raindancewx

The SOI has crashed massively in recent days after finishing August pretty low. Low SOI values in Sept somewhat favor a cold US in October. It is a pretty strong signal for a cold November in TX...and a warm signal east of the Rockies in Dec. Doesn't indicate much for January. Strong warm signal in the NW for February. Strong cold signal for SW in March.

Here the monsoon kind of blows everything else out of the water for long range winter precip trends, especially if filtered by ENSO. Modoki structure doesn't mean jack here for precip. 2006 and 2009 were both Modoki El Ninos, but of a different flavor. Albuquerque had near record record precip in 2006, a bit dry in 2009. We're closer to 2006 honestly, not a strong El Nino, and the warmest waters won't be as far west. Modoki stuff matters for temps and analogs here, not really for precip - long term correlation is very weak here. The monsoon has been average to above average in Western NM / AZ - and my analogs have near average to above average precipitation here and for AZ. Eastern NM has been dry, and the the rain/snow signals show up as dry there. Here are some long term looks at the Monsoon v. winter out here - I'm running +10% for 6/15-9/6 using 1981-2010 means, and +20% v. 1931-2017 means. Monsoon correlations to winter precipitation are incredible in Arizona for what its worth.

With September 1/5 over now, my mean high is 80.3F - that is 4th coldest here since 1931, behind only 1938, 1961, and 2006. The 1931-2017 mean high for Sep 1-6 here is 87.0F - so 6.7F below normal. The 30-year mean is lower, as the 1950s (warm AMO & high solar) had some absurdly hot Septembers, as did the late 1970s here. ABQ Sept 1-6 Mean Hi 1 9/6/1938 75.0 2 9/6/1961 76.5 3 9/6/2006 76.7 4 9/6/1988 80.8 - 9/6/1950 80.8 6 9/6/1975 81.2 7 9/6/2004 81.3 8 9/6/1972 81.5 9 9/6/1966 81.7 10 9/6/1977 81.8 11 9/6/1999 82.2 - 9/6/1935 82.2 13 9/6/2005 82.8 - 9/6/1970 82.8 15 9/6/1987 83.0 - 9/6/1965 83.0 17 9/6/1967 83.3 - 9/6/1963 83.3 19 9/6/1968 84.0 20 9/6/1942 84.2 21 9/6/2007 84.3 22 9/6/2003 84.5 23 9/6/2008 84.7 - 9/6/1994 84.7 25 9/6/1985 84.8 26 9/6/1981 85.0 - 9/6/1940 85.0 28 9/6/1984 85.2 29 9/6/2015 85.5 - 9/6/1976 85.5 31 9/6/1986 85.7 32 9/6/1991 86.0 33 9/6/1974 86.3 34 9/6/1996 86.5 - 9/6/1944 86.5 36 9/6/2009 86.7 - 9/6/1969 86.7 - 9/6/1934 86.7 39 9/6/1997 86.8 - 9/6/1993 86.8 - 9/6/1990 86.8 - 9/6/1941 86.8 - 9/6/1932 86.8 44 9/6/1992 87.2 - 9/6/1962 87.2 - 9/6/1957 87.2 47 9/6/1971 87.3 48 9/6/2011 87.5 - 9/6/1978 87.5 50 9/6/2016 87.7 - 9/6/1936 87.7 52 9/6/1989 87.8 53 9/6/1998 88.0 - 9/6/1973 88.0 - 9/6/1931 88.0 56 9/6/1953 88.3 - 9/6/1943 88.3 58 9/6/1980 88.7 - 9/6/1946 88.7 60 9/6/2001 89.0 61 9/6/1964 89.2 62 9/6/2014 89.5 - 9/6/1954 89.5 - 9/6/1937 89.5 65 9/6/2010 89.7 - 9/6/2002 89.7 - 9/6/1949 89.7 68 9/6/2017 90.0 69 9/6/1960 90.2 70 9/6/2000 90.3 - 9/6/1958 90.3 72 9/6/2013 90.5 73 9/6/1956 90.7 - 9/6/1933 90.7 75 9/6/1955 91.0 76 9/6/1947 92.3 77 9/6/1952 92.5 78 9/6/1939 92.7 79 9/6/1982 92.8 - 9/6/1959 92.8 81 9/6/2012 93.0 - 9/6/1945 93.0 83 9/6/1951 93.5 84 9/6/1995 93.7 85 9/6/1948 95.2 86 9/6/1983 97.0 - 9/6/1979 97.0 Mean Hi 87.0

