bluewave

The higher extent than recent years around the Chukchi Sea is a result of the record +EPO vortex since the late summer.

A recent study improved MJO forecasts out to four weeks using deep learning. https://www.nature.com/articles/s41467-021-23406-3 Producing accurate weather prediction beyond two weeks is an urgent challenge due to its ever-increasing socioeconomic value. The Madden-Julian Oscillation (MJO), a planetary-scale tropical convective system, serves as a primary source of global subseasonal (i.e., targeting three to four weeks) predictability. During the past decades, operational forecasting systems have improved substantially, while the MJO prediction skill has not yet reached its potential predictability, partly due to the systematic errors caused by imperfect numerical models. Here, to improve the MJO prediction skill, we blend the state-of-the-art dynamical forecasts and observations with a Deep Learning bias correction method. With Deep Learning bias correction, multi-model forecast errors in MJO amplitude and phase averaged over four weeks are significantly reduced by about 90% and 77%, respectively. Most models show the greatest improvement for MJO events starting from the Indian Ocean and crossing the Maritime Continent.

With all the record winter warmth over the last 20 years, getting intervals of blocking and a more favorable Pacific has been key to the record snowfall. This has been true for all the various ENSO states. It’s the few years when the Pacific didn’t relax and the NAO and AO blocking was lacking that snowfall went below normal. With the record blocking patterns since 2002-2003, we have seen an unprecedented number of seasons when stations from around Newark out to Eastern LI recorded 40” or more of snow. This dramatic increase in snowfall has occurred during a steady increase in winter temperatures. Our older winters with 40”+ peak snowfall amounts used to occur with a NYC average temperature around 32°. But many recent 40”+ seasons featured a NYC average DJF temperature in the 35°-40° range. So this is a new combination of warmth and heavy snow for our area. Seasons since 1950 when at least one station from Newark to Eastern LI recorded 40”+ and the NYC DJF average temperature. 35°+ average temperature seasons bolded 20-21….36.1 17-18….36.2 16-17….39.3 15-16….41.0 14-15…31.4 13-14…32.9 12-13…36.8 10-11….32.8 09-10…33.8 08-09…34.2 05-06…37.3 04-05…35.4 03-04…32.4 02-03…31.2 00-01….33.5 95-96…32.2 93-94…31.2 86-87….34.8 77-78….30.8 68-69….32.9 66-67….34.1 63-64…33.2 60-61….31.7 57-58…33.2 55-56…32.8 Yeah, that’s why we have been getting so much record warmth before or after our best snowstorms since the super El Niño in 15-16. Last December it was 60s before and after our best December snowstorm and -AO in years. May 2020 tied for our latest trace of snow following one of our warmest winters. 17-18 featured 30” of snow in March on Long Island after Newark hit 80° in late February. We had the blizzard in February 2017 a day after the 60s. 15-16 went +13.3 in December followed by the 30” snowstorm in January and new #1 NYC snowstorm. Even before the super El Niño, we were getting occasional patterns like this. Nemo in February 2013 following the very warm first few months of winter. The 11-12 lack of winter after the record snowstorm in late October. The February 2006 NYC #2 snow following one our warmest Januaries on record. The late winter 2005 snowstorms following one of the warmest first 2 weeks of January on record. This is a sea level version of spring in the Rockies where snow and warmth have often occurred together.

They didn’t ignore the NAO in their forecasts. But a forecast issued on November 8th for the 22-29 is going to miss a few things. Notice how good a job it did with the colder signal for the end of the month. So I would consider that EPS forecast very good for such a long range forecast. Just understand that it underestimated the big +EPO trough in Alaska. So if we don’t have a -NAO and the long range EPO verifies more positive, then the temperatures could beat expectations to the upside. That’s why you have to be careful when dealing with a record Pacific Jet and long range forecasts. Maybe someday the models will have AI built in and the forecast limit near 2 weeks can get extended further out in time.

