raindancewx

Looking back through my records, I have an elaborate "difference in proportions" analog system that tries to match overall conditions in Summer to those that precede snowy conditions in winter, and then score how likely a given outcome. The things I try to match are these, for Albuquerque, using about 90-years of snow data: >=3" Snow in March >=3" Snow in February >=3" Snow in January >=3" Snow in December >=3" Snow 'off-season', i.e. Nov or Apr >=10" in January-May >=6" in February-May Each of eight conditions in Summer is scored as -1, 0, or +1, with +1 ideal for snow overall, and -1 bad. Last year, the observed conditions in the Summer were not a great match for any of >=3" Snow ideals, and they were not a good match for a lot of snow late. This year is completely different. In essence, the scoring can be a total match (~0), or a total miss (~13) to conditions that are ideal historically. Last year, December had a Score=4 for 3"+ of snow, and that did happen, with 3.2" snow (and then we also topped 3" in February). All the other months were worse matches than December though. This year, Nov/Apr >=3" had a score of 2.5. Real strong match. We had 4.0" in November. This year, January has a score of 4, and there is a strong signal for a wet January at least. More importantly, there is a score of 3 for Jan-May seeing 10" of snow. Conditions in Summer were not a good match to heavy snow years in Dec/Feb/Mar. Will be interesting to see how we'd get to that 10"+ in Jan-May if it were to verify, given that April is somewhat favored too.

We'll have to see things shake out by 12/31, but tentative top SOI matches for Oct/Nov/Dec are a really interesting set of years. SOI Oct Nov Dec 2019 -5.2 -9.5 2.0 1932 -4.1 -4.6 1.8 1944 -8.5 -6.5 2.9 1985 -5.3 -1.5 0.8 1968 -1.6 -3.4 0.3 1978 -5.3 -2.1 -2.2 2016 -4.5 0.6 0.4 You can mimic the current result almost exactly by blending 1936 & 1944. SOI Oct Nov Dec 1944 -8.5 -6.5 +2.9 1936 0.3 -13.8 -0.7 Mean -4.1 -10.2 +1.1 2019 -5.2 -9.5 +2.0 Remember the October pattern? That blend says it is back. It's actually very difficult to get the SOI to match the values above given the huge Nov/Dec shift. Last year was +0.6 to +9.1, pretty different for Nov-Dec. The six year blend from above looks like a de-amplified version of January 1937/1945. I'm not suggesting this is "correct", and it isn't a forecast, since we don't know the December SOI yet. But it ~kind of looks like the Jamstec update...and that's interesting.

This setup reminds me more of a January pattern, at least from what I've seen over the past 10 years or so.

The latest Jamstec has trended to an El Nino again, and has it continuing for a while, like in the Euro update. The strongly positive Indian Ocean Dipole is also forecast to rapidly collapse, which may be happening now. If you blend the cold position in the US on the Nov map and the Dec map for US Winter, that's where I had the cold going for the overall period. Interior West, not getting into the South or NE, but a lot of the US slightly cold. It's taken a while but at least in the Northern Plains, the cold is winning out in December now in the middle of the US. Bismarck is around 2-3F below average, and the NE/SE anomalies have burned off a lot too already. The Canadian/CFS had the Northern Plains very warm for the month...which looks fairly wrong already, at least for 12/1-12/12.

