Climate Science Glossary

Term Lookup

Enter a term in the search box to find its definition.

Settings

Use the controls in the far right panel to increase or decrease the number of terms automatically displayed (or to completely turn that feature off).

Term Lookup

Settings


All IPCC definitions taken from Climate Change 2007: The Physical Science Basis. Working Group I Contribution to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Annex I, Glossary, pp. 941-954. Cambridge University Press.

Home Arguments Software Resources Comments The Consensus Project Translations About Donate

Twitter Facebook YouTube Pinterest

RSS Posts RSS Comments Email Subscribe


Climate's changed before
It's the sun
It's not bad
There is no consensus
It's cooling
Models are unreliable
Temp record is unreliable
Animals and plants can adapt
It hasn't warmed since 1998
Antarctica is gaining ice
View All Arguments...



Username
Password
Keep me logged in
New? Register here
Forgot your password?

Latest Posts

Archives

Roy Spencer's Junk Science

Posted on 5 March 2012 by bbickmore

This is a cross-post from Barry Bickmore's blog

Roy Spencer recently posted an article on his blog called “Ten Years After the Warming,” in which he argues that there’s no excuse for a decade without much warming, because the radiative forcing is supposedly higher than it’s ever been.  Steve Milloy has also reposted the article on his aptly titled blog, JunkScience.com.  (In case you don’t remember, Steve Milloy is a Fox News commentator who goes about labeling as “junk science” any environmental issues that might precipitate some government regulation.  Yes, that includes links between second-hand smoke and cancer.)  Spencer’s main point is this:

"You cannot simply say a lack of warming in 10 years is not that unusual, and that there have been previous 10-year periods without warming, too. No, we are supposedly in uncharted territory with a maximum in radiative forcing of the climate system. One cannot compare on an equal basis the last 10 years with any previous decades without warming."

This is the same Roy Spencer who is constantly claiming that he can explain most of the warming trend over the last 100 years by appealing to various modes of natural variation in the climate, e.g., the Pacific Decadal Oscillation and the El Niño Southern Oscillation.  These climate oscillations depend on complicated stuff like deep ocean currents that are hard to predict, given that we don’t have that many observations of what the state of the system is like at any given time.  (In other words, it’s expensive and hard to measure deep ocean temperatures and currents, so we don’t have that many observations.)  Since these kinds of things are hard to predict exactly with a model, climatologists usually talk about long-term trends caused by external “forcing” (by things like CO2 emissions and variations in solar output), overprinted by random “natural variation”.  The main difference between Roy Spencer and the rest of the climatologists is that he thinks that natural variation is important over much longer time periods, whereas the others generally think it’s mainly important over about a decade or less.  For example, he complained in his book, , The Great Global Warming Blunder,

"The IPCC has taken for granted that there are no natural variations in global average temperatures once one gets beyond a time scale of ten years or so.  (p. 16)

The United Nations’ Intergovernmental Panel on Climate Change does acknowledge that there is natural climate variability on a year-to-year basis, and maybe even decade-to-decade.  After all, we have clear evidence that events like El Niño and La Niña cause some years to be warmer than others.  Yet the IPCC refuses to accept that the global warming (or cooling) on time scales of thirty years or more can also be caused by Mother Nature.  That, apparently, is humanity’s job.  (p. 1)"

In this latest article, however, Roy seems to be saying that the temperature should have kept going up pretty steeply because the external forcing from greenhouse gases has continued to rise.   The problem is that this is true ONLY if you ignore natural variation that might temporarily offset the external forcing.

Spencer’s newfound suspicion of decadal-scale natural variations is unfounded.  Foster and Rahmstorf (2011), for example, statistically removed the effects of El Niño/La Niña cycles, volcanoes, and solar variation, to produce the temperature evolution that WOULD HAVE occurred if these random, natural variations hadn’t happened.  Here’s what they got (Fig. 1).

Fig. 1 Animated gif of the actual temperature record alternating with the one Foster and Rahmstorf produced after removing the signal from ENSO, solar variation, and volcanoes.

