Skip to main content

Winter is coming: NOAA’s 2017-2018 Winter Outlook

If it’s October, it must be time for my annual blog post detailing NOAA’s forecast for the upcoming winter. And as always seems to be the case (well not in 2015), we’re still trying to figure out what will happen across the tropical Pacific. Here I look ahead to provide some insight as to what is most likely to occur during the upcoming winter with regards to both precipitation and temperature.

Standard reminder: probabilities are not certainties

Before discussing the outlooks, I do want to remind readers again that these forecasts are probabilities (% chance) for below, near, or above-average climate outcomes with the maps showing only the most likely outcome (1). Because the probabilities shown are less than 100% (of course), it means there is no guarantee you will see temperature or precipitation departures from normal that match the color on the map. As we’ve explained in earlier blog posts, even when one outcome is more likely than another, there is still always a chance that a less favored outcome will occur (witness precipitation during the last two winters in California).

Cut to the chase: What’s the outlook for this winter

Map of U.S. precipitation outlook for winter 2017-18

Places where the forecast odds favor a much drier than usual winter (brown colors) or much wetter than usual winter (blue-green), or where the probability of a dry winter, a wet winter, or a near-normal winter are all equal (white). The darker the color, the stronger the chance of that outcome (not the bigger the departure from average). Click image for version that includes Alaska and Hawaii. NOAA Climate.gov map, based on data from NOAA CPC. 

Both the temperature and precipitation outlooks lean on typical La Niña impacts, particularly those of the past 30 years, and bear some resemblance to the outlooks issued for last winter (not surprisingly since the forecast guidance is similar – more on that below). In the image above, the winter precipitation outlook favors below-normal precipitation across the entire southern U. S., with probabilities greatest (exceeding 50%) along the eastern Gulf Coast to the coasts of northern Florida, Georgia, and southern South Carolina. In contrast, above-average precipitation is more likely across much of the northern parts of the country, in the northern Rockies, around the Great Lakes, in Hawaii, and western Alaska.

Map of U.S. temperature outlook for winter 2017-18

Places where the forecast odds favor a much colder than usual winter (blue colors) or much warmer than usual winter (red), or where the probability of a cold winter, a warm winter, or a near-normal winter are all equal (white). The darker the color, the stronger the chance of that outcome (not the bigger the departure from average). Click image for version that includes Alaska and Hawaii. NOAA Climate.gov map, based on data from NOAA CPC. 

The temperature outlook shown above indicates above-average temperatures across the southern US, extending northward out West through the central Rockies and all the way up to Maine in the eastern part of the nation. Above-average temperatures are also favored in Hawaii and in western and northern Alaska. Chances are greatest in an area extending from the desert Southwest to central and southern Texas and Louisiana (greater than 50%).

Probabilities are tilted toward colder-than-normal temperatures along the northern tier from the Pacific Northwest to Minnesota and also in southeastern Alaska. However, the likelihood of below-average temperatures across the North is modest, with no probabilities in these regions reaching 50%.

Both maps include blank regions where neither above-, nor near-, nor below-normal is favored. These areas (shown in white and labeled EC for “equal chances”), have the same chance for above-, near-, or below-normal (33.33%). This doesn’t mean that near-average temperature or precipitation is expected this winter in those regions, but rather that there’s no tilt in the odds toward any of the three outcomes.

The NOAA Climate Prediction Center (CPC) issues probabilistic seasonal temperature and precipitation forecasts so users can understand risk and opportunities when making climate-sensitive decisions. However, keep in mind that these outlooks will primarily benefit those who have the wherewithal to  play the long game. The maps show only the most likely outcome where there is greater confidence, but this is not the only possible outcome. As we’ve seen in recent California winters, even the less likely outcome sometimes does occur. 

Déjà vu all over again?

Don’t worry if the preceding paragraphs seemed familiar. You’re not crazy. Regular readers may remember that last year at this time, we were also anticipating the emergence of La Niña later in the fall.  Fast forward one year and, well, the situation doesn’t look much different. If La Niña were to develop this year, certain patterns of temperature and precipitation would be favored across the United States. Over the past few years, we’ve discussed the patterns preferred by El Niño and La Niña, and recently, Tom has presented figures showing what we’ve seen historically during all La Niña winters between 1950 and now (precipitation and temperature).

Six rows of small U.S. maps showing the temperature anomalies during every La Niña winter from 1950-2016

December-February temperatures compared to the 1981-2010 average during each La Niña winter since records began in 1950. Gray lines under the maps indicate event strength: strong (dark gray), moderate (medium gray), and weak (light gray). NOAA Climate.gov image based on climate division data from NOAA ESRL. 

