Lowering The Bar

Guest Post by Willis Eschenbach

Anthony has pointed out a new paper by McKinley et al. regarding the carbon sinks of the oceans (preprint available here , supplementary online information here). The oceans absorb and sequester carbon from the atmosphere. As usual in this world of “science by press release”, the paper has already been picked up and circulated around the planet. CNN says:

The ability of oceans to soak up atmospheric carbon dioxide is being hampered by climate change, according to a new scientific study.

A fresh analysis of existing observational data taken from locations across the North Atlantic Ocean recorded over a period of almost three decades (1981-2009) has revealed that global warming is having a negative impact on one of nature’s most important carbon sinks.

“Warming in the past four to five years has started to reduce the amount of carbon that large areas of the (North Atlantic) Ocean is picking up,” said Galen McKinley, lead author and assistant professor of atmospheric and oceanic sciences at the University of Wisconsin-Madison.

 

Figure 1. An estimate of the net CO2 flux into and out of the oceans, from Takahashi 1999. McKinley et al. say that the flux into the ocean is slowing.

The lead author says in the press release that things are getting worse … but since it is nearly guaranteed that the paper says something different from the spin the press release authors put on it, what does their paper actually say?

The first oddity about the paper is that they are discussing changes in the partial pressure of CO2 in the ocean (written as “pCO2”). But they’re not actually measuring the pCO2. They are calculating it from the dissolved inorganic carbon (DIC), alkalinity (ALK), sea surface salinity (SSS) and sea surface temperature (SST). Now, this is a standard scientific procedure used to estimate unknown variables in the oceanic carbon balance. But while it is generally a good estimate, it is still an estimate. It is calculated using an empirical formula, that is to say, a formula which is not based on physical first-principles. Instead, an empirical formula uses observation-derived parameters in an iterative goal-seeking algorithm to solve a complex formula.

As you might imagine, different authors use different parameters in the equation. There is a good overview of the function as it is used in the R computer language located in the “seacarb” package. If we take a look at the function “carb” in that package we see that in addition to the pCO2 depending on the variables they have measured, it is also affected by the levels of phosphate and silicate (which apparently the authors have not included). They give details of the different possible choices of values for the various parameters. From the description of the function “carb”:

The Lueker et al. (2000) constants for K1 and K2, the Perez and Fraga (1987) constant for Kf and the Dickson (1990) constant for Ks are recommended by Dickson et al. (2007). It is, however, critical to consider that each formulation is only valid for specific ranges of temperature and salinity:

For K1 and K2:

• Roy et al. (1993): S ranging between 0 and 45 and T ranging between 0 and 45oC.

• Lueker et al. (2000): S ranging between 19 and 43 and T ranging between 2 and 35oC.

• Millero et al. (2006): S ranging between 0.1 and 50 and T ranging between 1 and 50oC.

• Millero (2010): S ranging between 1 and 50 and T ranging between 0 and 50oC.

Millero (2010) provides a K1 and K2 formulation for the seawater, total and free pH scales. Therefore, when this method is used and if P=0, K1 and K2 are computed with the formulation corresponding to the pH scale given in the flag “pHscale”.

For Kh:

• Perez and Fraga (1987): S ranging between 10 and 40 and T ranging between 9 and 33oC.

• Dickson and Riley (1979 in Dickson and Goyet, 1994): S ranging between 0 and 45 and T ranging between 0 and 45oC.

For Ks:

• Dickson (1990): S ranging between 5 and 45 and T ranging between 0 and 45oC. • Khoo et al. (1977): S ranging between 20 and 45 and T ranging between 5 and 40oC.

As you might imagine, results depend on the choice of parameters.

In addition, McKinley et al. do not have observations for all input variables for all periods. Their study says:

For 2001-2007, ALK [total alkalinity] was directly measured. For 1993-1997, ALK was estimated from the ALK-SSS [sea surface salinity] relationship derived from 2001-2006 data (ALK = 43.857 * SSS + 773.8).

I bring these issues with the carbon calculations up for a simple reason—errors. Obviously, when you are estimating a critical value (pCO2) using an empirical formula with a choice of parameter values, with missing observations, and not including all of the known variables, you will get errors. How big will the errors be? It depends on the exact location being studied, the values of the various input variables, and your choice of parameters. As a result you will have to “ground-truth” the formula for the various biomes of interest. “Ground-truthing” is the process of comparing your calculations to actual measurements in the physical locations of interest. Once you have done that you can use the measured error, as well as any bias, in determining the significance of the results.

There is a discussion here of the oceanic carbon calculations, and some graphic examples of both calculated and measured pH, showing the size of the errors in another similar study. See in particular their Figure 1, which shows that errors in the calculation of pH, while generally moderate in size, are pervasive, unpredictable, and at times large.

Whatever the size of the errors resulting from the oceanic carbon calculations, they need to measured against observations in the regions studied, and then described and accounted for in the study. As far as I can tell the authors have not done either of these things.

The second oddity about the paper also involves errors. They have not (as far as I can tell) adjusted their error values for autocorrelation. Autocorrelation is a measure of how much tomorrow’s temperature is dependent on today’s temperature. As you know, warmer days are generally followed by warmer days, and colder by colder. It is unusual to see an ice-cold day in between two warm days.

Since when it is warmer it tends to stay warmer, and when it is cooler it tends to stay cooler (temperature records show positive autocorrelation), this means that the swings in the temperature will be larger and longer than we would find in purely random data. As a result, we need to adjust the calculations depending on the level of autocorrelation, in order to decide if the trends (or the difference between the trends) is statistically significant or not. As far as I can tell, the authors have not adjusted for autocorrelation.

The third oddity is one that I really don’t understand. The authors use a standard method (a “Student’s T-test”) to determine the uncertainty in the two trends, the trend in the pCO2 in the ocean, and the trend of CO2 in the atmosphere.

Then they use another test to determine if two trends (oceanic and atmospheric) are different. From their paper, here’s their description of the test, which contains the reason for the title of this piece, “Lowering the Bar”.

 

 

Figure 2. The description of the significance test used in  to determine if trends are significantly different or not.

The “p-value” that the authors discuss is a measure of how unusual a result is. For example, if we flip a coin five times and it comes up heads every time, does that mean that the coin is weighted to come up heads? Or is it just a random outcome? The p-value gives us the odds that it was just a random outcome.

In the hard sciences, people like to see a p-value that is less than 0.001 (written as “p<0.001”). This means that there is only one chance in a thousand (1 / 0.001) that it is just a random outcome.

In climate science, the bar is generally lower. A result with a p-value less than 0.05 is regarded as being statistically significant. A p-value of 0.05 means that there is one chance in twenty (1 / 0.05) that whatever you are looking at is just a random fluctuation.

(As a brief aside regarding the use of p=0.05 as significant , consider that a scientist may look at a variety of datasets trying to find the “fingerprint” of a hypothesized mechanism such as anthropogenic global warming. Suppose on the sixth dataset he examines, he finds an effect which is significant at p=0.05. What are the odds that this is a chance occurrence? The odds are not one in twenty, because he’s looked at several datasets, so his odds of hitting a random jackpot have increased. In this case, if he finds it on the sixth try, the odds are already one in four that it’s just random chance, not a real phenomenon. End of digression.)

Now, if I understand what McKinley et al. are saying above (which I may not, all corrections welcome), they are saying that in their study a p-value less than 0.317 is considered statistically significant. But at that level of p-value, the odds of what is observed being merely a random phenomenon, something occurring by pure chance, is about one in three. One in three? … what am I missing here? Is that really what they are claiming? I’ve read the paragraph backwards and forwards, and that’s how I understand it. And if that’s the case, they’ve lowered the bar all the way to the ground.

In mystery,

w.

Advertisements

100 thoughts on “Lowering The Bar

  1. How can the oceans get more acidic if they are giving off more CO2 than they take in?

  2. In the hard sciences, people like to see a p-value that is less than 0.001 (written as “p<0.001″). This means that there is only one chance in a thousand (1 / 0.001) that it is just a random outcome.
    In climate science, the bar is generally lower. A result with a p-value less than 0.05 is regarded as being statistically significant. A p-value of 0.05 means that there is one chance in twenty (1 / 0.05) that whatever you are looking at is just a random fluctuation.

    In biological sciences the “bar”is also generally P<0.05 (1 in 20). This is consistent with both climate and biology dealing with profoundly complex systems affected by nonlinear dynamics. Biological organisms are as variable and hard to predict as weather systems.
    As for climate modeling – thats clearly in a league of its own as to uncertainty and a bar of p<1 is perhaps fitting.

  3. What gets me is how CNN roars away from the start gate with the standard “It’s worse than we thought”. They grab for the meatiest, bouncyest, most eye-catching headline, probably with zero “investigative reporting”. I can just hear it being delivered in Mr. Blitzer’s cut-breath cadence as the tension rises higher and higher. Once one peels away all the layers of uncertainty, it becomes abundantly clear: this paper is reaching. Reaching hard. The most trusted name in news (itself an hubris-riddled statement like “consensus”) trying to be themost trusted name in science. Ah, I don’t think so.

  4. If this work was correct, wouldn’t the atmospheric CO2 levels show a signal that correlates with global SST? Has anyone looked?

  5. The adsorption of CO2 into sea water depends on water temperature and the partial pressure of the CO2. If temperature rises then the mass of adsorbed CO2 reduces, and conversely for a fall in temperature. If the partial pressure increases then so does the adsorption.
    At the moment sea temperatures are falling so ocean CO2 adsorption is now increasing.
    The only input climate has is temperature but this will not ‘hamper’ anything only change one parameter.

  6. I should have added above that the total adsorption also depends on the internal ocean biological processes which are using the adsorbed CO2 for food, algae, and as a building block for skeletal growth using the biocarbonate loop.( This prevents acidification of sea water). This uses the adsorbed CO2 thus leaving room for more.
    So this system, part of the carbon cycle, is continuous and ever changing in capacity night and day, summer and winter.
    To say that climate hampers this complex cyclic process is simplistic and wrong.

  7. Hmmm if I understand it well, they are testing hypothesis that the trends are different. And as long as the p value is not below 0.05 (0.317 sure is greater than 0.05) they conclude the hypothesis is disproven, i.e. the trends are not different.
    So they are not really lowering the bar, but I guess they are rather going around it.

