GaWx

Today’s Euro Weeklies aren’t what you want to see if you don’t want them to be warmer than yesterday: the most intense cold on the run is W and SW of Chicago and perhaps near ME. I personally hate them because the colder the better and I hate sweating in winter when I take my walks. Note that the warmest anomalies in the US are in the deep SE: Dec 1-7: very slightly warmer/strong SER: Dec 8-14: slightly warmer/stronger SER lingering: Dec 15-21: warmer/SER won’t give up Dec 22-28: SER finally weakened but slightly warmer NE vs yesterday Dec 29-Jan 4: SER staying weak but slightly warmer NE vs yesterday:

DJF MJO last 15 winters (2010-1 through 2024-5): # of days each phase *7: 286 days from 58 phase 7s or 4.9 days/phase *8: 95 days from 34 phase 8s or 2.8 days/phase 1: 76 days from 24 phase 1s or 3.2 days/phase 2: 96 days from 23 phase 2s or 4.2 days/phase 3: 167 days 4: 173 days 5: 212 days 6: 249 days ——————— Per the above data, the # of phase 8 days for the last 15 winters, combined, has been only 1/3 the # of phase 7 days. This was due to a combo of much fewer phase 8 periods than phase 7 periods (34 vs 58) and much shorter avg. duration for 8 vs 7 (~3 days vs ~5 days). So, there sometimes is a struggle in going from phase 7 to 8. The # of combined 8-1-2 days has been ~# of phase 7 days, alone! The coldest E US phases have been 8, 1, and 2 while the warmest have been 4-6 (see bottom). The # of phase 8-1-2 days, combined, has only barely been higher than the # of phase 6 days, alone! So, as we look at Dec MJO progs, keep in mind the relative difficulty in getting as many phase 8 days (as well as each of phases 1 and 2, for that matter) as phase 7 days (if any at all). Just because models suggest a 2+ week long phase 7 is very likely doesn’t necessarily mean the durations of each of phases 8, 1, and 2 immediately following will be anywhere near as long, if they even occur. Here are the 10+ day long phase 7s since 2010-1 along with the subsequent phase 8-1-2: - 1/14-30/13: 17 days followed by a combined 15 days in 8-1-2 - 2/19-29/2016: 11 days followed by a combined 10 days in 8-1-2 - 2/2-18/2018: 17 days followed by a combined 10 days in 8-1-2 - 2/7-21/2021: 15 days followed by 0 days in 8-1-2 and instead a combined 11 days in 6-5-6-7 - 12/19/2021-1/9/2022: 22 days followed by only 4 days in phase 8 before going back to 7; phases 7-6-5-4-3 then dominated the next 3.5 weeks. - 1/1-10/2023: 10 days followed by a combined 9 days in 8-1-2 - 2/16-28/2023: 13 days followed by 1 day in phase 6 and then right back to phase 7 for 5 more days…so, 19 days in phase 7 for all practical purposes…followed by a combined 19 days in 8-1-2 - 1/28-2/6/2024: 10 days followed by 2 days in phase 6 and then right back to phase 7 for 10 more days….so, 22 days in phase 7 for all practical purposes…followed by only 1 day in phase 8, which was then followed by a combined 31 days in 7-5-6-7-3-4-5-6 ——————— Analysis of the above: - Much more frequent 10+ day long phase 7 last 5 winters with 5 of them vs only 3 during the previous 10 winters! - The combined 8-1-2 duration was never longer than the prior phase 7 length (using the “for all practical purposes” phase 7 length in two cases). - Of the 8 long duration phase 7s, two (25%) of the following 8-1-2 periods lasted a notable length of 15-19 days and three (38%) lasted a mediocre 9-10 days. However, the other three (38%) were only a minuscule 4 or fewer days long (4, 1, 0) and these were all within the last 5 winters. - So, whereas there’s a decent chance to get a higher end duration of 15-19+ days in 8-1-2 following the upcoming 7, the chance doesn’t appear to be very high and the chance of getting only a very disappointing short period is probably at least about as high. - Thus, recent history tells us that the upcoming 2nd half of Dec MJO is practically un-forecastable as of now with dominance by the often cold 8-1-2 about equally as probable as a feared hardly any 8-1-2. @donsutherland1@bluewave ——————— Daily MJO: https://www.bom.gov.au/clim_data/IDCKGEM000/rmm.74toRealtime.txt

Holy cow! That 82 F at KATL is also the warmest so late in the season by 6 days! They missed the warmest for all of Nov by only 2 F! Here it wasn’t a record, but it was the 4th day in a row with low 80s. Bring back autumn! It will be arriving just in time to give us a cool Thanksgiving!