ENSO La Nina winters since 1950

December-February precipitation compared to the 1981-2010 average during each La Niña winter since records began in 1950. Gray lines under the maps indicate event strength: strong (dark gray), moderate (medium gray), and weak (light gray). NOAA Climate.gov image based on climate division data from NOAA ESRL. 

From these graphics, it’s easy to see that there’s been a fair amount of variability in the winter temperature and precipitation patterns during La Niña, but also that there are some clear tendencies for above or below normal temperature or precipitation in some regions. 

Another way to examine the common features of La Niña winters is to create a composite map (an average of all of these individual maps). This will highlight those regions that often have temperature or precipitation anomalies of the same sign. For temperature, there’s a strong tendency for temperatures to be below average across some of the West and North, particularly in the Northern Plains, with a weaker signal for above-average temperatures in the Southeast, as shown in the image below. 

U.S. map of winter temperature anomalies averaged over all La Niña winter from 1950-2016

Winter temperature differences from average (1981-2010) in degrees F during La Niña winters dating back to 1950. Temperatures tend to be colder than average across the northern Plains and warmer than average across the southern tier of the United States. NOAA Climate.gov image using data from ESRL and NCEI.

U.S. map of winter precipitation anomalies averaged over all La Niña winters from 1950-2016

Winter precipitation differences from average (1981-2010) in inches during La Niña winters dating back to 1950. Precipitation tends to be below-average across the southern tier of the United States and wetter than average across the Pacific Northwest and Ohio Valley. NOAA Climate.gov image using data from ESRL and NCEI.

The precipitation pattern, presented above, shows negative anomalies (indicating below-normal rainfall) across the entire southern part of the country with a weaker signal of above-average precipitation in the Ohio Valley and in the Pacific Northwest and the northern Rockies. 

However, these figures are based on about 20 different La Niña episodes, many of them from the 1950s, 1960s, and 1970s, and we have not removed the longer-term trends from the temperature and precipitation data used here. As Tom pointed out last month, the trend is an important component of seasonal temperature forecasts. It’s fairly trivial to break the sample size in half, and compare the temperature patterns for the older half to the more recent half. That provides a significantly different picture, with the average of the latest events much warmer than the earlier ones.  We can see this by comparing the right image below (more recent events) with the one to the left of it (older events). 

Two U.S. maps comparing winter temperature anomalies during La Niña winters at the start of the observation record to anomalies during the most recent La Niña winters

Comparison of winter temperature differences from average (1981-2010) in degrees F between the earliest and most recent ten La Niña winters dating back to 1950. Temperatures tend to be warmer across much of the country during the most recent ten La Niña events as compared to the earliest ten La Niña events . NOAA Climate.gov image using data from ESRL and NCEI.

This picture is consistent with long term warming trends over the United States.  These historical relationships along with guidance provided by a suite of computer models plays a strong role in the final outlooks. Differences between the two periods for the precipitation composites are much smaller and therefore are not shown here. 

Lead editor: Nat Johnson

Footnotes

(1) The terciles, technically, are the 33.33 and 66.67 percentile positions in the distribution. In other words, they are the boundaries between the lower and middle thirds of the distribution, and between the middle and upper thirds. These two boundaries define three categories: below-normal, near-normal and above-normal. In the maps, the CPC forecasts show the probability of the favored category only when there is a favored category; otherwise, they show EC (“equal chances”). Often, the near-normal category remains at 33.33%, and the category opposite the favored one is below 33.33% by the same amount that the favored category is above 33.33%. When the probability of the favored category becomes very large, such as 70% (which is very rare), the above rule for assigning the probabilities for the two non-favored categories becomes different. 

Comments

Legend title may be wrong for Precipitation figure.

Yep, you're right. It should be "difference from normal precip" in inches. We will get that updated first thing tomorrow.  

 

In reply to by Tyler

I have learned so much from reading through these graphs and articles. Thank you.

Any impact this winter from the NAO that's under discussion? The monthly time series was strongly positive but seems to be in some fli-flop pattern the past few months?

Thanks for the comment. The NAO is much harder to forecast this far in advance as compared to things like ENSO. However, we know that the sign of the NAO can mean the prevalence of certain patterns across the United States. In fact, an earlier ENSO blog post talked about what the phases of NAO mean for the United States. I am sure forecasters will be looking at forecasts of the NAO as we get closer to winter to see what type of influence it may have.

In reply to by Dave

Is there somewhere a more comprehensive reading that ties together ENSO and both NAO as well as PDO? Would be great if blogs like this included that information and how they think it might affect specific regions of the country (especially the boom or bust West Coast winter precip) in combination with ENSO. Last year many started off as being rather pessimistic about the precip outlook for California especially, but then we got hammered with several atmo rivers that made from one of the highest rain seasons ever. I still haven't been able to wrap my head around how that outlook was so off. The only thing I can see as far as SST anoms was that we suddenly got a very large and colder than average region in the North Pacific where previously the "warm blob" had setup in years prior, feeding the "ridiculously resilient ridge."