  8. The paper shows the chart from Takahashi 1999. There are charts for 1995 and 2000 in
    http://www.ldeo.columbia.edu/res/pi/CO2/carbondioxide/pages/air_sea_flux_1995.html
    http://www.ldeo.columbia.edu/res/pi/CO2/carbondioxide/pages/air_sea_flux_2000.html
    resp.
    All the 1995 and 2000 data is downloadable from these two pages.
    A simple arithmetic weighted-by-area averaging indicates SSTs nearly 0.1 deg C higher in 2000 (1995=15.77, 2000=15.86), and atmospheric CO2 (actually pCO2) about 7.4ppm higher in 2000 (1995=298.4, 2000=305.8). Consequently(?), the DELTA-PCO2’s in 1995 average -1.22 whereas in 2000 they average -1.11. [DELTA-PCO2 is the partial pressure pCO2 of the sea surface minus pCO2 of the air just above. I don’t know how Takahashi meaasured pCO2.]. If I have interpreted correctly, this means that the oceans were absorbing CO2 at a slightly greater rate in 1995 than in 2000 (assuming other factors are of minor importance). Presumably this means that the higher SSTs in 2000 slightly more than offset the higher atmospheric CO2.
    So when the paper makes the important-sounding statement that “the ability of oceans to soak up atmospheric carbon dioxide is being hampered by climate change“, they could have expressed it much more simply: “there was an increase in SST“.
    Come to think of it, there wasn’t anything else they could have meant, was there?
    What was the purpose of the paper again?

  9. The significance of Scientific Research today is the size of the Blast of the Press Release, measured in Megatons of Anxiety.

  10. Mike Bromley the Kurd:
    What gets me is how CNN roars away from the start gate with the standard “It’s worse than we thought”. They grab for the meatiest, bouncyest, most eye-catching headline, probably with zero “investigative reporting”.
    What’s to investigate? It’s been peer reviewed. [sarc/off]

  11. Here’s a link to more or less the definitive article on significance of results — John Ioannidis “Why Most Published Research Findings Are False”
    http://www.plosmedicine.org/article/info:doi/10.1371/journal.pmed.0020124
    And yes, Ioannidis is regarded as a serious guy. For the most part he is talking about his field of expertise — medicine. The situation there is especially bad because it turns out that someone who runs a study using the .05 criterion has one chance in 10 rather than one chance in 20 of getting a publishable result. One chance in 20 of “proving” that Twinkies cause cancer. And one chance in 20 of “proving” that they prevent it.

  12. Willis,
    “But they’re not actually measuring the pCO2. “
    I believe most of their pCO2 data is directly measured, by gas equilibration. That is , by actually measuring pCO2 above the sea water. That’s as direct as you can get.
    They describe the method on line 186, and I think that’s the source of the 1,116,539 datapoints mentioned on line 49. The DIC/ALK measures were taken from the SURATLANT data in a special region (line 193). Only 767 datapoints, but they can compute the complete dissolved C chemistry.
    The downside of direct CO2 measure is that you have to rely on empirical equations to attribute pCO2 variation to temperature (as opposed to, say, variation in alkalinity). That’s where working from the indirect measures is better.

  13. steveta_uk asks: “If this work was correct, wouldn’t the atmospheric CO2 levels show a signal that correlates with global SST? Has anyone looked?“.
    This is easy to verify for yourself. Just plot the annual CHANGE in atmospheric CO2 against Sea Surface temperature.
    Annual CO2 change: ftp://ftp.cmdl.noaa.gov/ccg/co2/trends/co2_gr_mlo.txt
    Annual SST: http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt
    As the sea surface warms it sheds CO2 to the atmosphere. You can see this as an increase in the annual change of atmospheric CO2.

  14. Willis:
    I think the explanation for the 0.317 criterion is earlier in the paper in a horrendously long paragraph that starts around line 42. It would take me all day to decode the thing. Assuming that I could do so at all. But apparently for one part of the analysis, they consider one sigma an adequate significance criterion. It could be reasonable … or not.
    – oceanic pCO2). All trends are presented with 1σ uncertainty bounds2, and as in previous
    – studies5-7, an indistinguishable difference between trends occurs when these bounds
    – overlap (see Methods).

  15. @david schofield

    How can the oceans get more acidic if they are giving off more CO2 than they take in?

    Good question. But even if they were taking in more CO2 all that would happen is that they would become less alkaline. At the moment they are slightly alkaline. If all the carbon in all the fossil fuels were burnt and turned completely into CO2. the oceans would just become a bit less alkaline.
    And that is why the correct term is ‘ocean neutralisation’ rather than the scare-mongering ‘acidification’ favoured by misguided and/or unscientific Warmists.

  16. The oceans are acidifying/not acidifying. The shellfish are dying/not dying. The coral reefs are bleaching/not bleaching. The interminable warmist BS is worse than we thought…

  17. Nature Precedings ( http://precedings.nature.com/about ) appears to be a sort of working paper series run by Nature Publishing Group, the subsidiary of MacMillan Publishing that also publishes Nature, Sci Am, etc. Papers are “screened” for appropriateness, but are explicitily not peer reviewed. Feedback is invited, but once a paper is posted it is permanently archived. Apparently posting here “precedes” eventual publication.

  18. Can’t have it both ways in the real world: Ocean ‘acidification’ cannot be the problem a significant portion of the AGW community claims it is if the oceans are not absorbing CO2 like they must to achieve it according to this part of the community’s fear mongering.
    We in effect have competing fear mongering scenarios and they cancel each other out.

  19. From what you say, and skimming the MS, it looks like a single standard deviation from the mean, encompassing 0.683 of the pdf, with 0.16 ish at both of the two tails.
    so the difference is significant if p<0.317.
    The faint justification would be the exceptionally noisy data. Presumably the confidence limits would be huge if they used 2sigma, which is what I always use (biologist…).

  20. Anthony’s original piece says that the article was published by Nature Geoscience on 7/10. It’s possible that the published NG version differs from the Nature Precedings archived working paper version, though I haven’t checked.

  21. The impact of global tempreture (SST for example) on atmospheric CO2 is very important issue. It happens on all relevant time scales.
    I don’t have time to study it in detail, but what I looked into so far, makes me think that the global temperatures determine atmospheric CO2, not the other way around. I also think that the seasonal CO2 variations are caused by the seasonal SST variations, for the most part.
    We don’t have accurate CO2 records and all “anomalous” data was ignored by the consensus.
    I predict that when the cooling really gets going, which is very likely (the sun), atmospheric CO2 will first stop rising and then decrease. That will be check-mate against CO2GW.

  22. UK Sceptic says:
    July 12, 2011 at 4:55 am
    The oceans are acidifying/not acidifying. The shellfish are dying/not dying. The coral reefs are bleaching/not bleaching. The interminable warmist BS is worse than we thought…

    Yeah, warm/cool will do that, don’t you know?
    I have noticed, however, that to the Warmists, it is “worse than we thought / as bad as we thought”

  23. @phlogiston
    Yes, in biological science and epidemiology p<0.05 is used. However, if I understand correctly, there is and additional methodological problems.
    Authors of this study use series of data that were merged to create a single dataset. In such instances first they should check for data set homogeneity using correction for multiple comparisons (like Bonferroni test for multiplicity of comparisons): in this case significance level should be set at 0.05/n (where n is the number of comparisons).

  24. BTW: did anybody check if cases are normally distributed with an appropriate normality test (linke KS)? If not any kind of t-test is a nonsense.

  25. CO2 level in the atmosphere has been dangerously low since the beginning of the ice age several million years ago. At 200ppm plants are being starved for it. Worrrying about it a gigantic waste of time. The earth is a huge lush garden at least up to 2000ppm. Rarely in deep time does it get as low as it is today. Low CO2 is a harbinger of starvation and death for the biosphere. We need more of it not less.

  26. Apparently, the cities can save us:
    Urban plants’ role as carbon sinks ‘underestimated’
    http://www.bbc.co.uk/news/science-environment-14121360

    Dr Davies said current assessments hold that once land in the UK is classified as urban, its biological carbon density is assumed to be zero.
    “Our study illustrates that this is not the case,” she explained.

    The findings will be published in the Journal of Applied Ecology. Are there any initial assumptions remaining, which have proven correct, in the CAGW hypothesis or model? GK

  27. steveta_uk asks: “If this work was correct, wouldn’t the atmospheric CO2 levels show a signal that correlates with global SST? Has anyone looked?“.
    A lot of people have done this including myself. There is little question that the Arctic ocean is the major sink for atmospheric CO2 and the strength of that sink is a function of ocean chemistry and temperature. In correlation, the question is which comes first (driving force),” the chicken or the egg” and what is natural change and how much does man’s contribution of CO2 to the atmosphere contribute to the warming of the Arctic ocean. I submit that CAGW is statistically insignificant. http://www.kidswincom.net/climate.pdf and http://www.kidswincom.net/CO2OLR.pdf.

  28. Next they will be saying that more coal fired power stations would cause global cooling if they can be made to produce the right soot.
    Is this what they mean by the beast eating itself, contradicting consensus.

  29. Are we seeing another example of pal reviewed science? If so, who are the pals and why are they still finding work?

  30. steveta_uk asks: “If this work was correct, wouldn’t the atmospheric CO2 levels show a signal that correlates with global SST? Has anyone looked?“.
    Yes, a lot of people have and there is little question that the Arctic Ocean is a major sink and the strength of that sink depends on both chemistry (including biology) and temperature. The question is which is the driver (SST or atmospheric CO2) and does man’s burning of fossil fuels have a significant effect on this relationship. I submit that changing SST is a natural process that is the driver and the CAGW effect is statistically insignificant. http://www.kidswincom.net/climate.pdf and http://www.kidswincom.net/CO2OLR.pdf.

  31. In low-level universities, there must be duplicates of a boilerplate programme tucked away which has all the correct boxes to fill in to build a case against us wicked humans exhaling; these ‘papers’ are not falsifiable science but naked Green environmental advocacy that, when ‘correctly’ completed, tell us we are all going to hell in a handcart if we persist in manufacturing plant food, and soon.
    Pure Baloney, no matter how it is sliced.

