On 18 April 2013, Professor Murry Salby, of Macquarie University, Sydney, Australia, gave a talk entitled Relationship Between Greenhouse Gases and Global Temperature at Helmut Schmidt University, Hamburg. His talk is available on Youtube.

Like many other people, I was impressed by Professor Salby's talk. I think, in times to come, it will be seen as a turning point and its significance will be great. He shows how the "climate science", that led to the belief in fossil-fuel use causing global warming, is simply wrong, being contradicted by observations. His reasoning is simple and straightforward and presents evidence obtained directly from observations.

By the way, he is not a crackpot by any measure whatever. He has a record of work with NASA and his new textbook "Physics of the Atmosphere and Climate" (~650 pages) provides a good measure of his stature.

I think it is not an exaggeration to state that he puts climate science on to a rigorous footing, where things are confirmed by comparison with observation. This is something it has lacked previously, even leading some people to question whether it should be accorded the term "science", because of its reliance on unvalidated models as "evidence".

He uses mathematical and statistical methods (Partial differential equations, Fourier integrals, cross-spectra and their relations - phase spectra, coherence spectra) that will be foreign to many readers - although they are familiar enough to engineers from some branches of engineering. But, as Professor Salby says, "the important consequences ... will be self-evident".

Here is a summary of what seemed to me to be the key results presented by Professor Salby in his talk.. I have annotated them [eg (16.36)] to show where, in the Youtube video, a particular result is discussed and presented.

Anyone who doubts these results should study the video carefully. In the cases I have attempted so far, I have had no difficulty reproducing Professor Salby's results by processing publicly available records.

[1] Modern changes in CO2 and temperature are not unprecedented

- Proxy CO2 and temperature underestimate the atmospheric magnitudes and rate of change of temperature and CO2. (25.45) Once the effect of diffusion of gasses in ice is accounted for, it becomes clear that modern changes in CO2 and temperature are by no means unprecedented - neither in magnitude nor rate of change. (27.52)

Why this is significant: Because recent rapid changes in CO2 and temperature have widely been taken as "proof" of human caused climate change and this has now been shown to be false.

[2] Humans do not detectably affect CO2 levels.

Net global emission of CO2 evolves independently of the human contribution, being driven by global temperature (slide 43.23)

Why this is significant: Increases in atmospheric CO2 have been assumed to be of human origin and therefore justify measures to curb CO2 emission. This is shown to be false.

[3] IPCC statements that observation of carbon isotopes confirm the human origin of CO2 are wrong.

- The IPCC has stated "All of the increase in CO2 concentration since pre-industrial times is caused by human activity. The increased CO2 is known to be caused by human activities because the character of the CO2 in the atmosphere, the ratio of its heavy to light carbon atoms has changed in a way that can be attributed to addition of fossil fuel carbon."
Professor Salby states "the observed sensitivity of native emission of CO2 and carbon 13 make this impossible. (52.29)

Why this is significant: Assumptions that the IPCC's statement was correct has led to costly measures to curb CO2 emission.

[4] On timescales of decades, climate models referred to by the IPCC have no predictive capability.

- Professor Salby gives a convincing argument that 24 IPCC climate models have no predictive skill on a decadal timescale - the same timescale responsible for twentieth century warming. (59.00)

Why this is significant: Belief that computer models can predict climate is taken, incorrectly, as proof of the reality of human caused climate change.

[5] The energy balance in IPCC climate models is wrong.

- Professor Salby shows that the mean of global temperature as predicted by 24 IPCC models tracks (to within a scale factor) predicted CO2 levels. This is contradicted by observed temperature, which has not tracked CO2 levels. But in addition, for this to happen in reality, numerous contributions to heat transfer in the atmosphere (eg reflection by cloud, mechanical heat transfer from the surface) would all have to change in direct proportion to CO2.

