donsutherland1

So far, its snowy outcomes have not been mentioned by the NWS. Representative is the extended forecast discussion from Mount Holly: The airmass will moderate as the flow becomes more southwesterly through the weekend and will warm from the 40s on Saturday to upper 50s Sunday and then 60s on Monday and Tuesday. The only hitch in the this part of the extended is that clouds will increase into Tuesday as a low pressure will track up through the Ohio Valley and into New England on Tuesday, dragging a cold front across our area Tuesday afternoon/evening, bringing another round of rain to the region.

The Governor needs to declare a virtual state of emergency ahead of this virtual snowstorm.

The 3/4/2020 run of the 12z showed what would be an unprecedented snowstorm. The model, using the more "conservative" 10:1 ratio, showed 20" or more snow in Philadelphia, New York City, and Boston. The closest storm to such a solution since regular recordkeeping began in the 19th century was the February 2003 President's Day snowstorm: The closest storm that includes the pre-National Weather Service era was the January 14-16, 1831 "Great Atlantic Snowstorm." That historic blizzard dumped 2 feet of snow in Philadelphia, 18"-20" in New York City, and around 2 feet in Boston. Almost certainly, one can expect widespread chatter on Social Media, but such a solution is highly unlikely, if not all but improbable. First, historic March snowstorms that brought 8" or more snow to all of the aforementioned cities typically had a much more expansive cold air mass and much colder air mass available to them. Second, the cold associated with the February 2003 snowstorm was severe even for January. Third, the GFS is alone in depicting what would be an almost unprecedented snowstorm. And a major caveat applies: "Several individual case studies illustrate the model still exhibits a cold bias and may produce excessive snow in the medium range." https://www.weather.gov/media/notification/pns19-09gfs_v15_1.pdf

From the memo announcing planned implementation of the FV3-based GFS: Several individual case studies illustrate the model still exhibits a cold bias and may produce excessive snow in the medium range. https://www.weather.gov/media/notification/pns19-09gfs_v15_1.pdf Nevertheless, the decision was made to go ahead despite the persistence of that issue. Bad decision making leads to bad outcomes. Bad outcomes can lead to a loss of confidence in the model's medium- and extended-range solutions. At present, medium-range and extended-range cold shots and snowstorms on the GFS are suspect when no other support is present. It should also be noted that the kind of cold that was present for the March 1956 (2nd snowstorm), 1960, and 1993 snowstorms (magnitude and expanse) is not shown on the GFS, which forecasts a storm similar to those with 10" or more snow in NYC and Boston in the Day 8-9 timeframe. The combination of an absence of support for such a storm and lack of sufficient cold argues strongly that that GFS forecast for a major March snowstorm is likely the product of the bias(es) that remained in place.

Today saw readings again reach much above normal levels. In the South, Atlanta picked up 0.49" rain today bringing its year-to-date precipitation to 20.03". That's the highest figure so early in the year. The previous record was 19.38" in 1990. Cooler air could return for a short time late in the week. As that happens, a storm could develop and deepen along the cold front with the potential to bring a little snow to parts of the Northeast, especially eastern New England where a measurable snowfall is possible. Recent model runs have kept the storm farther offshore lowering the probability of snowfall. Afterward, there is growing support on the guidance for an extended period of warmer than normal conditions. Based on the historic data, no significant snowfalls (6" or more) are likely in the major Middle Atlantic cities (Baltimore, New York City, Newark, Philadelphia, and Washington, DC) through the remainder of the 2019-2020 snow season. There is a greater but still low probability for Boston to see such a snowstorm. The ENSO Region 1+2 anomaly was +1.0°C and the Region 3.4 anomaly was +0.5°C for the week centered around February 26. For the past six weeks, the ENSO Region 1+2 anomaly has averaged +0.37°C and the ENSO Region 3.4 anomaly has averaged +0.42°C. Neutral ENSO conditions will likely prevail through March. The SOI was +2.11 today. Today, the preliminary Arctic Oscillation (AO) figure was +2.992. No significant stratospheric warming is likely through March 11, but the upper stratosphere above 5 mb will likely be warm. Wave 2 activity will likely be low. Overall, most of the stratosphere is forecast to remain cold on the EPS into the middle of the second week of March. On March 2, the MJO was in Phase 3 at an amplitude of 1.774 (RMM). The March 1-adjusted amplitude was 1.496. A sizable majority (>80%) of years during which the AO has been, on average, strongly positive during the first 15 days of February were followed by a warmer than normal March. The preliminary February 1-15 AO average was +2.758. Only 1989 (+3.336) and 1990 (+2.948) had higher AO averages during this period. Recent rapid warming in ENSO Region 1+2 has also typically preceded a warmer than normal March and spring in the Middle Atlantic region. The most recent C3S multi-system blend favors a somewhat warmer than normal spring in the region. A warmer than normal March and spring remain the base case. Based on sensitivity analysis applied to the latest guidance, there is an implied 74% probability that New York City will have a warmer than normal March.