My stuff works better seasonally then for months for snow. Look at Boston last year: Actual Analog Blend (1953-54, 1976-77, 1986-87, 1994-95 x2, 2006-07) Nov 0.1" (0.0+1.0+3.5+0.1+0.1+0.0)/6 = 0.8" Dec 0.1" (0.0+17.2+3.4+1.5+1.5+0.8)/6 = 4.1" Jan 2.1" (19.2+23.2+24.3+4.4+4.4+1.0)/6 = 12.8" Feb 11.6" (1.9+5.9+3.7+8.5+8.5+4.6)/6 = 5.5" Mar 13.5" (0.4+10.7+3.5+0.4+0.4+10.2)/6 = 4.3" Apr 0.0" (2.1+0.0+4.1+0.0+0.0+0.5) / 6 = 1.2" N-A 27.4" (0.8+4.1+12.8+5.5+4.3+1.2) = 28.7" At any given moment before the final total, it was out by 10" or so at some points. The math will snap toward the analog blend if the blend is right idea. This is what I had for Boston, in the raw blend - for 2019-20. Idea in the analogs was a lot of NE cities start strong then fade a lot late. Year Oct Nov Dec Jan Feb Mar Apr May Season 1953-1954 0 0 0 19.2 1.9 0.4 2.1 0 23.6 1953-1954 0 0 0 19.2 1.9 0.4 2.1 0 23.6 1983-1984 0 0 2.6 21.1 0.3 19 0 0 43 1983-1984 0 0 2.6 21.1 0.3 19 0 0 43 1992-1993 0 0.6 9.7 12.9 19.6 38.9 2.2 0 83.9 1995-1996 0 4.1 24.1 39.8 15.5 16.8 7.3 0 107.6 2009-2010 0.1 0 15.2 13.2 7 0.2 0 0 35.7 2009-2010 0.1 0 15.2 13.2 7 0.2 0 0 35.7 2009-2010 0.1 0 15.2 13.2 7 0.2 0 0 35.7 2018-2019 0 0.1 0.1 2.1 11.6 13.5 T 0 27.4 Blend 0.0 0.5 8.5 17.5 7.2 10.9 1.5 0.0 45.9

Euro trended warmer, finally, for Nino 3.4 in the December update. No members below -0.5C through July now. I like the high end of the plume through February, then transitioning toward the low end in Spring but we'll see. It did well last December -

Looking at local data, I'm expecting the type of unusual pattern in the Southwest US in the second half of November 2019, with tropical moisture coming up from the subtropical jet, but trailing, and not quite phasing with strong Northern stream storms, to return in January. I have a couple reasons for this. The SOI favors two paths on the correlation map - not one given the space in Arizona and Northern NM for the blue areas. The N. California and SE path for the Northern Branch looks similar to the Front Range blizzard path in late November. The Moisture from Baja stream NE across NM, but not really hitting the ski resorts in the north, is the path that brought the Rio Grande Valley a lot of snow - 4-8" as low as 4,000 feet. Statistically, the odds of heavy precipitation in November not making a difference in January precipitation are remote. The p-value is for the difference in proportions test hypothesis that the two groups are the same. You can think of the odds as the likelihood the groups are the same - anything below 0.05 (5.0%) you reject as being the same. Look at January - it's very unlikely that the big Novembers are not an indicator of a difference in January, which supports the SOI map. You can see the average for January after a very wet November is 0.78", more than double the long-term average, and you have roughly 3:1 odds that January will be wet after a top November for moisture. Looking even deeper, the top 11 wettest Novembers have several January clusters for precipitation for very wet days. I think these days will provide the opportunities for the late Nov-early Dec storms to repeat - heavy snow New England and maybe further into the NE or South this time, heavy snow Colorado front range and Rio Grande Valley, maybe further down to El Paso or Chihuahua this time. The most common day is January 4 in the top November years. Overall, Jan 4-5, 11-12,17-20, and 29 look like the windows for big storms like in late November. Once this window closes, it was roughly 11/21-12/4 before, I think the El Nino warm/signal will change in some way and the US will go into some kind of warm spell, kind of like early December was. I think the trigger was the MJO dying after being in phase 8, so that's probably an indicator that this part of the cycle is coming back. One mechanism for this cycle not coming back after January is I don't think the SOI is going to be able to drop 30 points in a day or two, like in November, with the Indian Ocean Dipole making low pressure difficult to achieve over Australia. Once that high-enhancement is gone, the pressure pattern over Australia should be much more stable. For now, the Indian Ocean dipole is probably still powerful to burp out another huge trade-wind reversal one-three times by Jan 10 or Jan 20.