Hopefully the MJO won’t turn sharply left along the 7/8 border like these: Dec of 2014 was mild: Dec-Jan 2019-20 was mild:

Whereas there can be a lag, the diagrams by phase are based on concurrent temps. So, phase 8 being the coldest in Dec is based on temps during phase 8 being the coldest of any phase rather than with a lag. Perhaps, it’s phase 7 in some cases being the cold generator that takes til phase 8 to show up in the E US.

12/9-14/91 also was a mild strong phase 8 in the E US though it cooled afterward:

The Weeklies have it mainly mild in the SE first 3 weeks of Dec, not the NE, thanks to a SE ridge. Anthony asked where’s the warmth? The maps he posted had SE warmth.

You don’t see any warmth, Anthony?

Why? I’m not going crazy. We look at long range guidance ITT every day. This is mainly a long range thread. You just posted weeks 3-4, for example. I’d rather have the most up to date.

Today’s Euro Weeklies 2m temp changes vs yesterday aren’t good for the E US overall if you prefer cold: Dec 1-7: warmer SE to Mid-Atlantic due to even stronger SE ridge: Dec 8-14: ~unchanged Dec 15-21: warmer SE/very slightly warmer NE Dec 22-28: slightly warmer Mid-Atlantic/very slightly warmer SE: Dec 29-Jan 4: warmer NE

Today’s Euro Weeklies update on the SPV compared to last run keeps the reversal (or near reversal) but it’s a fair bit stronger Dec-early Jan. Keep in mind that the prior two runs were already significantly stronger for that period: Today’s EW mean zonal wind at 10 mb (11/21): significantly stronger than 11/18 run with, for example, 12/12 at +29 vs +11 on 11/18 run: EW run from just 3 days ago (11/18): 12/12 had been at only +11

To add: -1983 doesn’t fit -1985 strong 2nd half of Oct -1989 was fairly stout last 1/3 of Oct -2000 wasn’t strong but was stronger than avg

The 6 coldest La Niña to neutral ENSO Decembers in the E US as a whole since 1974 have been 1983, 1985, 1989, 2000, 2005, and 2010: Here are the respective Dec MJOs: 1983: mainly inside COD or just outside/no strong 1985: almost all inside COD 1989: most of 8/1 just outside COD or inside it/no strong 2000:almost all just outside or inside COD 2005: all inside COD 2010: all just outside or in COD What do these 6 have in common regarding the MJO? No strong and most days just outside to inside COD. Below are the extended progs that have the Dec 1-21 MJO: look how strong they are, especially GEFS but even the Euro is solid compared to the six cold Decembers posted. My concern is that the MJO amplitude will be too strong, based on history, to allow for a widespread intense cold dominated E US in Dec. Thus, I’m hoping the progs remain going into 8, but weaken. My hypothesis as to why strong MJO has had a tendency (though not always, of course) to not be as cold as weaker MJO is because the higher amp may tend to bring up too much warmth from the tropics, where the MJO is measured: Any opinions?

Domination in the SE by mild/SE ridge in the first half of Dec has been forecasted by the Euro Weeklies for over a week. They’ve been steady as a rock. I’d be very surprised if it isn’t mild in the means. After that it is up in the air though recent runs have been suggesting a weaker version of a SE ridge/slightly warmer than normal for 12/15-21 lingering for the SE. We’ll see.