Temperatures rose well into the 50s today in the Middle Atlantic region. During mid-week, high temperatures could reach the 60s as ridging peaks along the East Coast at more than 150 meters above normal. Cooler air could return for a short time late in the week. As that happens, a storm could develop and deepen along the cold front with the potential to bring some snow to parts of the Northeast, especially eastern New England where a measurable snowfall is possible. Afterward, there is growing support on the guidance for an extended period of warmer than normal conditions. Based on the historic data, no significant snowfalls (6" or more) are likely in the major Middle Atlantic cities (Baltimore, New York City, Newark, Philadelphia, and Washington, DC) through the remainder of the 2019-2020 snow season. There is a greater but still low probability for Boston to see such a snowstorm. The ENSO Region 1+2 anomaly was +1.0°C and the Region 3.4 anomaly was +0.5°C for the week centered around February 26. For the past six weeks, the ENSO Region 1+2 anomaly has averaged +0.37°C and the ENSO Region 3.4 anomaly has averaged +0.42°C. Neutral ENSO conditions will likely prevail through March. The SOI was -7.80 today. Today, the preliminary Arctic Oscillation (AO) figure was +2.814. No significant stratospheric warming is likely through March 10, but the upper stratosphere above 3 mb will likely be warm. Wave 2 activity will likely diminish during the first week of March. Overall, most of the stratosphere is forecast to remain cold on the EPS into the start of the second week of March. On March 1, the MJO was in Phase 3 at an amplitude of 1.496 (RMM). The February 29-adjusted amplitude was 0.971. A sizable majority (>80%) of years during which the AO has been, on average, strongly positive during the first 15 days of February were followed by a warmer than normal March. The preliminary February 1-15 AO average was +2.758. Only 1989 (+3.336) and 1990 (+2.948) had higher AO averages during this period. Recent rapid warming in ENSO Region 1+2 has also typically preceded a warmer than normal March and spring in the Middle Atlantic region. The most recent C3S multi-system blend favors a somewhat warmer than normal spring in the region. A warmer than normal March and spring remain the base case. Based on sensitivity analysis applied to the latest guidance, there is an implied 72% probability that New York City will have a warmer than normal March.

For those who are interested, Coronavirus/COVID-19 Fact Sheets can be found here: https://www.cdc.gov/coronavirus/2019-ncov/communication/factsheets.html

For those keeping track of the coronavirus, here's a link to a dashboard created by Johns Hopkins University's Center for Systems Science and Engineering: http://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6

Historic warmth in Moscow for winter 2019-2020:

Milder air began overspreading the region today after a brisk start. Temperatures could surge into the 60s during mid-week as ridging peaks along the East Coast at more than 150 meters above normal. Cooler air could return for a short time late in the week. As that happens, a storm could develop and deepen along the cold front with the potential to bring some snow to parts of the Northeast, especially eastern New England where a measurable snowfall is possible. Afterward, there is growing support on the guidance for an extended period of warmer than normal conditions. Based on the historic data, no significant snowfalls (6" or more) are likely in the major Middle Atlantic cities (Baltimore, New York City, Newark, Philadelphia, and Washington, DC) through the remainder of the 2019-2020 snow season. There is a greater but still low probability for Boston to see such a snowstorm. The ENSO Region 1+2 anomaly was +0.3°C and the Region 3.4 anomaly was +0.6°C for the week centered around February 19. 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 +0.40°C. Neutral ENSO conditions will likely prevail through March. The SOI was -17.47 today. Today, the preliminary Arctic Oscillation (AO) figure was +1.991. No significant stratospheric warming is likely through March 9, but the upper stratosphere above 3 mb will likely be warm. Wave 2 activity will likely diminish during the first week of March. Overall, most of the stratosphere is forecast to remain cold on the EPS into the start of the second week of March. On February 29, the MJO was in Phase 4 at an amplitude of 0.951 (RMM). The February 28-adjusted amplitude was 0.439. A sizable majority (>80%) of years during which the AO has been, on average, strongly positive during the first 15 days of February were followed by a warmer than normal March. The preliminary February 1-15 AO average was +2.758. Only 1989 (+3.336) and 1990 (+2.948) had higher AO averages during this period. Recent rapid warming in ENSO Region 1+2 has also typically preceded a warmer than normal March and spring in the Middle Atlantic region. The most recent C3S multi-system blend favors a somewhat warmer than normal spring in the region. A warmer than normal March and spring remain the base case. Based on sensitivity analysis applied to the latest guidance, there is an implied 70% probability that New York City will have a warmer than normal March.