There is still random variation evident in Fig. 1, so we obviously haven’t removed all the noise, but the trend is much more consistently upward, including during the last decade.  The fact is that natural variation can EASILY account for a leveling off of the temperature rise for a decade or so, and what’s more, even though climate models aren’t very good at predicting when these natural variations will occur, they do at least predict that they WILL happen (Santer et al., 2011).

The bottom line is that Roy Spencer has been arguing all along that natural variation can cause the temperature to go up or down for a while no matter what the external forcing is doing, and no matter how long the time period, but now he suddenly can’t imagine that this could happen over a single decade!

It’s also funny that Steve Milloy passed on Spencer’s assertions, but then just two days later he was promoting a paper in which Spencer argued that standard climate models are uncertain because “alternative hypotheses for the cause(s) of the warming, such as natural climate cycles or indirect forcing by the sun, have seen relatively little research.”  (And of course, Roy cited his book, The Great Global Warming Blunder to support this point.)  Milloy likes to label as “junk science” any science that leads to conclusions that might precipitate government regulations, but the fact is that he doesn’t have the expertise to understand the science he pans or the “alternative” science he promotes.

0 0

Bookmark and Share Printable Version  |  Link to this page

Comments

Comments 1 to 15:

  1. Its very similar to the stock market, which can drop and even not rise over a 10 year period, but the trend is unambiguously up.
    0 0
  2. Consistency is not an option in some quarters and that's fortunate because it allows us to easily discriminate between good science and bad science. Of the unfortunate side of this story we know.
    0 0
  3. The Foster and Rahmstorf study, along with other attribution studies ar of course powerful evedence that Spencer is quite mistaken. Furthermore, the entire issue of deeper ocean heat content is not even considered. The greater thermal inertia and energy storage capacity of the deeper ocean makes it a much better metric for seeing what actually was occurring during the past decade with Earth's energy balance. The troposphere has a far lower heat capacity and lower thermal inertia and is far more subject to the noise of short-term natural variations, and thus requires filtering to see any underlying trend. In looking at the energy storage of the deeper ocean over the past decade we see it stored more energy than any 10 year period out of the past 40. Hardly a sign of a planet that is cooling.
    0 0
  4. Can natural oscillations explain the climate? Roy Spencer would seem to argue that this is the case. Recently (Scaffeta’s Widget Problem) you had an article on Scafetta’s claims to be able to model climate just using natural cycles.

    At that time I produced two versions of a very simple regression model: the first using only sun spots (as a proxy for radiation), optical mean depth (for aerosols) and the Atlantic Multi-decadal oscillation (as a representative oscillation): the second model added CO2 as a fourth independent variable. The first model was hopeless – it represented some of the variations but none of the trend. The second model was much better – it represented both the variations and the trend, and was actually more accurate than a 23 model IPCC ensemble. At that stage I made a tentative estimate of CO2 sensitivity which I now realise was in error (I multiplied the CO2 coefficient by the number of years of data (156) not the increase in CO2 (105 ppm)).

    Since then I have tried replacing CO2 with CO2-equivalent. This model has slightly improved accuracy (r2 = 0.90) and a sensible value for CO2-equivalent sensitivity (0.89 °C for a doubling of CO2-equivalent). You can see the model herehere.

    I fully realise that it is dangerous to read too much into a regression type model but it would be an interesting challenge to see if anyone can model the temperature from 1856 to the present as accurately as this model with a 4 independent variables and without invoking CO2.
    0 0
  5. RonManley:

    You said:"a sensible value for CO2-equivalent sensitivity (0.89 °C for a doubling of CO2-equivalent)"

    Isn't that too little? Most climate analysis found a climate sensitivity around 3ºC for a doubling of CO2.
    0 0
  6. Ron,

    I think the problem with your analysis is that you are predicting temperatures based on temperatures (AMO). That is, the AMO index is computed directly from observations of SST, so by including that as a term, you are by default forcing your model to conform to temperature measurements which in turn over the long term parallel global temperatures.

    It also, IMO, is not a valid proxy for ENSO events, because AMO and ENSO are not related (ENSO is a small-period Pacific cycle, while AMO is a longer period Atlantic cycle).

    It is also not a valid input because it clearly is not entirely global in nature... it's Atlantic.

