Oscillation (NAO)	+.38 to +.68	+.65	Verified

The absence of a strongly negative NAO month,  presumably due to the Modoki La Nina materializing during the westerly phase of the QBO in such proximity to solar maximum, is a notable distinction between the 2024-2025 winter season and the 2017-2018 analog. This undoubtedly played a major role in the relative dearth of snowfall this past winter, especially during the month of March, as compared to the 2017-2018 season. This especially true given the presence of stronger than average Pacific jet, which combined with +NAO in the seasonal mean made it extremely difficult to sustain a +PNA ridge in the proper location long enough to facilitate the development of a major east coast snow storm. 

 

 

Here is a prime example from last winter of the Pacific jet undercutting a +PNA ridge and this tilting it positively, leading to a failed phase attempt on the east coast.

 

This was a theme of the season.

Note how much weaker said Pacific jet was during the 2017-2018 winter season, which included several major east coast winter storms.

 

Nonetheless, being strong analog for a given season and having key differences from said season are not mutually exclusive concepts, and the 2017-2018 analog is a splendid illustration of this within the context of the 2024-2025 winter season. Given the similarity between the deviation periods from the prevalent MC background state of the past decade, it should come as no surprise that there were also some striking similarities between the respective cool ENSO events of 2024-2025 and 2017-2018.

East-Based La Niña Atmospheric Imprint Observed During Boreal Winter 2024-2025

The expectation last fall was for a hybrid-mixed type La Niña forcing regime to persist in the mean throughout winter 2024-2025.

 

This was correct, however, as has been discussed frequently on Eastern Mass Weather, there is more variance amongst the data sets of Modoki, hybrid and east-based events when ENSO remains fairly weak due to the larger role that the extra tropics play in modulating the intra seasonal pattern. Winter 2024-2025 represented a very good example of this variance. Note that the DM SST anomaly distribution is actually consistent with a Modoki type La Niña, with the greatest anomalies between 150W and the 180W dateline. 

This is near the western extent (-.7) of the -.5 to -.7 EMI forecast range.

EMI analogs included  2008, 2010, 2011, 2020 and 2021.

The bares a strong similarity to the primary ENSO analog of 2008-2009.

 

 

Here are the 500mb composites for all east-based, hybrid and Modoki La Niña events since 1950, followed by the DM 2024-2025 500mb composite below for comparison. It is clear that weather pattern last season was actually redolent of an east-based La Niña, despite the aforementioned Modoki SST configuration.

 

 

 

Clearly La Niña was not actually east-based, as per the SST anomaly configuration, thus this is a reflection of ENSO taking the proverbial "back seat" to extra tropical drivers as it so often does when it is weaker. In this case, those particular -EPO/+PNA drivers were the same as they were in 2017-2018. 

 

 

 

In hindsight, this is why the December to January 2017-2018 period ultimately proved to be the superior "mismatch" analog to the preferred January 2022 analog, despite the fact that that particular season was an east-based La Niña approaching solar minimum. Seasons of this ilk with more "cooks in the kitchen" so to speak, epitomize why weaker ENSO events average out colder than stronger events. A greater number of cooks (seasonal drivers) entails more opportunity for a variety of flavors.

 

Weak La Niña composites:

 

 

Clearly there is a resemblance between last season and the weak La Niña DM H5/temperature anomaly  composites over the US, despite the fact that subsidence over the dateline was slightly stronger last year.

 

 

Notice that subsidence in this area is also greater last season relative to the east-based composite, despite the weather pattern having evolved similarly in the DM mean.

 

 

This is likely due to a combination of factors, as if often the case with weaker events, which makes for a more difficult forecast. The ambiguity is the result of a plethora of competing forces, as opposed to when ENSO is more intense and thus a more prominent driver. Although La Niña was weak enough to allow for competing extra tropical forces to dictate the pattern to a fairly significant degree, it was in fact a more pervasive force throughout the hemisphere than implied by the meager ONI.

The Late Season Pattern Change Reveals Disparate Face of La Niña

Although this was not an official La Niña per the CPC and the peak ONI was a very paltry -0.6, the event peaked as borderline moderate per MEI (-1.0), and as low-end moderate when accounting for the ambient warming of the oceans. This is evident when considering the peak RONI value of -1.12, which essentially means that this was a moderate event when considering the increased gradient between the cooler ENSO waters, and the surrounding waters that are warming as a product of climate change. This was referenced earlier when it was explained that this past La Nina was of comparable intensity to the 2008-2009 analog when accounting for these factors. Clearly the resultant pattern in the DM mean resembled that of a weak La Niña, as it did an east-based event.

 

 

However, there was greater subsidence over the dateline than present in both the weak and east-based composites, which effectively means that this event was more proficiently pinning convective activity over the MC per La Nina climatology. This affords more opportunity for milder interludes and inland storm tracks, as well as less intense and protracted periods of NAO style blocking, which is precisely what was observed. Notice that the La Niña configuration during the month of December, when the -EPO/+PNA tandem was dominating, was more basin wide with weaker  subsidence over the dateline.

 

 

This was closer to the 2017-2018 configuration, in which the Hadley Cell was displaced to the east and subsidence over the dateline was less pronounced.

 

 

However, as the season progressed, La Niña shifted westward and assumed a Modoki configuration by March.

 

 

This is where the 1999-2000 analog took center stage and scripted the tame ending to winter 2024-2025, which included a "too little, too late" SSW per solar max/westerly QBO, cool ENSO climatolog and strongly positive NAO pattern.