The point about the Euro underestimating the +EPO long range is correct. But the EPS always had a colder +PNA -WPO pattern to end the month. It was never forecasting a torch for us. The reason this week was a little colder than the November 8th forecast was due to the stronger -NAO. Notice the stronger +EPO trough near Alaska than the EPS had back on November 8th. EPS forecasts from November 8th Nov 22-29 +EPO is stronger and -NAO more negative

The models can only do so much when we have seen record North Pacific Jet activity this fall. You posted the other day how this will be the most +EPO fall. I posted the record winds in that area during the fall a few days ago. So we see a model history of underestimating the Pacific Jet long range. As to my post yesterday, we have been in a -WPO pattern this month. And that is what the forecasts have been showing going forward. But if that Aleutians ridge is weaker or fades faster, that means the long range will correct warmer. That’s why getting a poleward Aleutians ridge is so important.

The JFK ASOS is in a marshy area that was filled close to Jamaica Bay and the 5 towns section of SW Nassau.

Northern Queens is right on the water which is a heat source along with one of the strongest UHI zones in the city. So they get a double whammy of warming. LGA temperatures are very close to the surrounding micronet sites in Northern Queens. The LGA ASOS is on a grassy strip larger than most residences in that area. So the average backyard sensor in that area would probably be closer to concrete. Also notice how close it is to the water when the flow is NW, N, or NE.

With these endless summer patterns lasting into October, NYC had a much earlier freeze relative to the means than POU did. Places that normally have an October freeze have been very late. But the delayed falls till November have allowed places like NYC to get their freeze closer to the recent averages. So the first freeze dates have been getting later faster in interior portions of the Northeast that used to have earlier fall freezes in the past. Latest first freezes 2nd latest date at POU and only 19th latest in NYC Frost/Freeze Summary for Poughkeepsie Area, NY (ThreadEx) Each section contains date and year of occurrence, value on that date. Click column heading to sort ascending, click again to sort descending. Year Last Value First Value Season Length 1995 - - 11-05 (1995) 32 - 1971 04-24 (1971) 30 11-05 (1971) 30 194 2021 04-06 (2021) 32 11-03 (2021) 29 210 1975 04-22 (1975) 25 10-30 (1975) 31 190 2007 04-21 (2007) 32 10-29 (2007) 27 190 1942 04-14 (1942) 27 10-29 (1942) 32 197 2011 04-22 (2011) 27 10-28 (2011) 28 188 2013 05-14 (2013) 30 10-24 (2013) 31 162 2005 05-13 (2005) 32 10-24 (2005) 31 163 1967 05-13 (1967) 30 10-23 (1967) 28 162 1998 - - 10-22 (1998) 32 - 1955 05-10 (1955) 30 10-22 (1955) 29 164 2014 04-25 (2014) 31 10-20 (2014) 30 177 1990 04-19 (1990) 27 10-20 (1990) 30 183 1973 04-30 (1973) 32 10-19 (1973) 31 171 2018 04-23 (2018) 28 10-18 (2018) 32 177 1948 05-02 (1948) 32 10-18 (1948) 32 168 2017 05-11 (2017) 32 10-17 (2017) 30 158 2015 04-26 (2015) 30 10-17 (2015) 29 173 Frost/Freeze Summary for NY CITY CENTRAL PARK, NY Each section contains date and year of occurrence, value on that date. Click column heading to sort ascending, click again to sort descending. Year Last Value First Value Season Length 1998 03-23 (1998) 32 12-22 (1998) 22 273 2001 03-28 (2001) 30 12-16 (2001) 32 262 1948 04-04 (1948) 32 12-11 (1948) 31 250 2011 03-29 (2011) 31 12-10 (2011) 32 255 2016 04-10 (2016) 31 12-09 (2016) 29 242 2009 03-24 (2009) 27 12-07 (2009) 32 257 1902 03-20 (1902) 31 12-05 (1902) 24 259 2010 03-27 (2010) 29 12-04 (2010) 32 251 2006 03-22 (2006) 31 12-04 (2006) 31 256 1907 04-03 (1907) 32 12-03 (1907) 30 243 1985 04-10 (1985) 28 12-02 (1985) 30 235 1963 03-23 (1963) 30 12-01 (1963) 29 252 1999 03-16 (1999) 32 11-30 (1999) 28 258 1979 04-08 (1979) 32 11-30 (1979) 29 235 1909 04-11 (1909) 24 11-30 (1909) 31 232 1952 03-18 (1952) 30 11-29 (1952) 28 255 1906 04-03 (1906) 32 11-29 (1906) 29 239 1958 04-09 (1958) 29 11-28 (1958) 32 232 2002 04-07 (2002) 30 11-27 (2002) 27 233 1991 03-31 (1991) 31 11-26 (1991) 31 239 1950 04-14 (1950) 26 11-26 (1950) 28 225 1981 03-21 (1981) 30 11-25 (1981) 31 248 1978 04-03 (1978) 31 11-25 (1978) 32 235 1941 03-31 (1941) 29 11-25 (1941) 31 238 2015 04-01 (2015) 32 11-24 (2015) 32 236 1938 04-07 (1938) 30 11-24 (1938) 21 230 1918 04-12 (1918) 32 11-24 (1918) 30 225 1898 04-07 (1898) 29 11-24 (1898) 32 230 2021 04-03 (2021) 32 11-23 (2021) 32 233