The mountains here are almost always between 0F-30F from mid-Nov to mid-Mar - nearly ideal snow conditions. Taos Ski Valley averages around 300 inches. Someone in New Mexico will just about always do well for snow in an El Nino, the ski resort 15 minutes away from Albuquerque had 30-40 inches by Dec 31 2015. 2014-15 had 70-100 inches of snow in all the resorts in a five day period in late February. 2009-10 was excellent for the eastern slopes. Most of the resorts had near record snow in 2004-05, which much of it in March. The only real bad El Nino in the last 30 years was 2002-03, and despite some similarities to this year, the PDO, Solar and ENSO conditions are very different. Most of the resorts will see a +20% type snow anomaly here this year. There are some years that are Ski Apache years (near MX), Ski Taos years (near CO), and Ski Sandia years (near Albuquerque) in El Ninos - but at least some of the resorts will do well. Late Nov, late Dec, mid-January, late February all look pretty promising to me for big mountain snows here.

The weaker El Nino is much more in line with my thinking. I lean toward +0.6C to +0.8C at this point for DJF. ASO is probably going to be around +0.3C given how August was at +0.0C. ONI of +0.3C in ASO is like 1958, 1968, 1979, 1990, 2006, 2014 among years that became Ninos late. Here is the new Canadian look that Griteater mentioned: I'm not particularly sold on the idea of a frigid NE like 2009 - the warmest waters were way west that year. That said, Nino 1.2 is still cold and I don't think the Canadian is really reflecting that yet. The Nino should start in October if the Canadian is right (idea probably isn't terrible, SOI is pushing warm water into Nino 3.4 up to the surface again). This is pretty similar to Larry Cosgrove's call from July. The Modoki index on the image on the top left is probably around +0.3, with the edges of the Pacific colder than the middle, but still warm. Here is what 2009-10 looked like for comparison: The look (Nino 3.4) is closer to 2006-07. The temperature outlook kind of reflects that too. I do think Nino 1.2 and the AMO are issues v. 2006, which is why I've been including 1934, 1976, 1986, 1994 with 2006 to cool the Atlantic and change the SSTs off the US West. I've been weighting 1994 twice, but it pushes the core of the Nino too far West, may end up weighting 1986 twice and 1994 once.

Local history and Sept 1-4 observations here suggest we're at a 70%+ chance September will be colder than average (1931-2017 basis) at least for high temperatures. I think this will be my first cold month on that basis since...August 2017. I really hate east-based La Ninas. I have this document called "controls" that looks at future outcomes after unusual conditions in a month. Cold September highs (2F or more below the 1951-2010 average) are associated with: - Cold January highs (2F or more below average) - Snowy February (>=3" in Albuquerque, average is 1.9") Those are the main effects. Cold DJF is more likely, as is >=3" in January. Also, other than El Nino, a cold September is the best early indicator of two months in Oct-May with over 3" of snow, and >=6 inches of snow in February to May. Rainy/cold July is associated with heavy snow in December, and December as the snowiest month. July was rainy, but not cold. I'm expecting a slightly colder than normal and slightly wetter than normal winter here at this point, with around a foot of snow (+25%) from Oct-May. We'll see.

The ONI value for JJA came in at +0.1C (same as MJJ). This actually makes sense, the ECMWF uses OISST, CPC uses a different SST data set. So both data sets showed a drop off from July. From CPC - http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/detrend.nino34.ascii.txt 2018 4 27.32 27.73 -0.41 2018 5 27.74 27.85 -0.11 2018 6 27.78 27.65 0.13 2018 7 27.42 27.26 0.16 2018 8 26.94 26.91 0.04 JJA (0.04+0.16+0.13)/3 = +0.11 MJJ (-0.11+0.13+0.16)/3 = +0.06 Still warming...just slowly, three steps forward two back. 2006 still looks pretty similar to me in an ONI sense (AMO is different). 1986 is fairly similar too. 1976 not bad. I like 1994 too, but for different reasons. I have 1934 in for completely different reasons too. Year DJF JFM FMA MAM AMJ MJJ JJA 1976 -1.6 -1.2 -0.7 -0.5 -0.3 0.0 0.2 0.4 0.6 0.8 0.9 0.8 1986 -0.5 -0.5 -0.3 -0.2 -0.1 0.0 0.2 0.4 0.7 0.9 1.1 1.2 2006 -0.8 -0.7 -0.5 -0.3 0.0 0.0 0.1 0.3 0.5 0.7 0.9 0.9 2018 -0.9 -0.8 -0.6 -0.4 -0.1 0.1 0.1