There hasn’t even been one 3+ day long Dec MJO phase 8 since 2017 and the one before that was way back in 2009! So, if the models are right with their 3++ day long phase 8, it would the first time in 8 years and only the 2nd time in 15 years! A whopping nearly 80% of 3+ day long Dec phase 8 periods averaged colder than normal at Raleigh though with lots of variation. When amplitude is <1.5, that rises even further to 85% along with an average of 5.0 F BN. If it’s a pretty strong amplitude (1.5+), it actually averaged only 0.5 F BN at Raleigh. At Baltimore, it’s pretty similar for Dec phase 8 with 4.8 F BN for amp <1.5 but only 0.3 F BN for amp >1.5. Fortunately, most Dec phase 8 are <1.5 amp.

Today’s Euro Weeklies 10 mb zonal wind mean is slightly weaker early and allows most to reverse unlike yesterday. However, afterwards it looks like yesterday’s significantly stronger run: Today’s: Yesterday’s: Two days ago: significantly weaker

Thanks, Don. One thing helping 2025 vs 1983 is that 1983 didn’t have a weak SPV as it didn’t get <25 m/s through Nov and Dec:

Today’s GEFS gets to phase 7 a day earlier than yesterday’s (11/28 vs 11/29) and three days earlier than the run from 2 days ago. The 11/19/25 Euro-ext gets it to phase 8 ten days earlier (12/5) than the prior run (12/15): Prior run (11/18/25):

Based on H5 height comparisons (not H5 height anomaly comparisons), I strongly doubt that H5 temps on 11/11/2025 were as cold as those of 11/26/1950: 11/26/1950 0Z: N NC ~520 dm per upper right map 11/11/2025 6Z: N NC 528 dm However, in terms of H5 temperature anomalies for any date in Nov back to 1950, I feel it’s likely because NC H5 ht anomalies on 11/11/2025 were as low as -45 dm vs 11/26/1950’s lowest being ~-41 to -42 dm per the comparisons of these two maps. Had the 11/11/2025 pattern instead occurred 2 weeks later, I believe it likely would have broken the Nov H5 temp. record. Keep in mind that H5 heights are directly related to air temps from H5 down to the surface.

Why not? Other long range models run daily like GEFS. Long range should always be taken with a grain, regardless, due to high levels of forecast uncertainty. But why not run them daily since despite unreliability they can at least hint at potential trend changes past 2 weeks. The EW have been halfway decent imho. For one thing, they had very early hints about the upcoming very weak SPV. For another, they’ve done pretty well with hints about general levels of Atlantic tropical activity weeks out.

Today’s Euro Weekly 2m temp anomalies vs yesterday’s run: Dec 1-7: ~same Dec 8-14: similar for most/slightly stronger cold in Lakes region Dec 15-21: similar for most/slightly colder NNE Dec 22-28: similar for most/slightly colder NE/slightly less cold Lakes region:

Today’s Euro Weeklies 10 mb 60N mean zonal wind isn’t as weak as recent runs: Today’s run (11/19): Yesterday’s (11/18): Does anyone else see what I’m referring to? I’m not just talking about whether it actually reverses. I’m talking more about the period afterward.

This is a good post imho and mirrors my general thinking:

12Z ensemble means at H5 just after the cold shot at the end of Nov: GEFS: EPS: more amplified including much stronger SE ridge than GEFS: We’ll see which verifies more closely. Lately, cold has had the edge.

Thanks Don, I see your point about the unpredictability of upcoming NE snow. Moreover, this thread isn’t strictly about NE snow (as you know). It’s of course also about NE and other areas’ temperatures, wintry precip in other areas, and many other things. Wintry precip, which is usually more variable/localized than the cold influence of Arctic airmasses, is thus often less predictable than temperatures. Therefore, regarding the potential lagged effects of an SSWE, I’m guessing that temperatures in the NE US as well as for other areas for whatever period would be less unpredictable than snowfall. From what I’ve learned, there’s a significantly better chance than normal for a multi week long period of cold domination in the E US usually starting 2-3+ weeks after an SSW. More often than not the cold will appear in the E US. Most likely that would naturally mean increased wintry precip in parts of the E US, but where is the question? Especially if there’s a decent sized sample, I do like to look back at specific cities’ stats (temps and wintry precip) to give me some idea of the potential, regardless.