  32. Mike Jonas says:
    July 12, 2011 at 3:09 am
    “Come to think of it, there wasn’t anything else they could have meant, was there?
    What was the purpose of the paper again?”
    This was my immediate reaction. What are they attempting to show that is different from SSTs are higher? Come to think of it, their claim is worse (more ridiculous) than we thought. Their claim is that the oceans have reduced capacity to absorb CO2. Well, if a sink has been absorbing CO2 then does it not have a reduced capacity to absorb CO2 and is that not true by definition?

  33. On top of using a 1-sigma confidence test (p-value<0.317), they perform the test three times (once on each of three data sets). They find only one of these datasets significant. The probability of having one dataset significant due to random chance rises from ~32% for one set to ~71% for one of three data sets.
    I would take thos odds in Las Vegas any day of the week.

  34. Don’t worry about being confused Willis, R.Gates said in another thread, that only about 1000 people in the whole world can understand this stuff. I work daily with information theory, and before that with measurement theory ( developing algorithms for calculating uncertainty in the test and measurement industry), and before that I would spend my weekends stomping around in the mud with my trust Hatch recording data. I guess I am just too stupid. I don’t understand what the authors work either.

  35. P=0.317 = non significant.
    If one of my students did a t-test and said that this P value was “significant” they would be marked down for such an obvious error.
    Or has “post-modern” science rendered all this stuffy and outdated obsession with P values of less than 0.05 obselete?
    Perhaps I should remark all the past assignments that I set and send extra marks and a grovelling apology to those students who clearly knew what they were talking about.

  36. I wonder if everone knows that warm water is more acidic than cold water?
    So as the oceans warm, if that is indeed what is happening, they become more acidic.
    And of course acidity is a property that exists even when the pH is more than 7, or what ever pH number happens to be neutral, as neutral boiling pure water at STP has a pH of about 6.
    So increasing the water temperature or increasing the amount of CO2 dissolved in the water increase the acidity of the water, or decrease the alkalinity of the water, which is the same thing.
    http://en.wikipedia.org/wiki/Dissociation_constant

  37. Having spent many years flogging drugs & the like to the medical profession, I’ve had some interesting discussions about p values. One cardiologist noted that the p<.05 was applicable to a study with only 11 subjects, the more subjects (ie data points) the lower the p value should be.
    If I were to take a study into see a consultant, claiming a p=<317, I'd be shown the door.
    Garbage, complete & under garbage,

  38. “For 1981-2009, trends in oceanic pCO2 are indistinguishable from trends in atmospheric
    pCO2 in all biomes (Figure 1a; Figure 1c, gray bars). Trends are due to changing chemistry of the surface ocean (pCO2-nonT) in all biomes”
    Eh? Does that mean what I think it does?

  39. As John Marshall notes, these authors have just found a round-about way of measuring ocean temperature. Higher ocean temps mean the ocean can hold less CO2. If they were looking at ocean temperatures directly, they would have to note that there has been no increase since 2003. Since they are using a proxy, they apparently feel they have an excuse not to mention this inconvenient truth and instead only analyze the trend from 1981-2009. The difference is dramatic, as NOAA’s ocean heat content graphic shows.
    If the sun stays quiet and we get a significant dip in temperature then CO2 absorption by the oceans will increase. Could atmospheric CO2 even begin to fall? At the least it will start increasing at a substantially slower rate. Might not have any effect on the level of anti-CO2 alarmism though.

  40. Don K says:
    July 12, 2011 at 4:34 am

    Willis:
    I think the explanation for the 0.317 criterion is earlier in the paper in a horrendously long paragraph that starts around line 42. It would take me all day to decode the thing. Assuming that I could do so at all. But apparently for one part of the analysis, they consider one sigma an adequate significance criterion. It could be reasonable … or not.
    – oceanic pCO2). All trends are presented with 1σ uncertainty bounds2, and as in previous
    – studies5-7, an indistinguishable difference between trends occurs when these bounds
    – overlap (see Methods).

    Thanks, Don. However, that’s not an explanation for why they are using one sigma bounds, just a statement that they are using them …
    w.

  41. My impression is that the researchers overfocused themselves on a part of the carbon cycle that isn’t important for CO2 sequestering. Due to ocean ion chemistry, a small decrease in pH is sufficient to give a tremendous increase in pCO2(aq). The net result is that the 30% increase of CO2 in the atmosphere only gives a 3% increase of CO2 mass in the upper oceans (the “mixed layer”). In quantities: the 240 GtC increase of CO2 in the atmosphere did increase the amount of C in the upper oceans with not more than 30 GtC.
    The average pCO2 difference between atmosphere and oceans is about 7 microatm, according to Feely e.a.:
    http://www.pmel.noaa.gov/pubs/outstand/feel2331/exchange.shtml
    But near the poles, especially in the NE Atlantic, the difference is much larger: 240 microatm. That is the place where the Thermohaline Circulation (THC) water sinks ( including extra CO2) into the deep oceans, showing up many centuries later near the Pacific equator. Thus the reduction in pH is not very important for CO2 sequestering, as the main sink places are hardly affected.
    Moreover, other research has shown that there is no reduction in overall sequestering of CO2 at all. It still is at about 55% of the emissions (45% if one includes uncertain land use changes).

  42. carlo napolitano says:
    July 12, 2011 at 6:29 am

    BTW: did anybody check if cases are normally distributed with an appropriate normality test (linke KS)? If not any kind of t-test is a nonsense.

    I thought of that, and didn’t find any comment in the paper indicating it was done, and the data aren’t available to do it.
    w.

  43. If the Arctic Ocean is cold and absorbing CO2 at a faster rate than oceans in the warmer southerly climes, where does it go?
    Is it not reasonable to expect that the polar currents moving towards the equator they carry that CO2 with them? Is it not also reasonable to expect that some of the CO2 will be released into the atmosphere as the water warms?
    If the water arrives in the equatorial zone and releases some of its CO2 as it warms, will this not put extra CO2 into the equatorial zone? Is this detectable? I think not – too much mixing in the atmosphere, but let’s continue…
    If there is a sustained adsorption of CO2 in the Arctic and release farther South, does this translate into detectable equatorial tropospheric heating?
    If not, perhaps the amount of CO2 added or subtracted from the ocean at (very) different water temperatures is not such a big deal.
    If the average temperature, all things considered, is constant, then there should be higher CO2 in the warming zones and lower in the cooling ones.

  44. Alec Rawls says:
    July 12, 2011 at 10:39 am
    If the sun stays quiet and we get a significant dip in temperature then CO2 absorption by the oceans will increase. Could atmospheric CO2 even begin to fall? At the least it will start increasing at a substantially slower rate.
    Quite unlikely. The (very) long term influence of (sea surface) temperature on atmospheric CO2 levels is about 8 ppmv/°C. That is over ice ages and intergalcials and over periods like the MWP and LIA. The current temperature induced variability is about 4 ppmv/°C around the trend, which itself is about 2 ppmv/year (for emissions at about 4 ppmv/year). Thus an in/decrease of 1°C will give you an in/decrease of 4-8 ppmv, depending of the duration, while the human contribution is at 4 ppmv/year, every year, of which halve the quantity disappears in the deep oceans and vegetation…

  45. In general one should not bother with normality tests. They’re much less robust than t (or F) tests, so it’s usually misleading to use them to evaluate the appropriateness of the latter. And if the distributions are wildly non-normal, one can apply a transformation before doing the tests.

  46. If the 12month difference in CO2 concentrations correlates strongly with tropospheric temperatures, as has been shown, this can mean that the 12month difference in our CO2 ouput directly drives temperatures, which would contradict AGW theory, or it could mean that temperatures drive CO2 increase/decrease; rendering it completely irrelevant how much or how little CO2 we emit.
    Thanks, i wasn’t aware just HOW irrelevant all the European renewables and cap&trade schemes were.

  47. Mike Jonas says:
    July 12, 2011 at 3:09 am
    ____________________________
    It seems the temperature point is probably correct. Then would the authors be assuming a homogeneous ocean, like my old favorite “assuming a spherical cow…”? What happens if the temperature in some locations increases such that biological mechanisms of increasing alkalinity is reduced. That is, the temperature exceeds an optimum temperature. Nearby locations that were below the optimum temperature would also warm and then be at the optimum.
    Is there an optimum temperature that is likely? Or with the achievable temperatures in the ocean, would not warmer actually tend to activate the biological processes needed to capture CO2 and increase alkalinity (e.g. nitrogen fixation)? There is quite a range of temperatures in the oceans. Although temperatures seem not to exceed the low 30s °C (http://www.emc.ncep.noaa.gov/research/cmb/sst_analysis/images/monsstv2.png), shallow locations might be significantly warmer. If we are only considering whether CO2 is absorbed or not, then warmer might be better in terms of biological productivity. However, population distributions of the species performing these functions might shift considerably as conditions change.
    It is strange that humans must manage every aspect of the planet on the one hand, and blame ourselves for interference on the other. Let nature adjust itself. Life finds a way to survive. There isn’t anything we can do that would make a difference anyway. I find it amusing how people propose solutions based on small-scale experiments or models, and don’t bother to calculate what would be required to actually do what they propose on a scale that would matter. For example adding iron to fertilize the ocean. A really silly idea when you look at the numbers, see my comment here: http://wattsupwiththat.com/2011/02/02/ocean-fertilization-to-affect-climate-have-a-low-chance-of-success/
    Having said that, a natural source of iron, that is dust, might influence ocean chemistry. A paper discussing a model (ick) of dust changes and possible effects on the ocean is here: http://adsabs.harvard.edu/full/2006TellB..58..560M. Although the authors describe a “positive feedback” for climate change, there are some interesting points made. One is increasing vegetation covering the land a result of CO2 fertilization reduces the dust reaching the ocean. As a result of decreased dust depostion the oceans outgas CO2. Well then, the observed increase in CO2 ppmv might not be due to combustion of fossil fuel after all. The paper also reinforces the importance of nitrogen fixation on CO2 uptake, that is, increasing alkalinity.
    What’s the take-home lesson? Ocean chemistry is not simple. Multiple game changers can invalidate simple models. The ability of biota to respond to changing conditions is one factor that can’t be swept under the rug in any realistic description of ocean processes. Once you realize that, models become recognizable as toys. They are interesting and useful for understanding interactions, but these very limited models should not guide policy making.