I really think this event will be an almost exact opposite of 2016-17 at this point. If you look at July 2016-Jan 2017 you had a La Nina in 3.4, but Nino 1.2 was very warm. This event looks very cold in Nino 1.2, but warm enough to be classified as an El Nino in 3.4 (although it is starting later than 2016-17 did). I went with 1934, 1976, 1986, 1994, 1994, 2006 as my Fall analogs based on ocean, solar and observed weather conditions locally. 1934, 1994, and 2006 were very cold starts to September here, which has happened this year. My Sept 1-3 highs in 2018 trail only 2006 and 1938 since 1931. Haven't had a September 2F or more below the 1951-2010 mean high here since 2006 - but we'll have a shot at it this year looks like and my analogs did have a much colder Fall for the SW v. 2017, around 5F colder overall (3F colder in Sept-Oct, 8F in Nov). Also: The NOAA PDO value for August fell from July, usually the NOAA PDO trends like the Mantua/JISAO values. My research suggests the PDO base state in Mar-Aug, plus the October temps in Nino 1.2, blended together, predict the Nov-Apr PDO state/intensity fairly well. I'm estimating 21.2C for Nino 1.2 in October and the March-August PDO to be around 0. Some that approximate those values: 1965, 2002, 2004 (PDO Only)

Sept 1-3 average high temperature (77.33F) is the third coldest start to September in Albuquerque since 1931. We're 10% of the way through the month, and more rain seems to be coming this week, so will be hard for highs to rebound much. Pretty decent odds of a cold September now.

Nino1+2 Nino3 Nino34 Nino4 Week SST SSTA SST SSTA SST SSTA SST SSTA 01AUG2018 21.2 0.1 25.3 0.0 27.1 0.1 29.1 0.4 08AUG2018 20.8-0.1 25.1-0.1 27.1 0.2 29.2 0.5 15AUG2018 20.9 0.2 25.4 0.4 27.3 0.4 29.5 0.8 22AUG2018 20.4-0.2 25.1 0.1 27.1 0.3 29.0 0.4 29AUG2018 20.3-0.3 24.9-0.1 27.0 0.2 29.1 0.4 No update from CPC yet, but June was +0.13, July was +0.17. August probably +0.2 to +0.3. So ONI likely +0.2 for JJA. Looking forward to the ECMWF update. The drop in August the model had did occur, now we have to see if the reversal happens.

Box C of the Modoki regions is a lot cooler than last year for the month of August. When Box C is cold, and you have an El Nino, the SW can be pretty cold, or at least as cold or colder than the rest of the US in winter. Depends on what the other zones do though too. Box C cold, when Box B is colder than A is typically a colder pattern for the West if it is an El Nino. Can be very warm if Box C is warm though. These are some years when the E Pac (tropics, Box B on the Modoki scale) were colder than Box A (Nino 3 essentially) but box C was cold (all for DJF): Right now, Box C is near average? So maybe you take all the El Ninos at low solar on both lists - 1953, 1963, 2009 Cold C Warm C B<A B<A 1953 1980 1954 1989 1956 1993 1958 1996 1963 2001 1966 2002 1967 2004 1968 2009 1977 2012 1979 2014 1990 2017 1992

61F at noon here in Albuquerque. We had a rare significant rainfall event in the morning - easy way to get a cold day here in the Summer. Likely our first day not hitting 80F since June 16th.

The vast majority of warming on Earth is at night (lows) and in cold uninhabited regions, the Arctic/Antarctic warming at 5x the rate of the other areas, etc. We can pretend that people care about temperatures in places they don't live and when they are sleeping, but the billions of people living better than they have before in any other age, suggests that actually, at least for now, it doesn't matter that much. It is part of why no one responds to these threads. The truth is pretty obvious.