Why this is significant: The output of climate models is the only "proof" that CO2 controls climate change. With this belief falsified, the whole "fossil-fuel use = global warming" hypothesis falls apart. In Professor Salby's words: "If the global energy balance is wrong, everything else is window dressing". (1.00.20)

[6] Atmospheric CO2 is proportional to the time integral of temperature.

Net rate of change of atmospheric CO2 is proportional to temperature (information from physical measurement), so total atmospheric CO2 is proportional to the time integral of temperature (confirmed processing observations of CO2 and temperature). (1.04.20)

Why this is significant: Its consequence is that human emissions do not significantly affect atmospheric CO2 levels.

_______________________________________________________________________________
Professor Salby's work has been rubbished thoroughly at Real Climate and at Skeptical Science - perhaps a hint at its significance.

Hi Missy,

Interesting question. I'll give it some thought, although my answer might well be "I haven't a clue", especially if it needs knowledge of carbonate chemistry to answer.

To help me, do you have a reference to the mass of CO2 estimated to have caused the pH changes, year by year, that you can point me to? (from chemistry calculations, not from conservation of mass arguments). And including year by year records of ocean pH?

I may be out of circulation for a day or two but, if so, that won't mean I am ignoring the question.

Hello Missy,

Land and ocean both emit and both absorb quantities of CO2 that dwarf the human contribution (slide at 36.39). I don't think that is controversial. But, as Salby said, the various mechanisms by which CO2 is emitted (other than fossil fuel use) are all poorly understood and have not been quantified, so it may be that no-one has the definitive answer to your question at present.

Salby's results showing the relation between global temperature and rate of change of atmospheric CO2 don't really seem to be open to debate, even if your question remains unanswered. His results are a matter of straightforward arithmetic applied to available observation records of CO2 and global temperature. I have had no problem reproducing his curves from the observation data, in the cases I have tried.

You said " ... I hear that the pH of the oceans is falling, implying that they are absorbing CO2."

I am guessing that you meant ".. implying that they are absorbing CO2 from the atmosphere". Am I right? If so, then I think that is an assumption rather than a fact. So I am not sure that you have actually identified a paradox or a contradiction.

The chemistry of carbonates in seawater is complicated and my understanding of the subject is minimal. It needs someone who understands ocean equilibrium chemistry and can do the calculations to give a definite answer. However...

"When water that contains dissolved calcium carbonate is warmed, CO2 is removed from the water as gas causing the equilibrium of bicarbonate and carbonate to shift to the right, increasing the concentration of dissolved carbonate"

Assuming this also applies to the carbonate in seawater, increasing global temperature will result in:
- CO2 being liberated in the ocean, and finding its way into the atmosphere.
- Additional carbonic acid in the ocean, lowering the pH.

As Salby mentions (around 17.11) increased atmospheric CO2 results in increased absorption rates on land. Presumably the CO2 fertilisation effect also applies to ocean plant life, so increased absorption by the ocean, simultaneous with increased emission, does not seem to involve any contradiction.

So I don't see that there is necessarily any contradiction between increased ocean emission and falling pH. Does that suffice as an answer? It's the best I can offer.

I did find some sources of data for ocean pH: http://www.epa.gov/climatechange/science/indicators/oceans/acidity.html and I found a source for some of the observational data used to produce the graphs.

A bit like atmospheric CO2 records, the pH shows a strong annual cyclic variation which is much greater than the annual trend. There is also significant variability of the year-to year change which needs to be explained and does not correspond to absorption from the atmosphere at constant rate. It would be interesting to compare the year-to-year variation of pH with annual temperature and see if they are related, as Salby did for CO2.

Hi Missy,

No probs - the question was interesting.

I'm surprised that you omitted the very next sentence:

"As the concentration of carbonate increases, calcium carbonate precipitates as the salt: Ca2+ + CO32- ⇋ CaCO3."