  48. DirkH says:
    July 12, 2011 at 11:39 am
    If the 12month difference in CO2 concentrations correlates strongly with tropospheric temperatures, as has been shown, this can mean that the 12month difference in our CO2 ouput directly drives temperatures, which would contradict AGW theory, or it could mean that temperatures drive CO2 increase/decrease; rendering it completely irrelevant how much or how little CO2 we emit.
    One need to be careful: temperature drives the variability in CO increase rate, doesn’t drive the increase itself (except for maybe 8 ppmv since the LIA), which is at about 55% of the human emissions over the past 50 years (and beyond).

  49. The assumption of such work is that of a process feedback.
    And that is just plain wrong.
    If the SST’s rise, the ocean sinks less CO2, which warms the planet and SST.
    An endless trend.
    But, CO2 warming is logarithmic, so the feedback is not linear, but subject to the law of diminshing returns.
    Another problem with the paper:
    “Warming in the past four to five years has started to reduce the amount of carbon that large areas of the (North Atlantic) Ocean is picking up,” said Galen McKinley, lead author and assistant professor of atmospheric and oceanic sciences at the University of Wisconsin-Madison.
    Wrong again.
    The process began not four or five years ago, but has been going on since the bottom of the Little Ice Age. At whatever rate CO2 warms at the level it was when temps began to rise, the logarithmic decrease curve was well underway. So warming oceans will not absorb as much CO2, but since CO2 levels were not zero 4 to 5 years ago, the ability of the oceans to absorb CO2 is even less affected by warming than 150 years ago, which was likewise not zero.
    Making a Mountain out of a moehill.

  50. Errata: The assumption of such work is that of a process feedback.
    Shour read: The assumption of such work is that of a process feedback endless loop.

  51. Dave Springer says:
    July 12, 2011 at 6:41 am
    CO2 level in the atmosphere has been dangerously low since the beginning of the ice age several million years ago. At 200ppm plants are being starved for it. Worrrying about it a gigantic waste of time. The earth is a huge lush garden at least up to 2000ppm. Rarely in deep time does it get as low as it is today. Low CO2 is a harbinger of starvation and death for the biosphere. We need more of it not less.
    Perhaps Mother Gaia with her great wisdom noticed that natural processes of carbon sequestration were proceeding toward a “tipping point” where all the life she struggles to maintain on the planet would be driven into a “death spiral” by carbon starvation and decide to support the rise of humankind because we are the only species capable of developing the physical and intellectual talents to seek out all that sequestered carbon and release it back into the cycle to restore the proper life supporting balance of the planet. All those tree hugging Gaia worshippers out there trying to reduce their carbon footprint to zero, may actually be acting to foil Mom’s clever plan. That could explain the present spate of earthquakes, volcanoes, and other natural nastiness we are experiencing. Mom is just giving us a good smack up side the head to tell us to snap out of it.
    semi-sarc/off

  52. Fred H. Haynie says:
    July 12, 2011 at 7:11 am
    steveta_uk asks: “If this work was correct, wouldn’t the atmospheric CO2 levels show a signal that correlates with global SST? Has anyone looked?“.
    A lot of people have done this including myself. There is little question that the Arctic ocean is the major sink for atmospheric CO2 and the strength of that sink is a function of ocean chemistry and temperature. In correlation, the question is which comes first (driving force),” the chicken or the egg” and what is natural change and how much does man’s contribution of CO2 to the atmosphere contribute to the warming of the Arctic ocean. I submit that CAGW is statistically insignificant. http://www.kidswincom.net/climate.pdf and http://www.kidswincom.net/CO2OLR.pdf.

    OK – Let us continue just a bit further: Geographically, CO2 is NOT well-mixed in the atmosphere – showing very strong peaks where there is no plant life (Gobi Desert, Saudi and Arabian Deserts westward through the Sahara Desert, Kalahari and central Australian deserts, etc.) and very strong troughs where plant life exceeds “normal” over large areas (Eastern US pine and hardwood forests, Amazon and Congo jungles, etc.)
    So, where do the prevailing winds blow across these “natural” – but un-plotted and un-analyzed!!! highs and lows of natural CO2 distribution? How did the “researchers” account for the sinks and sources of the CO2 they are supposedly plotting (er, photo-shopping from modelling) when they have not plotted so much of the world’s known variations in CO2 levels due to plant life?
    What have they correlated between massive differences in ocean plant life – not discussed apparently – compared to a temperature difference of ONLY 0.25 degree from “normal” temperatures?

  53. DesertYote said @ July 12, 2011 at 8:57 am
    “I guess I am just too stupid. I don’t understand what the authors work either.”
    It’s not meant to be understood. It’s meant to be believed

  54. Alkalinity, total hardness, carbonate hardness, pH, carbon dioxide, the carbonate-bicarbonate system, and calcium carbonate are all part of the buffering system present in seawater.
    Seawater contains carbon dioxide gas. When dissoved in water, carbon dioxide reacts with the hydrogen of water to form a weak acid, carbonic acid (H2CO3). If excess carbon dixoide is in the water, carbonic acid levels increase and pH levels drop. If too much carbon dioxide is taken from the water, carbonic acid decreases and pH levels quickly rise. Addition or deletion of carbon dioxide beyond atmospheric equilibrium temporarily changes pH but does not change the alkalinity (carbonate composition) of the water. Carbonic acid forms negatively charged carbonate (CO3 to 2nd -) and bicarbonate (HCO3-) ions. Carbonate in turn joins with calcium to form calcium carbonate (CaCO3). The chemical reactions that form bicarbonate, carbonate, and calcium carbonate are equilibrium reactions; that is they can go back and forth depending on conditions such as increase or decrease of carbon dioxide, pressure, and temperature.
    Under normal physical and biological conditions, the carbonate-bicarbonate ions act as a bank that automatically takes up excess carbonic or other acids, or forms more carbonic acid if carbon dioxide is lost. This is the buffer system that maintains the normal pH of seawater at about 8.2. Bicarbonate is most important in the normal pH range of seawater. Calcium is an important element present in seawater and is part of this system. Calcium is important for building coral skeletons, mollusk shells, and algal support, and has numerous other biological uses as well. As calcium carbonate is removed from seawater by chemical and biological processes (mostely utilized in the shells of tiny planktonic animals), more bicarbonate forms from carbonate, more carbon dioxide is assimilated into the buffer system, and pH remains constant. The carbonate-bicarbonate buffer system equilibrates back and forth and pH stays relatively constant.
    The only way to “acidify” the oceans is to deplete the Calcium, but since Calcium is the fifth most prominent element on earth, that is unlikely. Calcium Carbonate is not very soluble in water, so decreasing the pH (increasing CO2) of seawater will result in the increased precipitation CaCO3, thereby raising the pH.

  55. @Nick Stokes: “I believe most of their pCO2 data is directly measured, by gas equilibration. That is , by actually measuring pCO2 above the sea water. That’s as direct as you can get.
    The problem with that is that the rate of CO2 flux at the ocean surface is driven by the difference between pCO2 in the ocean surface and pCO2 just above it. So if your paper is addressing rate of CO2 flux, you can’t use one pCO2 for the other. See @Ferdinand Engelbeen below. [There are other factors, such as wind speed].
    @Ferdinand Engelbeen: “The average pCO2 difference between atmosphere and oceans is about 7 microatm, according to Feely
    Using data from Takahashi
    http://www.ldeo.columbia.edu/res/pi/CO2/carbondioxide/pages/air_sea_flux_1995.html
    http://www.ldeo.columbia.edu/res/pi/CO2/carbondioxide/pages/air_sea_flux_2000.html
    I calculate it at just over 1 (weighted by area, in both 1995 and 2000). In 1995 it ranged from about -200 in the Arabian sea to about +180 between Svalbard and Greenland. Given the high range, the way the average is calculated is probably rather significant.
    @Crispin in Waterloo: “If the Arctic Ocean is cold and absorbing CO2 at a faster rate than oceans in the warmer southerly climes, where does it go?
    I understand that it sinks into the deep, and is carted off by the Thermohaline Circulation. Several hundred years later, it gets back to the surface and can release the CO2 into the atmosphere again.
    @Hoser: “Ocean chemistry is not simple. … Once you realize that, models become recognizable as toys.
    Nicely put.

  56. Dave Wendt says (July 12, 2011 at 12:26 pm): “Perhaps Mother Gaia with her great wisdom…”
    Personally I think Gaia just got tired of all those nasty asteroid impacts and tried to develop a defense system (us).
    Unfortunately she got Al Gore instead.
    Epic FAIL.

  57. Mike Jonas,
    The direct pCO2 method is described here. It is done in flasks. Yes, in the sea there is disequilibrium, but that is fixed in the measurement process.

  58. RobR,
    That’s a nice summary of the buffering process. But it goes wrong right at the end.
    “Calcium Carbonate is not very soluble in water, so decreasing the pH (increasing CO2) of seawater will result in the increased precipitation CaCO3, thereby raising the pH.”
    No, as you’ve said, there is a chain of reaction. If you add CO2, down the chain free carbonate is converted to bicarb. Then CaCO3, tends to dissolve, not precipitate. That’s the problem that shell-building organisms have with increasing CO2.
    As you say, the pH is buffered. But what achieves that is the conversion of carb to bicarb (mainly) and bicarb to carbonic acid as CO2 is added. And at the end of the chain, adding dissolved carbonate by convertinging CaCO3, which is actually generally present in supersaturated concentrations.

  59. Does the ocean become more acidic when it warms and gives off CO2? Yes it does, until a new equilibrium is reached. This is illustrated well with the Bjerrum Plot for the carbonate system in sea water. It can also easily understood from
    more CO2 => water more acidic => more CO2 dissolved, in reverse.
    Now that the oceans are cooling, we can expect a reversal in the trend of CO2, but the ocean’s inertia is large so it won’t be noticeable immediately.
    The relationship between temperature and CO2 has been demonstrated by Lance Endersby and reproduced on the Seafriends web site http://www.seafriends.org.nz/issues/global/acid2.htm which deals with the CO2 chemistry in water. He arrives at 150ppm per degree C, more than enough to explain all ‘anthropogenic’ CO2 since the industrial age, all coming from the sea.
    There are many compelling reasons why CO2 cannot have an influence on climate, regardless of the views by many. Explained in: http://www.seafriends.org.nz/issues/global/climate.htm . Important educational reading!