My view is that there is a long-term sixty year cycle, or even a 120 year cycle. In the 60 year cycle, each of these lasts 10-20 years. In the 120 year cycle, that is true, but the magnitude of the sign for each ocean is much greater or weaker. So the AMO+/PDO- sign in the 1950s may have been more amplified given all the NE US hurricanes for instance, relative to 2011/2012 which only had Sandy and Isabel. The late 1950s, like now, were volatile for the two signs, first with the brief late 1950s PDO spike, and then the rapid AMO collapse in 1963...after Agung erupted. Which happened last year...and around the AMO flip in the 1840s too. AMO+/PDO+ AMO+/PDO- AMO-/PDO+ AMO-/PDO-

August 2018 US high temperature anomalies look fairly similar to some recent developing El Nino years - notably 1994, 1997, 2002, 2009, 2014, 2015. Main issues - not as hot as 1994, 2002, 2015 in the SW this August, or as cold in the Lakes.

NAM has snow in Wyoming in the coming hours. Awesome to see. Not unheard of at all.

2018 and some recent Augusts for context. Nino1+2 Nino3 Nino34 Nino4 Week SST SSTA SST SSTA SST SSTA SST SSTA 01AUG2018 21.2 0.1 25.3 0.0 27.1 0.1 29.1 0.4 08AUG2018 20.8-0.1 25.1-0.1 27.1 0.2 29.2 0.5 15AUG2018 20.9 0.2 25.4 0.4 27.3 0.4 29.5 0.8 22AUG2018 20.4-0.2 25.1 0.1 27.1 0.3 29.0 0.4 06AUG2014 22.2 1.2 25.6 0.4 27.0 0.0 29.2 0.5 13AUG2014 21.9 1.2 25.5 0.5 26.9 0.0 29.0 0.4 20AUG2014 22.1 1.4 25.5 0.5 27.1 0.3 29.1 0.4 27AUG2014 21.3 0.8 25.4 0.4 27.2 0.4 29.2 0.5 01AUG2012 21.6 0.5 26.1 0.9 27.6 0.6 29.0 0.3 08AUG2012 20.9 0.0 25.9 0.8 27.7 0.8 29.1 0.4 15AUG2012 20.8 0.1 25.6 0.6 27.4 0.6 29.1 0.4 22AUG2012 21.0 0.4 25.5 0.5 27.4 0.6 29.1 0.4 29AUG2012 20.3-0.2 25.6 0.7 27.7 0.9 29.1 0.5 02AUG2017 20.9-0.1 25.4 0.1 27.2 0.2 28.9 0.2 09AUG2017 20.5-0.3 25.1 0.0 26.7-0.2 28.7 0.1 16AUG2017 19.9-0.7 24.5-0.5 26.4-0.5 28.8 0.1 23AUG2017 19.6-1.0 24.6-0.4 26.7-0.1 28.9 0.2 30AUG2017 20.3-0.2 24.5-0.4 26.5-0.2 28.8 0.2

I've decided to roll with 1934, 1976, 1986, 1994, 1994, 2006 as my Fall analogs. Cold AMO ring, slightly positive PDO, El Nino developing, La Nina in prior winter, low-solar, similar Modoki structure, and similar Monsoon conditions. So we'll see how it does. We had near record November highs here last year (65.4F, average in Nov is 57.3F), so that is the main reason for the huge drop off expected. Going from +3.3F for Fall highs to -1.4F for Fall highs against 1951-2010 is what the analogs get for Albuquerque.

There is a lot more warm water below the surface in the 2N to 2S ENSO zone now. I don't know that Nino 1.2 is ever really going to get above average temperatures for winter, but the other zones look El Nino ish. My weather (Tmax) and precipitation (at a three week lag) has been very similar this Summer to 1934, 1976, 1986, 1994, 1994, 2006 blended together, so going to stick with that for September, probably into Fall. The AMO looked cold through Aug 24, but has warmed some by Africa lately. The PDO still looks near Neutral. Nino 3.4 warmth seems like it is going to be capped this year by warm waters from the West offsetting the cool waters to the East.

I can't find any data to support it...maybe the major cold push in the Northern Rockies? But I kind of think we get some snow here, even in the valleys this year in October. Happened in 2002 & 2009 even in the city. My tentative winter analogs (1934-35, 1976-77, 1986-87, 1994-95, 1994-95, 2006-07) have near identical rains in Albuquerque to observed rains this Summer, except that the rains are happening 2-3 weeks earlier than in the mean of the data. I am getting sick of how August always seems to fail here for major rains. You can get 3-4 inches of rain in August here, but we haven't had a meaningfully wet August since 2006. Wettest August since 2006 is 1.62", barely above average (<+10%). It is especially annoying since August + October rainfall here is a strong indicator for March snows, and low is bad. GFS has like 0.2" or something for the rest of the month, could realistically still finish August under an inch yet again if we don't get walloped tomorrow or during the weekend.