In other words heating the water leads to CO2 emission and the precipitation of calcium carbonate. So your conclusion that there would be, "Additional carbonic acid in the ocean, lowering the pH" seems not to hold up (the carbonate has precipitated). The text doesn't talk of carbonic acid at all, so I don't know how you reached your conclusion.

I omitted the very next sentence as it did not seem relevant.

I didn't mention carbonic acid because I assumed its relation to CO2 would be understood. Here is some added explanation:

If you dissolve CO2 in water, you get mainly undissociated CO2 molecules in the solution, which don't take part in acid/base reactions and do not affect the pH of the solution.

But a small proportion of the CO2 molecules (maybe 0.1%) react with the water to produce carbonic acid: H2O + CO2 = H2CO3. Carbonic acid is dissociated into negative bicarbonate ions and positive hydrogen ions. It is the latter that increase the hydrogen ion count, decreasing the pH.

So releasing CO2 in the sea gives:

- More dissolved CO2 which can find its way into the atmosphere.
- Additional carbonic acid in the water, resulting in reduced pH.

I hope that my previous reply now makes more sense.

However, bear in mind what I said before. Carbonate chemistry in seawater is quite complicated and working out in detail the consequence of a rise in temperature needs an expert. So my answer, although I believe it gives the true explanation, is no more than a plausibility argument until verified in detail by a carbonate-in-seawater chemist.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

"How can we say that our emissions have no effect?" Or insignificant effect, perhaps.

It's a dynamic equilibrium (or almost an equilibrium, as it is not precisely balanced, hence the net flow) easier to illustrate with equations than with words.

I'll have a think about how to put it into words that don't sound like either hand-waving bullshit or incomprehensible gobbledegook.

Missy,

As I said, that was the best answer I could offer. A chemist knowledgeable about carbonate-in-seawater chemistry could say whether or not what I said makes sense.

"If your mechanism were true, surely you could prove it in your kettle. If you were to boil a kettle of water you would hope to observe the remaining water becoming more acid as the kettle boils. And that acid should presumably dissolve the hard water scale from the heating element. On the other hand, you might see no change in pH and an accumulation of hard water scale. I know which I'd expect. Heating elements generally become furred-up, after all."

I may not have explained myself clearly if you think that I'd hope to see those things.

I am not sure why you seem to believe that boiling a solution (of calcium bicarbonate presumably), so that calcium carbonate is precipitated, is not going to change the pH of the solution.

I'm seeing what happens if you put actual data into them...

http://cdiac.esd.ornl.gov/trends/landuse/houghton/1850-2005.txt
http://cdiac.esd.ornl.gov/ftp/ndp030/global.1751_2009.ems
http://www.esrl.noaa.gov/gmd/ccgg/trends/#mlo_data

On equations, the mass-balance equation at the Skeptical Science page on "Murry Salby's Correlation Conundrum" seems convincing to me. What is wrong with it?

Hello Missy,

"What is wrong with it?"

Obviously, the additional CO2 in the atmosphere at the end of a 12 month period will be the CO2 added during that period, less the CO2 removed during that period. So, in that sense, no problem whatever with the mass balance calculation.

But beyond that, I get stuck.

"....the annual rise in atmospheric CO2 has been less than anthropogenic emissions every year for at least the last fifty...."

I think they are saying that this therefore shows the origin of the additional CO2 cannot have been natural. Whilst it sounds at least plausible, does it necessarily follow? I'm not saying what they say is wrong, but I can't get my head around it. Perhaps it's so obvious, I'll be embarrassed when it's explained to me.

I haven't followed their argument that Salby is wrong in overlooking the effect of average CO2 emission. Not saying they are wrong - just that I have not yet followed it. If Salby is correct that fluctuations of CO2 emission follow temperature, then I'd have thought average CO2 emission would also follow temperature. Perhaps I have missed a point.