    And last, but not least, there is no known mechanism behind the AMO, and not a long enough series of observations to determine if it is actually an independent cycle with a physical cause or merely an artifact of how things turned out in the past century.

    Bottom line... as you aptly point out, all it winds up being is climastrology. It is just curve fitting, but by choosing three known, physically relevant variables (solar irradiance, aerosols and CO2) and one direct-temperature-observation (AMO) as core components, you are of course getting a reasonable answer... if you put the right answer into the equation, of course you'll get that back.

    The trick would be to find some non-temperature based measure of long-term variability, and still have it all work out correctly.
    0 0
  7. As a quick point before anyone says it... the MEI as a measure of ENSO strength uses 6 variables: sea-level pressure (P), zonal (U) and meridional (V) components of the surface wind, sea surface temperature (S), surface air temperature (A), and total cloudiness fraction of the sky (C).

    Only a few of those are temperatures, and they are based on a very, very small area of the globe, so they are representative of the state of a known mechanism rather than simply being copies of global temperatures. Using the MEI is an entirely different matter than using AMO or some other completely temperature based proxy, and so is not a case of forcing the right answer in to make sure you get (or stumble into) the right answer at the end.
    0 0
  8. I posted this comment at Bickmore's blog too, but I thought I'd run it by everyone here to get more opinions and feedback about it:

    "Yet the IPCC refuses to accept that the global warming (or cooling) on time scales of thirty years or more can also be caused by Mother Nature.
    ...The IPCC has taken for granted that there are no natural variations in global average temperatures once one gets beyond a time scale of ten years or so."

    I think Dr. Spencer might need to learn the difference between "refusing to acknowledge" something and "having no good evidence" that something is real. Perhaps the IPCC "has taken for granted" that natural variability has little to no impact on longer timescales because there isn't any reason in the literature to assume otherwise, and their process merely reflects that reality.
    The IPCC can't very well take into account the vague notion of unquantified long-term noise that hasn't been established to exist at all. Given the lack of support for such things in the literature, even if we take Spencer's papers at face value, it's unlikely that his conclusions would be well-vetted, tested, and supported enough to make it into the IPCC's considerations. As I understand it these kind of things that are inconsistent with the existing literature would need more than one or two papers to be considered strong enough evidence. Spencer himself said that such alternatives have received relatively little research.
    Even if Spencer thinks he has uncovered such long-term natural variability (and so far it doesn't seem likely), it's not a fair criticism because his papers claiming it were first published after the last IPCC report. I think that counts as missing the deadline for inclusion. What exactly does he expect the IPCC to do? Invent a time machine to include papers published after their report, and credulously promote the very little bit of his research that claims to have found something to the status of other, better-supported conclusions that say the opposite?
    0 0
  9. Ron,

    Your method of estimating sensitivity assumes the system was always in equilibrium.
    0 0
  10. Thanks for the comments - most of which I accept - which is why my description of the 'model' is littered with caveats. For example, I state explicitly that the use of “natural climatic oscillations” as independent variables is “questionable”. Such a simple ‘model’ cannot of course trace the transfer of heat and, for that matter, CO2, into and out of the oceans. It cannot handle water vapour feedback. All I claim to have done is to show that temperature variations for the last 150 years can be easily simulated if you included GHGs; if you exclude them you have to postulate some other undefined alternative forcing mechanism.

    When Kevin Trenberth suggested that the null hypothesis should be that humans are changing the climate I did not agree with him; now I am starting think that he might be right.
    0 0
  11. Just made a post over at Roy's blog and spotted this priceless gem. There really are some opinionated ignorant people out there:

    "The physical principle behind the analysis lies in the Kirchhoff’s law of 19th century radiation physics, which can be restated in plain English as: an object that absorbs emits and an object that emits absorbs. Absorption and emission are two inseparable equivalent identities of the same physical essence. Carbon dioxide absorbs infrared therefore it emits as well thermal radiation. Nitrogen and oxygen do not absorb, THEREFORE DO NOT EMIT(my emphasis). CO2 approaches 0 K because of its emission if there is no radiation source; absorption of the thermal radiation from the earth ground surface rises CO2 temperature from -273.15°C to -78°C only. CO2 gains heat by colliding with warmer nitrogen and oxygen to rise its temperature further, which can be measured by spectroscopy. "

