donsutherland1

I think there's room for some marginal adjustments, but this very likely won't be a moderate, much less, significant snowfall for the Greater NYC area.

Some users of weather information such as the energy, transportation, and agriculture sectors, need such information for planning purposes, imperfect as it is. They understand the limitations of the information. The general public doesn't. Unfortunately, irresponsible actors on Social Media use the models to hype extreme solutions. In the past, I would never comment on such things. However, with almost an equal share of the public now getting its news/information from Social Media as from conventional sources and widespread perceptions that meteorologists essentially have no skill, I have decided to push back in recent years. Indeed, even at my own workplace, people ask me about some of the extreme stuff from Social Media (one, despite my advice, relies largely on X and TikTok for news/weather/information).

Just so there is no confusion, the evolving 500 mb pattern was not a classic KU storm pattern for January, particularly the first half of January. Instead, it was one that favored moderate and localized significant snow events if things came together. The possible heavier snow showers later today in parts of the Mid-Atlantic and the upcoming storm that will favor the Washington, DC-Baltimore areas shows that the potential of this pattern could be realized. Unfortunately, central NJ northward won't benefit through at least the next 7-10 days. The power of the strong and rapidly strengthening block will limit opportunities until that block begins to relax. Whether there will be sufficient short waves and how the synoptic details come together at that point will determine whether NYC to Boston could benefit down the road. Right now, that's not assured, which is frustrating for snow lovers given that parts of this region are experiencing their worst two-year snow drought on record. A week ago, things looked more promising for a larger part of the East when the AO was forecast to bottom out near or just above -3. Now, it's forecast to dive to near -4. Moreover, it's forecast to do so very rapidly. It's that factor that a lot of the commentators on social media missed in automatically assuming that this event would "trend" north. Things did shift gradually northward for a time over several model cycles, but IMO, as the models began showing the evolving deeper block overnight (the AO fell from -1.461 to -2.085), that shift rapidly corrected. Overall, the pattern for the Sunday-Monday storm has characteristics of Clusters 2 and 5 from Washington, DC's 6"+ January snowstorms in the eastern CONUS. It is not a perfect match for either cluster, but has enough similarity to provide insight. The trailing trough is too far south relative to NYC's 6"+ patterns. That's why I felt, and still feel, that snowfall amounts could fall off sharply once one gets north of Philadelphia and that NYC would be limited to a light snowfall. A lot would have to change for this idea to be invalid. The block is the big driver and it's the block that would need to change (weaker, farther north). At the current timeframe, the guidance is quite skillful when it comes to such big picture features. Finally, if New York City receives no measurable snowfall (still not a done deal, but certainly plausible), this wouldn't be the first January snowstorm to dump 6" or more in Washington, DC while bringing no measurable snowfall to NYC. The list is below: January 11-12, 1904: Washington, DC: 6.0"; New York City: None January 23-25, 1940: Washington, DC: 9.5"; New York City: Trace January 30, 2010: Washington, DC: 6.4"; New York City: None January 12-14, 2019: Washington, DC: 10.3"; New York City: Trace January 3, 2022: Washington, DC: 6.9"; New York City: None Highest Washington, DC snowfall with none in NYC: 8.4", March 9-10, 1999 Highest Washington, DC snowfall with a trace in NYC: 17.8", February 5-6, 2010, which also brought 28.5" to Philadelphia

NYC is at 12.6". I believe you missed the 1.0" that fell on 12/24. As to your question, it has on 11 occasions. The most recent such period was 1987-1989 with DCA's 3-year average coming to 24.8" and NYC's to 18.6".

Hartford is one example on the 12z with 0.4" of snow.

