raindancewx

The city had snow today. Very cold for most of the day, it was somehow 34F from like 11:45 pm to 12:05 am and then not above freezing at any other time on 12/31 or 1/1, so that kind of artificially raised highs on both days, oh well. Most of the snow today fell around 20F temps, so most places got 1-5" with only 0.05-0.20" liquid equivalent in town.

Weather.com has the Jan 1-5 high in Albuquerque at about 29F, which is second coldest in the last 100-years for that period if that ends up verifying. It's probably off a bit, but don't think it will be terribly wrong either.

Accuweather's '8%' event verified, it was 2.5" officially. This is what they have next -

I'm looking forward to seeing if this verifies in the favored zone or if Accuweather is way off.

Looks stormy here after Christmas. I'm still a bit skeptical that these systems will have much moisture with the SOI so positive for weeks now (+10.2 for 12/1-12/23), but we'll see. I do think the cold is fairly likely.

It feels so strange to have dew-points in the 30s again after such a prolonged period here with dew-points around 0-10. Just a trace so far in the city, but the best shot for us was always Friday anyway.

Models are kind of a mess with this system for timing, duration and location. Hopefully they become more consistent tomorrow.

CIPS Analogs has a lot of storms from 2002-03, 1987-88, 1984-85, 2000-01, and 2007-08 for this event in its top 15 analogs. SOI matches for Sept-Nov include 1990. I'm pretty sure 1987 and 1990 were good years for OKC in December.

Might get snow in the city Sunday morning. We'll have to see. I've never seen a CPC Outlook like this one for December -

Assuming we finish with almost no precipitation here for November in Albuquerque. I had to re-do my replication analogs. With the very dry November after a very wet October, you have a strong signal now for a wet December in Albuquerque. The interesting thing is almost all of these years have a lot of snow in Feb-Mar here, with little in January, and then average snow in December. I've been doing this replication method for a few years now, this was definitely the hardest blend I've ever had to re-create, as I limit myself to blends where each re-creation is within 0.2" of observed conditions. If you get every month from July-Dec within 0.2", it tends to predict Jan-Jun well overall, i.e. in three month aggregates. Also, 1974 and 2008 are low solar years, 1969 and 2008 have big hurricanes hitting the Gulf of Mexico. Every year in this blend is above average for December.

I had November Blizzards in the Plains for my outlook...so I'm on board. I think Gary Lezak's ideas this year are about right - https://www.kshb.com/weather/weather-news/gary-lezaks-winter-forecast-how-much-snow-will-kansas-city-get-this-year

I have a regression for El Nino winter highs here, and it seems to be working well. Low Solar + Big El Nino + After Big La Nina is the ideal for cold. I'd give solar a 10, the El Nino an 7.5, and the La Nina last year a 6.5 as a blend, it's like 24/30, or a solid 8/10 on an "ideal cold" scale. I don't really expect this blend to work at all for precipitation or national temps as it doesn't really incorporate the MJO, or how much colder the Atlantic is in these years, but I think its probably close for DJF highs here.

I went with 1953, 1976, 1986, 1994, 1994, 2006 as my blend for winter, but I think 1986 may end up being the top year, I'd consider that a better version of 2009 (low solar, after a La Nina, relatively cold AMO, etc). CPC is still stubbornly not showing anyone cold for the winter. But their November outlook from 10/18 is pretty terrible so far, so not really worried about it.

It is only November 12, but the high of 36F today was colder than every high since 1/26/2017. Looking forward to seeing if we really can drop to 15F or so tonight.

Thunder-snow by Los Alamos today.

A lot of the things I'm seeing (relatively) independently point to a wet and/or cold March in the SW US and Northern Mexico. Very Positive NAO in October: Hurricanes hitting the Gulf of Mexico in the hurricane season before an El Nino winter. Albuquerque 2F below average (1931-2017 basis) high for October (71.3F = average) and November (57.3F = average) in an El Nino year. We were 66.8F in October. November looks cold now too, with a high of maybe 38F on Monday?

