RealClimate logo

El Nino, Global Warming, and Anomalous U.S. Winter Warmth

Filed under: — mike @ 8 January 2007 - (Slovenčina) (Svenska)

It has now become all too common. Peculiar weather precipitates immediate blame on global warming by some, and equally immediate pronouncements by others (curiously, quite often the National Oceanic and Atmospheric Administration in recent years) that global warming can’t possibly be to blame. The reality, as we’ve often remarked here before, is that absolute statements of neither sort are scientifically defensible. Meteorological anomalies cannot be purely attributed to deterministic factors, let alone any one specific such factor (e.g. either global warming or a hypothetical long-term climate oscillation).

Lets consider the latest such example. In an odd repeat of last year (the ‘groundhog day’ analogy growing ever more appropriate), we find ourselves well into the meteorological Northern Hemisphere winter (Dec-Feb) with little evidence over large parts of the country (most noteably the eastern and central U.S.) that it ever really began. Unsurprisingly, numerous news stories have popped up asking whether global warming might be to blame. Almost as if on cue, representatives from NOAA’s National Weather Service have been dispatched to tell us that the event e.g. “has absolutely nothing to do with global warming”, but instead is entirely due to the impact of the current El Nino event.

[Update 1/9/07: NOAA coincidentally has announced today that 2006 was officially the warmest year on record for the U.S.]
[Update 2/11/08: It got bumped to second place. ]

So what’s really going on? The pattern so far this winter (admittedly after only 1 month) looks (figure on the immediate right) like a stronger version of what was observed last winter (figure to the far right–note that these anomalies reflect differences relative to a relatively warm 1971-2000 base period, this tends to decrease the amplitude of positive anomalies relative to the more commonly used, cooler 1961-1990 base period). This poses the first obvious conundrum for the pure “El Nino” attribution of the current warmth: since we were actually in a (weak) La Nina (i.e., the opposite of ‘El Nino’) last winter, how is it that we can explain away the anomalous winter U.S. warmth so far this winter by ‘El Nino’ when anomalous winter warmth last year occured in its absence?

The second conundrum with this explanation is that, while El Nino typically does perturb the winter Northern Hemisphere jet stream in a way that favors anomalous warmth over much of the northern half of the U.S., the typical amplitude of the warming (see Figure below right) is about 1C (i.e., about 2F). The current anomaly is roughly five times as large as this. One therefore cannot sensibly argue that the current U.S. winter temperature anomalies are attributed entirely to the current moderate El Nino event.

Indeed, though the current pattern of winter U.S. warmth looks much more like the pattern predicted by climate models as a response to anthropogenic forcing (see Figure below left) than the typical ‘El Nino’ pattern, neither can one attribute this warmth to anthropogenic forcing. As we are fond of reminding our readers, one cannot attribute a specific meteorological event, an anomalous season, or even (as seems may be the case here, depending on the next 2 months) two anomalous seasons in a row, to climate change. Moreover, not even the most extreme scenario for the next century predicts temperature changes over North America as large as the anomalies witnessed this past month. But one can argue that the pattern of anomalous winter warmth seen last year, and so far this year, is in the direction of what the models predict.

In reality, the individual roles of deterministic factors such as El Nino, anthropogenic climate change, and of purely random factors (i.e. “weather”) in the pattern observed thusfar this winter cannot even in principle be ascertained. What we do know, however, is that both anthropogenic climate change and El Nino favor, in a statistical sense, warmer winters over large parts of the U.S. When these factors act constructively, as is the case this winter, warmer temperatures are certainly more likely. Both factors also favor warmer global mean surface temperatures (the warming is one or two tenths of a degree C for a moderate to strong El Nino). It is precisely for this reason that some scientists are already concluding, with some justification, that 2007 stands a good chance of being the warmest year on record for the globe.

