Jump to content

ENSO State and Georgia Winters

Sign in to follow this  


The main focus on this article is not to make a winter forecast, but  to explore what effects El Nino's and teleconnections have on winter weather across the southeast. One of the main sources of data in this report comes from a weather friend of mine that I call "Brother Larry". Larry would prefer to remain anonymous, so from here on out you'll hear me refer to Larry as "Brother Larry". 🙂  Larry has a wealth of information about the weather history in Georgia, and I'll be using a lot of his findings to help give you an idea how this winter may turn out, based on the environment created by the El Nino, as well as several other factors. Again, this data is based on analog years, or those years that most closely identify with the current patterns, so keep that in mind. Analog's are not perfect, but they do give us a very good idea about how things have happened in the past and how they may happen again in the future. 

Again, almost all of the text below (other than a few of my own edits and additions) is from Larry, and he gets all the credit for the research and stats.


This data was compiled by taking a list of 26 “cold” US winters (Dec/Jan/Feb) since 1894 -1895 (i.e., the coldest 23%) for the eastern third of the US. This requires solid, widespread, below normal anomalies, and requires the southeast to be pretty chilly itself. The two maps to the right were created with data from the list of years below, but that dataset only goes back to 1948, so the maps I'm displaying are not 100% complete with the years in the list. Here's the list of those winters, and you can see the years I used on the maps themselves. 
Also, Larry's 26 coldest winters study was done the better part of 10 years back, since then, it is possible that some of 09, 10, 13, 14, etc. could be added, although he is not reassessing those now.
  • 2002 - 2003
  • 1995 - 1996
  • 1993 - 1994
  • 1981 - 1982
  • 1980 - 1981
  • 1978 - 1979
  • 1977 - 1978
  • 1976 - 1977
  • 1969 - 1970
  • 1968 - 1969
  • 1967 - 1968
  • 1963 - 1964
  • 1962 - 1963
  • 1960 - 1961
  • 1947 - 1948
  • 1939 - 1940
  • 1935 - 1936
  • 1917 - 1918
  • 1911 - 1912
  • 1909 - 1910
  • 1904 - 1905
  • 1903 - 1904
  • 1901 - 1902
  • 1900 - 1901
  • 1898 - 1899
  • 1894 - 1895
Temperature Anomalies


Nino Base State

ENSO Regions
ONI Chart from Golden Gate Weather Services - oni.png
So now that we have our list of base years, let's look at the base Nino state for those winters. Before we can do that, we need to take a look at the various Nino/Nina classifications. Graphs and Charts

The Oceanic Niño Index (ONI) has become the standard that NOAA uses for identifying El Niño (warm) and La Niña (cool) events in the tropical Pacific.  The ONI is defined as the running 3-month mean SST anomaly for the Niño 3.4 region (i.e., 50N-50S, 120W-170W).  Events are defined as 5 consecutive overlapping 3-month periods at or above the +0.5 anomaly for warm (El Niño) events and at or below the -0.5 anomaly for cold (La Niña) events.  The threshold is further broken down into: 
  • Weak (with a 0.5 to 0.9 SST anomaly)
  • Moderate (1.0 to 1.4)
  • Strong (1.5 to 1.9)
  • Very Strong (≥ 2.0) 
For the purpose of this blog post, for an event to be categorized as weak, moderate, strong or very strong, it must have equaled or exceeded the threshold for at least 3 consecutive overlapping 3-month periods. Here's an analysis of the 26 cold winters by ENSO state: 
  • Strong Nino: 0 of 26 (0%)
  • Moderate Nino: 1 of 26 (4%)
  • Weak Nino: 9 of 26 (35%)
  • Neutral positive: 4 of 26 (15%)
  • Neutral negative: 5 of 26 (19%)
  • Weak Nina: 6 of 26 (23%)
  • Moderate Nina: 1 of 26 (4%)
  • Strong Nina: 0 of 26 (0%)
Dec-Feb temperature anomalies during weak Nino's
Notice that out of all the cold years, the majority of them occurred during weak Nino's (35%). Also notice that out of all of those cold winters, none of them occurred with a strong Nino or a strong Nina

The map on the left depicts the temperature anomalies that occurred during a Weak Nino. Due to the data only going back to 1948, all of the years are not depicted, but this will give you a good idea. 

