bluewave

With the forcing in multiple location on the VP anomalies charts, the early to mid January forecast isn’t the canonical MJO 8-2 where the coldest temperatures are usually found in the Northeast. This time the coldest temperatures will be displaced further to the south and west. Could also be related to the record low sea ice around Hudson Bay with the warmth there.

This may turn into a game of how much spacing we get between the individual shortwaves with so many systems in the fast Northern Stream flow and strong 50/50 low.

Even slightly colder Decembers like this one are still close to 40° around NYC since 2011. Monthly Mean Avg Temperature for LAGUARDIA AIRPORT, NY Click column heading to sort ascending, click again to sort descending. Mean 41.4 41.4 2024 38.6 38.6 2023 44.3 44.3 2022 38.7 38.7 2021 45.0 45.0 2020 40.9 40.9 2019 38.9 38.9 2018 40.5 40.5 2017 36.2 36.2 2016 40.3 40.3 2015 50.8 50.8 2014 40.6 40.6 2013 38.2 38.2 2012 42.7 42.7 2011 43.3 43.3

When you read the study they recently published they only mention some improvement day 3-5 over the NWP models. https://deepmind.google/discover/blog/gencast-predicts-weather-and-the-risks-of-extreme-conditions-with-sota-accuracy/ https://www.nature.com/articles/s41586-024-08252-9 The largest improvements of GenCast are often at shorter lead times up to around 3–5 days,

Boxing Day 2010 was the last time we saw a heavy snowstorm after December 20th. Part of the issue is that the 2nd half of December has been warming faster than the 1st half. The other challenge has been the lack of solid benchmark storm tracks this time of year. Time Series Summary for NY CITY CENTRAL PARK, NY Click column heading to sort ascending, click again to sort descending 12-21 ro 12-31 snowfall 2024-12-31 2.8 2 2023-12-31 0.0 0 2022-12-31 T 0 2021-12-31 0.2 0 2020-12-31 0.0 0 2019-12-31 T 0 2018-12-31 T 0 2017-12-31 0.7 0 2016-12-31 T 0 2015-12-31 T 0 2014-12-31 T 0 2013-12-31 T 0 2012-12-31 0.4 0 2011-12-31 0.0 0 2010-12-31 20.0 0 2009-12-31 1.5 0 2008-12-31 0.5 0 2007-12-31 T 0 2006-12-31 0.0 0 2005-12-31 0.4 0 2004-12-31 2.4 0 2003-12-31 0.0 0 2002-12-31 5.0 0 2001-12-31 0.0 0 2000-12-31 12.8 0

The standalone AI models so far aren’t very good. We can remember the runs when they first launched a while back showing the hurricane going into New England. Then the system wound up curving well to the east. The one use of AI that may show some promise is using in a hybrid form where it corrects the NWP model biases.

I think the ultimate responsibility is on the modeling centers like ECMWF and NCEP for making operational model runs available beyond 120 to 168 hrs. We were joking a few months ago how extending the ECMWF OP to 360 hrs would go over. Once a product is out in the public domain there is nothing to stop someone from posting the image. We know that ensembles are the way to go beyond 120hrs to 168hrs and so do organizations like the NWS. Yet the modeling centers continue to run these OP runs out to 360 hrs and 384 hrs. It would be great if the only guidance products made available beyond 168hrs were the ensembles.

The one significant La Niña sign this December has been the much stronger than normal Northern Stream of the Pacific Jet. It lead to the ridge axis getting pushed further east than forecast with warmer temperatures than issued back on December 5th. Plus the dominant storm track through the Great Lakes was also classic La Niña. Very unusual to get such a strong Northern Stream with a +PNA. So even more of a La Niña mismatch December than we have seen on past occasions. Euro monthly forecast Verification

The mid 60s in NJ yesterday turned out to be a top 5 warmest for December 26th to 31st. Time Series Summary for NEWARK LIBERTY INTL AP, NJ Click column heading to sort ascending, click again to sort descending…warmest Dec 26th to 31st 1 1984-12-31 71 0 2 1982-12-31 68 0 3 2008-12-31 66 0 4 2024-12-31 65 2 5 1964-12-31 64 0 - 1936-12-31 64 0

We have had a dominant cutter, hugger, and suppression storm for most of the time since February 2022. So it may take some time to get a coastal track closer to the sweet spot for a 4”+ or 6”+ snow near NYC. You can see all 3 models with the windshield wiper effect from run to run the last 3 days. So it may just take some patience for something to click for us. Since there Northern Stream is still so active wave spacing and suppression could initially be an issue for us. I wouldn’t get invested on any individual system until 2-3 sets of models show it under 120 hrs. Plus we could always get a system that tracks too far north across the area and brings mixing issues near the coast if the vort ejecting from the West kicks up a transient ridge ahead of it.