  60. @Nick Stokes – Thanks for posting the link. What I was questioning was your earlier “measuring pCO2 above the sea water“, when they should be measuring pCO2 in the seawater. I see they aren’t measuring pCO2 above the sea water in situ – which is what your original statement sounded like – but in the lab. So there’s no problem (well not that one anyway).

  61. RobR is pretty much on the mark. However, in addition to the change on CO2 solubility as a function of temperature, there are very complex interactions with salt concentration. In complex CO2 buffering systems, especially with mixed mono/divalent cations, you cannot estimate the actual pCO2 with any sort of precision.
    The paper cited by Nick Stokes will work, as long as they measured both pCO2 and total acidifiable CO2.

  62. Dave Wendt says:
    July 12, 2011 at 12:26 pm

    The first sensible Gaian hypothesis I’ve ever read.
    😉
    I hope her aim is good when she “takes out” her opponents. The potential for collateral casualties is immense.

  63. About “problems for shell building organisms” … not unless they’re ancestor-worshipers. The bicarbs aid in new shell-building, and in dissolving dead shells. I.e., they accelerate the cycling. Which equals more shellfish, more life. Bigger corals, more migration of corals, faster reef-building, more clams and mollusks, etc. All the things that actually show in the geo-paleo-record.

  64. “Nick Stokes says:
    July 12, 2011 at 5:23 pm
    RobR,
    That’s a nice summary of the buffering process. But it goes wrong right at the end.
    “Calcium Carbonate is not very soluble in water, so decreasing the pH (increasing CO2) of seawater will result in the increased precipitation CaCO3, thereby raising the pH.”
    No, as you’ve said, there is a chain of reaction. If you add CO2, down the chain free carbonate is converted to bicarb. Then CaCO3, tends to dissolve, not precipitate. That’s the problem that shell-building organisms have with increasing CO2.
    =====
    That sounds plausible enough, but is it consistent with the formation of “cap carbonate” deposits when oceans warm after severe glaciations? For those who have never heard of them, cap carbonates are sometimes quite thick deposits of inorganically deposited limestone (Ca2CO3) or dolomite (CaMg(CO3)2). They are, for example, found worldwide in sedimentary sequences from the late Proterozoic — e.g. the 300 meter thick Noonday Dolomite in the Southwestern US.
    That suggests to me that the state equation (ignoring Mg++ which is sort of interchangeable with Ca++) might be something like:
    CO2 + H2O + Ca++ H2CO3 +Ca++ H+ + HCO3- + Ca++ 2H+ + CO3– + Ca++ 2H+ + Ca2CO3
    And that warming the ocean drives the equation to the right, simultaneously making the ocean more acidic and precipitating Ca2CO3. Note that this probably only works if one decreases pH (increases acidity) by warming the ocean, NOT if one simply pours in hydrochloric acid or some other acid.
    This is really complex beyond my ability to understand even though my much younger self received a degree in chemistry from an otherwise reputable university. Whatever the actual mechanism, it would seem to have to be consistent with precipitating lots of CaCO3 when cold oceans are warmed.

  65. I see that WordPress has thoughtfully removed the double ended arrows from the state equation I just posted which was already quite incomprehensible — artificial stupidity in action … It probably treated them as html comments which they did vaguely resemble. Let’s try again using === as a code for a double ended arrow (and removing an extraneous 2 from CaCO3).
    CO2 + H2O + Ca++ === H2CO3 +Ca++ === H+ + HCO3- + Ca++ === 2H+ + CO3– + Ca++ === 2H+ + CaCO3

  66. Nick Stokes says:
    July 12, 2011 at 3:45 am

    iWllis,

    “But they’re not actually measuring the pCO2. “

    I believe most of their pCO2 data is directly measured, by gas equilibration. That is , by actually measuring pCO2 above the sea water. That’s as direct as you can get.

    We do not disagree, Nick. My point remains. As I said:

    McKinley et al. do not have observations for all input variables for all periods. … I bring these issues with the carbon calculations up for a simple reason—errors. Obviously, when you are estimating a critical value (pCO2) using an empirical formula with a choice of parameter values, with missing observations, and not including all of the known variables, you will get errors. …
    Whatever the size of the errors resulting from the oceanic carbon calculations, they need to measured against observations in the regions studied, and then described and accounted for in the study. As far as I can tell the authors have not done either of these things.

    Recall also that this was only one of three problems with the error estimation in the study. The other two were lack of adjustment for autocorrelation, and an absurdly low bar for claiming results were significant. Between the three, the study is meaningless.
    Why do you think that they have picked such a ludicrously high p-value, 0.31, as their threshold for significance? Would they do that if their results were significant at the usual level?
    The real joke is that at a p-value of 0.31, when you look at three areas (say SP-SS, ST-SS, and ST-PS as in their study) you already have better than 50/50 chance of finding a “significant” result by nothing but pure, random chance.
    The odds of NOT getting a false result are (one minus p-value) to the Nth power, where N is the number of areas investigated. This gives us:
    (1 – 0.3) =.7 ^ 3 = 0.34
    That equation’s result gives the odds of NOT getting a bogus result by chance. The odds of it happening by chance are one minus that value, so McKinley et al. have about a sixty-five percent chance of spurious “significance” in their study … I’d take that bet.
    Thanks,
    w.

  67. Mike Jonas says:
    July 12, 2011 at 3:15 am
    steveta_uk asks: “If this work was correct, wouldn’t the atmospheric CO2 levels show a signal that correlates with global SST? Has anyone looked?“.
    I didn’t have SSTs available when I graphed this data in 2009, so I used satellite global lower troposphere temperature. Yes there is a connection.
    http://members.westnet.com.au/jonas1/deltaCO2vsTemp.JPG

    If you treat both the co2 and SST datasets the same way (taking the rate of change) and look at subsets of the curve you can see changes in the rate-of-change of co2 always lag the changes in the rate-of-change of sea surface temp by 6-9 months.
    http://tallbloke.wordpress.com/2011/06/26/which-causes-which-out-of-atmospheric-temperature-and-co2-content/
    Cause precedes effect.

  68. Willis:
    Thankyou for this but, with respect, your comment in this thread at July 12, 2011 at 11:26 pm
    is all that matters: everything else is mere detail.
    The paper by McKinley et al says;
    “Here, we study trends in observed surface ocean partial  pressure of CO2 (pCO2) in three gyre-scale biomes of the North Atlantic, considering decadal to multidecadal timescales between 1981 and 2009.”
    Simply, they analysed data from three regions.
    And the paper also says;
    “Convergence of the oceanic pCO2 trends to the atmospheric pCO2 trend for timeseries longer than 25 years is a robust feature, and the fact that temperature trends are largely indistinguishable from zero suggests that carbon accumulation is the primary driver of these trends. However, a long-term waning influence of pCO2-T is not entirely clear, given that timeseries starting in 1981 continue to be influenced by warming; and thus, multi-decadal climate variability may still be influencing subpolar biome pCO2 trends12,13,25 over the full period for which data is available. In the seasonally stratified subtropical biome (ST-SS, Figure 2b) oceanic pCO2 trends are also sensitive to the choice of years for short timeseries. Beyond 25 years, oceanic pCO2 trends are, with only one exception, indistinguishable from atmospheric pCO2 trends.”
    In other words, they discerned no affect of change to ocean take-up of CO2 from the air in any of the three regions when they analysed periods longer than 25 years (i.e. they find an expected relationship between changes in oceanic CO2 partial pressure and atmospheric CO2 partial pressure in each case).
    But they did discern an apparent effect of reduced ocean take-up of CO2 from the air in ONE of the three regions when they analysed the period from 1981 to 2005.
    This apparent effect was significant at one-sigma: i.e. the apparent significance of the effect had a chance of being wrong one-in-three times.
    SUMMARISING
    1.
    McKinley et al conducted six studies: i.e. three for longer than 25 years and three of 25 years.
    2.
    They used a definition of ‘significance’ that is statistically expected to provide a false indication of ‘significance’ one-in-three’ times.
    3.
    So, their results could be expected to provide one or two indications of ‘significance’ that are wrong.
    4.
    Their results provided only one indication of ‘significance’.
    5.
    Hence, the paper from McKinley et al provides no indication of anything except that it reports they failed to discern a change to ocean take-up of CO2 from the air.
    CONCLUSION
    The paper from McKinley et al is merely another example of pal review.
    Richard

  69. Willis,
    I think they shouldn’t have used the word significance. They aren’t doing a regular significance test, AFAICS. They are using 1 sd as a unit and plotting colored graphs like Fig 1. And they describe a separation of <1 sd as “indistinguishable”.
    This is a common use of an sd. It’s how measurement error is often expressed (+- 1 sd). It’s what we call sampling error in polling.
    I think they probably should have allowed for autocorrelation, but we don’t know how much there was. I’m familiar with patterns of surface temperature residuals, but not with CO2 residuals.

  70. Floor Anthoni says:
    July 12, 2011 at 5:25 pm
    He arrives at 150ppm per degree C, more than enough to explain all ‘anthropogenic’ CO2 since the industrial age, all coming from the sea.
    Sorry, but the conclusion of the late Lance Endersbee is completely bogus: he used a 21 year moving average for the temperature over a less than 30 year period, which is the period with the largest temperature increase. If you use the full period 1900-current without such smoothing, the “excellent” correlation between temperature and CO2 increase falls down, while the correlation between human emissions and increase in the atmosphere is excellent for any period (of at least a decade) for over 100 years…
    See: http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_co2_acc_31.jpg
    and
    http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_co2_1900_2004.jpg
    and
    http://www.ferdinand-engelbeen.be/klimaat/klim_img/acc_co2_1900_2004.jpg
    Further, the temperature-pCO2 curve is known for different salt/DIC contents and pH/alkalinity values of seawater. The temperature influence is only 16 ppmv/°C in average, far from the 150 ppmv “estimate” of Endersbee. In a complex world, the other side of the earth increases its uptake of CO2 with higher temperature, making that the real world effect of temperature on the CO2 level equilibrium is around 8 ppmv/°C.
    See:
    http://www.ldeo.columbia.edu/~csweeney/papers/NOAA_appxdD.pdf
    The pCO2 in surface ocean waters doubles for every 16°C temperature increase.
    For the average seawater temperature of 15°C, that means that the pCO2(aq) shifts from about 250 microatm to 500 microatm with a 16°C increase, or an increase of about 16 microatm/°C.