https://www.cpc.ncep.noaa.gov/data/indices/wksst8110.for Nino1+2 Nino3 Nino34 Nino4 Week SST SSTA SST SSTA SST SSTA SST SSTA 06NOV2019 20.7-0.6 25.2 0.3 27.2 0.5 29.3 0.7 13NOV2019 20.9-0.6 25.6 0.6 27.4 0.8 29.6 0.9 20NOV2019 21.7-0.1 25.6 0.6 27.4 0.7 29.5 1.0 27NOV2019 22.5 0.4 25.4 0.4 27.0 0.4 29.3 0.8 04DEC2019 22.5 0.1 25.3 0.2 26.9 0.3 29.4 0.9 Nino 3.4 / 3 should recover some in the coming weeks. My criteria for El Nino is 27.0C for DJF, and at least six months more than +0.5C above the 1951-2010 average for July-June. So far we have two of six (Oct-Nov) - but I'll re-classify it in my data set if we remain under 27.0C for the winter average. Local NWS thinks the pattern may do usual warm ENSO December thing later in the month. Dry northwest flow aloft returns for Thursday and Friday ahead of a a strong 190kt East Asian jet breaking off over the EPAC and nosing into the western U.S. during the weekend. GFS and ECWMF agree that rain with snow above 7000 feet or so breaks out across the northern quarter of NM or so sometime Saturday, continuing at times through Monday. It`s still way too early to nail down much at this point but this storm system/strong jet (135-140kt over northern NM Saturday) has the potential to drop heavy snow over the northwest highlands, Tusas Mountains and northern Sangre de Cristo Mountains over the weekend and into early next week. This pattern typically sets the stage for a positive PNA (think anomalous UL high pressure over/north of AK) and frequent troughs/lows moving into the western U.S. Let`s hope.

That system has SOI support - I think it may drift south from where it is depicted currently by 12/15 though. But you had a big SOI drop 12/3-12/5. So there should be something in the "SW domain" - I look for a low between 30-37N, 103-118W after those big drops. Snow pack / precip generally has been very low in Washington and SW Canada - part of me thinks that will snap hard the other way later, probably sometime in Feb-Apr if it happens.

Nino 4 (160E-150W) has really warmed up again back down to 200m over the last month. Nino 3.4 should re-warm too. It looks like the cold and warm water are both pushing East, so the Modoki look from Fall should come back. After that, probably a rapid collapse toward a La Nina sometime in Spring? Models have been showing that lately. It looks like the Indian Ocean Dipole look is collapsing or at least reversing too - rapid cooling of the waters by East Africa / Madagascar and rapid warming of the waters by Indonesia. Ideal +IOD

This winter should feature some good snow storms pretty far south. I think there are three mechanisms for that given what we've seen to date: 1) Exceptional dry air briefly overtakes an area, leading to very cold mornings...and then very wet subtropical/tropical air ends up on top of it. That 22F / -10 dew point air we had on Halloween Morning is probably going to combine with the subtropical jet at some point. We sort of had a version of it already here in late November, with the 20F / 12F dew-point morning turning into a 30F / 25F snow storm. 2) The Blue Norther of mid-November 2019 broke a lot of records from the legendary Blue Norther of 1911. I think its possible that setup returns one more time, but encounters more moisture from the subtropical jet. 3) We've had extended cloudy periods already in the SW this cold-season. If that continues in Dec/Jan, it will be eventually become very cold here, and that really puts the kibosh on any warm ups spreading to the east/north out of the Southwest. The cloudy periods are actually not great for valleys, we stay above freezing without the radiational cooling, but the mountains are only above freezing during the peak of the afternoon sun, so the cloudiness preserves the snow and cold higher up, which probably helps storms stay dynamic/cold as they pass through the Southwest.

I haven't really looked, but it seems like the PDO is a stronger correlation for the EPO than ENSO itself, even though the PDO is tied to ENSO. I think of the PDO/North Pacific as a dog on a leash, with ENSO as the master. At any given moment the dog may be ahead, behind, left or right of the ENSO signal, but never too far away from it, in terms of time. It's early days yet, but the warm Nino 4 = warm US December signal is holding up so far. There are signs the middle of the US will turn colder, but I'm fairly happy with my idea to warm up he green area on the correlation map from my raw winter forecast blend by 2F in light of Nino 4 coming in around 29.3C or so in December, not 28.6C as my analogs had. Somewhat different so far from last year. Composite on the left is the 12 warmest Nino 4 years since 1950.

So far, the December SOI has been slightly positive, but it's not at the level of last year yet. 2019 335 1012.31 1007.80 4.10 2019 336 1012.38 1007.35 6.80 2019 337 1012.67 1007.20 9.08 2019 338 1012.06 1009.10 -3.95 2019 339 1012.70 1010.55 -8.15 2019 340 1014.24 1009.75 4.00 +1.98 Dec 1-6 2019 2018 335 1014.01 1009.05 6.43 2018 336 1014.79 1008.40 13.86 2018 337 1014.10 1008.80 8.20 2018 338 1013.56 1009.35 2.54 2018 339 1012.80 1007.45 8.46 2018 340 1011.07 1006.20 5.97 +7.57 Dec 1-6 2018

I've been playing around with predictive variables for El Nino winter precipitation for Albuquerque. I've settled on a good blend. The average error in the hind-cast for precipitation is only 0.34 inches, with 93% of years within 0.80 inches (+/-). Forecast blend is for 2.10" in Albuquerque for Dec 2019-Feb 2020, give or take 0.8". That kind of makes sense in the wake of six days in November with over 0.10" precipitation. Five of those were over 0.20". Long-term average for precipitation in winter is about 1.30", so the confidence is pretty high that the city will be at least somewhat above average, without hitting record precipitation (3.46") in winter either.