As you can see, a weak Nino is what we'd like to see come Dec-Feb. If the current one stays too strong, it could severely limit our cold this winter based on past analogs. keep in mind, Larry's study is based on temperatures, not precipitation.  


Nino and the Pacific Decadal Oscillation (PDO)

+PDO or Warm State
-PDO or Cold State

PicturePDO Index (http://www.daculaweather.com/4_pdo_index.php)

While it's easy to look at one specific weather pattern, there are many factors that determine how a winter will turn out, with the ENSO state being just one of those. But there are other teleconnections and long term patterns that also have an effect on our winter weather, and they all work in tandem with each other.  Graphs and Charts

Now we are going to turn our attention to the PDO state or Pacific Decadal Oscillation. First, the definition from the National Center for Environmental Information:

"The Pacific Decadal Oscillation (PDO) is often described as a long-lived El Niño-like pattern of Pacific climate variability (Zhang et al. 1997). As seen with the better-known El Niño/Southern Oscillation (ENSO), extremes in the PDO pattern are marked by widespread variations in the Pacific Basin and the North American climate. In parallel with the ENSO phenomenon, the extreme phases of the PDO have been classified as being either warm or cool, as defined by ocean temperature anomalies in the northeast and tropical Pacific Ocean. When SSTs are anomalously cool in the interior North Pacific and warm along the Pacific Coast, and when sea level pressures are below average over the North Pacific, the PDO has a positive value. When the climate anomaly patterns are reversed, with warm SST anomalies in the interior and cool SST anomalies along the North American coast, or above average sea level pressures over the North Pacific, the PDO has a negative value (Courtesy of Mantua, 1999). " 

Here's an analysis of the cold 26 winters by DJF averaged PDO status:
  • + PDO: 18 of 58 (31%)
  • - PDO: 8 of 57 (14%)
Again, much as it was with weak Nino's, many of our coldest winters occurred during a + PDO state, while only 8 out of 57 occurred during a negative PDO state. 



Nino and the North Atlantic Oscillation (NAO)

In order for us to get long lasting cold air that stays locked in, we need some blocking. There are several teleconnection patterns that aid in developing this blocking, one of which is the North Atlantic Oscillation or NAO. Graphs and Charts
Negative NAO (-NAO)
Positive NAO (+NAO)
Here's the definition of the NAO:
"The North Atlantic Oscillation (NAO) is a climatic phenomenon in the North Atlantic Ocean of fluctuations in the difference of atmospheric pressure at sea level between the Icelandic low and the Azores high. Through fluctuations in the strength of the Icelandic low and the Azores high, it controls the strength and direction of westerly winds and storm tracks across the North Atlantic. It is part of the Arctic Oscillation, and varies over time with no particular periodicity."
Strong positive phases of the NAO tend to be associated with above-average temperatures in the eastern United States and across northern Europe and below-average temperatures in Greenland and oftentimes across southern Europe and the Middle East. They are also associated with above-average precipitation over northern Europe and Scandinavia in winter, and below-average precipitation over southern and central Europe. Opposite patterns of temperature and precipitation anomalies are typically observed during strong negative phases of the NAO. For us, negative is what we're looking for in the winter. 

Let's take a look at the analysis of the cold 26 Dec-Feb winters by averaged NAO status:
  • – NAO: 19 of 48 (40%)
  • + NAO: 7 of 67 (10%)
Again, a large number of the cold winters had a negative NAO. 


Piecing It All Together...