A few other issues are worthy of comment in the context of this discussion. A canard that has already been trotted out by climate change contrarians (and unfortunately parroted uncritically in some media reports) holds that weather in certain parts of the U.S. (e.g. blizzards and avalanches in Colorado) negates the observation of anomalous winter warmth. This argument is disingenuous at best. As clearly evident from the figure shown above, temperatures for the first month of this winter have been above normal across the United States (with the only exceptions being a couple small cold patches along the U.S./Mexico border). The large snowfall events in Boulder were not associated with cold temperatures, but instead with especially moisture-laden air masses passing through the region. If temperatures are at or below freezing (which is true even during this warmer-than-average winter in Colorado), that moisture will precipitate as snow, not rain. Indeed, snowfall is often predicted to increase in many regions in response to anthropogenic climate change, since warmer air, all other things being equal, holds more moisture, and therefore, the potential for greater amounts of precipitation whatever form that precipitation takes.

Another issue here involves the precise role of El Nino in climate change. El Nino has a profound influence on disparate regional weather phenomena. Witness for example the dramatic decrease in Atlantic tropical cyclones this most recent season relative to the previous one. This decrease can be attributed to the El Nino that developed over the crucial autumn season, which favored a strengthening of the upper level westerlies over the tropical North Atlantic, increased tropical Atlantic wind shear, and a consequently less favorable environment for tropical cyclogenesis.

If a particular seasonal anomaly appears to be related to El Nino, can we conclude that climate change played no role at all? Obviously not. It is possible, in fact probable, that climate change is actually influencing El Nino (e.g. favoring more frequent and larger El Nino events), although just how much is still very much an issue of active scientific debate. One of the key remaining puzzles in the science of climate change therefore involves figuring out just how El Nino itself might change in the future, a topic we’re certain to discuss here again in the future.

360 Responses to “El Nino, Global Warming, and Anomalous U.S. Winter Warmth”

  1. 351
    Steve Sadlov says:

    RE: #346 – It’s probably too soon to use the word “megadrought” however there are certainly precursors. I would not be surprised at all to see a 1950s repeat, and would not rule out a 1930s repeat.

  2. 352
    L. David Cooke says:

    RE: #349

    Hey Ike;

    One other question, if the collectors at ARM have only been in place since 1996 then how can you claim that the specific energy signatures for the various radiant atmospheric elements have been resolved a long time ago. What is even worse IMHO is your steadfast belief that you are correct in the light of what appears to be inaccurate data according to the ARM data collection team. Have you or anyone else gone back to the original works and calculated the corrections as deemed in the 2003, 2004 and 2006 work group recommendations?

    Finally, why does the desire to clearly define a direct measure seem to frighten you and Hank. Is it possible that the data will not support the theory? (That is very unlikely, whatever happened to crossing your Ts and dotting your Is? Is establishing a clean direct measurement a bad thing, I can assure you that would certainly silence the nay sayers once and for all. So why is it you want to avoid the pursuit of conclusive evidence?


  3. 353
    Hank Roberts says:

    David, you are asking these questions in various ways in several different threads scattered through, under topics set up for other matters. Can you get your material together on your own website (you say you have a server)? If you can display what you have and what you know, you have a much better chance of attracting the attention of someone who will help you.

    Eric Raymond will tell you in detail how to accomplish this:

    How To Ask Questions The Smart Way

    This is for people who asks question on the Net, especially on computer/technical questions. [Eric Steven Raymond]

  4. 354
    Ike Solem says:

    Numerous posts have pointed you to direct measures of longwave downwelling radiation; you should also keep in mind that the moon (which has no atmosphere) varies between 100C in the day to -150C at night, which is due to the lack of atmosphere; you can also try camping in the desert in the summer, when the low humidity results in hot days and freezing nights, all related to downwelling longwave radiation effects.

    As far as BOM and the record of Arctic temperature anomalies: their datasets match others, and are also available for inspection, unlike NOAA’s – for example, compare the Australian BOM’s January 2006 anomaly to that of the National Snow and Ice Data Center’s for Jan-Jul 2006 (see Figure 5 at bottom of page) – this temperature anomaly has also been reported elsewhere, and yet NOAA seems to be excluding this data from their SST analysis. Lack of transparency and refusing to reveal data are very bad signs in scientific inquiry…

    That’s not the end of it: NOAA just revised their CO2 atmospheric growth rate down because “December data should not have been published” – the whole story is told at “Surge in carbon levels raises fears of runaway warming, David Adam, Friday January 19, 2007 Guardian UK” – though I imagine they don’t mean Venus-style runaway warming, but rather an accelerating rate of the current global warming trend.