SOI has dropped some again after starting off pretty positive in December. Should be a big storm around 12/14 or 12/15 over the SW. New European ENSO plume from 12/1 should also be available publicly soon. I remain convinced that the +9 SOI last December is responsible for the warm-east/cold-west February look in 2019. If the SOI doesn't pop, a more canonical El Nino February is likely. 17 out of 18 years with an SOI over 8 in December see a warm February in at least parts of the South, all the way back to 1931. 5 Dec 2019 1012.70 1010.55 -8.15 -6.53 -8.09 4 Dec 2019 1012.06 1009.10 -3.95 -7.41 -8.12 3 Dec 2019 1012.67 1007.20 9.08 -8.42 -8.21 2 Dec 2019 1012.38 1007.35 6.80 -9.07 -8.48 1 Dec 2019 1012.31 1007.80 4.10 -9.40 -8.74

Here is monthly SST anomalies in Nino 3.4, through November 2019, using the stable 1951-2010 base I prefer. If you notice, the US roasted in Aug-Sept when El Nino conditions broke by this standard. Once they returned in October, ferocious cold returned to the pattern immediately - it's part of why I tend to look at 27.0C in the Oct-Feb as the 'on' switch for tropical forcing from Nino 3.4

If you weight the four Nino zones equally, Sept-Nov, these are the top matches. It's probably not the right approach since Nino 4 and Nino 3 make up Nino 3.4, and all the other zones are much bigger than Nino 1.2 Year Sept Oct Nov Sept Oct Nov Sept Oct Nov Sept Oct Nov Blend 2019 29.34 29.52 29.46 26.76 27.22 27.23 24.64 25.14 25.42 20.03 20.27 21.43 0.00 1977 28.95 29.04 29.02 27.12 27.35 27.19 24.87 25.25 25.30 19.61 20.54 21.19 3.23 1990 28.98 29.15 29.20 26.94 26.93 26.81 25.02 24.84 24.90 20.24 20.31 21.00 3.76 2003 28.94 29.12 29.07 26.99 27.14 27.04 25.05 25.38 25.75 20.28 21.25 21.94 4.41 2012 28.97 29.07 29.00 27.10 26.98 26.86 25.24 25.00 25.08 20.79 20.65 21.24 4.64 1968 28.57 28.69 29.10 26.72 26.75 27.20 24.82 24.96 25.15 21.05 20.87 21.56 4.88 2004 29.45 29.46 29.45 27.53 27.44 27.36 25.35 25.52 25.73 20.46 21.16 22.37 4.96 Left to right, the data is Nino 4, then 3.4, then 3, then 1.2. The blend of the years above is a warm East cold NW look for December, which I don't really buy. Here are the top Nino 4 matches - blended this is essentially what the CFS/Canadian had for December 2019 1994 2004 2002 2006 1987 2018 Here are the top Nino 3.4 matches - its literally the exact opposite look of the CFS/Canadian, as well as last year. 2019 2003 1951 1977 2014 1968 1969 My winter forecast had a cold middle of the US in December, with the coasts warm, and the '0' line around the mountains - Appalachians and Rockies. We'll see.