  71. If N(x) is the standard normal CDF, N(1) – N(-1) = .683, so this is apparently where the authors’ .683 and .317 are coming from, and so they are indeed using a 2-tailed p-value of .317 as their criterion for “significantly different”.
    In economics, RA Fisher’s arbitrary and admittedly rather high value of .05 is the usual threshold of “statistical significance.” In truth, this should be regarded as merely “noteworthy”, anything less being “not noteworthy”. I usually describe .05 as “significant” and give it one asterisk but hope for .01, or “highly significant” (two asterisks). If I’m desperate, 0.1 is “weakly significant” and rates a dagger note. In a process of star-inflation, many less-discerning economists give .10 one star and .05 two stars. But calling 0.3 significant would be considered laughable in any journal outside Nature Geoscience.
    A further problem (besides serial correlation) may be that in Willis’s quote they they describe their test as employing only one of the two standard errors (which one?) instead of incorporating both (resulting in approximately 1.41 times the single se if the regression errors are independent). But I haven’t looked at the quote in the context of the actual Nature Geoscience article.

  72. Don K:
    This is how I make a double arrow:
    Hope that solves the problem. Good post btw. GK

  73. Nick Stokes says:
    July 13, 2011 at 3:54 am

    Willis,
    I think they shouldn’t have used the word significance.

    And I think they shouldn’t have written the paper at all. But ignoring both of our wishes, they did write the paper, and they did use the word “significance”.

    They aren’t doing a regular significance test, AFAICS. They are using 1 sd as a unit and plotting colored graphs like Fig 1. And they describe a separation of <1 sd as “indistinguishable”.

    Sorry, that doesn’t agree with what they said. They are talking about statistically significant differences between trends. They say:

    If t*< T then the trends are not significantly different (p-values are greater than 0.317).

    So yes, they are definitely testing for significance, and they are using 0.317 as the critical value … you sure you want to continue to argue in favor of that?
    I don’t understand why you are looking so far afield to try to find something, anything at all right with this paper. It is garbage from start to finish. The fact that you are defending it speaks volumes about the weakness of your general claims. If you need to rescue this paper from the rubbish bin to make your case, you’re backing the wrong horse here …
    w.

  74. 2 sigma away from a mean (p = 0.05) is what is required in statistics to be considered significant (only 5% of the normal population distribution around a mean remains at this point; so if your mean is below the 0.05 marker, it’s likely to be a new mean and not simple random points from the mean you are testing against). This is a STATISTICAL fact, and is regardless of the field of science being used. You can make the criteria for significance even higher, such in material sciences and engineering, from the outset as is needed (by stating it outright), but for statistics itself, 0.05 is the accepted minimum value for a p to be beneath for significance to be established.
    1 sigma is NEVER significant.

  75. Engelbeen, let’s check some reality here.
    1) Lance Endersbee used reliable data which unfortunately does not cover a long time span. He then applied a strong averaging filter, which is entirely appropriate since this simulates the ocean’s inertia. The outcome is surprisingly linear and is the only inconvenient fact giving an indication of the ocean’s overall rate of outgassing. What Takahashi et al are trying to do with local CO2 ocean fluxes will not lead to any idea what the oceans are doing in their totality – never. Just look at their graphs and the doctored false colour scale. Look at the huge range of values both positive and negative, in their publications.
    2) what you are doing is taking a correlation serious, a correlation based on half century of reliable CO2 data, spliced onto a multiple-century curve of doctored data (ice cores), doctored and spliced precisely (as a hockey stick) to make the CO2 to temperature correlation look perfectly linear, and therefore convincing. This is Voodoo science.

  76. Floor Anthoni says:
    July 13, 2011 at 1:41 pm
    I don’t think this is the proper place to repeat all the arguments again, but a short reply:
    1. Lance Endersbee used reliable data which unfortunately does not cover a long time span.
    While the reliability of the SST data is subject to some discussion (see CA today), my points are: why did Endersbee use only SST since 1980 and not since 1850 or (if he didn’t trust the ice core CO2 data) since 1960 to compare with CO2? Even the filtered temperature data are negative in the period 1960-1970, with increasing CO2, so where leads that for the correlation? And such a heavy smoothing makes even the worst correlation better between two upgoing variables.
    2. a correlation based on half century of reliable CO2 data, spliced onto a multiple-century curve of doctored data (ice cores)
    I have far more confidence in ice core CO2 data than in historical CO2 data taken at places where the variability of CO2 levels within one day varies between 250 and 600 ppmv. But to show you that there is nothing wrong by splicing in this case, here the separate comparisons for the periods 1900-1959 (ice cores) and 1960-2004 (Mauna Loa):
    Temperature – atmospheric CO2 1900-1959:
    http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_co2_1900_1959.jpg
    I have not made the 1960-2004 comparison for temperature and CO2 but 1900-2004 is here:
    http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_co2_1900_2004.jpg
    Note that no matter what the temperature does, even with drops and increasings of halve the temperature scale, CO2 simply continues to go up.
    Now accumulated emissions 1900-1959 vs. atm. CO2 increase (ice cores) 1900-1959:
    http://www.ferdinand-engelbeen.be/klimaat/klim_img/acc_co2_1900_1959.jpg
    Not bad, despite relative larger errors in CO2 emission estimates and CO2 measurements.
    And based on far better inventories and measurements (Mauna Loa and South Pole) here for 1960-2006:
    http://www.ferdinand-engelbeen.be/klimaat/klim_img/acc_co2_1960_2006.jpg
    My conclusion is that temperature has a small, fast influence (about 4 ppmv/°C) on the variability around the trend, but that the trend itself is almost completely caused by the emissions. Or give me a simple explanation why nature would follow the emissions in such a fascinating extremely linear way…

  77. Engelbeen:
    1. The oceans contain 40,000 Gt carbon against an atmosphere of 700 (round figures here). Henry’s law insists that about 3% outgases for every degree C. For the whole ocean thus 1200 Gt = 700ppmv, or 700/50 = 14 ppmv per degree C for each 100m of ocean depth. The regularly perturbed layer is at least 600m which means that 80ppmv is easily released for each ºC. Nobody knows how much deeper perturbation goes, but these figures are BIG even though some oceans breathe in whereas others breathe out. (We’re not taking into account the biological side)
    2. Reliable data for average ocean temperature is still not available, but certainly not before 1984. Now that over 3000 drifters do the rounds, more accurate data must eventually become available. Even so, it will be difficult to establish what the ‘effective’ average ocean temperature is in terms of ocean-CO2 breathing, also since wind speed plays a role. Now that the world began cooling, we may see what is really going on. Even so, the Mauna Loa and other CO2 data will first be ‘corrected’ and ‘adjusted’ to support the orthodoxy, as was/is done with land temperatures and satellites.
    3. The linearity of CO2 in air vs cumulative emissions is indeed creepy but it is only a correlation. It is equally creepy that CO2 rises linearly for an exponential increase in emissions. To explain it as ‘the sea/environment absorbs a fixed percentage’ is perhaps just too simple. For instance, a higher concentration of CO2 encourages plants to grow more vigorously but it takes a while before carbon enters the soils from where it stimulates plant growth further. In a carbon-starved environment, which Earth really is, plant growth should react exponentially with increasing levels of CO2. Thus by leaving a linear increase in air, the environment is capable of sequestering an exponential increase in emissions. Remember, increased sequestration is not possible without increased concentration in air. So this is a simple explanation for you. (Note that experimental growth measurements on plants have been too short of duration to allow the soil to sequester carbon. Also many experiments were conducted in pots.)
    For me the jury is still out. Time will tell.

  78. tallbloke-“… you can see changes in the rate-of-change of co2 always lag the changes in the rate-of-change of sea surface temp by 6-9 months.
    My comment was made in some haste, purely in answer to the question of whether there was a connection. Yes there is a lag of around 6 months, with temperature preceding CO2, in my graph the CO2 curve was moved back 6 months to line up visually.

  79. In figure 2, I notice that in the quoted text, it’s not “Student’s t-test” but “a student t test”. This is, I believe, an unconventional way if not incorrect way to describe in a scientific paper the proper name of the test. Goes to the writers knowledge as well as the reviewers and editors. GIGO