Brief update: Now that the forecast period is within the range of the EPS and GEFS, it is possible to get a reasonable idea of whether the extreme forecast is in line for possible verification. In fact, BAM's forecast appears headed for a spectacular bust (>20°F in all four highlighted cities). Some will argue that I shouldn't treat it as BAM's forecast, but when one posts deterministic output to the general public with no clarifications or caveats, that's exactly what one is doing. Again, what the social media site ignored or did not understand is that: 1. Amount of deep cold in the Northern Hemisphere matters (it was limited then and it remains limited now) 2. Historic climate provides useful benchmarks for assessing the credibility of model output. One can't blindly take model output verbatim stripped of historic context, especially when one is dealing with potentially extreme events. 3. Model skill at the timeframe involved (> 5 weeks!) is essentially non-existent. Thus, one should typically focus on patterns and probabilities, not details and values 4. Basing one's forecasts on cherry-picked model runs is especially risky. To date, no other single model run has come close to the cherry-picked one featured by BAM. On a side note, in late December, the site cherry-picked another extreme model run to invoke "historic cold." Day's later, the local NBC affiliate called it out. In addition, NWS Atlanta called out the posting of extreme snowfall maps on social media. In the end, the failure to consider or understand (or both) the four points raised above are bad practices. Indeed, if one is tempted to even consider a model at the timeframe involved, one should ask, "Is the output reasonable?" One then needs to look at whether such events have occurred before (rarity) and what was involved. As the kind of cold shown was extremely rare or unprecedented for the climate record, one should wait for strong evidence before embracing the outlier guidance. Of course, the outlier guidance did not survive to the following model cycle, but that wasn't mentioned. Hype does not serve the public interest. Social Media today, which is now relied upon by the public almost as much as conventional sources for news/weather/information, provides a powerful platform for hype. Not surprisingly, the general public is confused about the skill of meteorologists, because hype is psychologically more memorable and the busts that occur skew their perceptions of the profession. The next update will be the verification here and on X.

A quick follow-up: Truly bitter cold air remains in very limited supply in the Northern Hemisphere: Based on the latest data, even as the cold will be sustained, it likely won't be notable in many places. My own thinking from six days ago will be too cold. NYC and Philadelphia might see some teens, but single digits are unlikely. There's a chance that Boston may not even see teens courtesy of the powerful block suppressing the coldest air to the South. Detroit, Chicago, and Indianapolis could still see a subzero low, but for Detroit that is less likely than for Chicago or Indianapolis. Moreover, most of the below cities likely will not match or exceed their coldest low temperatures or coldest 7-day period since 2020. All this does not mean that the ECMWF weekly guidance will be incorrect. It's becoming clearer, that the guidance's cold anomaly is more the result of persistence of cold than severity of cold.

A gradual transition toward sustained colder than normal weather is ongoing. Temperatures could fall to below normal levels during or after the coming weekend. An upper level disturbance will likely bring some snow flurries and snow showers from Washington, DC to central New Jersey. Some areas could pick up a small accumulation. A more important storm will bring parts of the Middle Atlantic region a moderate to locally significant snowfall Sunday night into Monday. Washington, DC, Baltimore, and Philadelphia have the potential to see 3"-6" of snow with locally higher amounts. The highest amounts will likely be centered over the Washington, DC-Baltimore area. Some degree of uncertainty persists. Once north of Philadelphia, accumulations could fall off quite rapidly. A light amount remains likely for the New York City area. With the storm appearing a little stronger and tracking slightly farther to the north on the guidance, the probability of Richmond's seeing an end to its current record streak of 1,082 days without daily 1" or greater daily snowfall has decreased. The risk of sleet and freezing rain there has increased. The second and third weeks of January will likely feature widespread cold anomalies in much of the eastern half of the CONUS and Canada south of the Hudson and James Bays. The potential exists for widespread cold anomalies of 6°-10° below normal during that week in parts of that region, particularly the Southeast. Severe cold still appears unlikely in the New York City and Philadelphia areas through at least mid-month. The ENSO Region 1+2 anomaly was -0.2°C and the Region 3.4 anomaly was -1.1°C for the week centered around December 25. For the past six weeks, the ENSO Region 1+2 anomaly has averaged +0.10°C and the ENSO Region 3.4 anomaly has averaged -0.53°C. Neutral ENSO conditions may still evolve into a La Niña event during the winter. Uncertainty as to whether a La Niña will actually develop persists. On the December 19 outlook, 56% of dynamical models but 0% of the statistical models forecast the development of a La Niña. The SOI was -5.36 today. The preliminary Arctic Oscillation (AO) was -1.461 today.

Yes, that is true. But it's still worth trying to identify those who routinely make irresponsibly extreme calls if the public is to be truly weather-aware.

Apparently the local NBC outlet called out one site that has been hyping coming extreme to historic cold for several weeks. Viewers were probably asking why the station wasn’t forecasting such cold. IMO, only when meteorologists start to call out their own peers for irresponsible claims will incentives for hyping events for clicks and engagement diminish. This is a welcome development.