    Quote: “The US National Oceanic and Atmospheric Administration (Noaa) has now told us that the story below is based on preliminary data for December, which it should not have published. It has withdrawn the data pending further analysis.”

    I think that now that there is no question about the accuracy of radiation models, the last refuge of the contrarians is to claim that ‘there isn’t enough data to be sure of anything’ – yet there are many satellite and other measurements (you cite ARM, but here is the Atmospheric Radiation Measurement program’s page on Downwelling Longwave Radiation – look for yourself! – here is yet another example:skyrad20’s data plot – and here’s a picture of the instrument in place. Claiming that the data can’t be trusted because it doesn’t extend back far enough is nonsense.

    When it comes to radiation models, the use of computers and the advancement of physical theory allows one to place a good deal of confidence in the models – for example, solar physicists tell us that it takes some 10,000 years for photons released from nuclear fusion reactions in the sun’s core to reach the surface – but would you say that statement is untrustworthy unless an actual ‘in-situ measurement of photon migration rates’ was made?

    See also the comments about Sherwood Idso’s discredited 1980 attacks on the basic radiation models in the Human Hand on Climate thread, #44.

  5. 355
    L. David Cooke says:

    RE: #353

    Hey Ike;

    Again you missed that my concern is the radiative downwelling related to GHG. The point I had been making was in relation to night time clear sky readings as a means to minimize confouding variables. The next step would be the implementation of discrete detectors, meaning highly frequency selective.

    If the detection capability across the bandwidth were sweepable such as using a variable diffraction grating on a detector with a sensitivity of 3 db across the entire range would be fantastic. (It would seem that a super conductor junction of yttrium-barium-copper-oxide (Y1Ba2Cu3O7) and a surface-stabilized ferroelectric liquid crystals (SSFLCs) that could sweep the frequency band would seem to be an optimal detector, if the detector chamber could be kept cool enough.) If such a detector could be devised then the discrete values from a single detector would reduce the standard deviation of error from multiple variables from multiple detectors.

    Now if you can see your way to define a means to establish the ratio of water vapor emission spectra at around 920nm to the emission spectra in the 1.2 um range we can get to a 1/2 wave harmonic of the CO2 shortwave value. Then if we can establish the ratio of short wave to long wave we might be able to remove the value associated with water vapor based on Lidar measurements leaving the radiative value of the CO2. If this were possible then we might be able to extract the direct value of CO2 attributed downwelling radiation. However, that leaves a lot of ifs…

    Dave Cooke

  6. 356
    aaron says:

    i reall really really need some one who is a expert in this field to answer there question
    its for school project
    huge grade
    pleseeeeeeeeeeeeeeeeee answer

    Person Interviewed: ____________________________
    Personâ??s Role: ________________________________


    1. Whatâ??s your opinion on Global Warming?

    2. Do you think recycling paper would help?

    3. Why should people recycle?

    4. Do you think there should be a recycling program at school?

    5. Should people be worried about Global Warming?

    6. Are there other people that believe Global Warming is important?

    7. If nothing were done, how would Global Warming affect our future?

    8. Is the government doing anything about global warming?

    9. Are there any negative things about recycling?

    10. Can recycling make a difference?

  7. 357
    Ike Solem says:


    You really do need to go camping in the desert; then you will feel the effect of low GHG (i.e. low H2O) at night. I have a feeling that your only experience with spectroscopic measurements is in a controlled laboratory setting looking at isolated chemical species.

    I notice that you are no longer attempting to question the radiative models of the atmosphere and are now only attempting to attack the issue of data collection – which you don’t seem to understand either. For example, where in the atmosphere is the main effect of CO2 and other infrared-absorbing gases most noticeable? What is the effect of temperature and pressure on the width of spectroscopic absorbtion curves? Your notions of emission spectra seem to be taken from an introductory physics textbook, but those spectra are collected in laboratories under conditions of low pressure and temperature.