There was no evidence of the fabled “March lion” today. After a cold start, temperatures rose into the lower 40s under bright sunshine and a cloudless sky.

March 13-15, 2018 would probably be a better case. March 2001 saw 3.5" in NYC. The March 2018 storm dumped 14.9" in Boston with just a trace in NYC and 0.3" in Newark.

This might be useful for illustrating the magnitude of the GFS's cold bias. For comparison, you noted that the 6z GFS has a 3/1-17 mean temperature of 43.0 degrees for NYC. The multi-model blend I use in my sensitivity analyses shows 44.4 degrees for the same period.

During the second half of March, the EPO is much less a factor due to the shorter wave lengths. For example, during the March 16-31, 1981-2019 period, New York City had a mean temperature of 45.1°. During that period, the mean temperature for EPO cases < 0 was 45.0° and the average for those with an EPO of -1.000 or below was 44.7°.

Despite today's chill, New York City is on track to finish February with a mean temperature of 40.1°. That would tie 1954 for the 7th warmest February on record. 2020 was also the fourth year since 2010 where the monthly mean temperature was 40.0° or above and the 10th such year on record. February 2020 also saw near record to record low monthly snowfall in much of the Middle Atlantic and southern New England areas. Monthly snowfall totals through 4 pm included: Allentown: Trace (tied monthly record low) Boston: 0.5" (9th February case with < 1" snowfall and 3rd since 2010) Bridgeport: Trace (tied monthly record low) Islip: 0.0" (old record: Trace) New York City-Central Park: Trace (record: 0.0", 1998) New York City-JFK: Trace (tied monthly record low) New York City-LGA: Trace (tied monthly record low) Newark: Trace (tied monthly record low) Philadelphia: Trace (tied monthly record low) Providence: Trace (tied monthly record low) Starting tomorrow milder air will move into the region. Temperatures could surge into the 60s during the middle of next week as ridging peaks along the East Coast at more than 150 meters above normal. Based on the historic data, no significant snowfalls (6" or more) are likely in the major Middle Atlantic cities through the remainder of the 2019-2020 snow season. There is a greater but still low probability for Boston to see such a snowstorm. The ENSO Region 1+2 anomaly was +0.3°C and the Region 3.4 anomaly was +0.6°C for the week centered around February 19. 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 +0.40°C. Neutral ENSO conditions will likely prevail through March. The SOI was -16.14 today. Today, the preliminary Arctic Oscillation (AO) figure was +1.557. The preliminary February average was +3.231. Only 1990 (+3.402) and 1989 (+3.279) had February averages of +3.000 or above. The preliminary winter average for the AO was +1.960. 82% days were positive, including 31% that had a figure of +3.000 or above. The exceptional strength of the polar vortex and AO+ highlighted the uncertain nature of linkage, if it exists at all, between a solar minimum and Arctic blocking. No significant stratospheric warming is likely through March 8, but the upper stratosphere above 3 mb will likely be warm. Wave 2 activity will likely diminish during the first week of March. Overall, most of the stratosphere is forecast to remain cold on the EPS into the start of the second week of March. On February 28, the MJO was in Phase 3 at an amplitude of 0.434 (RMM). The February 27-adjusted amplitude was 0.226. A sizable majority (>80%) of years during which the AO has been, on average, strongly positive during the first 15 days of February were followed by a warmer than normal March. The preliminary February 1-15 AO average was +2.758. Only 1989 (+3.336) and 1990 (+2.948) had higher AO averages during this period. Recent rapid warming in ENSO Region 1+2 has also typically preceded a warmer than normal March and spring in the Middle Atlantic region. The most recent C3S multi-system blend favors a somewhat warmer than normal spring in the region. A warmer than normal March and spring remain the base case. Based on sensitivity analysis applied to the latest guidance, there is an implied 67% probability that New York City will have a warmer than normal March. Overall, the Middle Atlantic and southern New England regions will likely see temperatures run 3° to 5° above normal for March.

The 18z run has amnesia and doesn’t remember the snowstorm from the 12z run.

I have not seen anything publicly posted toward that end. My guess from everything I have seen is that a sort of sunk cost fallacy applies. The decision to push ahead with the FV-3 core is irrevocable.

The 2/29/2020 12z GFS shows what would be a rare March snowstorm where Boston, New York City, and Philadelphia would pick up 8" or more snow. Since 1890, there have been just six such snowstorms. The composite 500 mb and temperature anomaly data for those cases illustrates that the solution shown on the 12z GFS is very unlikely to verify. In addition, the historic data based on cases when New York City had less than 6" seasonal snowfall and Philadelphia had less than 2" seasonal snowfall through the end of February argue strongly against such a scenario. None of those cases featured the kind of snowstorm shown on the GFS. In the end, the GFS scenario is likely a product of the significant cold bias that has plagued that model's extended range forecasts, a bias that has been particularly notable since the FV-3 core was introduced, delayed, and then implemented.

