CO2 and temperature both increasing: D’Aleo’s attempt at falsification of AGW debunked

by

Below are two graphs of global average temperature and CO2 concentration. First I show the temperature anomaly from the three major datasets of surface temperature together with the CO2 concentration as measured at Mauna Loa since 1958:

The CO2 concentrations are plotted on a logarithmic axis because the temperature effect of CO2 is logarithmic. The 11 year running mean through the yearly temperature anomalies is given by the thick colored lines.

Before the 1970’s, the temperature trend was more or less flat for a few decades (see also the graphs in this earlier post). The strong increase in cooling aerosols (resulting from e.g. SO2 emissions) counteracted much of the greenhouse warming over that period. Since that time however, greenhouse forcing has been dominant, resulting in the temperature and CO2 trends following a similar pattern (at least over the multi-decadal timescale; short term variability is heavily influenced by e.g. El Nino/La Nina, major volcanic eruptions and other natural phenomena). A graph of the time evolution of relevant known climate forcings over the past 130 years can be found here.

A very popular graph that purportedly falsifies the whole “AGW dogma” is the following, showing unrelated trends of temperature and CO2 for a recent 11 year period. It’s been carefully crafted to create a certain impression:

However, this graph is entirely misleading:

- There are more factors than only CO2 that influence global average temperature.

- The expected trend in temperature does not necessarily rise above the expected level of yearly variability over the course of a decade.

- The graph purposefully starts at a record high temperature (1998) to maximize the visual impression of “falling temperatures”. It also strongly depends on the specific datasets used. This is a clear example of cherrypicking.

Using the same logic as this graph is based on, one could also falsify the theory of gravity by pointing to a bird in the sky (conveniently forgetting that there are more forces than gravity and that the bird has wings).

About these ads

Tags: , , , , , , , ,

23 Responses to “CO2 and temperature both increasing: D’Aleo’s attempt at falsification of AGW debunked”

  1. Scott Mandia Says:

    So is D’Aleo ignorant or deceitful? Either way he continues to make a fool of himself and he is doing science education a huge disservice. It is a shame.

  2. Nir Says:

    Not too long ago, I spent a morning playing around with R, and as a little experiment, I created a scatterplot of CO2 concentration vs the (30 year smoothed) HadCRUT dataset

    http://s852.photobucket.com/albums/ab90/jhudsy/?action=view&current=co2vsHAD.png

    The scatterplot used CO2 concentrations rather than log(CO2 concentrations), but the fit was still remarkable. Especially given that the axis of the fitted line to log(CO2) had a slope of around 2, which is close to the lower estimate of CO2 sensitivity according to the IPCC.

    I’ll leave discussions of whether this little experiment was at all useful to the statistics experts, but I think that single plot is a rather powerful debunking of D’Aleo’s claims in itself.

  3. Arthur Smith Says:

    Note that D’Aleo’s graph also multiplies the CO2 scale by a factor of 4 in comparison to the temperature scale, relative to your graph. I.e. he is implicitly claiming that climate scientists expect the transient sensitivity to be 8 or more degrees C per doubling!!!

  4. Tom Fuller Says:

    Great! Let’s all condemn cherry-picking and vow not to do it again. Circulate a petition and I will sign and recommend signatures. Also, let’s go back and identify other instances of this dubious behaviour and create a Hall of Shame… any takers?

  5. Scott A. Mandia Says:

    Tom,

    No need for petitions. Make this your next post and send us the link.

    I have a few examples here:

    http://profmandia.wordpress.com/2010/02/20/chopping-down-the-cherry-tree/

    and Bart in his last thread has more.

  6. MP Says:

    Tamino also had a post on this one in 2008 :

    http://tamino.wordpress.com/2008/04/16/perjury/

    In the course of the thread Leif Svalgaard produced two graphs:

    http://www.leif.org/research/DAleo.png

    and this one where he standardized (z-scores) both data sets:

    http://www.leif.org/research/DAleo2.png

    The latter is the probably the more objective representation for showing correlation because the mean of the z-score products (covariance) is the same as the Pearson correlation coefficient.

  7. MP Says:

    aargh spamfilter again! :)

  8. Kelly Says:

    I have a similar posts here. and here..