I saw. My analog blend of 1953-54 (x2), 1983-84 (x2), 1992-93, 1995-96, 2009-10 (x3), 2018-19 had Boston at 8.5 inches in December, and about 40 in total. So I don't see any big issues for what I had yet. CPC has ONI at +0.3C now for SON. They raised October as I expected, from this - 2019 8 26.91 26.91 0.00 2019 9 26.77 26.80 -0.03 2019 10 27.19 26.75 0.44 To this - 2019 8 26.91 26.91 0.00 2019 9 26.77 26.80 -0.03 2019 10 27.22 26.75 0.46 2019 11 27.23 26.75 0.48 Not much. Keep in mind, long-term averages for Oct-Feb in Nino 3.4 temps are 26.5C - so this borderline event in ONI sense is warmer than several events considered El Ninos historically. 1977 9 27.11 26.51 0.60 1977 10 27.34 26.48 0.86 1977 11 27.18 26.46 0.72 1969 9 27.15 26.34 0.81 1969 10 27.34 26.33 1.01 1969 11 27.11 26.35 0.76 1968 9 26.72 26.34 0.38 1968 10 26.75 26.33 0.42 1968 11 27.20 26.35 0.85 1958 9 26.40 26.15 0.25 1958 10 26.45 26.03 0.41 1958 11 26.75 26.10 0.65 1953 9 27.00 26.14 0.85 1953 10 26.87 26.01 0.86 1953 11 26.88 26.06 0.82 1951 9 27.22 26.14 1.08 1951 10 27.20 26.01 1.19 1951 11 27.25 26.06 1.19 In more recent times, we're pretty close to 2004, 2006, 2014. 2014 9 27.01 26.80 0.21 2014 10 27.16 26.75 0.40 2014 11 27.46 26.75 0.71 2006 9 27.32 26.80 0.52 2006 10 27.42 26.75 0.66 2006 11 27.70 26.75 0.95 2004 9 27.52 26.80 0.72 2004 10 27.44 26.75 0.68 2004 11 27.36 26.75 0.61

Official measurement for Boston is 2.3" through 12/2.

I should say, I don't actually expect the tropical forcing map I posted to verify for December - it's just what you get from blending the years that match. I think my blend of 1953 (x2), 1983 (x2), 1992, 1995, 2009 (x3), 2018 should hold fairly well. November behaved, with the warm West look, and a colder East. I did warm up the raw blend 2F because my blend is too cold in Nino 4 - around 28.6C and it's close to a degree above that. Anyway, snow pack is good here too: PDO still looks somewhat negative to me with the warmth east of Japan and NOT by Alaska & Canada on the immediate coast. SOI has been going positive past couple days - have to watch that. First time we've had two positive days above +4 in six weeks I think.

SOI has popped positive... Date Tahiti (hPa) Darwin (hPa) Daily Contribution 30 day Av. SOI 90 day Av. SOI 2 Dec 2019 1012.38 1007.35 6.80 -9.07 -8.48 1 Dec 2019 1012.31 1007.80 4.10 -9.40 -8.74 QBO is still positive. Looks like 2002 more than 1995 now. 1995 8.38 8.01 8.79 11.79 14.92 15.62 11.74 9.53 6.98 3.43 -0.77 -4.57 2002 4.64 8.00 9.32 14.03 14.16 13.26 10.05 10.60 8.90 7.66 4.46 -0.50 2019 9.02 9.25 11.82 13.36 14.59 14.36 10.96 9.97 8.25 7.27 5.07 -999.00 Solar activity for the six months to Nov 2019, was the lowest six month average (0.8/month) in any six-month period since Jun-Nov 1823 according to the SILSO data. The 12-month average to November 2019 was down to 3.8, lower than the 5.5 in the period ending November 2008. PDO still looks slightly negative in November. Warmth east of Japan shouldn't be there in a warm PDO, and the ring of warmth right along the Western North American coast isn't there, the warmth is in the 'east of Japan' area and it fades toward Alaska & Canada. Compared to last November, I'd say the PDO is more negative this year, and it was -0.05 in Nov 2018. Nino 1.2 has popped positive, right in time for Christmas, as our Peruvian friends have been observing for hundreds of years. The reading for Nino 3.4 is probably going to be about 27.25C on the monthly data, which I would consider an El Nino reading, given the 1951-2010 November average SST in Nino 3.4 is 26.5C. CPC uses 26.75C as their baseline (it is the 1986-2015 average) in November. So it's real close either way. The little cold pocket of water that has been showing up below Nino 3.4 does seem to be surfacing as the warmth to the east surfaces, so this is becoming less of a Modoki El Nino by the day at this point for now. Nino1+2 Nino3 Nino34 Nino4 Week SST SSTA SST SSTA SST SSTA SST SSTA 25SEP2019 20.0-0.5 24.8-0.1 27.2 0.5 29.7 1.1 02OCT2019 20.0-0.6 25.1 0.3 27.2 0.5 29.7 1.0 09OCT2019 19.7-1.0 24.8-0.1 27.1 0.4 29.5 0.9 16OCT2019 20.6-0.2 25.3 0.4 27.5 0.8 29.7 1.1 23OCT2019 19.7-1.3 25.0 0.1 27.3 0.6 29.7 1.0 30OCT2019 20.8-0.4 25.4 0.5 27.4 0.7 29.6 0.9 06NOV2019 20.7-0.6 25.2 0.3 27.2 0.5 29.3 0.7 13NOV2019 20.9-0.6 25.6 0.6 27.4 0.8 29.6 0.9 20NOV2019 21.7-0.1 25.6 0.6 27.4 0.7 29.5 1.0 27NOV2019 22.5 0.4 25.4 0.4 27.0 0.4 29.3 0.8