There appears to be a significant disconnect between the snowfall forecast and QPF forecast on the National Blend of Models. Even where zero QPF is forecast, snow is forecast. I'm not aware of this issue having occurred with the prior version (4.1). I suspect that there is an adjustment algorithm concerning the edge of the snow shield that is leading to this disconnect. Taken literally, NYC would see 95:1 snow-liquid ratios. More than likely, either the QPF will prove to be higher or the snowfall will prove to be lower. Perhaps the disconnect will also disappear as the event grows nearer.

Phoenix Caps Off its Warmest Year with Its Warmest December on Record Phoenix capped off its warmest year on record with its warmest December on record. During 2024, Phoenix tied or broke 39 daily record high maximum temperatures, 42 daily record high minimum temperatures, and 43 record high mean temperatures. December has seen remarkable stability in the average monthly high temperatures. Its overall warming has been driven largely by increasing low temperatures. December 2024: Annual: During 2024, Phoenix also had an average high of 90.5° (old record: 89.8°, 1989) and an average low of 66.7° (old record: 65.5°, 2014).

A gradual transition toward sustained colder than normal weather is getting underway. Temperatures could fall to below normal readings during or after the coming weekend. The second and third weeks of January will likely feature widespread cold anomalies in much of the eastern half of the CONUS and Canada south of the Hudson and James Bays. The potential exists for widespread cold anomalies of 6°-10° below normal during that week in parts of that region, particularly the Southeast. Overall, January will likely be colder than normal in such cities as Philadelphia and New York City. Should the AO dive below -3.000, which now appears likely, the prospects for a significant or major snowfall in the New York City area could be reduced. Since 1950, a January AO ranging from -2.600 to -1.400 accounted for the largest share of 6" and 10" or above January snowfalls for storms that dumped 4" or more snow in New York City. For the 4" or above storms, 67% saw 6" or more while 56% went on to see 10" or more. When the AO was -3.000 or below, 60% of 4" or above snowstorms saw 6" or more snowfall, while 20% of those storms saw 10" or more. The exact synoptic details will be crucial to determining whether that potential will be realized. As a result, a storm that brings parts of the Middle Atlantic region a moderate snowfall Sunday night into Monday could bring only a light amount to the New York City area. It is possible that all of the snow could miss to the south of the City. Richmond will very likely see its current record streak of 1,081 days without daily 1" or greater snowfall come to an end. The NAO fell to a preliminary -2.751 on September 24th (all-time September record: -2.371, September 12, 1971). That was the 9th lowest value on record. La Niña winters following September cases where the NAO fell to -1.900 or below featured a predominantly positive NAO. The most recent such winters were 2016-2017 and 2022-2023. The mean temperatures for those winters in New York City were 39.3° and 41.0° respectively. The 1991-2020 normal value is 36.2°. A warmer to much warmer than normal outcome was favored by the November run of the ECMWF for Winter 2024-2025. Even as the NAO outlook appears on course to be relatively accurate, Winter 2024-2025 will likely wind up being notably colder than the subset of winters listed above. The ENSO Region 1+2 anomaly was -0.2°C and the Region 3.4 anomaly was -1.1°C for the week centered around December 25. For the past six weeks, the ENSO Region 1+2 anomaly has averaged +0.10°C and the ENSO Region 3.4 anomaly has averaged -0.53°C. Neutral ENSO conditions may still evolve into a La Niña event during the winter. Uncertainty as to whether a La Niña will actually develop persists. On the December 19 outlook, 56% of dynamical models but 0% of the statistical models forecast the development of a La Niña. The SOI was -1.97 today. The preliminary Arctic Oscillation (AO) was -1.282 today.

The AO is now forecast to drop toward -4.000 in coming days. That can lead to suppressed solutions. Indeed, the EPS 500 mb height anomalies at 132 hours illustrates the issue. I've compared the map with a cluster of 6" or greater January snowstorms for the NYC area during AO-/PNA+ regimes. That's why the best chance for accumulating snow--perhaps even a moderate snowstorm--will likely be from Philadelphia and southward. A light snowfall is possible in the NYC area, but one can't rule out the prospect that all of the snow misses to the south at this point. Put simply, the forecast pattern through at least January 10th is not conducive for KU snowstorms even as there will be sufficient cold. Richmond's ongoing record-breaking 1,081-day stretch without a 1" or above snowfall will likely end, perhaps on Sunday or Monday.