    Showing 2 series with dual axes is not advisable from a data visualization standpoint, see Stephen Few here.

    To me, a scatterplot of both series against each other or rescaling both series to the same scale are much better than dual axis scale plots.

    D Kelly O’Day
    http://chartsgraphs.wordpress.com

  9. Länkar 2010-03-18 Says:

    [...] CO2 and temperature both increasing: D’Aleo’s attempt at falsification of AGW debunked [...]

  10. Peter Wilson Says:

    “Before the 1970’s, the temperature trend was more or less flat for a few decades (see also the graphs in this earlier post). The strong increase in cooling aerosols (resulting from e.g. SO2 emissions) counteracted much of the greenhouse warming over that period”

    Really? Given the very large uncertainties over the effects of the many different types of aerosols, as documented in AR4 as well as elsewhere, how can you possibly offer this as a statement of fact? Even in AR4 the error bars for our understanding of aerosols crosses zero (meaning we don’t even know for sure the direction of change), nor do we have any accurate measures of the amount, type and distribution of aerosols over the 20th century. To say the aerosols in the 50′s and 60′s cooled the earth, but suddenly stopped doing so in the 80′s, is [edit]

    [Reply: Uncertainty is not the same as knowing nothing. I explained before how the historic aerosol forcing is estimated. Yes, it's uncertain. No, it's not entirely unknown. Look at some of Martin Wild's papers on global dimming - global brightening for a start. BV]

  11. McCloud Says:

    So you are certain that AGW is caused by increasing CO2 but are uncertain about the role of SO2?
    [Reply: There's no black and white "certainty" and "uncertainty". But the influence of CO2 is indeed much better constrained/known than the exact magnitute of the influence of aerosols. BV]

  12. adriaan Says:

    I may become boring, but I come back with one of my favourites:
    Frank et al 2010, Nature. They do exactly the inverse analysis to show how much CO2 will be released due to increasing temperatures. This is also debunked? And this is all physics, remember? Pretending to know all about everything is something reserved for divinity. As a scientist, you should always remain skeptical, even on your own playing ground. Someone can hurry in and tell you something new.

    [Reply: Nobody I know pretends to know it all. Strawman argument. BV]

  13. adriaan Says:

    @Bart,

    I do not follow your strawman qualifications. You take the second, general part of my comment, mark it as strawman, without addressing the first part. The remark I made has so far not been touched by you or dhogaza. Do you think that Frank et al is based on proper physics? If so, why did they find a much lower feedback than previously found? If not, where does their physics falter?
    {strawman on} As a physicist, you should be aware of the (devastating) effects of positive feedback in systems theory? They always result in out of control run-away situations. Time constants of the system are very important to estimate the effect, along with systemic limits, which may lead to oscillating systems. Remarkable in this context is the fact that the IPCC reports state that 21% of all released CO2 remains forever in the atmosphere, and that they use this value in their models. {strawman off}

    [Reply: positive feedbacks can lead to stable situations as long as the ‘gain’ factor is less than one (i.e. for every initial change in the quantity, the feedback change is less than the original one). See here. BV]

  14. adriaan Says:

    @Bart,

    You are supposed to be the one knowleadgeable on atmospheric physics. If so, than please answer my question. And if you like, try once more to adress my previous remarks about CO2 and the Bern model.

  15. Scott Mandia Says:

    adriann,

    RC has a post on Frank et al. here:

    http://www.realclimate.org/index.php/archives/2010/02/good-news-for-the-earths-climate-system/

    An interesting comment toward the end of that thread:

    Even if the mean estimates as high as say 20 ppm/ºC are more realistic, this feedback rate still does not compare to the rate of increase in CO2 resulting from fossil fuel burning, which at recent rates would exceed that amount in between one and two decades.

  16. Paul_K Says:

    Adriaan,
    Climate scientists borrowed the term “positive feedback” from systems control engineering, but didn’t take the definition with it. This has caused a lot of confusion. In the hope of avoiding another runaway blog debate on the issue, maybe Bart will allow me to clarify this just one-time here.