Now let's take the combination of the ENSO state (in our case, Nino), and factor in the PDO and NAO and let's see what we get. Here's the analysis of the 26 cold winters by a combination of Dec-Feb averaged PDO and NAO status:
  • + PDO/-NAO: 12 of 25 (46%)
  • - PDO/-NAO: 7 of 23 (30%) (all 7 had (PDO – NAO) > 0)
  • + PDO/+NAO: 6 of 33 (18%)
  • - PDO/+NAO: 1 of 34 (3%)
Things start to change a little. Obviously, the combination of +PDO and -NAO are the best combination, and that makes perfect sense. During the positive phase of the PDO, the wintertime Aleutian low is deepened and shifted southward, warm/humid air is advected along the North American west coast and temperatures are higher than usual from the Pacific Northwest to Alaska but below normal in Mexico and the southeastern United States. Add the effects of the blocking provided by the NAO and you lock in the cold air instead of having it rush out to sea. 

Now, let's really lay it out.
Here's "Brother Larry's" analysis of the 26 cold winters by a combination of ENSO state and Dec-Feb averaged PDO and NAO status:

Strong Nino:
  • +PDO/-NAO: 0 of 6 (0%)
  • -PDO/-NAO: 0 of 1 (0%)
  • +PDO/+NAO: 0 of 7 (0%)
  • -PDO/+NAO: 0 of 1 (0%)

Moderate Nino:
  • +PDO/-NAO: 1 of 2 (50%)
  • -PDO/-NAO: 0 of 0 (N/A)
  • +PDO/+NAO: 0 of 0 (N/A)
  • -PDO/+NAO: 0 of 2 (0%)

Weak Nino:
  • +PDO/-NAO: 6 of 7 (86%)
  • -PDO/-NAO: 2 of 2 (100%)
  • +PDO/+NAO: 1 of 2 (50%)
  • -PDO/+NAO: 0 of 4 (0%)

Neutral Positive:
  • +PDO/-NAO: 2 of 4 (50%)
  • -PDO/-NAO: 0 of 3 (0%)
  • +PDO/+NAO: 2 of 10 (20%)
  • -PDO/+NAO: 0 of 5 (0%)

Neutral Negative:
  • +PDO/-NAO: 1 of 3 (33%)
  • -PDO/-NAO: 2 of 7 (29%)
  • +PDO/+NAO: 2 of 10 (20%)
  • -PDO/+NAO: 0 of 5 (0%)

Weak Nina:
  • +PDO/-NAO: 2 of 2 (100%)
  • -PDO/-NAO: 3 of 5 (60%)
  • +PDO/+NAO: 1 of 4 (25%)
  • -PDO/+NAO: 0 of 7 (0%)

Moderate Nina:
  • +PDO/-NAO: 0 of 1 (0%)
  • -PDO/-NAO: 0 of 3 (0%)
  • +PDO/+NAO: 0 of 0 (N/A)
  • -PDO/+NAO: 1 of 4 (25%)

Strong Nina:
  • +PDO/-NAO: 0 of 0 (N/A)
  • -PDO/-NAO: 0 of 2 (0%)
  • +PDO/+NAO: 0 of 0 (N/A)
  • -PDO/+NAO: 0 of 6 (0%)