In reply to by tom.diliberto

In the last temperature map comparing the earliest 10 year La Nina's (1950-1960?) to the most recent 10 years. Are the most recent ten years 2000-2010 or 2006-2016?

The earliest 10 La Nina winters were determined by taking the first 10 cases where the DJF values on this chart were blue marking a La Nina event. The most recent 10 La Nina winters were determined by taking the last 10 cases on the same chart.

In reply to by mike shapley

Looks like a lot of money is spent on this technical mumbo jumbo that does not equate to more accurate forecasts. Glad the current White House admin recognizes as much and has proposed significant budget cuts to both the NOAA and National Weather Service in the upcoming fiscal year.

Hi Hal, we've been trying to avoid using too much "technical mumbo jumbo" in our explanations as we know scientific jargon can be too dense. If you are actually curious about how our probabilistic forecasts perform, we provide two sites where you can look at the numbers: one and two. And if you'd like an explainer on how to properly verify our forecasts to see if they truly are accurate or not, I wrote a three part series on just that topic. Part one, part two and part three.

Making judgements about scientific study from a ideological point of view is both silly and dangerous. You think a public budget for weather and climate forecasting is not worth the money? Do you have any idea how many lives have been saved by accurate hurricane forecasts?

In response to the comment from the Disney villain, it's actually a well-calibrated mix of mumbo jumbo, some balderdash, and a pinch of hocus-pocus. (Technical tidbit: the European model has all of the aforementioned plus an additional weighting of poppy-cock but exchanges Imperial hocus-pocus for the German Hokuspokus, which some believe is the key to its generally superior predictions).

In reply to by Hal Jones

I live in the High Cascades region of the Pacific Northwest. Last year’s prediction for the west coast was spectacularly off. (https://www.climate.gov/news-features/blogs/enso/what-expect-winter-noaa’s-2016-17-winter-outlook) I wonder if any study has been made of how far off the predictions have been from the actuals each year, and is there a trend that predictions are less and less accurate, due to the effects of climate change?

Hi Lisa, I am sure there are research papers being written as we speak about the winter rains across the west coast last year.

As a note, last year's winter outlook had a slight tilt towards wetter conditions for much of Washington but Oregon and much of California had an equal chance for above, normal or below-average rainfall (southern California had a slight tilt towards drier than average). An equal chance forecast isn't necessarily an incredibly far off prediction. It's just an acknowledgment that sometimes signals for winter precipitation are conflicting or difficult to determine months in advance.

In reply to by Lisa

I LIVE IN GOLDENDALE WASHINGTON AND OUR TOWN IS EAST OF MT, ADAMS AND NORTH OF THE COLUMBIA RIVER. THEY HAVE SAID WE ARE IN THE BANANA BELT, NOT SURE WHAT THAT MEANS. I SEE THAT EACH YEAR BRING SOMETHING DIFFERENT . THIS IS SOME GOOD READINGS AND THE MAPS THAT SHOW OUTCOME. I WILL BE FOLLOWING EVERY DAY THANK YOU

So, I wanted to ask how do you figure in what you have seen predicted vs the interference (or "assistance") of what the public has come to know has "weather management", in the past? In other words do you see any correlation between the change of weather patterns over the years and controlled weather changing previous predictions. If so do you feel it's for better or worse? From a homeschooling family using weather and science currently we are using various sources to discover earth changes. Are they naturally occurring or human made. For example earthquakes in Oklahoma related to nearby fracking and oil refinery or the fault on Mississippi river.

Thanks for the summary of previous La Nina events and forecast for the 2018 winter season. There's always going to be some variability with seasonal forecasts, and some people are always going to struggle with that. My question is how you do think the active hurricane pattern we observed this year in the Atlantic will affect the anticipated La Nina event and if NOAA has evaluated any links between the two. Thanks again!

Hi Alex,

The hurricane pattern in the Atlantic was certainly active this year! However, while ENSO affects hurricanes, hurricanes don't affect ENSO. Sort of like how allergy season affects sneezes, but sneezes don't affect allergy season--ENSO changes the background state in which hurricanes form.