Final December numbers:

December is closing on a warmer than normal note, but a big change toward sustained colder weather lies ahead for January. The pattern will likely become more favorable for light or moderate snowfalls as the cold returns during the first week of January. The second and third weeks of January will likely feature widespread cold anomalies in much of the eastern half of the CONUS and Canada south of the Hudson and James Bays. The potential exists for widespread cold anomalies of 6°-10° below normal during that week in parts of that region, particularly the Southeast. Overall, January will likely be colder than normal in such cities as Philadelphia and New York City. Should the AO dive below -3.000, which now appears likely, the prospects for a significant or major snowfall in the New York City area could be reduced. Since 1950, a January AO ranging from -2.600 to -1.400 accounted for the largest share of 6" and 10" or above January snowfalls for storms that dumped 4" or more snow in New York City. For the 4" or above storms, 67% saw 6" or more while 56% went on to see 10" or more. When the AO was -3.000 or below, 60% of 4" or above snowstorms saw 6" or more snowfall, while 20% of those storms saw 10" or more. The exact synoptic details will be crucial to determining whether that potential will be realized. On account of a lack of snowfall, 2023-2024 will set new records for the lowest snowfall over a two-year period for Boston and New York City. Boston: 2023-2024 Total to Date: 26.9"; Record: 38.2", 1979-1980 New York City: 2023-2024 Total to Date: 12.6"; Record: 17.4", 1997-1998 The NAO fell to a preliminary -2.751 on September 24th (all-time September record: -2.371, September 12, 1971). That was the 9th lowest value on record. La Niña winters following September cases where the NAO fell to -1.900 or below featured a predominantly positive NAO. The most recent such winters were 2016-2017 and 2022-2023. The mean temperatures for those winters in New York City were 39.3° and 41.0° respectively. The 1991-2020 normal value is 36.2°. A warmer to much warmer than normal outcome was favored by the November run of the ECMWF for Winter 2024-2025. Even as the NAO outlook appears on course to be relatively accurate, Winter 2024-2025 will likely wind up being notably colder than the subset of winters listed above. The ENSO Region 1+2 anomaly was -0.2°C and the Region 3.4 anomaly was -1.1°C for the week centered around December 25. For the past six weeks, the ENSO Region 1+2 anomaly has averaged +0.10°C and the ENSO Region 3.4 anomaly has averaged -0.53°C. Neutral ENSO conditions may still evolve into a La Niña event during the winter. Uncertainty as to whether a La Niña will actually develop persists. On the December 19 outlook, 56% of dynamical models but 0% of the statistical models forecast the development of a La Niña. The SOI was -7.58 today. The preliminary Arctic Oscillation (AO) was -0.855 today.

DCA -5.0 NYC -2.5 BOS -1.0 ORD -2.0 ATL -5.5 IAH -0.7 DEN +0.8 PHX + 3.2 SEA +3.0

Yes. It illustrates that even within a week, the ensembles can be incorrect.

The weekly outcome:

Today was an unseasonably mild day in the Mid-Atlantic region. High temperatures included: Atlantic City: 63° Baltimore: 65° Bridgeport: 56° Islip: 56° New York City: 60° Newark: 65° Norfolk: 72° Philadelphia: 66° Reading: 65° (old record: 64°, 1982) Richmond: 69° Trenton: 65° Washington, DC: 67° December will close with above to much above normal temperatures and additional periods of rain. The opening days of January will likely also begin with above normal temperatures, but it will start to turn colder. Snowfall prospects will remain limited through the end of December. However, the pattern will likely become more favorable for moderate or perhaps larger snowfalls as the cold returns during the first week of January. The second week of January will likely feature widespread cold anomalies in much of the eastern half of the CONUS and Canada south of the Hudson and James Bays. The potential exists for widespread cold anomalies of 6°-10° below normal during that week in that region. Should the AO dive below -3.000, the prospects for a significant or major snowfall in the New York City area could be reduced. Since 1950, a January AO ranging from -2.600 to -1.400 accounted for the largest share of 6" and 10" or above January snowfalls for storms that dumped 4" or more snow in New York City. For the 4" or above storms, 67% saw 6" or more while 56% went on to see 10" or more. When the AO was -3.000 or below, 60% of 4" or above snowstorms saw 6" or more snowfall, while 20% of those storms saw 10" or more. The exact synoptic details will be crucial to determining whether that potential will be realized. On account of a lack of snowfall, 2023-2024 will set new records for the lowest snowfall over a two-year period for Boston and New York City. Boston: 2023-2024 Total to Date: 26.9"; Record: 38.2", 1979-1980 New York City: 2023-2024 Total to Date: 12.6"; Record: 17.4", 1997-1998 The NAO fell to a preliminary -2.751 on September 24th (all-time September record: -2.371, September 12, 1971). That was the 9th lowest value on record. La Niña winters following September cases where the NAO fell to -1.900 or below featured a predominantly positive NAO. The most recent such winters were 2016-2017 and 2022-2023. The mean temperatures for those winters in New York City were 39.3° and 41.0° respectively. The 1991-2020 normal value is 36.2°. A warmer to much warmer than normal outcome was favored by the November run of the ECMWF for Winter 2024-2025. Even as the NAO outlook appears on course to be relatively accurate, Winter 2024-2025 will likely wind up being notably colder than the subset of winters listed above. The ENSO Region 1+2 anomaly was -0.2°C and the Region 3.4 anomaly was -0.8°C for the week centered around December 18. For the past six weeks, the ENSO Region 1+2 anomaly has averaged +0.17°C and the ENSO Region 3.4 anomaly has averaged -0.35°C. Neutral ENSO conditions may still evolve into a La Niña event during the winter. Uncertainty as to whether a La Niña will actually develop persists. On the December 19 outlook, 56% of dynamical models but 0% of the statistical models forecast the development of a La Niña. The SOI was +3.06 today. The preliminary Arctic Oscillation (AO) was +0.226 today. Based on sensitivity analysis applied to the latest guidance, there is an implied near 100% probability that New York City will have a colder than normal December (1991-2020 normal). December will likely finish with a mean temperature near 38.0° (1.1° below normal).

The fireworks of the December 29 0z ECMWF and GFS runs have now faded. Even as those runs aren't cast in stone, they provided a glimpse of the potential that lies ahead. A colder pattern will develop starting around January 2-3. As an AO-/PNA+ pattern develops, the potential for moderate or perhaps significant snowfalls will increase in the northern Mid-Atlantic region, including the New York City area. The cold will be in place. There will be some short waves that offer the possibility of blossoming into storms. Synoptic details will be crucial. From this far out, they can't be resolved with reliability. Individual ensemble member snowfall forecasts maintain a wide spread among dates (January 6-12). Typically, they converge on a single narrow timeframe in advance of a significant or major event. In the meantime, below are 500 mb clusters for New York City's January 6" or above snowstorms during AO-/PNA+ regimes, which are most favorable for significant events relative to climatology, for reference.

I believe we are now leaving the Holocene and perhaps have already left. It’s probably only a matter of time before geologists reach consensus.

Week ahead outlook: The coming week will be a transitional period. Very mild weather will give way to progressively colder weather. Snowfall will still likely be limited in the Mid-Atlantic and southern New England areas.

I suspect that it's a combination of both. It's too soon to be sure whether NYC is in the very early stages of a structural decline in snowfall from a warming climate. By the mid-2030s, it will probably be clear e.g., if NYC's seasonal snowfall has fallen to 20" or below (as suggested with regression analysis). Here's how NYC's 30-year mean snowfall stacks up against its 30-season winter mean temperatures: Here's Washington, DC's where seasonal snowfall has become structurally lower: Notice that both cities saw an increase as they approached mean winter temperatures of 36° that continued shortly afterward. NYC's decrease will likely be lower simply because NYC is closer to Canada and even in a warmer climate will see stronger shots of cold than Washington, DC. Both cities will remain capable of seeing big snowstorms, even as Washington's seasonal snowfall has declined and even if New York City's is in the early stages of a structural, non-cyclical decline. Both cities can still see very snowy winters. Winter 2009-10 in Washington, DC provides an example of multiple big snowstorms there and a very snowy winter despite a climate that had warmed by 1.4° over the prior 20 years.