  • Weak Nino’s give the highest percentage chance for cold of any of the ENSO states by far (with weak Nina’s second); however, a combo of -PDO/+NAO seems to make it difficult even for weak Nino’s.
  • A very impressive 13 of 16 (81%) of the aggregate of weak Nino's and weak Nina's with -NAO were cold.
  • Don't ever bet on cold with either a strong Nino or a strong Nina since none of the 23 were cold.
  • +PDO about doubles the percent chance for cold versus a -PDO.
  • A -NAO more than doubles the percentage chance for cold versus a +NAO and a somewhat higher chance than a +PDO gives. So, I give small edge to –NAO over a +PDO regarding cold prospects. Regardless, both are very important.
  • A +PDO/-NAO combo gives close to twice the percentage chance for cold versus the percentage chance for all PDO/NAO combos in the aggregate.
  • A -PDO/-NAO is the next best combo for cold prospects, but mainly if NAO is more negative than PDO.
  • Don't ever bet on cold with a combination of -PDO/+NAO, since only 1 out of 34 were cold.
  • If there is a +PDO, the chances for a –NAO appear to be high for only weak to moderate Nino’s. The chances appear to only be about 50-50 for strong Nino’s. For neutral ENSO, the chances seem to be surprisingly low (partial negative correlation suggested).
  • The best shot at a +PDO/-NAO combo appears to be with a weak to moderate Nino's. On the other hand, only 3 of 34 (9%) Nina’s had a +PDO/-NAO.
  • A pretty high 17 of 34 (50%) Nina’s had a –PDO/+NAO.
  • A +PDO is difficult with a moderate to strong Nina. Only one out of 16 (6%) had one.
  • A –PDO seems rather difficult with a strong Nino, only 2 out of 15 (13%) had one.
  • +PDO and –PDO winters are about evenly split. But +NAO winters have been a bit more common than -NAO in long term: 58% vs. 42%.


Winter Precipitation

Regarding wintry precipitation for Atlanta, when looking at the three standalone super Nino's (1972-1973, 1982 -1983, 1997-1998) as well as the six strong to super strong 2nd year Nino's (1877-1888, 1888-1889, 1896-1887, 1905-1906, 1940-1941, 1987-1988), Atlanta more often than not, had one major winter storm, but not always:
  • 1877-1888: 2.5" 1/3/1878 & a non-major freezing rain followed on 1/9/1878
  • 1888-1889: 6" 2/21/1889
  • 1896-1967: 6.2" 12/2/1896
  • 1905-1906: 6.2" for the season including 3.5"+ major 1/26/1906, measurable snow D, J, and F
  • 1940 -1941: only T of snow
  • 1972 -1973: historic ZR 1/7-8/1973, which included 1" mainly ice pellets; (also, historic snow hit central GA in Feb though Atlanta missed that one)
  • 1982 -1983: 10.3" S/IP for season including 7.9" 3/24/1983 (heaviest since 1940) and measurable snow J, F, and M
  • 1987- 1988: 4.2" of mainly IP 1/7/1988 (would have been ~8" if all snow)
  • 1997- 1998: only 0.6" 12/29/1997



View the full article


Sign in to follow this  


Recommended Comments

There are no comments to display.