HI, I'd like to commend you guys on your website. I read it a lot, and while I am sad that it's getting warmer where I live (NM USA), it's helpful to understand what I can expect. I an an avid small farmer, and knowing that I can plant early/later than the old averages help! I have also seen a lot of changes in the consistency of spring weather, going from two weeks of quite warm (and the trees start blossoming) to sub zero temps for a few days (especially in April) that just knock out the fruit. So it's helpful to know what to expect as far as that goes. Our last moderate La Nina (10-11) saw many days of sub zero weather in Feb that took everyone by surprise, burst pipes and ended up causing a natural gas shortage because the producers and the state weren't ready for it. I am grateful for the work you do, to keep myself informed about what i can expect, keep it up!

Looking at the next-to-last map titled, “Precipitation patterns averaged across all La Niña winters since 1950,” I see a demarcation at the northern border of California and the southern border of Oregon that doesn’t look natural because it strictly follows the border. There are also similar changes across state boundaries for Tennessee & North Carolina; Kentucky & Virginia; and West Virginia & Virginia. Are these changes in average precipitation at state boundaries due to extrapolating data because of a paucity of stations gathering data in those areas? Similar demarcations for average temperature do not appear at those boundaries in the preceding map titled, “Temperature patterns averaged across all La Niña winters since 1950.”

Hi James, 

Yes, your observations are spot on!  There are some transitions and gradients in some of the maps that are not physically realistic.  The reason is that these maps are based on data that are averaged within the U.S. Climate Divisions: https://www.ncdc.noaa.gov/monitoring-references/maps/us-climate-divisio…. As you can see, there are large variations in the sizes of these climate divisions.  In regions where you have two neighboring climate divisions covering a large area, it's possible for the average precipitation to be quite distinct between the two divisions.  This would lead to unnaturally sharp contrasts in the maps you mention.  If the maps instead were based on data averaged in small, regularly spaced grids, then we would expect the contrasts to be more smoothly varying.

It would be a great saving on gas bills if NYC has a normal snowfall 25" or so with near normal tempertures.

Thanks for providing such great information. Regardless of the exact accuracy every year, seeing these trends does provide insight.

The La Niña graphics seem to show New England as being historically colder than average (and tend towards wetter), but this winter's forecast predicts it will be warmer than usual and last year was really warm too. Is there something else causing that?

Hi CJ,

You're right that in the average of all La Nina winters since 1950, there is a very slight tendency for below average temperatures in New England. However, one of the figures above shows that the tendency is reversed in the last 10 La Nina winters, indicating that the long-term warming trend is counteracting that slight tendency for cooler temperatures in New England.  The forecast models of the North American Multi-Model Ensemble (NMME) (http://www.cpc.ncep.noaa.gov/products/NMME/seasanom.shtml) also have been fairly consistent in forecasting a tendency for warmer than average across the entire eastern U.S., so they seem to be honing in on this La Nina + trend signal.  Because no two El Nino or La Nina episodes are the same, it's also possible that there are particular features of this forecast event that may lead to deviations from the classic La Nina signal, but it is difficult to tease out exactly what those features may be without additional study.

The precipitation signal over New England is quite weak, and NOAA's Winter Outlook reflects this.  That doesn't mean New England won't see a wet (or dry) winter - either option certainly is possible!  New England is impacted by other important large-scale climate patterns such as the North Atlantic Oscillation (NAO), but those patterns are much more difficult to predict more than a couple weeks in advance.

 

In reply to by CJ

The Southeast Winter outlook was way off. Warmer than normal? Quite the opposite, Ive been in North Florida for 50+ years, I dont recall it ever being as cold down here as it has been thus far. 2 snowfalls in Florida in one year is rare but thats what weve had. And did I forget to say: It's COLD! What happened?

Well, we can't call it yet, winter's not over! These forecasts are for the winter average, which means December-February. And in fact, December's average temperature in Florida was 61.9°F, which is 3.1 degrees Fahrenheit warmer than the 1981-2010 average. We won't know for another ~6 weeks whether the recent weeks of very cold temperatures will be enough to cancel out December's warmth, not to mention what might happen in February.  The latest monthly outlook still favors (though not super strongly) much warmer than average temperatures across the South in February.  [[{"fid":"34508","view_mode":"default","fields":{"format":"default","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_caption[und][0][format]":"full_html"},"type":"media","field_deltas":{"1":{"format":"default","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_caption[und][0][format]":"full_html"}},"link_text":null,"attributes":{"class":"media-element file-default","data-delta":"1"}}]]

So as far as whether this winter's seasonal outlook was a bust for Florida or anywhere else, we still have to wait and see. 

 

In reply to by Gregg

Is there anything to the year to year rebound effect? Had an extremely cold winter in Jan. and Feb. 2018, in S.E. CONUS. Could it be the opposite situation in 2019?

This post is on our older winter outlook from last year.  Check back with the blog in mid-late October for our article about winter 2018-19!  

In reply to by Tony S.

Add new comment