Nov 1991: 1.93" precip in Albuquerque. Nov 2019: 1.99" precip in Albuquerque. (1991 was the record for Nov here for the past 100 years). I think because the IOD is enhancing the response you'd get from the subtropical jet-stream for an El Nino this weak, 1991 is probably a good precipitation analog nationally. Not really convinced for temperatures. The models depicting a positive SOI for December may not be too wrong - it looks neutral to positive for most of the next ten days on the European, although their might be one big drop in a a couple days which would support a big storm in the SW mid-month, and then maybe a pattern change nationally. The Canadian may actually be trying to do 1965, 2018 for its forecasts because of the SOI pattern. SOI Sept Oct Nov 1965 -13.5 -11.0 -16.7 2018 -8.5 +2.6 +0.6 Blend -11.0 -4.2 -8.1 2019 -12.7 -5.2 -9.5

My basic idea for December for New Mexico & Colorado & West Texas is that the storminess of November dies off somewhat, but we still finish wetter than average, with temperatures probably not that far from average either way, despite impressive cold at times later in the month. In recent El Ninos, the US temperature pattern for Oct 16-Oct 31 has tended to become the December temperature pattern nationally, but shifted somewhat East. We had near to record cold in that period this year. My objective high temperature & precipitation matches for the July-November period include six very cold Decembers locally, including the record and near record cold of 1943, 1952, 1953, 1992. Only 1939 was warmer than normal, and that was the weakest of the ten objective matches. Objective matches for the subsurface in the Tropical Pacific, years with a big warm up Sept to Oct, and then a cool off in Oct to Nov also support somewhat cold temps in the SW. Additionally, if you blend all years when Nino 4 is above 29.3C (there aren't too many of them) in Fall/Dec, the composite looks a lot like the map below. The top SOI matches for Sept-Nov also look like the map below. Years with positive NAO readings in December. I do expect the pattern in early November to repeat at some point later in the winter - probably January, but its worth remembering most of the severe cold was destroyed pretty quickly. It's obviously amplified, but the very warm coast/colder middle thing repeated last Nov 1-15 in January. It's just it came in much more severely, even though it was similar spatially.

The 10/31 forecast from the Canadian for November, as well as the 11/30 forecast for December both look like a blend of 1965 & 2018 on that model. The Canadian forecast from 11.30.18 for December 2018 was actually pretty bad, unlike the winter forecast. The subsurface data for November came in for 100-180W, top 300m. https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ocean/index/heat_content_index.txt Best blend I could find was this - 100-180W Sept Oct Nov 1986 0.65 0.95 0.52 1986 0.65 0.95 0.52 1991 0.60 1.41 1.22 1991 0.60 1.41 1.22 1991 0.60 1.41 1.22 1991 0.60 1.41 1.22 1998 -2.15 -2.35 -2.33 2005 -0.33 -0.14 -0.57 2005 -0.33 -0.14 -0.57 Blend 0.10 0.55 0.27 2019 0.00 0.70 0.26 Crazily enough...it looks identical to last year? So I am skeptical. I tried to respect the subsurface features though - big warm up in October and then the big cool down in November.

The Canadian Model has the El Nino for winter now. I don't think this is right, but it has the "Nino 4 warm" pattern, with the US roasting in December, except the SW. Interestingly, it has the NE somewhat cooler. I think there will be more cold in the middle of the US than shown, a lot of snow pack is coming to the middle of the US and western Canada from what I can see. 11.30 run for Dec 10.31 run for Dec The winter look is a drier 2004. Last year, the November idea for Dec-Feb wasn't completely wrong, for what its worth.