I had to adjust my precipitation replication analogs with the rains in late October here - we had 1.99" for the month. The simplest blend I could come up with in an El Nino is 1941, 1957, 2004. That gets July, August, September and October within 0.2" each month v. 2018 observations. Those years all have major hurricanes hitting the Gulf of Mexico too. It is interesting to note that all of those years were part of multi-year warm events (though 2003-04 was a Warm Neutral). These years also all feature pretty major snow storms in March, which essentially never happens here (in the valleys) without high sunspot support. Will be interesting to see if maybe the sun rapidly starts getting active again, or if this is some kind of grand exception to the March sunspot snow rule. 28% of years (15/53) with over 55 sunspots from July-June see heavy March snow, but in all others, only 3% (1/34) do. There are actually at least two other blends that work for July-Oct, but they have more years. November should be clarifying - all three replication blends have an active November, around 0.8-1.0 inches of precipitation. I think the blend must still be off a bit - mostly because I think we'll get pretty good snows in December this year.

Total precipitation in August & October has a lot of predictive power here for when the snowiest month will be within the Oct-May cold season. If you throw out 1948-49 when April was snowiest and the three years with "ties" for snowiest month, some pretty distinct patterns emerge for what to look for if a certain month is going to be snowiest. Snowiest August October Nov 1.74" 0.51" Dec 1.71" 0.94" Jan 1.47" 0.61" Feb 1.13" 0.70" Mar 1.44" 1.40" 2018 0.95" 1.64" Prior to the heavy rains over the past two days, February's composite was closest to what we observed in August & October. With the huge rains in October, MARCH, which hasn't been wet in Albuquerque since 2007, is now the closest composite. That being said, some years when December is snowiest have big Octobers too, but the data at least implies a secondary stormy period in February-March. My main issue is Albuquerque has only had over three inches of snow in March one time since 1931 in a year without high sunspot activity (3/1975). "High" I define as an average of at least 55 sunspots per month from July-June. Every 100 sunspots on an annualized basis is worth about 0.11" in Albuquerque, so the March number for precipitation would be 1.10" if the blend below holds. Snow would also be lower if the sunspot idea is correct, closer to four inches instead of six, and even four inches of snow would be the highest March snow total in Albuquerque in a low-solar year since 1931. The other (low) possibility is that the solar minimum is now, i.e. Oct/Nov 2018, and we will rapidly ramp up in activity by March allowing a big snow total as shown below. If the composite idea is correct, 1957, 1972, 2004, 2004, 2004, 2015 blended together is a near perfect match to July-Oct precipitation by month here, and all of those years but 2015-16 had their top snowfall in March. Snow Oct Nov Dec Jan Feb Mar Apr May Total 1957 0.0 0.0 2.9 2.3 0.5 7.3 0.0 0.0 13.0 1972 0.0 2.9 1.2 9.5 1.8 13.9 8.1 0.0 37.4 2004 0.2 1.7 0.3 0.0 0.0 4.2 0.5 0.0 6.9 2004 0.2 1.7 0.3 0.0 0.0 4.2 0.5 0.0 6.9 2004 0.2 1.7 0.3 0.0 0.0 4.2 0.5 0.0 6.9 2015 0.0 0.0 7.2 2.2 0.4 0.0 0.0 0.0 9.8 Mean 0.1 1.3 2.0 2.3 0.5 5.6 1.6 0.0 13.5"

Against 1931-2017 means, the 10/1-10/23 mean high here is still running 5.6F below average, behind only 1984, 1986, and 2000.

Phoenix, for 10/1-10/20, is having its second coldest observed mean high since 1895, of 81F. That is around 20F below the record heat in October 1991. Albuquerque is seeing the 6th coldest high for 10/1-10/20, at around 16F below the record heat for the period in October 1979.

Going forward, this is probably a decent way to estimate final El Nino / La Nina strength in winter. The 100W-180W numbers come from here - http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ocean/index/heat_content_index.txt Implies around 27.6C for Nino 3.4 this DJF, give or take 0.55C or so, peak ONI of right around +0.5 to +1.5 against the 1985-2014 base NOAA uses (26.58C), just like the models show.

This October cold spell here is already basically as cold as it got all of last winter for any sustained period. My analogs had an October high of 68F here. That may end up too warm....since through 10/17 we're already down to 68.4F for the October high. The high of 68.4F for 10/1-10/17 is actually the 4th coldest in Albuquerque since 1931.

Looks like the high here was 43F today. That is colder than...99.6% of all October highs since 1931 here. Record coldest high is 36F.

Increasingly looking like October 2018 will be the first cold month here since August 2017 on a mean high basis, relative to 1931-2017.

This region does pretty well for snow in my analogs - I have snow anomaly maps for each analog toward the back if you all are curious. This is what I have - https://www.scribd.com/document/390797995/Winter-2018-19-Outlook