Increasingly, it appears that credible meteorological organizations and meteorologists are being drowned out by Social Media hype by amateurs and some of their own peers. The amateurs may not necessarily know better, but one would expect that the professionals would. But exaggerated posts (snowstorms, SSWs, and severe cold) in recent years hints otherwise. Last night, the 12/28 0z ECMWF operational model forecast a historic-type Arctic blast. Several professional accounts were among Social Media accounts trumpeting the model verbatim. At least one even specifically referred to "historic cold." A turn toward colder weather is likely in early January and the second week of January could be much colder than normal. But there is a large distinction between much colder than normal, a significant Arctic blast, and a historic-type Arctic blast. If one examines the bigger picture one finds: 1. Historic-type Arctic blasts have become less common as the Arctic region has warmed and the Northern Hemisphere's deep cold pool has shrunk. The last true historic large-scale Arctic blast occurred in February 11-20, 2021. 2. The climate has warmed, particularly since 2000. The ongoing warming is a big reason why the Contiguous U.S. has not seen a winter where excessively cold minimum temperatures have exceeded areal coverage of excessively warm minimum temperatures (winter average) since 2013-2014: Here are the Northern Hemisphere temperature anomalies 2 weeks and 1 week before the start of the February 2021 historic Arctic outbreak: 2 weeks: 1 week: 3 days: And here's the latest map (December 26, 2024): Key Difference: There was much more very cold air, both on our side of the Northern Hemisphere and the Northern Hemisphere overall prior to the February 2021 Arctic outbreak. It is unclear whether those accounts have much understanding of how to use ensembles, even as they almost certainly have awareness of ensembles. IMO, those accounts would do well to visit and study WPC's outstanding ensemble training page: https://www.wpc.ncep.noaa.gov/ensembletraining/ The WPC guide notes of operational model runs: "...in a chaotic system like an atmospheric model, non linear errors (although inconsequential in short lead times) will grow - sometimes rapidly. Eventually these growing errors cause the model forecast ouput to eventually become useless." That lesson is ignored or not well-understood by the accounts hyping historic cold outbreaks some 10 days or more in the future. If one looks at the latest ECMWF Extreme Forecast Index, one sees significant, but not historic cold is currently forecast: The very cold ECMWF weekly outlook for January 6-13: That's very cold. But it doesn't match the coldest shots since 2000 in Boston, Chicago, Detroit, Indianapolis, Minneapolis, New York City, Philadelphia, and Washington, DC. All said, the recent ECMWF weeklies, including today's run, is plenty cold to communicate the message that it will likely turn significantly colder in the long-range with the potential of a severe shot of cold. Exaggeration and hype are not needed. There is no need to invoke historic Arctic outbreaks that still have very little support (occasional runs on a single model, no run-to-run continuity, no widespread ensemble or other model support), especially at timeframes when operational model skill scores are low. Speaking of very little support, the 12/28 12z ECMWF still contains an Arctic outbreak during the second week of January, but it is nowhere near as frigid as the 12/28 0z run featured and pushed on Social Media by multiple accounts. And for the Social Media accounts that hyped the 0z run, no mention whatsoever about the 12z run? What is likely? January 10-20, will likely see the temperature fall well into the teens and perhaps even single digits in New York City and Philadelphia. Boston will likely see single digits with perhaps a chance at a subzero low. Detroit , Chicago, and Indianapolis could see one or more subzero lows. None of those cities will likely see historic-type cold e.g., widespread subzero lows from Washington, DC to Boston and double-digit subzero lows in Chicago, Indianapolis, and Detroit. Some of those cities might see their coldest readings of the 2020s, but most very likely won't exceed their coldest readings since 2000. Most or all of those cities likely won't experience their coldest 7-day period since 2000. Historic-type low temperatures and 7-day mean temperatures are unlikely. And if the cold fails to measure up to the coldest Arctic outbreaks since 2000, one can expect the Social Media accounts to still claim that they were "correct" in their assessment. However, in fact, their assessment would be wrong, as the cold wasn't even three-decade cold, much less historic-type cold. The general public will then come away with yet another example that skews their perceptions of the meteorology profession, lumping in outstanding public and private sector meteorologists with those who seek Social Media clicks and engagement.