    Also this article would be worth a read. It really does look like Roy's option #3 is the biggest culprit.
    0 0
  12. RonManley

    "Can natural oscillations explain the climate...". Yes they can to some extent if by Climate we mean just what happens to air temperatures, and just shorter term 0 the oceans are 90%) changes like a decade or so. However if we work with all the components that make up the climate system - Air, Land, Cryosphere, Oceans - then any variations in the sum of all of these has to come from variations in something else. But there is no other something else apart from the Earths radiative energy balance with the Sun & Space.

    Arguments like Roys are based on just looking at a small part of the climate system (the Air accounts for only 3% of the added heat due to AGW) and not asking whether the other internal parts of the climate are changing to cause the Air to warm, or whether all the components are changing together due to an external source of heating.
    0 0
  13. Dr. Bickmore,

    Re Spencer's comment that:
    "You cannot simply say a lack of warming in 10 years is not that unusual, and that there have been previous 10-year periods without warming, too. No, we are supposedly in uncharted territory with a maximum in radiative forcing of the climate system. One cannot compare on an equal basis the last 10 years with any previous decades without warming."

    This is a very disingenuous statement for someone like Dr. Spencer to make. It seems to be designed to feed fodder to those who deny the theory of AGW or those who claim that climate sensitivity is low. He has to know that what he is claiming is misleading. Either that our he does not understand the physics as well as he claims or should.

    His argument seems to rely on the following assumptions:
    1) CO2 is the only driver of global temperature.
    2) The resultant warming will be montonic.
    3) That trends calculated over a period of 10 years are statisically significant.

    Assumption 1 is of course false, need to elaborate on that. Assumption 2 is also false, and Dr. Spencer argues that intenal variability is real and important. Assumption 3 is critical, trends over periods of 10 years for such a noisy data series as global temprature do not yield statiscally significant trends.

    It is important to note that the interannual variability because of oscillations or volcanic eruptions can be as large as +/- 0.2 C. By comparison the long-term rate of warming of global temperature is about 0.15-0.20 C per decade.

    It is well-established that the interannual and even decadal variability affect global temperatures, and that decadal long slowdowns are real (e.g., Meehl et al) that is not under dispute. But as shown above, we have processes acting on very different time scales.

    It could be circa 2060 when CO2 values have doubled and the resultant radiative forcing is about +3.7 Wm-2 and a super El La Nina will still reduce global temperatures by ~0.2 C. But the global temperature that year will be much higher than an equivalent event occuring in the early 21st century.

    Similarly, when another slowdown happens down the road, say circa 2050, the plateau during that period will be substantially higher than the one being currently experienced.

    This process is already evident (see Figure below) and explains why 2011 was the warmest La Nina year on record, and one of the warmest years on record rather than one of the coldest.



    [Source]

    Your observation that, "The bottom line is that Roy Spencer has been arguing all along that natural variation can cause the temperature to go up or down for a while no matter what the external forcing is doing, and no matter how long the time period, but now he suddenly can’t imagine that this could happen over a single decade!", beautifuly exposes the inanity of Dr. Spencer's claim.
    0 0
  14. Ron,

    A question of methodology here: Did you divide your work so that you have a training data set (say prior to 1985) and a prediction set? If not then your comparison with IPCC on anything is meaningless, yes, no?

    Dave
    0 0
  15. Dave123

    I tried splitting the data in two. The parameters were very similar for both periods.

    As I've said I don't make claims for this 'model' as a model. Think of it more as an educational toy.
    0 0

You need to be logged in to post a comment. Login via the left margin or if you're new, register here.



The Consensus Project Website

TEXTBOOK

THE ESCALATOR

(free to republish)

THE DEBUNKING HANDBOOK

BOOK NOW AVAILABLE

The Scientific Guide to
Global Warming Skepticism

Smartphone Apps

iPhone
Android
Nokia

© Copyright 2014 John Cook
Home | Links | Translations | About Us | Contact Us