It’s a warmer NW flow off the water today at LGA since the ASOS is right near the water in the NW corner of the airport. Same reason that they are cooler than Newark and sometimes JFK during heatwaves with offshore flow. If you look at the NYC micronet, the temperatures at LGA usually match the other observations in northern Queens. CITY SKY/WX TMP DP RH WIND PRES REMARKS Central Park CLEAR 30 19 63 CALM 30.34R LaGuardia Arpt CLEAR 33 19 56 NW15G21 30.32R WCI 23 Kennedy Intl PTCLDY 31 18 58 NW16 30.33R WCI 20 Newark Liberty PTCLDY 32 19 58 NW8 30.33R WCI 25 Teterboro Arpt CLEAR 29 18 63 NW8 30.32R WCI 21 Bronx Lehman C N/A 32 19 59 N5 N/A WCI 27 Queens College N/A 32 19 59 N9 N/A WCI 24 Breezy Point N/A 32 N/A N/A NW13 N/A WCI 22 Brooklyn Coll N/A 32 19 59 NW12 N/A WCI 23 Staten Island N/A 32 21 64 NW8 N/A WCI 25

As long as NYC gets 3” or more in a coupled La Niña December since the 1995, it usually does well in the seasonal snowfall department. It’s the few years like 2011-2012, 2007-2008, 1999-2000, and 1998-1999 that came up short. Coupled La Ninas since 1995 bolded Monthly Total Snowfall for NY CITY CENTRAL PARK, NY Click column heading to sort ascending, click again to sort descending. Year Oct Nov Dec Jan Feb Mar Apr Season 2020-2021 0.0 0.0 10.5 2.1 26.0 T 0.0 38.6 2019-2020 0.0 0.0 2.5 2.3 T T T 4.8 2018-2019 0.0 6.4 T 1.1 2.6 10.4 0.0 20.5 2017-2018 0.0 T 7.7 11.2 4.9 11.6 5.5 40.9 2016-2017 0.0 T 3.2 7.9 9.4 9.7 0.0 30.2 2015-2016 0.0 0.0 T 27.9 4.0 0.9 T 32.8 2014-2015 0.0 0.2 1.0 16.9 13.6 18.6 0.0 50.3 2013-2014 0.0 T 8.6 19.7 29.0 0.1 T 57.4 2012-2013 0.0 4.7 0.4 1.5 12.2 7.3 0.0 26.1 2011-2012 2.9 0.0 0.0 4.3 0.2 0.0 0.0 7.4 2010-2011 0.0 T 20.1 36.0 4.8 1.0 T 61.9 2009-2010 0.0 0.0 12.4 2.1 36.9 T 0.0 51.4 2008-2009 0.0 T 6.0 9.0 4.3 8.3 T 27.6 2007-2008 0.0 T 2.9 T 9.0 T 0.0 11.9 2006-2007 0.0 0.0 0.0 2.6 3.8 6.0 T 12.4 2005-2006 0.0 T 9.7 2.0 26.9 1.3 0.1 40.0 2004-2005 0.0 T 3.0 15.3 15.8 6.9 0.0 41.0 2003-2004 0.0 0.0 19.8 17.3 0.7 4.8 0.0 42.6 2002-2003 T T 11.0 4.7 26.1 3.5 4.0 49.3 2001-2002 0.0 0.0 T 3.5 T T T 3.5 2000-2001 T 0.0 13.4 8.3 9.5 3.8 0.0 35.0 1999-2000 0.0 0.0 T 9.5 5.2 0.4 1.2 16.3 1998-1999 0.0 0.0 2.0 4.5 1.7 4.5 0.0 12.7 1997-1998 0.0 T T 0.5 0.0 5.0 0.0 5.5 1996-1997 0.0 0.1 T 4.4 3.8 1.7 T 10.0 1995-1996 0.0 2.9 11.5 26.1 21.2 13.2 0.7 75.6