DCA _ NYC _ BOS __ ORD _ ATL _ IAH ___ DEN _ PHX _ SEA 4.1 4.3 4.5 5.0 3.0 1.5 1.0 0.0 -1.1

Not in NYC. March 1990 had a mean temperature of 45.1°. March 2012 had a mean temperature of 50.9° (just below the 51.1° record from 1945).

I expect a warmer to much warmer than normal March (probably 3°-4° above normal overall) with below to much below normal snowfall. I don't think March will quite reach the extremes seen in such cases as 1945 or 2012. In terms of probabilities, my guess is that 2012-type warmth would have about a one-in-four chance this March (exceptionally high relative to climatology, but still not the most likely outcome).

Back on February 17, there was chatter that 1960 was an analog for the upcoming March. That's an example of the kind of extreme suggestions that periodically show up on Social Media despite a lack of evidence (the most common extreme idea remains calls for imminent sudden or major stratospheric warming events). For illustrative purposes, I posted maps showing that the area of cold for the first half of February was much less expansive than it was during the same timeframe in 1960 and that there was an absence of a mechanism to deliver the cold air into the CONUS for an extended period. These were two enormous and highly visible "red flags" that argued against the kind of severe cold seen during March 1960 e.g., monthly average temperature of 33.3° in New York City where the coldest March since then was 1984 with a mean temperature of 36.7°. Nationwide, March 1960 was the 3rd coldest March on record in the contiguous United States with a mean temperature nearly 5.6° below the 1901-2000 average. February 1960 was also the 18th coldest February on record in the contiguous United States. In terms of the Arctic Oscillation (AO), the February 16-29, 1960 average was -2.566. The preliminary February 16-28, 2020 average is +3.906. Put another way, had one asked, "What would it take to produce a March outcome similar to that in 1960?" one would have been hard-pressed to invoke such a comparison. With February nearing a conclusion, it's useful to examine and compare how the second half of February has fared relative to that in 1960. Spoiler alert: there's no meaningful similarity in terms of the development of widespread and severe cold.

Those are three month average anomalies. Broad-brushing or imposing such anomalies on a single month can lead one to incorrect conclusions. The risks of error are particularly high in a month like March when the wave lengths are changing (shortening). Here's how things worked out for the first and second half of March for NYC (1981-2019): March 1-15, 1981-2019: Mean temperature: 40.2°; MJO Phase 5 (amplitude 1.000 or above): 43.0° or 2.8° warmer than the mean for the whole period March 16-31, 1981-2019: Mean temperature: 45.1°; MJO Phase 5 (amplitude 1.000 or above): 46.3° or 1.2° warmer than the mean for the whole period I don't think it's any surprise that the EPS weeklies (and I expect the evolution to continue when the newest data becomes available later today) and CFSv2 have now moved aggressively toward a warm March outcome. One need only look back to late January/early February to see the perils of ignoring or discounting critical details. Discounting what appeared likely to be an exceptionally strong AO+ regime for February led to some forecasts calling for severe and sustained cold to develop. At last word, the severe cold that was touted over social media by some never materialized. Instead, cities such as Philadelphia, New York, and Boston are about to record among their 10 warmest February cases on record. The warm February was not unexpected. It was not unforeseeable. In tennis, it would be the equivalent of an unforced error. A similar situation now exists concerning March, though the changing wave lengths make the forecast more complex. Nevertheless, it's a higher confidence than usual one. IMO, pointing out such matters is useful for learning purposes. The general lesson is that one needs evidence that is reliable for drawing sound conclusions. While such an approach won't eliminate forecasting error, it can reduce the risk of error.

Phase 5 is not cold in March (one can't simply impose a three-month temperature anomaly onto a single month). Here's the 1981-2019 data for New York City: All dates: 42.7° All Phase 5 dates: 43.7° Phase 5 dates (amplitude 1.000 or above): 44.6° It's also warmer than normal across the Middle Atlantic and southern New England areas.

Morning thoughts... 1. Prospects for a warmer to much warmer than normal March in the Middle Atlantic and New England areas have continued to increase. 2. The base case has long favored a warmer than normal March on account of such factors as the exceptionally strong polar vortex in February (which has a lingering influence on the March pattern), past strong AO+ regimes in February, and the recent warming of ENSO Region 1+2. 3. Modeling has evolved quite rapidly toward the warmer March idea (EPS weeklies and the CFSv2). Below is the evolution of the CFSv2 forecasts from February 19 through February 27 that illustrate this dramatic trend. 4. Evidence for an alternative cold scenario for March is almost non-existent. 5. March snowfall in the Middle Atlantic region remains on course to be below to much below normal.