 

 

In conclusion, the variability of winter 2024-2025 was likely due to a combination of La Niña being weak enough to allow for greater extra tropical influence than experienced in stronger events, such as 1999-2000, yet strong enough to compete with these extra topical drivers and assert itself during intervals in which their influence waned. During December and January, the slight eastward displacement of La Niña relative to the balance of the season combined with the -EPO/+PNA extra tropical drivers to consistently deliver seasonably cold air to the east, which was prolonged after the New Year given the only significant -NAO period of the season. However, given the antecedent -EPO/+PNA pattern in place, the addition of the negative NAO also limited precipitation. February was a month of flux and competing forces, as the -EPO/+PNA tandem continued and teamed with the "bottom up" PV disruption to deliver ample cold to the nation's midsection; however, said arctic surge was met with considerable resistance when approaching the east coast per La Niña climatology. This battle combined with seasonable cold to produce the only month of above average snowfall of the season for the southern New England forecast area. March was predisposed to significant warmth and a relative dearth of precipitation due to a confluence of factors. First of all, the instant delivery of cold during the month of February was not the only implication of the February PV disruption originating in the troposphere and not having to propagate downward. It also meant that the cold that was displaced southward by the Feb PV disruption was exhausted by the onset of March. Compounding matters was the fact that La Niña had evolved into a full-fledged Modoki and potentially played a role in the neutralization of the PNA for the first time all season, along with the maintenance of a strongly positive NAO. The final polar vortex disruption was of the traditional SSW variety, which meant that it took a few weeks to propagate downward, and thus did not impact the mid latitude weather pattern until April. This was also very consistent with the 2000 analog, in addition to the ultimate evolution of La Niña, thus the DM season ended in very meek manner, much as it did that year. 

The 2024-2025 winter season was a lesson in both the limitations in the use of analog seasons, as well as the limitations of seasonal forecasting in general. Regardless of how strong an analog is, it only takes one subtle difference over a relatively small timeframe to drastically alter an outcome that is inherently subject to a high degree of variance, such as snowfall. This understandably plays a large role in shaping how society perceives a given season. Moving forward, it will be important to remain mindful that analogs such as 1999-2000, which offered a great deal of utility in forecasting winter 2024-2024, were on the precipice of a major shift in the extra tropical Pacific and what that may or may not mean moving forward.

 

 

This is one of many questions that will guide forecasting efforts moving forward later this year.

December-March 2024-2025 Snowfall Outlook Verification

The snowfall forecast for winter 2024-2025 was one of the better Eastern Mass Weather efforts, but there still remained a slight positive bias overall. This despite the fact that this season's forecast was ever so slightly too warm (+1 to +3F vs +2 to +4F DM forecast) due in large part due to the -EPO/+PNA pattern that persisted for much of the season, as opposed to the forecast +EPO/slight -PNA. 

 

 

 

 Snowfall was over forecast by anywhere from 2.6% to as much as 30.7% across southern/central New England and New York State. This was primarily due to a combination of unfavorable, inland storm tracks attributable to the +NAO/WPO tandem during the months of  December, February and March, and -NAO combining with the prominent seasonal -EPO/+PNA to suppress the storm track during the heart of winter in January. This is evinced by the fact that DM precipitation was slightly drier than what was a fairly dry forecast across much of the east.

 

 

Ironically enough, February was still the snowiest month across southern New England because the warmth was not prohibitive and the +NAO allowed for a more active storm track. However, the majority of events were mixed and there was no KU type events, which limited potential. The suppressed mid winter storm track is why the forecasting error was even more minimal to the south and east, across the mid Atlantic and portions of coastal southern New England, where the it verified at three locales and was actually even under forecast by 6.4% in Washington, DC. Across Northern New England, the forecast was very accurate, having verified in Portland, ME and been under forecast by 0.9% in Burlington, VT, presumably due to the inland/northerly storm track being less prohibitive. 

 

City	Predicted Snowfall	Actual	Forecast Error
Boston, MA	24-34"	28.1"	Verified
NewYork, NY(Central Park)	10-20"	12.9"	Verified
Philadelphia, PA	8-18"	7.7"	3.9%
Baltimore, MD	6-16"	12.7"	Verified
Washington, DC	4-14"	14.9"	6.4%
Albany, NY	48-58"	43.9"	9.3%
Hartford, CT	24-34"	23.4"	2.6%
Providence, RI	20-30"	15.6"	28.2%
Worcester, MA	50-60"	41.2"	21.4%
Tolland, CT	40-50"	32"	25%
Methuen, MA	41-51"	32.5"	26.2%
Hyannis, MA	14-24"	10.8"	30%
Burlington, VT	65-75"	75.7"	0.9%
Portland, ME	48-58"	54.1	Verified
Concord, NH	49-59"	37.5"	30.7%

Final Grade for 2024-2025 Season: B

How This Snowfall Forecast Compares To Past Efforts

The mean snowfall forecasting error for the 2024-2025 winter season throughout the 15 select Mid Atlantic and northeast locations was 12.3%. This was the 2nd most accurate forecast amongst the eleven forecast seasons in terms of mean forecasting error. Only the 2014-2015 (10.9%) effort fared better. The seasonal forecast ranges verified in four of the cities and was within 10% in 9/15 cities. 

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

2024-2025: 12.3%

2023-2024: 120.3%

2022-2023: 900.6%

2021-2022: 36.2%

2020-2021: 14.1%

2019-2020: 678.3%

2018-2019: 90.3%

2017-2018: 13.2%

2016-2017: 35.3%

2015-2016: 71.4% 

2014-2015: 10.9%

The ten season running mean error: 180.3%

31/161 (19.3%) of snowfall outlooks have been under forecast.

100/161 (62.1%) of snowfall outlooks have been over forecast.

30/161 (18.6%) of snowfall outlooks have verified within the forecast range.