It was easier to get an occasional NYC December average temperature near or below freezing when the average December temperature was in the mid 30s. NYC Dec….1991-2020…..39.1…..1981-2010…37.5…..1971-2000…..37.3…..1961-1990…..36.6………+2.5 Jan……………………….33.7…………………….32.6………………………32.1………………………..31.5……..+2.2 Feb……………………….35.9…………………….35.3………………………34.6………………………33.6……..+2.3 Avg……………………….36.2…………………….35.1……………………….34.7……………………….33.9……+2.3

NYC December average temperatures since 2010 with coupled La Niña years bolded Time Series Summary for NY CITY CENTRAL PARK, NY - Month of Dec Click column heading to sort ascending, click again to sort descending. Year Mean Avg Temperature Temperature Departure 2020 39.2 +1.7 2019 38.3 +0.8 2018 40.1 +2.5 2017 35.0 -2.5 2016 38.3 +0.8 2015 50.8 +13.3 2014 40.5 +3.0 2013 38.5 +1.0 2012 41.5 +4.0 2011 43.3 +5.8 2010 32.8 -4.5

Like we are seeing this month, a -WPO and +EPO can average near to below normal. But it all depends on how the other teleconnections line up. It’s more a back and forth pattern than any record cold. You would want a -EPO for a daily cold record. But smaller cold departures can occur with a well placed -WPO. -WPO +EPO this November

Our only colder than average Decembers since 2010 were the La Ninas in 2017 and 2010. La Nina winters usually have the coldest departure month of the winter in December. This year may be easier to get a colder December since December had the biggest winter monthly temperature rise in the new 1991-2020 climate normals. So the bar could be lower for a colder December with the right pattern. But since December has seen the greatest temperature rise of the winter with the new climate normals update, factors other than La Niña and El Niño have had a big influence. So we always have to watch for this warmer December tendency to have a bigger influence than La Nina or El Niño expectations. New 1991-2020 climate normals compared to 1981-2010 EWR Dec….38.0….36.5….+1.5 Jan….32.8….31.6…..+1.2 Feb….35.1….34.6…..+0.5 Avg…..35.3….34.2….+1.1 NYC Dec….39.1…..37.5….+1.6 Jan….33.7…..32.6...+1.1 Feb….35.9….35.3….+0.6 Avg…..36.2…35.1….+1.2 LGA Dec…40.0….38.2….+1.8 Jan….34.4….32.9….+1.5 Feb….36.3….35.3….+1.0 Avg….36.9….35.5…..+1.4 JFK Dec….38.3….37.7….+0.6 Jan…..32.8….32.7….+0.1 Feb….34.5….34.9…..+0.4 Avg….35.2….35.1…..+0.1 ISP Dec….37.1…..35.6…..+1.5 Jan….31.9….30.6…..+1.3 Feb...33.3….32.8…..+0.5 Avg….34.1….33.0…..+1.1 BDR Dec…37.0….35.4….+1.6 Jan….31.4….30.1….+1.3 Feb….33.1….32.4….+0.7 Avg…..33.8…32.6….+1.2 HPN Dec…35.1….33.6….+1.5 Jan….29.8….28.3…+1.5 Feb….31.9….30.9….+1.0 Avg…..32.6….30.9…+1.4

The day 10 EPS looks more like a -WPO based near the Aleutians than a -EPO. New run Old run

We need few more corrections stronger with the Aleutians ridge to get the gradient south of us. But today is an improvement over yesterday on the GEFS. Our better La Niña winters for snowfall usually have more of a poleward extension of the Aleutians ridge. New run Old run

It’s good that the strong winds wait until a day after Thanksgiving.