    The atmosphere is a very dense and warm environment in comparison to such textbook emission spectra, and emission bands in such an environment are certain to overlap – but I hope you’re not trying to explain the observed downwelling longwave radiation as being due to something other than greenhouse gases; i.e. H20, CO2, CH4, N2O, CFC’s, etc. (notice that the definition of a ‘greenhouse gas’ is that it absorbs infrared, unlike O2 and N2). There is a ‘window’ in the CO2/H2O absorption (8-11 um, I believe) where CFC’s absorb, which is why they have a strong effect. See
    Infrared Radiation Parameterizations for the Minor CO2 Bands and for Several CFC Bands in the Window Region, Kratz et al 1993

    Hopefully that settles the issue…though I doubt it, somehow.

    You also seem to be ignoring the very real trends in the warming arctic SST temperature record, after attempting to defend NOAA’s questionable SST anomaly graphs – a topic that is far more closely related to the original post on El Nino and anomalous winter warmth. For an even more dramatic example of the differences between NOAA’s SST temps and others, see

    NOAA June 2006 (I managed to find their archives from 2000-2006, but not prior to that)

    Australian BOM June 2006

    The large (4C) temperature anomalies in the Arctic help explain the dramatic decline in Arctice sea ice over the past two years, don’t they? Why does NOAA’s graph not reflect this very real trend (see also Abrupt decline in the Arctic winter sea ice cover, Josefino C. Comiso – 2005-2006 winter periods). This matches the data on Arctic SST anomalies at the BOM archival site., which show a dramatic increase starting in 2005. Long-standing predictions of climate models are that the first major effects of global warming will be felt at the poles and at high altitudes, as well.

  8. 358
    Jodro says:

    Above I asked why all the climate feedback mechanisms appear to be ‘positive,’ ie ice cap melting results in more absorption of heat as there’s no more snow to reflect sunlight, etc. Is this just bad luck, or there reasons for them all being positive? The next question is, of course, can we develop ‘negative’ feedback mechanisms? Today the US government suggests that scientists develop giant mirrors etc to reflect sunlight… They really do have totally closed, one-track minds… any solution rather than cutting emissions… I suppose the idea is that in a 100 years from now Americans can still drive their SUV’s while we all live in pitch darkness…

    But here’s a simple solution I thought of… it sounds ridiculous, but perhaps someone is capable of doing the maths on whether it would compensate for the loss of snow cover; it’s also a nice participatory event, ie everyone can do something and feel they’re making a difference… what if everyone decorates the roof of their homes white or with some sun-reflecting paint? Ie what if every building on the planet reflects sunlight back into space? Would it make (enough of) a difference?

  9. 359

    Re “what if everyone decorates the roof of their homes white or with some sun-reflecting paint? Ie what if every building on the planet reflects sunlight back into space? Would it make (enough of) a difference?”

    Good idea in principle, but urban areas only make up 1-2% of Earth’s land surface, and land in turn is only 29.2% of total surface area. Might help regionally, though.

  10. 360
    Patrick Henry says:

    Our last two months in Greeley, Colorado have been two of the coldest on record. In January we were below normal by an average of 12 degrees, and high temperatures were below normal for 28 out of 31 days. It is the coldest winter we have had in about 30 years.

    [Response:The 90 day running average for the entire U.S. (Nov 1-Jan 30) is available from NOAA here. Most of Colorado was above normal (and this is based on a rather warm 1971-2000 baseline climatology to begin with). Only far eastern Colorado was below normal, and in this case only slightly so. Much of the U.S. is still running well above the 1971-2000 average. This is all despite the reason cold spell. If you think NOAA’s numbers are wrong, you should consider contacting them. Here at RC, we’ll go w/ NOAA’s hard numbers over the anecdotal, imprecise assertions of some of our readers. This post seems to have outlived it usefulness, so we’re closing the comment thread on this note. -mike]