My other problem is that with CO2 in the atmosphere, being exchanged in both directions between atmosphere and ocean+land, we have a system in dynamic equilibrium (or almost in equilibrium if there is a small and changing human input). So far as I can see, to understand what goes on, it necessarily has to be analysed as a dynamic system.

I analysed it, using a linear 1st order differential equation for the quantity of carbon in the atmosphere as a function of time - essentially the same as Salby's at 18.28. I used numbers similar to those from his CO2 budget graphic 36.39 (presumably of IPCC origin). I assumed constant human emission of up to 10 GtC per year and calculated the resulting percentage of atmospheric CO2 of human origin. The result I got was hugely different from SkS. I've asked a friend to check my calculations.

So... I would not say that the SkS reasoning is wrong - just that I have not so far got my head around it and the results of the calculations I did, treating the system as one in dynamic equilibrium, differ greatly from their numbers. So a bit inconclusive, so far, I'm afraid.

To me it was a delight to watch because, a couple of existences or more ago, I lived and breathed Fourier theory and subsequently I was heavily involved in estimating cross-spectra (and from those, coherence spectra and phase spectra) of pairs of time series, so that part of his talk instantly made perfect sense to me.

His presentation is a talk, not a research paper. And he covers an awful lot of stuff in the time, so it's not surprising that the details of his working are essentially "an exercise for the student". I've tried reproducing some of his graphs and have had no difficulty doing so (other than not knowing the precise specs of the lowpass filters he used to produce smoothed curves). Not a complete verification so far, but enough to convince me that it was certainly not a work of fiction.

The idea that CO2 emissions follow temperature and that the rise in CO2 concentrations over the last 50 years is _because_ of a rise in temperature implies to me that, were temperatures not rising, the planet would be absorbing all anthropogenic emissions. Is that the proposal?

I think that Salby provided evidence that CO2 concentration follows the integral of temperature (w.r.t. time) - not temperature itself.

"...the planet would be absorbing all anthropogenic emissions. Is that the proposal?"

Well I think it probably is an implication of what he said. And I think it may well be what would happen. You have very large flux of CO2 between atmosphere and land+ocean (flux proportional to CO2 concentration) and very large flux between land+ocean and atmosphere (flux proportional to temperature). So you have (in the absence of human emission and temperature changes due to eg solar variation) a system in dynamic equilibrium, with the two large fluxes equal and in opposite directions.

If you now inject a small additional flux (small compared with the large existing fluxes), you don't change the equilibrium condition much. That applies for any system of two reservoirs in dynamic equilibrium - climate systems, large scale queueing systems or whatever. My calculations (mentioned in a post above), using the equilibrium fluxes and atmospheric concentration given in Salby's 'carbon budget' slide (origin IPCC, I imagine) said that a constant 10GtC annual emission would raise the atmospheric CO2 content by about 6%. That would not be a very significant increase by comparison with doublings of CO2 which are often discussed.

Ignoring the issue of what caused the current temperature increase (if it not CO2), if the linear rise of 0.6 degrees in surface temperatures in 50 years is enough to raise CO2 levels by 100ppm (again linear) then a fall of several degrees (5?) during the last ice age would have meant a fall into negative territory. Does that not make you wonder a little? ?

As I mentioned above, I think Salby's hypothesis is that it's the integral of temperature that counts, not the temperature itself.

The lack of a theory for the initiation and termination of ice ages is one of the things that convinces me that 'climate scientists' don't know all that much about the subject - 'science is settled' notwithstanding. On BH we have discussed the possibility that climate is a chaotic system that hovers around in a warm period and then, for no very special or obvious reason, quite abruptly transitions into an ice age.

I wont be joining the Salby fan club - I found the guy creepy - especially those long pauses and intense stares. But, being sacked from two universities is certainly an achievement. I don't rate his chances on making a hat-trick.

Well whether you fancy him or not probably does not have much bearing on the question of whether his hypotheses will stand the test of examination and further investigation.

The advice often given is to avoid ad-homs, for several good reasons.