The only thing I lean on is what I think the actual pattern will be. But colder patterns have been few and far between over the last decade. So it may seem to you that me reporting this is a bias. But it’s the actual pattern which has been biased warm. When we have had really cold patterns over this interval I was there.

That isn’t correct. I had a back and forth temperature pattern for this month. The only warm up that I highlighted back in late November and early December was the 55°+ day which occurred on the 17th. All I said about the cooldown before Christmas was that it probably wouldn’t be a cold as the one in 2022 for NYC which was correct. But commented on the similar timing both years. Then I was discussing that the warm up this week would probably wait until closer to the New Year. I said probably over 50° but it came in a little earlier and was around 10° warmer in spots. My guess for this month for the 7 station average was -1 to +1. So my forecast for this month wasn’t that warm compared to what the final will be. The reason you think I may lean too warm is related to the fact that most of the time we have actually been much warmer than normal. Since portions of the northeast have had over 50 top 10 warmest months since 2010 to only 1 top 10 coldest month. So it’s the actual pattern which has been warm and I have just been describing what has been happening.

We can see a back and forth shift from run to run with set ups like this. The good news is that the AO blocking has corrected much stronger in recent days. So it will probably take time for the Pacific flow to back off enough for a nice East Coast snowstorm. But we often have to be patient in situations like this since it’s hard to tell which shortwave will be the first to establish the favorable storm track.

The ridge today was a little weaker over Oregon with the fast Pacific flow lowering heights from yesterday for the storm on the 6th. So the storm ends up further south today under the block. We probably have to be patient with patterns like this since it can take time for the Pacific and Atlantic get in sync for a nice east coast storm. New run Old run

It may take some time for the fast Northern Stream to back off a bit. We saw it was a big player in December. The EPS is just a few degrees too far east with the systems. So it has an active storm track into British Columbia and just off the East Coast. Slow the Pacific flow just a bit and we will be in business here.

The breaks of sun allowed Central Park to jump up to 61°.

That’s why we get such big variance between the collection of OP Euro, GFS, and CMC runs beyond 120 hrs. The ensemble means try to smooth out the spacing and amplitudes of each individual shortwaves. So ensemble means probably are good starting point for us.

NAO and AO volatility continue to be the big story since the summer. Big swings both using 500mb anomalies and the raw teleconnection indices. Very challenging to do a long range forecast of these indices with such record swings. We have seen unprecedented volatility since 2009-2010. New EPS run first week of January Old run

This rapid increase in AO blocking over the last 10 days of runs may be getting and assist from the record warmth around Hudson Bay with the lowest ice there on record for this time of year. New EPS run first week of January Old run

Low 60s today from CNJ to Newark with 50s most other spots.

It’s patterns like this that make us want to see OP runs only be made available through 120-168 hrs. Everything beyond that point should just be ensemble means. People posting these individual long range OP snowfall charts on social media will actually cause issues for the NWS. We saw this several years ago when some long range OP snowfall charts were posted around Philly and the NWS had to issue a statement due to a poster on social media. I think it’s the responsibility of the modeling centers to do this since once a model forecast is in the public domain it’s going to be posted by somebody.

Yeah, it can snow pretty much anytime in January or February regardless of the monthly average temperatures with the right storm track. We got this with the snow near the 80° day in February 2018. Then again multiple times in winters like 16-17 with blizzards the day after 60s record warmth.

Storm track is more important than average temperatures for NYC snowfall in January. Since it can always snow even when it’s warm this time of year with the right storm track with the average temperatures most Januaries.The temperatures only matter on the day of the event along with the storm track. NYC January average temperatures and snowfall last 10 years 2024….37.0°….2.3” 2023….43.5°…T 2022….30.3°…15.3” 2021….34.8°…2.1” 2020….39.1°…2.3” 2019….32.5°…1.1” 2018….31.7°….11.2” 2017….38.0°…7.9” 2016….34.5°…27.9” 2015….29.9°…16.9”

The OP and ensemble mean are different sides of the same coin. One is overamped on the northern stream and the other is a weaker system moving through in the fast flow. Hopefully, we can slow the Northern Stream influence enough to allow us to maximize the window of opportunity we get this month.

From JFK to ISP and Shirley it’s in the low 50s.