  80. Floor Anthoni says:
    July 13, 2011 at 6:43 pm
    Floor, here a short (?) answer to your points:
    1. Henry’s law insists that about 3% outgases for every degree C.
    Henry’s Law is about pressure, not about % of a content. That is an essential difference. According to solubility parameters of CO2 in seawater (including all dissociation reactions), the temperature influence on the water phase CO2 pressure is about 16 microatm/°C. Thus a 1°C increase in temperature of even the upper few meters of the oceans increases the pCO2 of the ocean water at the surface with 16 microatm. If the pCO2 pressure of the atmosphere is increased with 16 microatm (about 16 ppmv), the net result is zero CO2 transport between the oceans and the atmosphere (if there was a previous _dynamic_ equilibrium between atmosphere and ocean surface).
    16 ppmv increase of CO2 in the atmosphere means that a release of 32 GtC (as CO2) into the atmosphere is sufficient to compensate for 1°C increase in seawater temperature, no matter if that comes from the (deep) oceans or from humans. Thus it doesn’t matter at all how much carbon is in the (deep) oceans or what the temperature in the deep is, only the ocean surface temperature matters with a relative small influence on atmospheric CO2 levels.
    The human emissions nowadays are about 8 GtC/year, or about 4 ppmv increase per year. Even only after 4 years, that surpasses the effect of 1°C increase of the oceans surface. From a lot of studies, it is clear that both vegetation and oceans are net sinks for CO2, which together remove about halve the amount of CO2 injected by humans (in quantity, not in individual molecules). Oceans remove about 2/3rd, vegetation 1/3rd of 4 GtC/year.
    2. Reliable data for average ocean temperature is still not available, but certainly not before 1984.
    Agreed, but I think that the “consensus” (including most of the sceptics) agree that there was a cooling in the period 1945-1975, which is hardly compatible with increasing CO2 levels, if the oceans were the main source.
    Even so, the Mauna Loa and other CO2 data will first be ‘corrected’ and ‘adjusted’ to support the orthodoxy
    Wow, that is a firm accusation. Any proof of that? All raw data (calculated from 1 hour of 10-second samples and calibration) for 4 stations (including Mauna Loa and the South Pole) are on line, up to 2009 at:
    ftp://ftp.cmdl.noaa.gov/ccg/co2/in-situ/mlo/
    These are all the data, without any correction or adjustment. For daily, monthly and yearly averages, only “selected” data are used, which are not contaminated by local air masses (volcanic or valley vegetation in the case of Mauna Loa). But no matter if you use all raw data or only selected data, the final result is the same: levels and trends are within 0.1 ppmv of each other.
    BTW, the rules for calibration, calculation and inclusion of the data for averaging are here:
    http://www.esrl.noaa.gov/gmd/ccgg/about/co2_measurements.html
    3. The linearity of CO2 in air vs cumulative emissions is indeed creepy but it is only a correlation. It is equally creepy that CO2 rises linearly for an exponential increase in emissions.
    Have a better look at the increase in the atmosphere, that also increases slightly exponentially:
    http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_co2_acc_31.jpg
    That means that the sinks also increase exponentially, which results in maintaining a more or less fixed percentage of the emissions. The whole carbon cycle reacts as a simple linear first order process to the disturbance in the atmosphere, caused by the emissions…
    BTW, increased plant growth over the past decades is measured by the oxygen balance. The 100+ ppmv increase since the industrial revolution (60+ ppmv since Mauna Loa started 50 years ago) did result only in 1.5 GtC/year (0.7 ppmv/year) extra sequestering of CO2 by vegetation.

  81. Engelbeen,
    1. Henry’s law states that “the solubility of a gas in a liquid at a particular temperature is proportional to the pressure of that gas above the liquid.” In practical terms it is also about how much gas dissolves in a liquid at a certain temperature. For all practicality, it has been measured for CO2: http://www.seafriends.org.nz/issues/global/global16.gif. Which leads to the stated 1200 Gt per ºC for the whole ocean. This is a pretty hard fact from which my further calculations follow. How else do you think the oceans contain that much CO2? Remember that it is quite valid to apply this to the whole ocean because when oceans warm, the places breathing out will do more so while the places breathing in will do less so, in equal measure.
    2. I did not say that the CO2 measurements of Mauna Loa et al. were inaccurate NOW but the daily variations leave enough leaway to doctor the results LATER when they become inconvenient. Just wait; we’ve been there before, WITHOUT EXCEPTION.
    3. the CO2 growth curve indeed curves upward slightly but negligibly so, and in accordance with warming seas. If uncertainty were plotted (as it should BUT NEVER IS), the curve remains well within linear bounds. When you read the esrl.noaa link you provided, carefully, you will notice that their selection criteria are entirely arbitrary. Which leaves me with justifiably deep suspicions.
    4. The oxygen balance has been discontinued for a long time for good reasons – it has deep flaws. Rather than explaining these here, I refer to http://www.seafriends.org.nz/issues/global/climate4.htm#missing_oxygen
    Just wait and see.

  82. Floor Anthoni says:
    July 14, 2011 at 5:03 pm
    1.Henry’s law states that “the solubility of a gas in a liquid at a particular temperature is proportional to the pressure of that gas above the liquid.”
    A more complete definition is at Wiki:
    “At a constant temperature, the amount of a given gas that dissolves in a given type and volume of liquid is directly proportional to the partial pressure of that gas in equilibrium with that liquid.”
    At a given pressure of CO2 in seawater, CO2 will come out if that pressure is higher than the partial pressure in the atmosphere (as happens near the equator), or get in, if the pressure is lower than in the atmosphere (as happens near the poles). With sufficient time and constant temperature, as much CO2 is going into the oceans as is coming out, a dynamic equilibrium is at work. At that moment, the (weighted) averages of pCO2(aq) and pCO2(atm) are equal.
    If you increase all surface seawater with average 1°C, pCO2(aq) increases with about 16 microatm. Thus more CO2 will be emitted from the sea surface into the atmosphere than reverse. This increases the amount and thus pCO2 of the atmosphere, until after some time the increase reaches 16 microatm (~ 16 ppmv). At that moment a new dynamic equilibrium is reached and the streams in and out between atmosphere and ocean surface will be equal again. Only 32 GtC is needed to give the necessary increase in partial pressure in the atmosphere to reach a new dynamic equilibrium. No matter how much CO2 is in the oceans, no matter what the temperature of the deeper layers is.
    The figure you provided is for CO2 in pure water, which is completely inapropriate for seawater, which contains Ca and Mg ions, mainly in the form of carbonates and bicarbonates. This makes that seawater contains far more CO2 than according to Henry’s Law, because only 1% of all CO2 in seawater is free CO2 (or H2CO3), which obeys Henry’s Law. See:
    http://www.eng.warwick.ac.uk/staff/gpk/Teaching-undergrad/es427/Exam%200405%20Revision/Ocean-chemistry.pdf from page 5 on.
    2. Mauna Loa is not the only station where CO2 is measured. There are some 70 places in “background” areas, maintained by different groups in different countries, independent of each other. Don’t you think that at least some of them will protest if someone tries to manipulate the figures?
    3. The uncertainty of the emission estimates (based on fuel sales and burning efficiency) nowadays is not too bad (-0.5 to +1 GtC), the uncertainty of the global CO2 levels (average of several sealevel stations, Mauna Loa not included), is +/- 0.4 GtC (+/- 0.2 ppmv) around a trend of +120 GtC (60 ppmv) in the past 50 years. Simply negligible.
    In the period 1945-1975 (or 1960-1975, if you don’t trust the ice core data), ocean temperature trend and CO2 trend are opposite.
    Even including all outliers of the CO2 data doesn’t change the average or trend. Only the variability increases.
    4. Your calculation of oxygen use from fossil fuel burning does only account for the carbon in the fossil fuel, not for the hydrogen. That makes that you underestimate the oxygen use for other fuels than coal: a factor 1.5 for oil and a factor 2 for natural gas. The oxygen measurements just are going on, see:
    http://bluemoon.ucsd.edu/images/ALLo.pdf

  83. Floor,
    Some addition to your point 4. From your website:
    During that decade humans burnt 65GtC or 30ppmv of which 57% or 17ppmv remained in air. During that same period, about 200 per meg = 200 / 4.8 = 41ppmv oxygen went missing, or nearly 2.5 times as much. We can account for 17ppmv oxygen but not for the remainder
    In addition to the underestimating of oxygen use for oil and natural gas, what is used as oxygen is for 30 ppmv CO2, not 17 ppmv. It is not relevant for the amount of oxygen used how much CO2 remained in air, only what was emitted. Thus the final sum is:
    30 ppmv O2 for the carbon in fossil fuels + x ppmv O2 for the hydrogen in oil and gas.
    As oil uses 1.5 times O2 compared to its carbon and gas 2 times, that makes that the 41 ppmv O2 use is plausible, as oil and gas are a large part of fossil fuel use.
    If you want, you can figure it out exactly how much O2 is used, as the inventories of the use of the different types of fuel are available. See further where these calculations lead us:
    http://www.bowdoin.edu/~mbattle/papers_posters_and_talks/BenderGBC2005.pdf

  84. Engelbeen,
    1. You don’t seem to grasp the CO2 situation:
    a. there is no equilibrium. CO2 flows constantly from the sea through air to the land when the seas warm, and back when they cool.
    b. during ice ages the land is poor, whereas in warm interglacials it is rich, sustaining and powering growing civilisations.
    c. plants would grab more CO2 from air if they could, but they are starved. Only a higher level in air allows them to grab more. They don’t care where that CO2 comes from.
    d. you cannot have global warming without also an increase of CO2 in air and slight acidification of the sea.
    e. human CO2 plays a negligible role.
    f. the effect of temperature far exceeds that of human contributions.
    g. concentrations in air have a negligible effect on the sea.
    2. Your last point taken. I have to check this further. Still, we are talking about correlations, trying to figure out whether the tail wags the dog or the other way round.
    3. HOWEVER, why are we worried about more CO2 wherever it comes from? It is a highly beneficial gas which does NOT affect temperature, as has extensively been proved by many criteria:
    a. none of the model predictions have come true, thereby decisively disproving the global warming hypothesis. The models have proved themselves wrong.
    b. temperature always leads CO2.
    c. many thermometer stations disprove the hypothesis.
    d. in distant epochs there were periods with high CO2 and low temperatures.
    e. temperature on Venus between 1 and 0.2 bar (our troposphere) follows exactly that of Earth. http://www.seafriends.org.nz/issues/global/climate1.htm#Venus_atmosphere. In other words there are no greenhouse gases, or any gas is a greenhouse gas, mainly nitrogen.
    f. never has empirical evidence been found for the hypothesis.
    4. what society MUST do is just:
    wait and see.