In terms of what I expect for winter (oceans/solar/ENSO order/matching my Summer weather) I still like 1934-35, 1976-77, 1986-87, 1994-95, 1994-95, 2006-07. The blend is a weak El Nino, warmer anomalies to the west than by Peru, with a warm/wet Summer in my area, the cold AMO ring, relatively neutral PDO readings, and pretty low solar conditions, and when blended, those years follow a healthy (east-based) La Nina, like this year will. The weather last winter was also similar to 1933-34, 1975-76, 1985-86, 1993-94, 2005-06 in various ways. In terms of the "values", it looks like this - Year ONI DJF AMO N-A SUN Jul-J ONIp DJF Mod DJF PDO N-A Monsoon 1934 -0.3 -0.025 27.6 -1.0 0.19 1.02 4.08 1976 0.7 -0.315 23.2 -1.5 -0.14 1.04 3.10 1986 1.1 -0.190 19.1 -0.6 0.36 1.91 6.84 1994 1.1 -0.004 36.9 0.0 0.51 -0.37 4.80 1994 1.1 -0.004 36.9 0.0 0.51 -0.37 4.80 2006 0.6 0.208 20.1 -0.8 0.14 -0.04 9.42 Mean 0.71 -0.055 27.3 -0.65 0.26 0.53 5.51 That is pretty close to the values I'm estimating for 2018-19 currently (and the SSTA map looks similar) 2018 0.8 0.000 11.0 -0.9 0.20 0.50 5.50

Every meteorological agency in the world has a different method for El Nino/La Nina designation. These are the coordinates from the Sintex/Jamstec for Modoki designation: Region SSTA Anomalies by Box Global Area Box A 165E to 140W, 10S to 10N Trop Pac Middle Box B 110W to 70W, 15S to 5N S. America waters Box C 125E-145E, 10S to 20N Japan to Equator The formula to get the Modoki value is Box A - (0.5*Box B)-(0.5*Box C) Nino 3.4 is 120-170W, 5S to 5N, so somewhat close to box A. Box is somewhat close to Box B. Box C is kind of its own thing since it is far away from the equator. It doesn't seem to matter for the Eastern 2/3 of North America, but Box C warming/cooling does have implications in the Western US/Canada/Mexico.

The weeklies showed continued warming in all ENSO regions. Very different from last August. Also, the SOI crashed again and the warm subsurface blob (at least at the equator) remains healthy. Nino1+2 Nino3 Nino34 Nino4 Week SST SSTA SST SSTA SST SSTA SST SSTA 18JUL2018 21.3-0.3 26.1 0.6 27.6 0.4 29.1 0.3 25JUL2018 21.1-0.3 25.8 0.3 27.4 0.3 29.0 0.3 01AUG2018 21.2 0.1 25.3 0.0 27.1 0.1 29.1 0.4 08AUG2018 20.8-0.1 25.1-0.1 27.1 0.2 29.2 0.5 15AUG2018 20.9 0.2 25.4 0.4 27.3 0.4 29.5 0.8 19JUL2017 21.4-0.1 25.7 0.2 27.6 0.4 29.2 0.4 26JUL2017 21.3 0.0 25.5 0.1 27.1 0.0 28.9 0.2 02AUG2017 20.9-0.1 25.4 0.1 27.2 0.2 28.9 0.2 09AUG2017 20.5-0.3 25.1 0.0 26.7-0.2 28.7 0.1 16AUG2017 19.9-0.7 24.5-0.5 26.4-0.5 28.8 0.1

The SOI has flipped positive again lately. Looking at the Box A, Box B, and Box C modoki numbers for last year, 1985-86 is a pretty decent match - will be interesting to see if 1986-87 ends up being similar to this winter. I kind of doubt it, although the AMO cold ring was there.

Sea Ice Extent on 8/17/18 was 5.568 million square km. That is above 2017 (5.371), 2016 (5.398), 2012 (4.691), 2011 (5.495), and 2007 (5.322) for the same date (8/17). Will be curious to see how the annualized AMO v. ice extent on 8/1 plot looks after this year is over.