Add a comment...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Posts

    • I want to trust the pioneer model, but it is the outlier of the models and has been so wrong last couple of years! We just can’t seem to get out of this La Niña like pattern and before you know it we will be in a La Niña down the Road! Have really enjoyed last couple of days! 
    • Saturday, September 21, 2019 Model Diagnostic Discussion NWS Weather Prediction Center College Park MD 257 AM EDT Sat Sep 21 2019 Valid Sep 21/0000 UTC thru Sep 24/1200 UTC 00Z Model Evaluation with Preferences and Forecast Confidence ...Synopsis... The synoptic pattern will feature a rather strong deep layer ridge across a large portion of the eastern U.S. this weekend, as a pair of fairly robust northern stream troughs traverse the western and central U.S. The first trough will advance east across the central/northern Plains and upper Midwest Saturday and Sunday along with an attendant cold front and a multiple waves of low pressure riding up along it. This energy will break down the ridge over the East by later Sunday and will allow a cold front to arrive across the OH Valley and lower Great Lakes region by Monday. The height falls are expected to sharpen a bit over the Northeast by Tuesday, and there is support for a somewhat stronger surface low to develop over the interior of northern New England as the cold front pushes off the East Coast. Meanwhile, a second upstream trough will cross the West Coast and move into the Great Basin by Monday before then cutting off from the westerlies and dropping into the Southwest on Tuesday. This upper trough will be a key player in also steering the moisture and energy associated with Hurricane Lorena up across areas of northwest Mexico and into parts of the Four Corners region on Monday and Tuesday. By the end of the period, a new northern stream trough will begin to advance across the northern High Plains along with a cold front. ...Deep trough crossing the Plains/Midwest this weekend... ...Energy closing off over the Northeast early next week... Preference: General model blend...through 60 hours             Non-GFS blend...after 60 hours Confidence: Slightly above average Model differences with the first trough are rather small through most of the weekend as the energy crosses the Intermountain West and advances across the Plains and Midwest, but by later Monday and Tuesday as the energy crosses the Great Lakes and moves into the Northeast, the 00Z GFS becomes a bit more progressive than the remaining guidance. The 00Z non-NCEP guidance is now all very well clustered and close to that of the 00Z NAM. The 00Z GEFS mean is a tad more progressive than the non-GFS suite of models toward the end of the period, but not as progressive as the deterministic GFS, which suggests the GFS is a bit of an outlier at least by the end of the period. Will recommend a general model blend through about 60 hours, and then a non-GFS blend thereafter. ...Upper trough digging into the West by Monday... ...Closed low evolution over the Southwest by Tuesday... ...Moisture/energy associated with Hurricane Lorena... Preference: Blend of the 00Z NAM/GFS and 00Z UKMET Confidence: Average...becoming below average by Mon/Tues Regarding the second trough, the model guidance is well clustered and in good agreement with respect to timing and depth as the system arrives on Monday in across the Intermountain West. However, by Tuesday as the energy cuts off and drops south into the Southwest, the 00Z ECMWF is positioned west of the 00Z NAM/GFS and 00Z UKMET solutions which are now pretty well clustered. The 00Z CMC is now the farthest east solution by the end of the period. The spread in the guidance does still play a role in how the moisture and energy associated with Hurricane Lorena evolves which is expected to lift north through the Gulf of CA and across adjacent areas of northwest Mexico through the period. The 00Z GEFS mean is farthest east with its trough/closed low at the end of the period and is even a little east of the CMC solution. The 12Z ECENS mean tends to support the majority cluster of solutions, but it does suggest the 00Z ECMWF may be a tad too far west. Based on the latest clustering of solutions, a blend of the 00Z NAM/GFS and 00Z UKMET will be preferred. Confidence becomes a bit limited late in the period though based on the relatively large ensemble spread seen with the height fall details. ...Upper trough over the northern High Plains by Tuesday... Preference: General model blend Confidence: Above average The models have good agreement overall with the details of a northern stream trough and associated cold front crossing the northern High Plains by Tuesday. The 00Z GFS is just a tad sharper than the remaining guidance, but the model differences are modest enough that a general model blend can be preferred. Orrison Model Diagnostic DiscussionPreferred Height Pattern500 mb Height TrendsForecast issued by the Weather Prediction Center
    • Gooooood morning! Are you ready for some football? 🙂 It's going to be a great day for it!   There are big differences in temperatures between the Euro and GFS as we move out a little further, and I'm siding with the GFS right now. The Euro wasn't seeing this current cool down in advance, and it's like it's having a difficult time seeing any cool at all.  In case you were wondering, here is Weatherbell's look at fall and a preliminary look at winter temperature anomalies.   They have lots of reasoning behind what you see here, and these maps are based on August/early September inputs. Other than that... crickets.... As long as we stay under a dome og high pressure... nothing changes and nothing happens. I noticed in the forecast that for the next several days, the winds would be 0-5 mph, or for all practical purposes, non-existent.  Hang in there... watch some football, and have a cold one for me! 🙂  Cheers! 🍺 I hope everyone has a GREAT weekend! 🍺 It sure is nice to be home!   
    • Rip Current Statement issued September 21 at 3:44AM EDT until September 21 at 8:00PM EDT by NWS View the full article
    • Rip Current Statement issued September 21 at 3:04AM EDT until September 21 at 9:00PM EDT by NWS View the full article
  • Create New...