  85. Floor,
    1.a. I am very aware of the CO2 situation: CO2 levels are dynamic: that means if the temperature fluctuates, so does the inputs and outputs of the atmosphere. In opposite direction for ocean surface and vegetation.
    For seasonal changes over the full globe, the variation is about 5 ppmv for 1°C. A double amplitude in the NH, a halve amplitude in the SH. The variation is mainly by vegetation, not by the oceans, as the 13C/12C ratio shows.
    The (weighted) average of pCO2(aq) is lower than the pCO2(atm), thus the oceans are a net sink for CO2, not a net source (averaged over a year).
    1.b. From ice cores, we know that there was a firm correlation between temperature and CO2 levels on earth over the past near million years: about 8 ppmv/°C over glacials and interglacials. If you don’t trust the ice cores, similar changes were found in foramins (and roughly in stomata data). Even in current times, the variability around the trend caused by temperature variations is about 4 ppmv/°C. Thus worst case, the increase in temperature since the LIA (maximum 1°C), can have increased CO2 levels with 8 ppmv. That is all. The rest of the 100+ ppmv increase is from human emissions.
    1.c. Plants use more CO2 when it is available. But not 100% more with a doubling of CO2. Extra growth is mainly between 20-80%, average 50%. The same for the oceans, where diffusion speed is very low and most absorption and release is thanks to wind stirring. That means that any extra CO2 above the (temperature mandated) equilibrium resides rather long in the atmosphere (~ 40 years half life time).
    1.d. The direct effect of temperature is small: 16 ppmv/°C, but as vegetation acts in countercurrent, the real global effect is only 4-8 ppmv/°C. If temperature was the cause, the pH of the oceans would increase, not decrease, as CO2 escapes and relative more Ca en Mg ions stay in solution. Total inorganic carbon content (DIC) would decrease. But we see exactly the opposite: pH lowers and DIC increases. See:
    http://www.bios.edu/Labs/co2lab/research/IntDecVar_OCC.html
    1.e. Human emissions are responsible for near all of the increase of CO2 in the atmosphere.
    1.f. Simply not true, temperature has a minor role in the increase.
    1.g. If you have a bottle with carbonated water and you close the bottle, at a certain moment the pressure of the CO2 in the air above the liquid will be in equilibrium with the pressure of CO2 in the water: as many molecules enter the water as molecules escape. No matter how much more CO2 still is in the water.
    If you change the temperature of the bottle, so will change the pressure above the water, again until a new equilibrium is reached. Thus the difference in (partial) pressure between CO2 in seawater (and plant alveoles) and in the atmosphere determines the net direction that CO2 will flow and with what speed (besides wind mixing speed, sink/upwelling speed,…).
    3. I haven’t much problems with that point, but my fear is that if you don’t accept that the extra CO2 (in mass, not in origin) in the atmosphere is from human emissions, while all observations do agree with that, you undermine the position of the sceptics where the “consensus” is far weaker: the effect of 30% more CO2 (or a CO2 doubling)… One can discuss that point to every degree (my view here is anyway less than the “consensus”, thus maybe a beneficial 1-1.5°C for a CO2 doubling). At that point, the models have been proven wrong…

  86. Engelbeen, I begin with point 3, where you think that I “undermine the position of the sceptics”. It may come as a surprise to you that the skeptics do not have a consensus as the warmists have. So my statements are just looked at and evaluated on their merit. No more, no less. I hope though, that they will give them serious thought, because nature does not work the way we think.
    Our face-off is becoming a bit messy, particularly for outsiders. I do not claim that the increase in CO2 in air does NOT come from human emissions, but I am skeptical about it, because, as I have shown, there are other explanations that are not considered by the mainstream. Your replies also clearly show that the science is not settled, and you didn’t read my points carefully enough. So here we go again.
    Our main difference, I think is that you look too much at the reservoir concept of the atmosphere, like a bean counter considering the asset inside rather than the invisible flow (“black market”) which is quite significant but almost impossible to measure. You mentioned the word ‘residual’ which is exactly what we see, but not the process. You mention a residual half life time of 40 years. Well, true residuals are infinite. The point is that most CO2 rotates very quickly, inside the soil and inside the ocean, in a matter of weeks to one year. That is the main flow. Similarly the main CO2 flows are from land to ocean and back, with a residual in air. In other words, the air part is the insignificant part, as if it is just a boundary. Remember that so many scientific measurements are wrong because they are done on ecological boundaries (glaciers, sea ice, coastal thermometers, bristle cone pines, and many more). Boundaries always change, and because of their nature, by factors belonging to either or both habitats (more factors than thought).
    1a. rethink that from the flow perspective. Your constant referral to pCO2 is not good because it is something very local and not very accurate, and again remember that boundary statement above? We need to look at the overall picture; not the boundaries.
    1b. same as 1a.
    1c. you didn’t read my point carefully. Growth measurements have been done in pots and otherwise over a short time, without allowing carbon enrichment of the soil. It is not representative of natural ecosystems. New research from satellite data show surprising sequestration by some natural ecosystems. Thus wait and see.
    1d. again the bean counter approach as 1a. looking at the residual rather than the flow.
    1e. quite a religious statement here. More doubt would grace you.
    1f. as I have shown from the solubility of CO2 in water, temperature is a MAJOR factor. You can’t go around this with pCO2.
    1g. the classroom experiment, but where is your buffering? Do you really think that while CO2 was over 4000ppm in air, the oceans were highly acidic (10 times or perhaps one pH unit)? Don’t poopoo ocean chemistry and biochemistry. This remains a huge area of massive doubt. And here too, none of our predictions and fears have come true.
    I think the bottom line of our exchange is that we agree to disagree. For me the jury is still out, and I am neither swayed your way, nor my own way.

  87. Floor Anthoni says:
    July 18, 2011 at 10:54 pm
    Our main difference, I think is that you look too much at the reservoir concept of the atmosphere
    While I look at the atmosphere as a reservoir, what happens to the CO2 mass in the atmosphere is a matter of difference in flows: no matter how much CO2 is circulating back and forth between oceans and vegetation via the atmosphere, only the difference at the end of the year is important, not the height or variability of the (individual) flows, neither the amounts in the other reservoirs.
    The individual flows are only roughly known (or even unknown), but the net result after a year is well known: a loss of 4 +/- 2 GtC out of the atmosphere for all natural flows together. Thus all natural CO2 flows out of the atmosphere together are larger than all natural CO2 inflows together. It is that simple…
    Again…
    1.a. The only driving force for CO2 to go out of the oceans or into the oceans is pCO2. Without a difference in pCO2 between the oceans surface and the atmosphere, there is simply no (net) CO2 flow between them.
    Thus if the oceans warm up, the pCO2(aq) increases, more CO2 is released from the oceans and enters the atmosphere. That increases the pCO2 of the atmosphere until pCO2(aq) and pCO2(atm) are equal, at which point there is again no net flow. No matter how much CO2 resides in the oceans or the atmosphere.
    For each ocean water temperature (and salinity, DIC,…) the pCO2(aq) curve is available. The curve says that for 1°C increase of the water, pCO2(aq) increases with 16 microatm. Thus an increase of ~16 ppmv in the atmosphere is sufficient to compensate for 1°C increase in seawater temperature. You don’t need to know any individual CO2 flow or any individual pressure difference over the oceans, only an additional 32 GtC in the atmosphere is sufficient to stop the (net) CO2 outgassing of the oceans. And as humans have added some 350+ GtC in the past 160 years, that is more than sufficient…
    1.b. Here too pCO2 is important, not the flows (that is just a matter of time), neither the quantities in the (deep) oceans, as the good correlation between temperature (proxy) and CO2 levels over 800.000 years shows.
    1.c. As already said, carbon sequestration by vegetation is measured nowadays by the oxygen balance: in the period 1900-2004 some 1.2 +/- 0.6 GtC/year. With 30% (200 GtC) more CO2 in the atmosphere. A rather meager result, even if it is 1/6th of the human emissions.
    1.d. The physico-chemical result of an increase in ocean temperature is a pH increase and DIC decrease, but we observe the opposite, that means that more CO2 is going into the oceans than comes out. Thus the net flow is opposite to what you expect from the temperature increase.
    1.e. After several years of discussion, I may be quite confident, as I never heard an alternative explanation for the increase of CO2 which doesn’t violate one or more observations, as 1.d. shows. If a theory fails the observations, the theory is rejected…
    1.f. See 1.a.
    1.g. At 4000 ppmv, the oceans contained far more Ca and Mg ions at that time, which means that the pH might have been not so far from today. But foramins have eaten away lots of CO2 and Mg and Ca as can be seen in the white cliffs of Dover and lots of chalk deposits all over the world where once was the seafloor… But I have no fear that fish or foramins or the Great Barrier Reef will die from “acid” oceans…

  88. Engelbeen, I have reworked the missing oxygen experiment of Bender et al. and Battle et al. and found strong evidence supporting your claims. When evaluating the 1990s decade for which we know humans burnt 60GtC, equivalent to 30ppm CO2 and (+45%) 43ppm O2, against Bender’s 200 permeg O2, divided by 4.8 = 41 ppmv O2, it is reasonable to conclude that these two figures match and that the residual CO2 in air is caused by human burning of whatever. The residual part of 30 ppm CO2 is 17 ppm while the other 13 ppm has vanished. There is no other place for it to go, without affecting O2, than the sea. So the sea has been absorbing CO2 rather than expelling it, and for at least 60 years or longer. That it may have been precipitated by human burning, is not relevant.
    So, yes you are right. It remains a mystery of course why the sea absorbs precisely 43% of human emissions, but something else may be more worrisome. For at least 10,000 years the sea has been expelling CO2 since last ice age ended, and now has stopped doing so. It suggests that we have now arrived at the end of the warm interglacial, about to enter the next ice age. With the sun doing a nap now, we may be on the cusp of a steep roller coaster ride, not into a Dalton or Maunder minimum but into a true next ice age.
    Thanks for your comments.

  89. Floor Anthoni – There is another possible explanation, which I think is actually correct: Absent MM CO2, the oceans would have been outgassing CO2, but MM CO2 has tipped the balance the other way.
    Says nothing about where things go next – eg, a severe cooling period could quite possibly see all MM CO2 being absorbed for a while.

  90. Floor Anthoni says:
    July 21, 2011 at 2:37 pm
    So, yes you are right.
    That is what real science needs to be: weighting the evidence and adjusting or rejecting the theories if the observations don’t fit the theory… One can only hope that one day, the CAGW people will become true scientists again…
    If we may take the ice core CO2 levels as real, then we are now some 100 ppmv over the temperature mandated equilibrium of CO2 level. The (weighted) average difference in pCO2 is not that large (some 7 microatm), but that is for the oceans mixed layer, which simply follows the atmospheric CO2 level. More important are the poles, where the waters sink into the deep, but with more CO2 now than in the past, as the ocean pCO2 at the THC sink place should be for near freezing water, but the atmospheric pCO2 increased with about 100 microatm. Thus more CO2 enters the deep in ratio with the increase in the atmosphere (the same may apply for increased CO2 in plant alveoles). Which may explain the fixed ratio vs. man made CO2…
    I don’t have that fear for a new glacial period, but a new Dalton minimum may be coming, the sun is doing strange things nowadays…

Comments are closed.