This year the La Niña is having a strong influence on the atmosphere. Last winter was much more El Niño-like. Maybe next winter we can finally get a fully coupled El Niño with a raging STJ pattern. https://www.climate.gov/news-features/blogs/enso/did-northern-hemisphere-get-memo-years-la-niña

We will need the Aleutians ridge to put up a fight against the record Pacific Jet in December. Notice how the record jet anomalies have been parked over the Aleutians this fall. So models may struggle more than normal beyond 5 days with such a supercharged Pacific Jet.

Big model war on the 0z guidance for early December. The EPS and GEPS are colder with the Aleutians ridge building toward the Arctic. The GEFS is flatter and warmer with a faster Pacific Jet. That poleward Aleutians ridge matches December La Niña climatology more closely since 2000.

The colder day 15 EPS last night is a great example of how MJO sensitive the early December pattern may be. The 0z EPS weakens the MJO rather than the weeklies stronger wave. So this allows the Aleutians ridge to poke up closer to the Arctic. The newer EPS is actually closer to La Niña climo than what the weeklies snowed yesterday. So another example of how we really have to wait until December gets started to see the true atmospheric response will be. The stronger La Niña 5-6 phases usually hold off until after December. New colder run Weaker MJO Older weeklies stronger MJO

As long as NYC gets 3” or more in a coupled La Niña December since the 1995, it usually does well in the seasonal snowfall department. It’s the few years like 2011-2012, 2007-2008, 1999-2000, and 1998-1999 that came up short. Coupled La Ninas since 1995 bolded Monthly Total Snowfall for NY CITY CENTRAL PARK, NY Click column heading to sort ascending, click again to sort descending. Year Oct Nov Dec Jan Feb Mar Apr Season 2020-2021 0.0 0.0 10.5 2.1 26.0 T 0.0 38.6 2019-2020 0.0 0.0 2.5 2.3 T T T 4.8 2018-2019 0.0 6.4 T 1.1 2.6 10.4 0.0 20.5 2017-2018 0.0 T 7.7 11.2 4.9 11.6 5.5 40.9 2016-2017 0.0 T 3.2 7.9 9.4 9.7 0.0 30.2 2015-2016 0.0 0.0 T 27.9 4.0 0.9 T 32.8 2014-2015 0.0 0.2 1.0 16.9 13.6 18.6 0.0 50.3 2013-2014 0.0 T 8.6 19.7 29.0 0.1 T 57.4 2012-2013 0.0 4.7 0.4 1.5 12.2 7.3 0.0 26.1 2011-2012 2.9 0.0 0.0 4.3 0.2 0.0 0.0 7.4 2010-2011 0.0 T 20.1 36.0 4.8 1.0 T 61.9 2009-2010 0.0 0.0 12.4 2.1 36.9 T 0.0 51.4 2008-2009 0.0 T 6.0 9.0 4.3 8.3 T 27.6 2007-2008 0.0 T 2.9 T 9.0 T 0.0 11.9 2006-2007 0.0 0.0 0.0 2.6 3.8 6.0 T 12.4 2005-2006 0.0 T 9.7 2.0 26.9 1.3 0.1 40.0 2004-2005 0.0 T 3.0 15.3 15.8 6.9 0.0 41.0 2003-2004 0.0 0.0 19.8 17.3 0.7 4.8 0.0 42.6 2002-2003 T T 11.0 4.7 26.1 3.5 4.0 49.3 2001-2002 0.0 0.0 T 3.5 T T T 3.5 2000-2001 T 0.0 13.4 8.3 9.5 3.8 0.0 35.0 1999-2000 0.0 0.0 T 9.5 5.2 0.4 1.2 16.3 1998-1999 0.0 0.0 2.0 4.5 1.7 4.5 0.0 12.7 1997-1998 0.0 T T 0.5 0.0 5.0 0.0 5.5 1996-1997 0.0 0.1 T 4.4 3.8 1.7 T 10.0 1995-1996 0.0 2.9 11.5 26.1 21.2 13.2 0.7 75.6

Leaf clean ups will probably run into early December around the area.

Yeah, these delayed falls have become the new normal. The first one I remember seeing this late was 2007. Beautiful fall color now in Central Park.