    Consider a climate forcing X which results in a direct temperature change Dt. Dt itself then generates a further (aDt) in a feedback. aDt generates a^2 times Dt and so on. We end up with a geometric progression:

    X -> Dt (1+a+a^2+a^3+…ad infinitum)
    The geometric series is convergent if and only if abs(a) Dt (1/(1-a)) if abs(a) infinity if abs(a)>= 1.
    Both climate scientists and systems control people would recognise this so far as a feedback system.
    Where the difference lies is that the climate scientists use the term “positive feedback” to describe a situation where a>0, and “negative feedback” to describe the situation where a0 (equivalent to abs(a) >=1 i.e. unconditionally unstable), and a negative feedback as log(abs(a))<0, unconditionally stable.
    Thus, when a climate scientist talks about a positive feedback, it does not imply runaway properties – other than in conversations on the subject!

  17. Paul_K Says:

    Adriaan,
    Sorry, my above post got completely scrambled because of a postscript problem, I think.
    3rd para should read:

    The geometric series is convergent if and only if abs(a) is less than 1 and divergent if abs(a) is greater than or equal to 1. Summing the series, we obtain: X-> Dt (1/(1-a)) if abs(a) is less than 1, or to infinity if abs(a)>= 1.

    Both climate scientists and systems control people would recognise this so far as a feedback system.
    Where the difference lies is that climate scientists frequently use the term “positive feedback” to describe a situation where a is greater than 0, and “negative feedback” to describe the situation where a is less than 0. The control systems guy on the other hand would describe “positive feedback” as being when log(abs(a)) is greater than 0 (equivalent to abs(a) >=1 i.e. unconditionally unstable), and a negative feedback as log(abs(a))<0, unconditionally stable.
    Thus, when a climate scientist talks about a positive feedback, it does not imply runaway properties – other than in conversations on the subject!

  18. adriaan Says:

    @Paul_k,

    Thanks for illuminating me, I thougth we were sharing the same language. As has occurred before, things are (not always) but often redefined in climate science. Residence time is another, related misused term.

  19. adriaan Says:

    @Scott Mandia,

    I followed your RC link. I must say that I have never read something more unfounded than this piece. It looks very impressive, with all its refs, but it does not tell me anything new. And they do not address the main issue: since they (Frank et al) are using all the “approved” hockeysticks, they still come with an estimation of gamma far below what was previously considered correct. Would they have used temperature references (calibrations) like Patterson (PNAS 2010), they would to have reduce gamma even further, probably into the negative. (since they seem to have also timing problems). Notice that all the discussion about timing in the Frank paper is based on the hockeysticks, which show very poor time resolution. What is not mentioned at RC is that the timeseries of the CO2 core samples had to be spliced filtered with different splines than those used for the temperature records in order to be able to compare them to the temperature proxies. Would Frank et al have bothered to also carry out a calibaration against the 1100-1330 timeframe, they might have come to a negative gamma.

  20. adriaan Says:

    @All,
    And as far as modelling and simulation goes…

    Have you ever considered how accurate the models used by Intel and AMD must be? And these models deal with millions of transistors.

    I think the main reason for Intel and AMD to be succesful, is that they hired the right engineers.

    In AGW, you do not hire, you get hooked.

  21. Eli Rabett Says:

    Actually Adriaan, when Frank et al 2010, Nature. calculate how much CO2 will be released due to increasing temperatures, it’s not much physics, it’s mostly chemistry, having to do with the carbonate equilibria in sea water, biology, having to do with the balance between growth and rotting of organic material, and a bit of geology, having to do with weathering. The physics part of it is pretty trivial

  22. Charles Higley Says:

    [edit: OT (trying to claim that the whole of science is bunk) - please move to open thread. BV]

  23. adriaan Says:

    @Eli Rabett:

    And that is where we biologists come in. Its not physiscs, its rotting. Please provide me with a more on topic answer. You have been advocating that everything is physiscs, I got slammed and it is rotten?

    I made very precise, physics based comments on the Frank et al paper, you should be able to do better than this. They used proxies for atmospheric temperature, and calculate the effects on atmospheric CO2 (measured by ice core proxies) as consenquence of temperature changes. It is the overall response of the system, regardless of whether is is physical, chemical or biological in origin.

    That your physics can not deal with this result indicates that your model is too simplistic. And this is a general statement, models are (by definition) simplistic. Otherwise you would not use a model.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Follow

Get every new post delivered to your Inbox.

Join 124 other followers

%d bloggers like this: