Murry Salby has now completed his tour of Britain and there have been a few posts discussing his presentations. Some positive, others not. I’ll be quite honest and say that I think Salby’s ideas are scientifically incorrect and that it’s not difficult to show this. Hence, I do have an issue with those who seem to think that he might have a point. It implies, in my opinion at least, either too much wishful thinking or a lack of willingness/inability to actually determine whether or not his ideas do have merit. I apologise to those who think this may be an unfair way to start this post, but I am more than willing to be proven wrong.
Scottish Sceptic has, however, written a report on Salby’s lecture in Edinburgh. Given my never-ending optimism, I thought I might write a few brief posts addressing some of what is said in Scottish Sceptic’s review of Salby’s talk. Maybe I should comment directly on his post, but it does seem easier to simply write this post of my own. A pingback should appear there and I’ll leave a comment pointing out that I’ve addressed some what is raised in the post.
What I’m also going to try and do is keep it fairly simple, by only addressing one point at a time. If this seems to work and if the outcome is positive/constructive, I may write more. If not, I can simply stop. What I thought I would address in this post is what Scottish Sceptic regards as one of the key graphs. It’s shown below and is a plot of atmospherice CO2 concentrations and temperature anomaly, from 1960 to today. The claim is that this graph shows that there isn’t a good correlation between rising atmospheric CO2 levels, and rising surface temperatures.
CO2 concentrations and temperature anomaly from 1960 to today (credit : Murry Salby)
There are two main issues with this interpretation. One is that rising atmospheric CO2 concentrations act to reduce the amount of outgoing longwave radiation, increasing the amount of energy in the climate system. Only a few percent of this energy is associated with heating the surface. You really should be comparing the rise in atmospheric CO2 with the rise in total energy, not only with the rise in surface temperature. Furthermore, because the rise in surface temperature is only associated with a few percent of the energy excess, it is very sensitive to internal variability and other factors. Hence, we don’t really expect the surface temperatures to rise smoothly, even if the atmospheric CO2 concentration is rising smoothly. Additionally, there are factors (such as anthropogenic and volcanic aerosols) that can also influence how much energy is being trapped in the climate system, and so we wouldn’t even expect a perfect correlation between the rise in atmospheric CO2 and the rise in total energy (on short timescales at least).
The second issue relates to how much we might have expected surface temperatures to have risen given a certain rise in anthropogenic CO2. The change in adjusted forcing, ΔQ, when the CO2 concentration changes from Co to C is
The figure above considers the period 1960 to today, during which time CO2 concentrations changed from 310ppm to about 390ppm. This implies a change in adjusted forcing of 1.23 Wm-2. The expected change in temperature would be determined by the transient climate response (TCR) which is expected to be about 1.5oC per doubling of CO2. A doubling of CO2 also produces a change in forcing of ΔQ2x = 3.7Wm-2. The expected temperature change is then
This suggests that the expected change in surface temperature since 1960 is about 0.5oC. Therefore, if you really want to compare changes in atmospheric CO2 and surface temperatures, you really should ensure that the two y-axes are scaled properly. The right-hand y-axis scale in the figure above is 1.3oC and makes it appear that surface temperatures have risen slower than expected when, in fact, it’s pretty close to what would be expected.
So, essentially I don’t think this figure really illustrates what Murry Salby is claiming. Not only do we not expect a particularly good correlation between the rise in atmospheric CO2 and surface temperatures, the actual rise in surface temperatures since 1960 is entirely consistent with a TCR of about 1.5o. There’s my comment number 1. Feel free to comment here and I will also do my best to address constructive comments either here or on Scottish Sceptic’s original post.
...and Then There's Physics 
I actually think that graph shows a fairly good correlation but I’m not particularly good at reading graphs.
Interesting to read ScottishSkeptic’s “Thanks Prof Salby” post. It did seem a bit like he was relieved not to have to worry about climate change anymore. Salby has relieved him of this fear. It reminded me of that recent research which found the conservative brain has a larger fear centre – http://abcnews.go.com/blogs/health/2012/09/03/conservatives-and-liberals-have-different-brains-studies-show/
I am glad you are writing a critique, because I was beginning to think I would have to write it!
Salby gave a correlation of “greater than 0.5” with a maximum when temperature leads CO2 by 10months. This is worse that the correlation between tempreature and net-emissions (0.63) and when the component of soil moisture is included it becomes very good at around 0.9 for CO2 CH4 and Carbon 13/12 ratio.
However, there is an important issue about the phase of correlation between the current and proxy record. In simple terms this means that current temperature looks more like the differential of CO2 and proxy temperature looks more like proportional. Much of his original talk in Hamburg was about this.
For information, I have reproduced the CO2 differential (net-emissions) graphs and I can vouch that net-emissions is a better fit than the level of CO2. It does look much better (I got a much better looking graph) and I could match the features of the temperature record from the graph.
Rachel, yes, if you were to change the y-axis scale on the right-hand side so that it more properly represented the trend, then it would be pretty good.
ScottishSceptic, you’ve kind of moved to other points. I was hoping to try and address this graph first and then we could move on to other aspects of what Murry Salby is suggesting.
Assuming a TCR of 2 C per doubling of CO2, and with a 24.6% (77.85 ppmv) increase in CO2 concentration since observations began at Mauna Loa (1959), we expect a 0.64 C increase in temperature due to CO2 over that period. Thus, where the scale to match prediction to observation appropriately, the scale ratio would be 0.82 C per 100 ppmv CO2. Instead, Salby uses a scale ratio of 1.5 C per 1000 ppmv of CO2.
As an alternative approach to the correct ratio, we can simply use Salby’s Ratio as determined from his comparison of predicted temperature and predicted CO2 increases over the 21st century in his lecture at the Sydney Institute. That ratio requires a scale ratio 0.69 C per 100 ppmv CO2. Using either the Salby Ratio, or that predicted from the IPCC predicted TCR, there is clearly no mismatch between the long term trend in temperatures and the rise in CO2 levels. Indeed, temperature evolution can be almost exactly modeled as the predicted TCR to CO2 plus an ENSO adjustment.
Looking at Scottish Sceptics post, I noted a familiar graph at the top of the page, ie, the second graph discussed in my post on Salby’s Ratio. That graph has a scale ratio of 4.55 C per 100 ppmv. A quick search showed that indeed Salby continues to trot out that basic lie in his European tour, as seen from 1:00:00 on this video of his talk at Hamburg. Given that he definitely showed the second graph at the Scottish Climate and Energy Forum, he almost certainly showed the first as well. That means that Salby was not content to use just two different scale ratios when comparing temperatures and CO2 concentrations. Rather,he used at least three. Apparently he was confident that AGW “skeptics”, scottish or otherwise, would not call him on this patent bullshit. His apparent low opinion of his audiences shows no evidence of having been disconfirmed.
Wotts, I know you are inclined to give people the benefit of the doubt, but anybody unwilling to acknowledge the basic dishonesty of Salby’s cavalier manipulation of scale ratios is so blinkered or dishonest as to be not worth the time it takes for discussion.
Reading the post I have to say that you really are not addressing the issues raised by Salby but instead you are addressing those raised by the previous post: Hermann Harde: greenhouse effect 30% smaller than IPCC says.
If your intention is a critique of Harde, then perhaps I should point out that his main finding is that when you use the latest HITRAN database, the CO2 greenhouse effect is about 30% smaller. I think he assumes the rest is as per pro-IPCC dogma.
But critiquing Salby on the greenhouse effect is quite unethical. As far as I can remember all Salby said about the greenhouse effect is “this arrow here is the incoming radiation and this is the outgoing radiation”. I don’t think he disputed it at all.
Rather than attempting a refute an issue which Salby was not addressing, it would help if you get to the point and address the issue of whether the correlation between net-emissions and temperature – with the time lag showing temperature leads CO2 emissions, indicate these features are CAUSED by temperature and humidity on the surface.
Tom, I tend to agree with your last paragraph and would be interested in ScottishSceptics views. It is so obviously wrong that it is indefensible. In case ScottishSceptic is unaware of this issue, it is discussed very clearly here and, as Tom points out, the second of the two graphs in the link is the graph shown in the photograph of Salby while giving his talk in Edinburgh, shown on ScottishSceptic’s post. If Salby showed the first of the two graphs, then what he presented was trivially wrong and it is indefensible.
ScottishSceptic, was your comment directed at me? I was addressing the point I thought he was trying to make with the figure you presented in your own post. Your caption says This shows poor correlation. That’s really all I’ve addressed. Why is that figure a key figure if not for the reasons I presented?
There are two graphs on ScottishSceptic’s post. One is identical to the one you’ve got here Wotts and the other is the one Tom refers to, with Salby standing in front of it. They’re both quite different. How do you explain that ScottishSceptic?
scottishskeptic the correlations between net-emissions and temperature is well known, and was first pointed out by Bacastow back in the mid 1970s and is essentially the result of changes in the terrestrial biota in the americas due to ENSO.
However, Salby’s analysis is fundamentally flawed as statistical correllations are not sensitive to the average value of the signals over which they are computed (i.e. they measure only the similarity of the wiggles in the two signals around their average values). Sadly it is the average value of net emissions that gives rise to the long term trend, not the wiggles. This means the correllation tells you nothing about the cause of the long term rise in atmospheric CO2, it just explains the very small wiggles of atmospheric CO2 concentrations around its long term trend.
I wrote an article on this for Skeptical Science, which you can find here: http://www.skepticalscience.com/salby_correlation_conundrum.html I emailed Prof. Salby a copy of this article prior to publication for his comment, but he failed to reply.
Thanks Dikran, that’s clearer than I could have put it, but essentially what I would have said anyway.
why can’t salby compare those two datasets using cross-correlation functions? Why the wiggle-matching? And talk about correlations?
Anonymous, I think he has produced correlation coefficients and they do correlate well, but as Dikran points out, he’s correlating the wiggles on top of long-term trends, and not correlating the long-term trends themselves.
I was going to avoid talking about the correlations until we’d discussed the graph I presented here, but since Dikran has presented a good argument I thought I would add something more. What Dikran says is pretty conclusive, so one doesn’t need to consider what I’m going to present here, but I thought I would add it anyway.
What Salby is correlating is the temperature with the rate of change of atmospheric CO2 (ppm/yr). He’s suggesting that the temperatures sets the rate at which the atmospheric CO2 concentrations will rise. One obvious problem with this is that if the temperature were to remain as it is today, we would outgas the oceans in about 20000 years.
Furthermore, this is completely at odds with Henry’s law. Henry law’s tells us that the concentration of CO2 in a liquid is set by the partial pressure of the gas in the air above the liquid and by a coefficient that depends on temperature. So, what Salby is plotting is the change in the rate at which CO2 concentrations are rising, but this is because the temperature does vary with time (on short timescales) which results changes in the atmospheric concentration of CO2 producing the wiggles that we see on the long-term trend. So, the wiggles do indeed correlate with temperature because it is temperature that’s driving them. The relationship is, however, something like 10ppm per degree and so this cannot explain the 120 ppm rise that we’ve seen since the mid-1800s.
I think it is worth pointing out that the image at the start of the report at ScottishSceptic is a far worse example of the problem pointed out by Tom Curtis’ excellent “Salby’s ratio” article than the one shown in Wotts; article.
I don’t know if the site allows me to embed images, so if that hasn’t worked, the URL of the image in question is here: http://scef.org.uk/images/salby/fig0.jpg
What Waslby should have done is to estimate the rate of warming we should expect to see immediately following an increase of CO2 such as that we have observed and used that to scale the two plots, so that there was some actual physics involved. There is no statistical or physical reason for scaling the two y-axes the way he did.
Dikran, indeed. In fact that was why I did the calculation at the end of the post. To show that we would have expected around 0.5oC of warming from 1960 to today and yet Salby chose to make his right-hand y-axis have an extent of 1.3oC.
“but anybody unwilling to acknowledge the basic dishonesty of Salby’s cavalier manipulation of scale”.
The whole Salby lecture was about the relationship of CO2 and temperature.
To put it very simply, he was asking “what is the correct relationship between CO2 and temperature.”
Then you say “he’s wrong because he’s using the wrong relationship”
In case that isn’t completely obvious … just look at one side where it says: “Carbon dioxide” you will see the value ppm. That is the scale for the Carbon dioxide curve. Then if you look a the other where it says: “anomalous temperature” you will see “K” that means “Kelvin” which is like centigrade.
If you still have problems, I suggest you print out the graph twice,. get some tipex, and remove one curve from each graph.
I see that Salby also tries to deceive with the SCIAMACHY satellite CO2 data (Fig 8 in ScottishSceptic’s blog post). As the researchers that produced that graph explained:
“The resulting annual composite averages of atmospheric XCO2 after quality filtering for the years 2003–2009 are shown in Fig. 2 exhibiting similar patterns for all years superposed by a steady quite homogeneous increase with time. A significant part of the CO2 spatial variations shown in Fig. 2 results from the irregular sampling of the SCIAMACHY XCO2. For example, the mid- and high-latitudes of the Northern Hemisphere are strongly weighted towards late spring, summer, and early autumn, where CO2 is known to be much lower than for the (true) yearly average. This uneven weighting is due to the significantly higher cloud cover in winter but also because of larger solar zenith angles and snow coverage. As a result, most of the mid- and high-latitude measurements in winter are automatically filtered out by the implemented quality filtering scheme.
ScottishSceptic, let me try and clarify. In a previous video presentation Salby presented a graph showing temperature and atmospheric CO2 from 2000 to 2100. It was essentially a model of how temperature should change with time as atmospheric CO2 concentrations changed with time. Immediately after that he presented a second figure that showed the period 1980 – 2012 and showed how the temperature rise did not follow the CO2. However, in the second figure he did not have the same ratio of the two y-axis scales. If he had chosen then same ratio as in his first figure, the match would have been remarkably good. He also should have chosen to keep the same ratio. Not doing so is either a mistake or dishonest. That is what people are referring to as dishonest.
Let’s go back to the figure I present here. Various lines of evidence suggest a TCR of between 1.5 and 2oC per doubling of CO2. The left hand y-axis in the graph above goes from 310ppm to 390ppm. Such a change in atmospheric CO2 would produce a change in forcing of about 1.23 Wm-2. The right-hand y-axis goes from -0.4 K to 0.9 K (a range of 1.3K). If we use the second equation in the post above, that would suggest that Salby is implying a TCR of around 3.9oC. That’s at least twice as big as any reasonable estimate. If he’d scaled his right-hand y-axis to be something comparable to what we’d expect, the comparison would be quite good. Again, it appears that Salby has chosen a y-axis scale that is completely unmotivated and that makes the fit appear much worse than it really is. This appears to be either incompetence or dishonesty.
Lars, that appears to be an illustration of why one should become quite expert at a topic before making remarkable claims.
Scottish,
I think the objection to the graph raised is that it’s visually misleading.
The y axis covers a CO2 increase of 80ppm, from 310 to 390. As Wotts suggest, using a reasonable TCR, say 1.5 degC, would predict a corresponding temperature rise of 0.5 degrees C. However, Salby uses a temperature axis covering 1.3 degrees C. Visually, this makes the rise in temperature seem less well correlated than it really is.
Scottish
And according to Salby’s belief system (‘Salby Ratio’ = 0.69 C per 100 ppmv CO2), atmospheric concentrations of CO2 went negative during glacials.
Salby appears to be wrong because he is using the wrong relationship.
If you wish to defend Salby’s belief system you will need to explain this. Please do – I am very interested to hear your reasoning.
Lars Karlsson says: …. “It’s not fair the map doesn’t say what I want it to say and the author agrees”.
Another distraction – the report says that the distribution is not correlated with geographic area – you are saying “it isn’t correlated with geographic area” … we do not disagree, let’s move on to the important issue: that net global emissions are related to global temperature and that the 10 month delay shows that temperature is driving net emissions.
Scottish, do you accept that Salby’s graph is misleading? Wotts has explained it really well. Even I get it and I’m not a scientist or an engineer like you are.
Scotttish, no as Rachel suggests, let’s stick with this graph for now and discuss whether or not you think it is misleading. Ideally we’d also discuss the two graphs that I mention in my previous comment. All others commenting agree that it is important to use an appropriate ratio between changes in CO2 and changes in temperature. Do you agree or not?
Rachel: “Scottish, do you accept that Salby’s graph is misleading?”
Did it mislead you? Or are you totally incapable of reading a simple graph?
So please can we get onto the point that there is a relationship between surface conditions and net CO2 emissions and that together with the 10month lag this is overwhelming evidence that temperature is “at least in part “driving CO2 emissions.
Scottish, you certainly haven’t discussed the issues highlighted here. You said
Not only is that an explanation, it’s also rather condescending. You said it was a key graph. Myself, and others, are claiming that it’s misleading because the choice of y-axis ranges make the correlation seem much poorer than it actually is. Could you at least try to explain whether you agree that it is misleading or, if not, why it isn’t misleading.
Scottish,
Yes, it did mislead me. I have already said I’m not particularly good at reading graphs. The correlation is stronger than Salby would have us think. It is actually quite dishonest of him to do this and I feel duped. Now because you’re smarter than I am you already realise the problem with his graph or alternatively you were mistaken too. Which is it?
wottsupwiththatblog
“Scotttish, no as Rachel suggests, let’s stick with this graph for now and discuss whether or not you think it is misleading.”
The graph shows two values which have no clear relationship on the same page in order to show how the two curves compare. THERE IS NO OTHER WAY TO SHOW THIS
I am absolutely appalled that you even asked such a question because in no sense at all could showing two curves which are being compared on the same graph be considered “misleading”.
Now please get to the point. The lecture was about the observed relationship between net-CO2 emissions and global temperature and the 10month lag shows temperature is driving the CO2 emissions.
Oh for goodness’ sake Scottish. Just answer the question in good faith:
“Could you at least try to explain whether you agree that it is misleading or, if not, why it isn’t misleading.”
Scottish, is that a serious response? Are you sure you don’t want some time to think about it? There are plenty of different ways in which to present that graph. The right-hand y-axis could have gone from -1000.0 to 1000000.0, or from -0.2 to 0.4. The ranges on the two y-axes have a significant influence on how that graph is interpreted. I think it’s misleading. I think everyone else commenting thinks it’s misleading. Unless you can provide an explanation as to why the chosen ranges of the two y-axes are appropriate, I will simply have to conclude that it is indeed misleading (which is what I think anyway). Maybe then we can move on to what you think the main point is (which Dikran has actually addressed above anyway).
wottsupwiththatblog “Scottish, is that a serious response? ”
Of course it is.
The graph shows two curves. The two curves are being compared. The text which none of you appear to have read makes it clear that the figure is there to show how the change in gradient with one does not match the other and how they generally look and appear different.
Even a five year old kid can understand this concept.
If they have different shapes they do not match!!!
Now please can we move on to the important issue which is that net-CO2 emissions appear to be driven by ground conditions.
Yeah, okay, Scottish, we’ll deal with that: interannual variability cannot drive a long-term trend. Got that? Right, back to the graph.
Scottish,
You state in your report
also noting
and other similar comments on the short term variations.
Far more important than it’s misleading visual appearance is the question as to whether the graph supports your contentions.
There’s no reference, so I used GISS data to test this, starting at 1960 and ending at 2011, and using 310 and 390ppm as the relevant CO2 concentrations (I’ve not checked those exactly)
What do the numbers show?
A least squares best fit gives a slope of 1.48 degrees C/century with a standard deviation of the residuals of 0.097 degrees. The slope translates to a TCR of 2.3 per doubling using Wott’s formulae above*
Now, this would seem to support a hypothesis that CO2 gives a rising background temperature consistent with TCR around 2, superimposed on natural “noise”. Which is exactly the IPCC position.
Your position that the sharp variations don’t correlate is irrelevant, it is predicted that they will continue in the future as they have in the past.
Your position that the trends don’t correlate is incorrect, they do.
The data supports the IPCC.
* disclaimer – fitting and calcs done in a rush in excel. Corrections welcome!
ScottishSceptic wrote “The graph shows two values which have no clear relationship on the same page in order to show how the two curves compare. THERE IS NO OTHER WAY TO SHOW THIS”
Yes, there is, the default way of drawing the graph would be to scale the data so that both sets of axes were fully utilised, rather than most of the temperature axis being left unused. That is what most scientific software (e.g. MATLAB) would do.
Alternatively, you could just normalise both time series, and plot them together, in which case they do look as if there is some relationship between them.
http://woodfortrees.org/graph/gistemp/from:1980/mean:12/normalise/plot/esrl-co2/from:1980/mean:12/normalise
However, in this case, there is an even better way, which is to scale the axes so that the two plots would be superimposed on each other if we observed the amount of warming that phsyics suggests should result from the observed increase in CO2.
scotishskeptic wrote “Now please can we move on to the important issue which is that net-CO2 emissions appear to be driven by ground conditions.”
I have already addressed this earlier in the thread, Salby’s method is fundamentally flawed, see the article I pointed out previously http://www.skepticalscience.com/salby_correlation_conundrum.html where I explain it in detail.
Scottish, seriously give this a little more thought, because at the moment you appear to making the kind of mistake that even a high-school student shouldn’t make. The apparent gradient of each line/curve in the graph depends on the choice of range of the two y-axes. If the right-hand y-axis had a different range, the temperature graph would appear different. The reason they appear so different is because of the choice of y-axis ranges. Furthermore, as I try to point out in the post, internal variability means that you wouldn’t expect the slope of the surface temperature profile to be constant with time (even if CO2 was rising steadily). Sometimes it would change faster than expected. At other times slower. If the right-hand y-axis range had been more appropriate for a TCR of 1.5 degrees, then this would have been apparent. Hence, the graph is misleading.
Scottish, the graph is misleading as several posters have pointed out because the range of temperature variation in response to CO2 forcing is considerably smaller than the range depicted. One could determine that for a 2 oC climate sensitivity a change in [CO2] from ~ 320 ppm to 400 ppm should yield at equilibrium a temperature change of ~ 0.64 oC (use delta T = (ln (400/320))*2/ln2).
So in fact we’ve had virtually all the warming expected from the rise in [CO2] since 1960 even though the Earth has quite a bit of warming to do to come to equilibrium with the current CO2-induced forcing (the rapid heat uptake of the oceans tells us so).
Clearly the rise in [CO2] can’t come from increased temperature since we know:
(1) that the temperature rise is far too small to yield much of a rise in [CO2] since 1960 and
(2) because we know that there has not been a temperature-induced release of CO2 from the vast ocean reservoir. This is obvious since we know that the oceans have absorbed a truly humungous amount of CO2 during the period of interest. Obviously the oceans (and terrestrial environment) can’t at the same time be both a sink for CO2 (mopping up around 1/2 of our emissions), and a sink.
Something is obviously very,very wrong with Dr. Salby’s logic. In fact his flaws are obvious, and rather embarrasingly puerile and it’s not surprising he hasn’t published this silly stuff.
It’s very surprising (a) that you can’t see the howlers in Salby’s presentations and (b) don’t seem very interested in understanding something that’s extraordinarily simple and (c) think you can press home you “argument” by SHOUTING in capital letters!
I have to do various family things this evening, so will probably be unavailable for moderation. Please bear that in mind when commenting 🙂
ooops, Scottish, you have probably noticed that my sentence should say:
Obviously the oceans (and terrestrial environment) can’t at the same time be both a sink for CO2 (mopping up around 1/2 of our emissions), and a source.
Scottish,
I don’t think it’s all that difficult to mislead someone with a dodgy graph. Did you click the link Tom provided earlier to his post about the Salby ratio? It shows rather well what can happen when you change the scale.
If I had gone to Salby’s talk and been duped by this graph, I think I would feel quite cross.
Wotts,
I will be around for moderation if necessary.
verytallguy – good comment about the sharp variations. What this actually means is that there is a lot of the power spectrum at higher frequencies that cannot be explained by the CO2 curve, but as you rightly point out there is the whole question of noise.
What the comment is trying to say is that there is much variance within the frequency band we are interested (about 1-5years) which cannot be explained by a similar variation in the CO2 curve. So this is evidence that CO2 is not (significantly) directly driving temperature within this frequency band.
In other words, if you had e.g. a 2yearly oscillation on one, then if there is a coupling between the two values, then we should expect to see a similar two yearly oscillation on the other. If however we see a lot of variation within the 1-5year frequency band we would expect to see this reproduced on the other. The lack of correspondence shows that either such a relationship doesn’t exist or if it does, that the response is severely damped.
That would be OK,. except the overall trend is not that different (30 year warming followed by 15 year pause versus 1-5year). So, claiming correspondence for things happening over 30 years without at least some correspondence for shorter term variations is highly dubious.
But it means much the same if I say “they don’t look the same”.
I hope that makes sense.
Guys, gals, seriously? You think Scottisch Sceptic will admit to anything? He’d be admitting that he invited someone that misled his audience! A personal failure – and notably one we’d already informed him would happen.
What can I say, Marco. I’m an optimist.
scottishsceptic, much of those “sharp variations” in the temperature data are actually rather well correllated with ENSO, which is essentially responsible for the correlation between temperature and the *rate* of net emissions. The fact that those sharp variations are not present in the CO2 concentration (which is the time integral of net emissions) is an indication that the correlation Salby discusses doesn’t explain the rise in amtospheric CO2 itself, just a minor modulation in the rate at which it increases from year to year. See the SkS post I wrote that explains the mathematical flaw in Salby’s reasoning, which you can find here:
http://www.skepticalscience.com/salby_correlation_conundrum.html
This is now the third time I have pointed this out.
Rachel: “Scottish,
I don’t think it’s all that difficult to mislead someone with a dodgy graph. Did you click the link Tom provided earlier to his post about the Salby ratio? ”
The graph shows two curves that are being compared. Both are scaled to fit within the box. There is absolutely nothing dodgy about this graphic.
So please, has anyone here read the report and if so do you understand that the report is about the relationship between net-CO2 emissions and global temperature and that the 10month lag shows that global temperature leads the CO2 emissions and therefore is likely causing them?
dikranmarsupial: “scottishsceptic, much of those “sharp variations” in the temperature data are actually rather well correllated with ENSO”
What’s the correlation?
scottishsceptic wrote “The graph shows two curves that are being compared. Both are scaled to fit within the box. There is absolutely nothing dodgy about this graphic.”
the are scaled to fit within the box, but they are not scaled to fit the box. There is a big difference between the two. This is what you get if you scale the data to fit the box
http://woodfortrees.org/graph/gistemp/from:1980/mean:12/normalise/plot/esrl-co2/from:1980/mean:12/normalise
and the relationship between temperature and CO2 concentration suddenly *looks* stronger than in Saby’s diagram. Now those of us with a scientific background might reasonably be expected to not be fooled by the scaling, but was Salby speaking to a scientific audience on his tour, or his Sydney Institute talks? No.
Scottish Sceptic says: “”temperature data are actually rather well correllated with ENSO”
If I understand you correctly, you are saying that ENSO is causing** CO2 changes …. which is what Salby was saying without specifically naming it.
**that’s my inference given the time lag.
Can I restate it this way. Salby is saying: “there is variation in temperature” … there is strong evidence that this causes CO2 to be evolved due to surface conditions.
It really isn’t rocket science, and nor does it say anything about radiative warming or the initial cause of warming.
scottish sceptic, It is very widely known that there is a correlation between ENSO and surface temperatures, to see how much it affects climate see e.g. Foster and Rahmstorf (http://iopscience.iop.org/1748-9326/6/4/044022). If you want to know the correlation coefficient, over some particular time period, download the datasets and calculate it for yourself (on your blog you describe youself as a climate scientist, so this should not be difficult for you).
What’s the internet meme for when somebody completely avoids admitting an error after repeatedly being shown that error by many different people in many different ways? There has to be a name for this. To me, it was obvious in Tom’s comment way at the top. By the time Rachel posted an embedded image…well, seriously, come on.
ScottishSceptic “If I understand you correctly, you are saying that ENSO is causing** CO2 changes …. which is what Salby was saying without specifically naming it.”
You do not understand correctly. ENSO is causing changes in the **RATE** of increase in CO2. If you integrate the rate of increase over time you get the increase in CO2 itself, and you will find the influence of ENSO on CO2 itself is very small as it is the mean value of the rate of increase of CO2 that gives rise to the long term increase, not the variation in the rate of increase due to ENSO. See the SkS article for details.
dikranmarsupial – “You do not understand correctly. ENSO is causing changes in the **RATE** of increase in CO2.”.
So I am saying an increase in temperature leads to an increase in the rate of emission of CO2. And you are saying that “ENSO is causing changes in the **RATE** of increase in CO2.”
what Salby did not address was the size of this temperature-driven evolution of CO2, so as we seem to largely agree I’m not sure what else I can add.
It is well known (I’ve even checked for myself) that a 1 degree rise in temperature results in about a 10 ppm increase in atmospheric CO2 concentrations. All that Salby has “discovered” is the well known result that variations in temperature will cause variations in atmospheric CO2. This is however small and short timescale. It cannot explain the 120ppm rise since the mid-1800s.
dikranmarsupial: “It is very widely known that there is a correlation between ENSO and surface temperatures, to see how much it affects climate see e.g. Foster and Rahmstorf (http://iopscience.iop.org/1748-9326/6/4/044022).”
What Salby says is that the rate of emissions is correlated with temperature across the globe and this increases if you add “soil moisture”. There is also a corresponding increase in methane and decrease in Carbon 13/13 ratio.
These are all pretty indicative of some form of decomposition. But I suppose it could be decomposition in the sea.
What Salby says is that this temperature driven surface decomposition explains “at least in part” changes in CO2 in the atmosphere.
If I were going to critique Salby, that would be where I started — how much is “at least in part”. Is this just ENSO, is it more, is it less.
And it all goes back to another conversation where I was attacked because on the sceptic view we had said: “we do not know how much of the rise is human and how much is natural”. That now looks a very sensible conclusion compared to the IPCC all this CO2 rise is human.
scottishsceptic,you did not say “increase in temperature leads to an increase in the rate of emission of CO2.” you said “, you said “ENSO is causing** CO2 changes”. There is a big diference between the rate of change and the change itself. Just because ENSO modulates the rate of change of CO2, that does not mean it is the cause of the “evolution of CO2”.
If I am driving down the motorway and at each junction alternate between depressing the accelerator all the way to the floor and depressing it only half way down, then the locations of the junctions will be very well correlated with the rate of change of distance (i.e. speed) but not with the distance I have travelled, which depends on the average position of the accellerator pedal, in the long run, the wiggles up and down have little effect.
It is easy to demonstrate that the natural environment is a net carbon sink, taking in more carbon each year than it emits and that the strength of the net sink has been gradually increasing with time. That fact alone dispenses with Prof. Salby’s hypothesis.
Wotts, It’s more accurate to state that in a world with rising atmospheric [CO2] resulting from manmade emissions, that variations in temperature will cause year on year variability in the rate at which manmade emissions accumulate in the atmosphere. As you say, we understand this very well in terms of very short term ENSO-induced changes in tropical terrestrial productivity and ocean responses. However at no time are the oceans (or terrestrial environment) making a nett addition to atmospheric CO2. None of the massive accumulation of CO2 in the industrial era has come from natural sources. That’s obvious since we know very well that both oceans and terrestrial environment have been a nett sink for CO2.
A nett sink for CO2 (i.e. the natural environment) cannot at the same time be a nett source of CO2. It’s surprising that Scottishskeptic fails to address this rather obvious point!
Bobby asks:
It’s called intellectual dishonesty.
And I’ve seen enough for one night.
Exhibit:
There is no possibility of reasoning with someone who is incapable of acting in good faith.
Wotts, I think you are missing the point of the objections from Scottish. He seems not to understand your explanation that the axes are wrong. To clear up his misunderstanding you must redraw the graph with the correct axes. It will then all be clean to him. A picture is, as they say, worth a thousand words 🙂
I have two questions for Scottish Sceptic.
Firstly, you and Salby claim that increasing temperature is causing an increase in CO2 concentration. What exactly is causing this increase in temperature?
Secondly, where is this extra CO2 coming from if not from human emissions caused by burning fossil fuels?
Please answer these questions and don’t ignore them since they are critical. You have to be able to account for all factors, not just the ones you cherry pick.
Geoff Harris, I already did just that, but no response from scottishsceptic so far.
http://woodfortrees.org/graph/gistemp/from:1980/mean:12/normalise/plot/esrl-co2/from:1980/mean:12/normalise
Let’s play spot the difference.
Here’s Murry Salby graph #1
Here’s Murry Salby graph #2
Here’s the woodfortrees version thanks to dikranmarsupial
Ah, ok – I have to admit to not reading the whole thread, else I would have seen your post :$
I would be very keen to hear your answers to Ian Forrester’s questions, ScottishSceptic.
No problem, Geoff, it helps to make sure scottishsceptic hasn’t missed it!
As the oceans are an important part of the equation, I would like to point to Ocean Acidification Summary for Policymakers, Third Symposium on the Ocean in a High-CO2 World. The oceans are net recipients of CO2, with dangerous consequences for marine life.
Rachel, that graph #2 is the one that made John Nielsen-Gammon realize that Salby was way out there in lala-land: his “CO2 from native emission (Temperature induced)” would mean that CO2 concentrations are negative during ice ages.
Oh, and CO2 concentrations would be somewhere around 180 or even lower in the LIA, if my rough guestimate is in the right ballpark.
Marco,
John Nielsen-Gammon’s negative CO2 is the dinner party rebuttal isn’t it? I think Doug Bostrom coined the phrase. It’s so easy to understand and doesn’t require any graphs.
Nope, that was me!
😉
Oh, I’m so sorry, BBD! Mea culpa, mea culpa!
Shocked, I tell you…
But seriously – this is a killer for Salby, which is why I raised it some way upthread – there’s just *no way* the “Salby Ratio” can be right. As Marco says, we are clearly in la-la land here.
Note that we are all three in agreement 😉
And there’s no denying that 🙂
Scottish
Ah, so you’re claiming that if we can’t show that the wiggles in the CO2 record are reproduced in the temperature record, then that proves the trend isn’t causal either.
Let’s have a think about that, and have a very amateur go at testing it out.
You mention one cause yourself:
But you don’t investigate further. Let’s have a look…
Ah = it is severely damped. A simple two box model proposing time constants and expansion coefficients for this is presented here, so you can check exactly what that means numerically.
The other cause is that the magnitude of high frequency noise of other forcings and internal variability are much higher than the magnitude of the forcing from the interannual CO2 cycle. We can check that by looking at the data from Mauna Loa: http://scrippsco2.ucsd.edu/data/in_situ_co2/weekly_mlo.csv
In 2011, the difference between annual min and max CO2 was 6.4ppm. Plugging in TCR=2 to Wotts formulae above, this translates to a temperature difference of 0.047C, well below the observed noise of 0.097, see my post above. And that’s before even considering the damping. And that southern hemisphere interannual CO2 changes are smaller.
Scottish
This is what is most obviously wrong and easily shown to be. We have emitted about double the CO2 needed to cause the atmospheric increase. A simple mass balance.
Not merely 100% of the rise, but actually about 200% of the rise is human.
The sources you and Salby claim to be emitting CO2 are very obviously absorbing it.
If CO2 rises are caused by temperature induced emissions of CO2, how do you claim to resolve the mass balance issue?
verytallguy: “we do not know how much of the rise is human and how much is natural”. That now looks a very sensible conclusion compared to the IPCC all this CO2 rise is human.
This is what is most obviously wrong and easily shown to be. We have emitted about double the CO2 needed to cause the atmospheric increase. A simple mass balance.
Not merely 100% of the rise, but actually about 200% of the rise is human.
But it all depends on the nature of those sources and sinks. If the sink is what we engineers call “an infinite sink” then no matter how much CO2 you dump into it, it will not increase the amount of CO2. But if the nature of that sink changes even marginally, then it will significantly impact the level.
A system like this is a sewerage system going into the sea. (within practical limits) … you can flush as many toilets as you will and it will not raise the level of the sea … but just a small movement of the earth will change the tide level.
If the carbon sinks approach the behaviour of an infinite sink then they may be very insensitive to the amount of carbon that is available to them to gobble up, but a small change in temperature may dramatically change the level to which they can reduce CO2.
Not to drive off-topic, but further to Lars Karlsson here’s an update on acidification, for COP19.
http://www.bioacid.de/upload/downloads/press/OceanUnderStress.pdf
Surprisingly difficult to find an actual copy, despite all the publicity around the report.
Oops. Above link is busted. Trying again… That’s Hot, Sour & Breathless – Ocean under stress
Scottish,
Firstly, the sinks aren’t infinite. Also, I think your toilet flushing analogy fails on the basis that the water that we flush is part of the system anyway. If the water that we were using had been buried for millions of years, then I suspect what we flush into the oceans would indeed cause the sea level to rise.
The other problem is that we have Henry’s Law. It tells us how sensitive the atmospheric concentration is to changes in temperature. It’s not sensitive enough by at least a factor of 10. Plus, as others have already pointed out if it was this sensitive, there would have been periods in the past when the atmospheric CO2 concentration would have been negative (or zero).
You really can’t just make up new physical laws just because what the current ones are indicating is inconvenient.
Scottish,
Your argument is nonsense. The anthropogenic CO2 is being dumped into the atmosphere which has a known amount of CO2 in it. You are dumping into it 2x the amount that the level is increasing by. You can hook the atomosphere up to any one way rate constant to simulate your infinite capacity sink and you will never get that the 2x amount you are dumping into the atmosphere is not responsible for increasing the atmospheric CO2. Your magical one way rate constant has no way to know whether the CO2 is from human or nature sources. If humans weren’t putting 2x the amount that the level is increasing by into the atmosphere you would have a valid argument.
And applying Salby’s claims to the LGM still yields negative atmospheric CO2.
I asked Scottish to explain this hours ago. Nothing. But then it cannot be explained, only ignored, so no surprise.
Mind you, Scottish is now making up absolute nonsense about “infinite sinks” in his attempts to defend the indefensible, so perhaps he will concoct some fable or other in a doomed attempt to save face. Although given where he is now, it’s not worth the bother.
To Lars’s comment (and Doug’s link), there’s obvious proof that the ocean is not the source of CO2. If it was, CO2 ocean concentration would not increase, but it is increasing and in a significant manner. Never mind the fact that we’re adding 200% of the CO2 needed to increase the atmospheric concentration, which is completely self-consistent with the ocean CO2 concentration point. Never mind Henry’s law, demonstrating that the recent temperature increase could only cause a ~10PPM effect instead of the required 120PPM effect. Never mind the fact that the physics of global warming via anthropogenic CO2 is self-consistent, fact-based and the CO2 growth correlates well to temperature increases. Never mind…do I need to go on?
This one is so obvious that we should be talking about the positive reception to Salby’s “work” as a psychological study in confirmation bias if I’m being generous (incompetence or intellectual dishonesty if I’m not).
Scottish Sceptic, I ask once again:
What is causing the temperature to rise?
Where is the increasing CO2 coming from?
The fact that you appear either unwilling or unable to answer such simple questions tells us all we need to know about your “theory”. It is rubbish.
Jeez-o, brick wall/head interaction springs to mind. As does “face” and “palm”. As a Scot, I can only apologise!
Bobby:
“This one is so obvious that we should be talking about the positive reception to Salby’s “work” as a psychological study in confirmation bias if I’m being generous (incompetence or intellectual dishonesty if I’m not).”
Yes, I’m almost done with one, about 450 pages long given the ~400p ages of annotated, cateogrized and summarized blog comments that started in July. Not just confirmation bias, but near-total non-skepticism, and lots of paranoia and conspiracy ideation, tribal groupthink (HT/T JoNova)
Kit: no need to apologize: every place has some, and Monckton notwithstanding, I don’t think Scotland has a bigger percentage than similar places. I had the pleasure of escorting Edinburgh Principal Professor Sir Timonthy O’Shea and his wife around the Computer History Museum earlier this year, and they were sharp and sensible, so I know that not all Edinburgh residents reject science.
The oddity in this case is something I’ve seen elsewhere. Nt U of Edinburgh is a major research university and has multi-disciplinary efforts around climate change, with well-published experts like Gabrielle Hegerl and many more.. They have frequent seminars, 5 in next 4 weeks.
If this is anything like most schools, such are open to the public. Not everyone is handy to such a school, but one would think any skeptic (in the classic sense) wanting to learn about a topic would attend seminars, assess speakers and be able to ask questions. Not everyone is so fortunate to have such access, and people in more isolated places might have a little more excuse.
But instead of talking at U Edinburgh, Salby spoke for the Parliament and the SCEF, on a tip sponsored by the publishers of the “Slayers” who reject the Greenhouse Effect.
In Hamburg, he did not speak at the U of hamburg (which has a large climate science effort), but at Helmut Schmidt U, the German military academy, which has no such department. He was sponsored by the EE department (and Prof. Dr. Hermann Harde, Emeritus Professor of Physics/Materials Science. He “was invited by Fritz Vahrenholt” / EIKE, even though the latter is affiliated with U of Hamburg.
In 2011-2012, rather than giving seminars around Australia, the usual way to expose new ideas to experts, he spoke twice for general audiences at The Sydney Institute, not normally a science venue.
All this might be a hint.
SS says:
“That now looks a very sensible conclusion compared to the IPCC all this CO2 rise is human.”
Hmmm.
Wottsupwiththatblog says:
“It is well known (I’ve even checked for myself) that a 1 degree rise in temperature results in about a 10 ppm increase in atmospheric CO2 concentrations. All that Salby has “discovered” is the well known result that variations in temperature will cause variations in atmospheric CO2. “
Seems to me that these two statements are mutually exclusive – unless the scientists working in the IPCC ignore something that is “well-known.”
Seems to me that someone must be wrong here. I wonder who…
It is interesting to note that not one of the three most notorious “Scottish” “sceptics”, Monckton, Montford and Haseler are, in fact, not Scottish, all three being English imports.
I’m glad that I decided to go to Edinburgh to further my education when St. Andrews would have been much closer to home.
Oops, I got too many negatives in that first sentence but you will get what I mean;-)
On the lighter note of flushing toilets, and ground water contributions to sea level – in fact more detailed studies do show a contribution to sea level rise from the taping of ground water, and the draining of the Aral Sea. They also show a reduction in sea level rise due to the large scale construction of dams. All three terms are small relative to the big three contributors of thermal expansion, glacial melt and ice cap melt.
More seriously, it is absurd for Scottish Sceptic to be talking about infinite sinks when the partial pressure of CO2 in the ocean is known to be rising, proving beyond question that the sink is not infinite.
Scottish Sceptic’s excursion into infinite sinks merely demonstrates that there is no limit to the absurdities you can believe if you keep your discourse carefully fact free.
Still waiting for SS to answer Ian Forrester’s questions below:
“I have two questions for Scottish Sceptic.
Firstly, you and Salby claim that increasing temperature is causing an increase in CO2 concentration. What exactly is causing this increase in temperature?
Secondly, where is this extra CO2 coming from if not from human emissions caused by burning fossil fuels?
Please answer these questions and don’t ignore them since they are critical. You have to be able to account for all factors, not just the ones you cherry pick.”
Joshua @2:54 AM, there is not contradiction between the claims that:
1) The observed temperature rise in the twentieth century would have caused CO2 concentration in the atmosphere to increase by about 10 ppmv, all else being equal; and
2) The observed increase in CO2 concentration over the twentieth century is entirely the result of anthropogenic emissions, with natural emissions constituting a net sink.
Indeed, both statements are true.
The resolution is that with the human emissions and no temperature increase, the CO2 concentration would have risen by about 10 ppmv less – but that in either case (temperature increase or no temperature increase) the increase atmospheric CO2 in each year would have been less than anthropogenic emissions in that year on average, and hence on average in each year the natural contribution is a net sink.
Thus, we do not even need to invoke the obvious point that most of the temperature increase is due to increased CO2 concentrations to show the compatibility.
Further to toilet flushing, Eli Rabbet has just made a new post, that refers to estimates of the net contributions of groundwater and dams to sea level over the twenty first century, with a net contribution of 31 +/- 11 mm by 2050, and ~50 mm by 2100 (see Aslak Grinstead).
My working hypothesis is that Rachel has put everybody – including Wotts – into moderation. Somehow I alone was spared.
Tom –
Thanks for the explanation. Unfortunately, with a relatively quick read I couldn’t follow it.
I don’t understand how increased atmospheric CO2 from natural temperature rise plus increased atmospheric CO2 from ACO2 emissions plus increased atmospheric CO2 from ACO2 emissions-induced temperature rise makes increased atmospheric CO2 from natural temperature rise = a net sink. If all the temperature rise were ACO2 emissions-induced, then I could understand how 100% of the increased atmospheric CO2 would be attributable to ACO2, but my understanding of the IPCC perspective is that only “most” of the temperature rise is thus attributed.
I’ll try to read what you wrote more closely later. Hopefully, I’ll be able to wrap my mind around it – but I may have just reached the boundaries of my (clearly limited) knowledge and intellectual capacity.
Ha, ha, yes, everyone but you dear, BBD.
Joshua, I think it is a subtlety that I don’t explain particularly well. I’ll have a go though.
The relationship between temperature and concentration is such that if the temperature was to rise by 1oC – without any anthropogenic CO2 emissions – then the atmospheric CO2 concentration would rise by about 10 ppm.
Similarly, if anthropogenic CO2 emissions were not accompanied by a rise in surface temperatures, the atmospheric CO2 concentration would be lower than if it was accompanied by a corresponding rise in temperatures (by something like 10ppm per degree).
So, yes the atmospheric CO2 concentration does depend on temperature but this doesn’t mean that we should interpret the rise in atmospheric CO2 as being partly natural and partly anthropogenic.
The basic reasons for this is that if there was a natural component, we would expect one CO2 reservoir (soil, biosphere, oceans) to show a reduction in CO2 concentrations. We don’t see this. All the different reservoirs show a rise in CO2 concentration. Hence they’re all absorbing some of our emissions and they can’t – at the same time – also be a source of natural CO2 emissions. At least, I think that’s the basic argument that Tom was making. Happy to be corrected or clarified by those who can explain this more clearly than I can.
Joshua: “I don’t understand how increased atmospheric CO2 from natural temperature rise plus increased atmospheric CO2 from ACO2 emissions plus increased atmospheric CO2 from ACO2 emissions-induced temperature rise makes increased atmospheric CO2 from natural temperature rise = a net sink.
I find this a bit of a brain-twister as well, but I’m going with one of two possibilities, the most likely being that the increase in atmospheric CO2 has simply swamped the predicted CO2 emissions due to temperature change. Or, the increased concentration of dissolved CO2 pushes more CO2 towards the carbonic acid side, where it’s not directly affected by temperature increases like dissolved CO2 is. The decrease in ocean pH tells us that the latter is certainly occurring, although how that affects the relationship of atmospheric to dissolved CO2, I don’t know,
Joshua, Tom wrote “1) The observed temperature rise in the twentieth century would have caused CO2 concentration in the atmosphere to increase by about 10 ppmv, all else being equal; ”
The important thing here is the “all else being equal”, but in this case all things are decidedly not equal. One of the reasons for this is that while the solubility of CO2 in the oceans decreases with increasing temperature, the exchange of CO2 between atmosphere and oceans also depends on the difference in partial pressure of CO2 in the atmosphere and the surface ocean (Henry’s Law). If CO2 increases in the atmosphere, then it becomes more difficult for CO2 to go from the ocean to the atmosphere, but easier to go from the atmosphere to the oceans. The 10ppmv takes into acount the effect of Henry’s law due to the CO2 emmitted from the oceans due to the temperature related drop in solubility, but not the increase due to anthropogenic emissions. We have emitted CO2 at a sufficient rate for Henry’s law to dominate the themperature related reduction in solubility, which means that the oceans are actually taking in more CO2 than they emit. Even though solubility has dropped, Henry’s law has more than compensated.
The key thing to remember is that the action of the carbon cycle is the result of an equilibrium between phsyical processes acting in different directions. If we peturb the system from its equilibrium state, the balance between these processes is also disturbed and they tend to act to bring the system back into equilibrium. It is that fact that has mean that CO2 levels have been as stable as they were prior to industrialisation. Our fossil fuels emissions have pushed the carbon cycle away from its pre-industrial equilibrium, and the balance of feedback mechansims is trying to restore it by opposing the rise in CO2, quite a bit of that is due to Henry’s law.
Hope this helps.
That’s MR BALL to you, I’ll add to me previous comment by saying that another way to see this is to recognise that just because there is a temperature dependence (in Henry’s Law) doesn’t imply that we should regard this temperature dependence as a natural component.
Following on from Wotts:
How does this apply during deglaciation? We know that CO2 increases from ~180ppm to ~280ppm across glacial terminations, and the estimated global temperature change from the LGM to the Holocene is ~5C. Henry’s Law gives us ~ 50ppm CO2 from the oceans, so presumably the rest is accounted for by increased wetland area, permafrost melt, even clathrates?
Any pointers to the literature about hypothesised sources?
Joshua,
I see this a bit more simply:
We have emitted enough CO2 to account for 200% of the atmospheric increase. Ergo, all of it is anthropogenic; The rest of our emissions go into sinks – the ocean mainly
If temperature hadn’t risen eg because of unprecedented volcanic activity reducing forcings, the CO2 rise would likely be 10ppm less. However, all of that rise would also be anthropogenic.
BBD, the 10ppm per degree was just an estimate. However, I think I read somewhere that you’d expect around 10ppm per degree on short timescales but that there is an equilibrium with the deep ocean that the means that on longer timescales (centuries) it rises to 16 – 20 ppm. Overall, a 5 degree rise in temperature would then produce something like a 100ppm increase in CO2. I could, of course, be talking complete nonsense 🙂
Thanks Wotts – it does seem as though the oceanic source was much larger than the terrestrial contribution (CH4/CO2) to the total change. It’s something I’ve always been vague about, mainly because I haven’t come across any quantifications in the limited selection of the literature that I have seen. Wetland expansion does seem to be a major – perhaps the major – terrestrial source of CH4 according to the textbooks (eg Cronin 2010). I was rather hoping some of the omniscients here might chip in with a few references 😉
Wotts, the deep ocean does not warm by anywhere near 5 C between glacial and interglacial. My understanding is that a number of factors are involved, including the increased production of methane from wetlands, with the methane then decaying to CO2. Having said that, I have not read anything recent on the topic; nor anything in great detail.
That’s my albeit rather hazy impression too, Tom.
Tom, I wasn’t suggesting that the deep ocean did warm by 5oC. I shall try and find it somewhere, but I remember reading something that suggested that there was a mixing with the deep ocean that meant that it took quite a long time (centuries I think) to reach a new equilibrium atmospheric concentration after a temperature rise. My understanding was that if one only considered the upper layer of the ocean one would expect about a rise of about 10ppm per degree but that including the slower mixing with the deep ocean gives about 20ppm per degree. Of course, I may have misread or misunderstood this so shall try and find it again.
Tom and BBD, I may have been over-interpreting this
I urge people to take a look, at Ruddiman, Kutzbach, Vavrus(2011) as cotext for all this CH4/CH02 discussion.
See FIgure 2, which shows the CO2 and CH4 curves for The Holocene (red) vs 7 past deglaciations, via “insolation alignment” (and they explain why that’s the correct way).
Without humans, the Earth would have been down to ~250ppm about the time the I.R. started,
At the right edge of the graph, draw a thick line going straight up to 400Pppm, about ~650ppb on the top graph. Doing the same for CH4 (~1800ppb) is off the top of the page into middle of the next).
See Fig 4 for Holocene versus average earlier interglacials. Remember, these graphs stop *before* the I.R. All this is perfectly consistent with the physics discussed here. Humans have been affecting the climate for ~8,000 years, although not much earlier. This is from 2011, research since has strengthened it. By the way, this is a fine example of the way real science works when there is a truly interesting new hypothesis, not a dumb old idea being repackaged for the gullible.
Thanks for this, John. I have to admit being rather sceptical about the Ruddiman hypothesis when I first came across it a few years back, but it’s looking stronger all the time. Reading the study you kindly link.
> Given my never-ending optimism I thought I might […]
Do anything you please, as long as you keep at it never-endingly.
BBD:
Yes, of course, proper sceptics at the time thought it was pretty radical, but had serious evidence, plus some issues, but was worth further research, unlike Salby. Hypotheses got poked at and refined, and the major complaints against it gave slowly been resolved, often by other researchers (always a good sign).
The insulation alignment works, the glacial alignment that Broeker used, doesn’t.
The biggest challenge was the astonishment of idea that relatively small numbers if people could have had any impact millennia ago, but incorporation of better land-use data helped resolve that, ie, that early farming methods created s much bigger footprint/person.
The CO2/CH4 jiggles over last 2000 years have also gotten much more support.
We better understand the 50M person dieoff in the Americas that caused most of the unique ~9ppm CO2 drop into 1600AD, and other researchers have found good evidence for the effects of human wars/plagues on the more reactive CH4 levels.
That pair is actually quite useful, since they act as filters with different time constants, in a useful way, unlike Salby’s manipulations.
When I get bsck home I can post another URL or so.
I like all this because “climate has always changed” is not a very useful generalization in understanding the past. Bill’s work is totally consistent with the science, but is auch better explanation.
Note: Ruddiman has a fine new book Earth Transformed that pits all together and us giving the Tyndall Lecture @ AGU next month, ie a lot more people take thus quite seriously.
Assuming RKV11 is correct, CH4/CO2 evolution during the Holocene does look odd compared to previous interglacials. I’m ever-more persuaded that the Ruddiman hypothesis has nads.
* * *
I know it’s a bit paleoclimate-nerdy, but the argument in RKV11 that MIS11 (400ka) wasn’t an unusually prolonged interglacial catches the eye because, as the authors state, MIS11 is sometimes proposed as an analogue for the Holocene.
BBD: yes nerdy, but that’s what Figures 3 and 5 are about. MIS5, 9, 19, and 7 are pretty good analogs, but 15, 17 and 11 aren’t. 11’s deglaclation took unusually long.
Meanwhile, CH4, which is important also. See:
0) Mitchell, eta al (2011), Multidecadal variability of atmospheric methane, 1000–1800 C.E. in JGR. IT says:
‘Correlations with temperature are dominated by changes in Northern Hemisphere high latitude temperatures between 1400 and 1600 C.E. during the onset of the Little Ice Age. Times of war and plague when large population losses could have reduced anthropogenic emissions are coincident with short periods of decreasing global methane concentrations.’
(Again the idea of different-duration filters:
Yearly variations in CO2?CH4 ~seasons in the N. Hemisphere.
A few decades: CH4 can show sharp jiggles, but CO2 usually does not, given longer concentration lifetime. It takes really extreme events to change CO2 very fast.
1) Scince Daily, work at Bohr Inst in DK in 2012.
2) Celia Julia Sapart, U Utrecht in Netherlands led a team writing, Natural and anthropogenic variations in methane
sources during the past two millennia for Nature in 2012.
Here is Law Dome CO2, 1AD-1900AD., with a drop into 1600 that is unique in terms of slope. That data is converted to a “slope chart”, in which the slopes (change in CO2 ppm/year) are graphed for 25, 50, and 75-year regressions centered on the year.
This shows (of course) that shorter periods allow for wilder swings, so focus on the green line, 75-years. During any 75-year period through 1300AD, the yearly change in CO2 stayed within the +/- 0.05 band. Then it started to go up and down faster , but the drop of more than 0.15ppm/year centered on 1600AD is unique during last 2 millennia. (Bill ascribes this to human populations finally large enough for plagues to affect CO2 on less than century time-scale.)
Of course, by now, this line would be off the scale at the top, ~2.0 ppm/year.
The point of all this is that not only does physics constrain various levels (as in different CO2 reservoirs), it also constrains rates of change. From RKV11, Figure 2, one can see that the major state change of the Younger Dryas (YD) hammered the CH4 down, but only dropped CO2 a few ppm. It took a 50M-person die-off in strongly-biomass-productive areas to drop CO2 ~9/10 ppm in 75 years..
But, post I.R. rises in both are all natural 🙂 Right.
Thanks for those explanations. It is interesting to see that while they are somewhat different, they nonetheless overlap significantly. That is as it should be.
I think that I sorta, kinda understand now.
John, do you have a quick conversion factor for converting cal/cm2/day into watts/m2?
Thanks.
Once again, John, thanks for all this material. I *hope* you are working from your notes and not having to hunt all this down afresh – I am not worthy, etc.
Absent anthropogenic influence, it is very difficult to think of a physical mechanism that accounts for the drop in CO2ppm centred on 1600CE. Yes. Taken alongside the CH4 curve for MIS1, this is further convincing support for Ruddiman. I really should say that I have intended to read Earth Transformed since I first became aware that Ruddiman had followed up Plows, Plagues & Petroleum but the hefty price tag has deterred me (~£100). IIRC it’s been out since early this year, so second-hand copies will start to appear soon and I will strike 😉
I am benignly predisposed towards Ruddiman because he wrote the best climate textbook I have yet come across (Earth’s Climate P&P) which improved my understanding of the subtleties of orbital forcing and more besides. Perhaps this is partly why I have been a little cautious about the long-term anthropogenic hypothesis – I know I am biased…
BBD
Thanks, yes, I’ve written on this before.
ET pricing: yes, I’m really not sure that W.H.Freeman is doing this right, so maybe the thing is to get library to get it, if that’s possible. I’m also not really sure that it was out earlier this year, as sometimes Amazon shows “phantom books” that didn’t actually exist. this entry for Wegman’s book first appeared in 2007. Deep Climate found it in 2010, and he and I both tried to order it. I even tried various other places that claimed it was in stock, but when I did it suddenly became out of stock or not yet published. Now it;’s there with a 2020 date.
In this case, I *know* the book didn’t exist earlier this year, because I got one of the first copies.,
Maybe W.H.Freeman will fix the pricing, because it is a fine book. I am of course biased,l especially as I’m quoted in it 🙂
It is always good to be cautious about new hypotheses, and this set of them (since there are really several interrelated ones), has required an especially strong interdisciplinary approach, in some cases depending on accidents, like happening to meet somebody at conference from a different discipline.)
By the way, in response tofact that Salby is no longer a Professor, Scottish Skeptic says:
‘I’ve checked and anyone can make anyone a professor,. so we’ve decided to make him and honorary professor of the Scottish Climate and Energy Forum.’
HAHA!! That was hilarious. OK, that thread was awesome in demonstrating his dishonesty and thin skin so clearly. Any rational person will come to a similar judgment simply based on his behavior. I hope he never deletes that thread. Somebody should archive.
@ John
What I think I may do is email WHF next week and just ask. Presumably Ruddiman’s publisher will have a clue…
😉
* * *
“Professor” Salby rides again, eh? You can’t keep a good man down. Er.
Bobby: if you check the chronology Defamation By Internet? Part 1 – Murry Salby’s Short-Lived Blog Storm”, you may note a few WebCite links (~50).
So, I’ve been WebCiting this one, as in this. However, it’s been cached ~16 times, and only a few of those were mine, so it seems others are interested in saving that thread for posterity. 🙂
In the 1980s and early 1990s, Salby authored with distinguished scientists from CU, NCAR, NOAA, and was a tenured Professor at a school strong in atmospheric sciences, and his work was well thought of by people who I respect as knowledgeable. He stopped working with such authors, and nothing submitted after 2002 got much attention from the field, compared to earlier work. (I found almost every paper since ~1990 and checked. One has to be careful comparing citation counts across fields, but there was a clear downtrend from his peak in the mid-1990s. The field had moved on, and Salby’s career had waned long before he left for MQ.)
Now, he is down to giving lectures for:
The Sydney Institute, EIKE, Klimarealistine, then on this tour:
sponsored by Stairway Press, publisher of “Slaying the Sky Dragon – Death of the Greenhouse Gas Theory “, i.e., the Slayers, who even Anthony Watts bans, and Mike Haseler.
After his bait-and-switch in Melbourne IUGG 2011, he’s not likely to get many Australian university invites. It’s unclear how many more European trips he can get. Maybe GWPF can sign him up.
For those who are interested, William Connolley comments on Scottish Sceptics claim that “anyone can make anyone a professor” as justification for continuing to call Salby “Professor Salby”. His conclusion is similar to mine, if expressed more politely than I would, or than Scottish Sceptic deserves.
Thanks Tom, interesting. I don’t know if you’ve noticed, but ScottishSceptic has asked (in response to one of your comments) if you’d be willing to do a proper critique. That would seem to suggest that none of those that have already been done are suitable. Can’t really imagine what else there is to say though.
You go go back, Jack, do it again
Wheels turning round and round
Actually, there is quite a lot more that could be said. Regarding ice cores, for example, I could note that:
1) Salby in deriving his formula for damping in ice incorrectly assumes that the rate of descent of the column is constant with depth (in fact, the ice compresses resulting in a slower rate of descent than near the surface);
2) Salby does not distinguish between CO2 in air bubbles, in clathrates and dissolved in the ice, all of which diffuse at different rates.
2a) In particular, Salby ignores that beyond a certain depth, the ice bubbles collapse and the CO2 becomes chemically bound to the ice in the form of clathrates, significantly reducing the rates of diffusion at lower levels.
3)Salby treats diffusion in firn and diffusion in ice as equal in value, an absurd proposition on the face of it.
4)Salby nowhere feeds empirical data into his equation to estimate damping in icecores. rather he sets the values for the equation to reproduce the cross coherence in the proxy record. His explicit words are (26:50, Hamburg Lecure) “The effect is illustrated by damping that recovers the cross coherence in the proxy record”. This is equivalent to assuming that the cross coherence is entirely a function of damping and that the sources of variation in CO2 are effectively white noise. Thus all that Salby proves is that if damping by diffusion is large enough, the ice core CO2 record is non-informative about past levels of CO2; but he makes no estimate of the actual rate of damping. Rather, having established the truism, he merely assumes that damping is sufficiently large to make that record non-informative even for events only 1000 years ago.
5) Salby ignores a large literature on diffusion of CO2 in ice (which I have only briefly looked into). For example, he ignores an empirical estimate on the Sipple Ice Core, which shows that even at 80 kyr, CO2 diffusion is limited to within +/- 60 years of the sample date, and is therefore of little consequence for determining past CO2 concentrations in the atmosphere. Likewise he ignores model based estimates (ie, estimates based on mathematical calculation as his is, but with empirical inputs to estimate constants, which Salby’s lack). One recent model based estimate (Bereiter et al, 2013) shows that even at 1.5 million years, there is only expected to be a 5% damping of a 40 kyr glacial/interglacial signal. Damping for more recent times, and 100 kyr signals are, of course, much less.
6) Finally, Salby ignores the extensive work on the take up of CO2 by the oceans and weathering that show a significant CO2 spike will only be absorbed in time periods much longer than the putative damping period even by his “estimate”.
I am sure that more, even, than this could be said on just this one topic. It would no doubt be entertaining to see an actual expert on ice core gas diffusion, and or an expert on CO2 take up go to town on Salby.
However, while I disagree with you about how much more could be said, I certainly agree that no more need be said. Salby has been thoroughly refuted on many occasions such that only those who are ignorant, or determined to believe what he says because it is convenient to them, can take him seriously anymore.
I have made a post which has simply disappeared. Would somebody check the spam folder.
Found it. Sorry, Tom.
John, Watts does not ban the Slayers. Just count the number of blogposts of Tim Ball that he has put up in the last few weeks. There are at least two I know of, one in which Beck’s CO2 analysis made a return to the blogosphere.
Marco: thanks, I’ll have to go back and dig out the quotes where Watts had said he was tired of the Slayers and banning them, (or maybe it was just some) or words to that effect, but either I misinterpreted the original or he has changed his mind.
Nothing would surprise me, and if accepts Slayers, all to the good I guess 🙂
Tom, wow there is indeed much more that could be said. An area I’m not particularly familiar with, though. I should probably have written Can’t see much point in writing more, rather than Can’t really imagine what else there is to say though. 🙂
John, I think Watts has banned discussion of the “dragon slayer” hypothesis, rather than dragon slayers per se. Of course he is happy to entertain hypotheses equally absurd, eg, Salby on CO2; so that hardly shores up his intellectual credentials.
Wotts, my point was, of course, not that much more could be said only in that one area. In fact, I am certain an expert in the field could say much more than I have, and that his other claims are equally fragile to close inspection.
Tom, I’m sure you’re right. Someone (possibly yourself) has made the point somewhere that those experts who have been exposed to Salby’s ideas have found them so absurd that few (if any) have bothered to critique them.
Tom: thanks
I had seen something where Watts had complained of constant comments from Slayers, and stopped that, so given his propensity to ban people, assumed the latter.
I mostly ignore WUWT and slayers, so didn’t see they were still around. I will fix the report in progress.
WOTTS: yes, experts, starting with John N-G who heard Salby in 2011 and commented at RealClimate, Colin Prentice at MQ, and Eric Wolff at Cambridge.
Marco & tom:
I found what I’d misinterpreted::
‘ Anthony Watts says:
July 9, 2013 at 11:16 pm
I gave Mr. Courtney a two week time out for distributing my email address without my permission related to a comment here he was defending. As a result my inbox was filled with junk from the “slayers”.
I get a lot of mail every day, I don’t need more, especially lectures telling me the greenhouse effect is “bogus”. ‘
Jeff Id (Condon) also gor tired of them. and the comment thread there (where John O’Sullivan, Doug Cotton appear) make SS’s defense of Salby look like sanity.
John, there was also this comment, here:
Of course, he had another anti-slayer thread just 16 days later, in which he did not close comments. It appears that he merely inconsistently moderates the slayers, sometimes banning their defense of their anti-greenhouse effect poppycock, and sometimes not – while always being happy to promote their other forms of poppycock.
From the Scottish Skeptic blog (http://scottishsceptic.wordpress.com/about/)
I am proficient in dozens of programming languages and now write PHP/MySQL websites as a hobby.
All I will say here is that a claim of proficiency in “dozens of programming languages” should raise the eyebrows of anyone with a programming background…
I agree. There’s not much modesty anywhere on that about page.
“a claim of proficiency in “dozens of programming languages” should raise the eyebrows of anyone with a programming background…”
…especially when they go on to code in PHP/MySQL by choice 😉
Pingback: Watt about the CO2 residence time? | Wotts Up With That Blog
Tom: In that thread,, Watts did finally close comments, with the succinct ending:
Anthony Watts says: November 20, 2013 at 10:13 am
This thread has degenerated to another war between “slayers” who think CO2 has no effect on anything and people that live in the real world. I’m closing it as it serves no purpose to continue.’
FWIW, I now have a follow on blog about Salby’s ratio in Hamburg. No major additions to the prior blog. Just an update examining Salby’s defense of his choice of scale ratio in the comparison of observed CO2 and Temperature.
Scottish Sceptic could be a little busy at the moment http://www.heraldscotland.com/politics/referendum-news/ukip-wiped-out-north-of-the-border-after-its-scots-leader-is-sacked.22833054
Oh dear, he must be bitterly disappointed. Presumably that’s why he hasn’t been posting recently.
Now, what’s that word again? Oh, yes: “schadenfreude”.
Maybe he will move back to England and take the other non-Scottish “sceptics” with him.
Ian, I know the Scots don’t like the English but surely you can’t hate them that much?
OK if you don’t want them back maybe we can send them to Gruinard Island or St Kilda, they wont bother too many people there 🙂
For the edification of all, I note a firsthand report by a PSI “slayer” on Sallby in Edinburgh, with surprise appearance by Monckton.
Pingback: A challenge? | And Then There's Physics
Pingback: Murry Salby in London | …and Then There's Physics