Rocket Scientist’s Journal
… UNDER CONSTRUCTION …
THE ACQUITTAL OF CARBON DIOXIDE
by Jeffrey A. Glassman, PhD
ABSTRACT
Carbon dioxide in the atmosphere is the product of oceanic respiration due to the well‑known but under‑appreciated solubility pump. Carbon dioxide rises out of warm ocean waters where it is added to the atmosphere. There it is mixed with residual and accidental CO2, and circulated, to be absorbed into the sink of the cold ocean waters. Next the thermohaline circulation carries the CO2‑rich sea water deep into the ocean. A millennium later it appears at the surface in warm waters, saturated by lower pressure and higher temperature, to be exhausted back into the atmosphere.
Throughout the past 420 millennia, comprising four interglacial periods, the Vostok record of atmospheric carbon dioxide concentration is imprinted with, and fully characterized by, the physics of the solubility of CO2 in water, along with the lag in the deep ocean circulation. Notwithstanding that carbon dioxide is a greenhouse gas, atmospheric carbon dioxide has neither caused nor amplified global temperature increases. Increased carbon dioxide has been an effect of global warming, not a cause. Technically, carbon dioxide is a lagging proxy for ocean temperatures. When global temperature, and along with it, ocean temperature rises, the physics of solubility causes atmospheric CO2 to increase. If increases in carbon dioxide, or any other greenhouse gas, could have in turn raised global temperatures, the positive feedback would have been catastrophic. While the conditions for such a catastrophe were present in the Vostok record from natural causes, the runaway event did not occur. Carbon dioxide does not accumulate in the atmosphere.
I. INTRODUCTION
Carbon dioxide, a benign gas, is now the hyper–volatile fuel of public policy, media hype, and world politics. Climatologists, undeterred by their inability to predict even the dominant features of the earth’s climate record – the ice ages and the glacial periods – have nonetheless scored a political coup by cobbling together three selected bits of science into a cataclysmic prediction: man is on the verge of destroying life on the planet.
The three cobblestones are (1) a smattering of greenhouse gas physics, (2) half a million years worth of data from Vostok ice cores and (3) half a century of data from Mauna Loa atmospheric CO2 monitoring. Presented here are new results from analysis of the second, the Vostok data, reductions which have a profound effect on the other two legs of the global warming stool, on the role of carbon dioxide, and ultimately on public policy.
{Rev. 6/29/08.} The IPCC said,
One family of hypotheses to explain glacial/inter-glacial variations of atmospheric CO2 relies on physical mechanisms that could change the dissolution and outgassing of CO2 in the ocean. The solubility of CO2 is increased at low temperature, but reduced at high salinity. These effects nearly cancel out over the glacial/inter-glacial cycle, so simple solubility changes are not the answer.
IPCC, Third Assessment Report (TAR), Box 3.4, Causes of glacial/inter-glacial changes in atmospheric CO2, p. 202. Contrary to the IPCC conclusion, "changes in solubility" and second order effects of salinity are irrelevant. Changes in CO2 concentration due to classical temperature effects on solubility between ice age epochs account for the measured variations. These are intra-epoch effects, and whether they "nearly cancel out" on a larger scale is immaterial. {End Rev. 6/28/08.}
II. VOSTOK DATA
A. CLIMATOLOGISTS’ VIEW OF VOSTOK DATA
Figure 1
"CO2, temperature, and dust concentration
measured from the Vostok, Antarctica ice
core as reported by Petit et al., 1999."
[Dust record deleted.] http://en.wikipedia.org
/wiki/Image: Vostok-ice-core-petit.png#file.
Figure 2
Extraneous traces deleted, http://www.realclimate.
org/index.php?p=221">.
Climatologists show the Vostok ice core data of temperature and carbon dioxide graphically on a frequently reproduced and well‑known chart like that in Figure 1. These data reveal a compelling correlation between the concentration of CO2 and temperature.
An aside: Recently published, new ice core data extend the carbon dioxide trace back an additional 200,000 years. Figure 2. This extended record cannot contribute to this analysis until someone reduces and publishes corresponding temperature data.
The author of Figure 1 employs a bit of marginally acceptable, subjective chartsmanship to underscore a point. He selected scale factors and data ranges to emphasize the correlation between carbon dioxide and temperature. The peak to peak swings in the chart traces are arbitrarily made to look alike. This is subjective and artificial, but harmless here.
What is not harmless, though, is climatologists seizing on the lock-step rising and falling of temperature and carbon dioxide as evidence, if not proof, of their greenhouse gas theory: increased CO2 allegedly causes increased temperatures. (A tacit assumption is that the ice core temperature swings represent the global swings, an assumption adopted for this analysis, too.)
When other analysts examined the data, they found that the CO2 trace lagged the temperature curve by about a millennium. This confounds the greenhouse theory prediction. CO2 couldn’t be the cause of past global temperature increases!
The climatologists were quick with an offense and a defense. They labeled the discoverers of the lag as contrarians. And Carbon dioxide while not initiating the temperature rise surely amplified it:
CO2 changes parallel Antarctic temperature changes during deglaciations (citations). This is consistent with a significant contribution of these greenhouse gases to the glacial–interglacial changes by amplifying the initial orbital forcing (citation).
Bold added, IPCC [2001], ¶2.4.1. http://pame.arctic-council.org/climate/ipcc_tar/ wg1/072.htm. That was a close call for the catastrophists!
B. VOSTOK REMAPPED
Figure 3
Vostok CO2-temperature pairs.
Figure 4
Vostok CO2/Temperature
Constellation.
In Figure 3, each pair of simultaneous readings of temperature and CO2 concentration is a dot on the graph, connected in sequence just to show that the time relationship is not lost. For example, the graph has labels for the ages of the first and last points. Without the paths, the dots form a constellation of data, as shown in Figure 4.
This analysis has no further call for the start and end marks. The graphs are just for human visualization of the data. At its roots, the information in the data is arithmetical.
III. MODELING VOSTOK CO2 CONCENTRATION
Figure 5
Best fit mathematical lines to the Vostok data.
Zero temperature refers to the current
global temperature. http://www.
ferdinand-engelbeen.be/klimaat/
correlation.html.
Engelbeen’s result is shown in Figure 5. He shows a best linear fit and a best quadratic fit, also known as the first and second order fits, respectively. Mathematics guarantees that increasing the order of the fit improves (or at least can’t worsen) the fit.
Mr. Engelbeen found this important Vostok relationship “surprisingly linear”. (Comment #2, http://www.realclimate.org/index.php?p=13#comment-69.) More importantly, his analysis confirms that the curvature in the data is not an optical illusion.
Curves like Engelbeen’s are purely mathematical fits. They indicate correlation, a mathematical relationship, but he gives them no connection to physics. The goal here is to uncover the physical relationship between the historic CO2 concentration and temperature. What causes the concentration effect to be curved as it is? In other words, can a cause and effect model be developed which might account for the correlation seen in the Vostok data?
A. CLIMATOLOGISTS CAN’T ACCOUNT FOR ATMOSPHERIC CO2
According to at least one report, climatologists are at a loss to explain the source of the CO2:
Where did the carbon dioxide come from? “This is one of the grand unsolved puzzles in climate research,” said Thomas Stocker, a climate modeler at the Physics Institute of the University of Bern. Schoen [1999].
Moreover and to the contrary, climatologists dismiss the oceans as the source. Gavin A. Schmidt (NASA Goddard Institute for Space Studies (GISS), New York, New York; and Department of Earth and Environmental Sciences, Columbia University, New York, New York.) and his blog group at RealClimate believe …
The oceans cannot be a source of carbon to the atmosphere, because we observe them to be a sink of carbon from the atmosphere.
RealClimate, the Group, http://www.realclimate.org/index.php? p=160. Instead, this new analysis establishes that there is no contradiction in the oceans being simultaneously both a source and a sink.
The International Panel of Climate Change (IPCC) seems to agree with RealClimate:
[T]he observed increase in CO2 is predominately due to the oxidation of organic carbon by fossil-fuel combustion and deforestation.
IPCC [2001], C.1 Observed Changes in Globally Well-Mixed Greenhouse Gas Concentrations and Radiative Forcing. http://pame.arctic-council.org/climate/ipcc_tar/wg1/016.htm But predominantly means not completely. So IPCC concedes:
Thus, the terrestrial biosphere does not cause the difference in atmospheric CO2 between glacial and interglacial periods. The cause must lie in the ocean, and indeed the amount of atmospheric change to be accounted for must be augmented to account for a fraction of the carbon transferred between the land and ocean.
IPCC [2001], 3.3 Palaeo CO2 and Natural Changes in the Carbon Cycle, 3.3.1 Geological History of Atmospheric CO2. http://www.grida.no/climate/ipcc_tar/wg1/107.htm. That fraction Stocker estimates is about half:
“About 50% of the 80-ppm glacial-to-interglacial increase can be explained by a change in the solubility of carbon dioxide.”
Schoen [1999], above, continuing her Stocker quotation. The phrase “change in the solubility” can be read several ways. Regardless, the analysis here shows that the well–known, fixed and constant physics of the temperature–dependent solubility of CO2 in water accounts for all the Vostok CO2 concentration measurements.
B. SCIENTIFIC OBSERVATION: SOLUBILITY PHYSICS APPEARS TO ACCOUNT FOR ATMOSPHERIC CO2 CONCENTRATION
Figure 6
Solubility, X_1, of CO2 in water.
Handbook of Chemistry & Physics,
34th ed., 1953, Solubility of Gases
in Water, p. 1532. The curve is the
best–fit, fifth order by the author.
Figure 7
Vostok CO2 concentration
appears to be imprinted by
the solubility pump.
Figure 8
Straight line fit to Vostok
constellation of relative CO2
concentration and temperature
data pairs.
Figure 9
Vostok data represented
by alternative straight lines.
The complement of solubility, 1-X_1, represents the relative amount remaining in the air. (More precisely, the amount remaining in the atmosphere would be C-X_1, where C is an arbitrary constant. The constant C is immaterial to the slope of the curve, so does not enter into the fitting to the Vostok data. Therefore without loss of generality, C is shown as 1.)
As chartsmanship underscored the correlation between Vostok data traces, chartsmanship can make clear the correlation between the Vostok CO2 samples and CO2 solubility in water. Correlation is the key observation underlying this analysis. It is shown in Figure 7 by artful plotting of the complement of the solubility curve atop the Vostok data.
C. FITTING SOLUBILITY PHYSICS TO VOSTOK MEASUREMENTS
To measure this apparent effect of the solubility pump, the concentration of CO2 may be expressed in relative terms, too. In the following, where relative CO2 concentration is shown, it is in percent of the midpoint of the Vostok concentration, and gets the new label CO2r. Also for convenience, the temperature difference gets the popular nickname “Del T”, short for the conventional “Delta T”.
The straight line fit to the constellation of data in relative CO2 concentration is shown in Figure 8.
Correlation and straight line fits share some important properties. The straight line is the unique line that minimizes the total (sum square) error between itself and, in this case, the CO2 concentration ratio samples. That straight line has a slope of 3.42% per degree Centigrade. As shown below, this result places the Vostok data squarely on the solubility curve, showing a physically meaningful operating point.
D. THE OTHER STRAIGHT LINE FIT AND CORRELATION
The analysis could as easily have found the best fit straight line that minimizes the error between the fit and the temperature samples instead of the CO2 concentration. Conventionally, the independent variable is graphed on the x-axis, called the abscissa. But to this point, determining which of the variables might be independent and which dependent, is an objective of the analysis.
The choice of which is the dependent and which is the independent variable is often subjective, reflective of a presumed cause and effect model. Climatologists by their Greenhouse Catastrophe Model assume, and attempt to prove, that temperature is the dependent variable. The straight line fit corresponding to dependent temperature is shown alongside that for independent temperature in the next chart, Figure 9.
The catastrophe model has a slope of 21.6 degrees Centigrade per 100 percent change in CO2 concentration, or 0.216ºC/%.
The product of the two slopes is the mathematical “coefficient of determination”, conventionally labeled r2, with r being the “correlation coefficient”.
This dual line–fitting method unmasks some of the mystery of correlation. The smaller the angle between the lines, the stronger the correlation between the two variables. Here the product of the slopes is 0.740. Since the maximum is one, it is subjectively a fairly strong correlation (r = 0.860).
Others, however, have reported a lag in the CO2 data with respect to the temperature. Equivalently, temperature events lead or precede CO2 concentration changes. Good analytical techniques require quantification of that lead or lag, and offsetting the data traces to an optimum.
The adjustment is readily made because the graphing steps above preserve the information in the Vostok records. The offset has no effect on the conclusions reached, but does provide a small increase in accuracy.
E. MEASURING AND MODELING THE LAG IN THE CO2 DATA
Figure 10
Cross-correlation function
for all Vostok data.
Figure 11
Cross-correlation function
for most recent 4,000 years
of Vostok data.
Figure 12
Vostok sample record with CO2
offset to maximize its correlation
with the temperature record.
Figure 13
Vostok offset, relative CO2
and temperature pair constellation.
Figure 14
Best linear fit to Vostok data
pairs of offset, relative CO2
and temperature.
Figure 15
Best linear fit pair to Vostok
data pairs of offset,
relative CO2 and temperature.
Figure 16
The first order Vostok CO2
concentration varies with
temperature according to the
solubility curve at
0.247 g/100 g water,
corresponding to a temperature
of 8.26ºC.
Figure 17
The 3.49%/ºC slope of the Vostok
CO2 concentration fits the slope
of the solubility curve at 8.26ºC
for 3rd, 4th, or 5th order fits.
Figure 18
The Vostok CO2 concentration
best fits the solubility curve
in the domain of 0ºC to 14ºC.
Figure 19
Vostok CO2 concentration varies
according to the physics
of the solubility of CO2 in water.
By convention, the Greek tau (t for time) stands for lag. The relation between correlation and tau is the correlation function. Auto–correlation is correlation of a record with itself, and cross–correlation is the correlation between two different records. Figure 10 contains the cross–correlation function of CO2 and temperature for the entire Vostok record of 400,000 years. (The graph is more dense on the left because of an intentional computational artifact. Sample intervals increase exponentially to simplify the computation load. The correlation method wraps the data on itself, analogous to a 420,000–year long tape loop.)
Zooming in by a factor of 100 shows the fine structure in the near term. This is Figure 11.
Three or four nearly equivalent peaks appear where carbon dioxide has the greatest correlation with temperature. The fact that the correlation is relatively poor at zero temperature offset emphasizes that the lag is real, and that any model should account for the lag. Subsequent analysis is offset to the nearest local peak in the correlation at 1073 years. As already stated, the correlation shift has no effect on the qualitative result, namely that CO2 is not responsible for but is a response to global temperature. Applying the lag to the model does improve the accuracy of the results by a few percent.
F. LAG–COMPENSATED CO2 RECORD
Offsetting the CO2 trace by 1073 years has the scientifically desirable effect of sharpening or flattening the constellation of data. This is an improvement in signal to noise ratio. It makes the curvature more apparent, as shown in Figure 12.
Again dropping the sample paths and representing the CO2 concentration in percentage produces the new constellation of ice core data, offset for maximum correlation, shown in Figure 13.
The best fit straight line through these points shows that the average variation of CO2 concentration is 3.49% per degree Centigrade, shown in Figure 14. The complementary, catastrophe straight line fit is 21.8ºC per 100% change in CO2 concentration, or 0.218ºC/%, included in Figure 15.
The offset for lag increased the slope from 3.42%/ºC to 3.49%/ºC with temperature as the independent variable, and the catastrophe slope from 0.216 ºC/% to 0.218 ºC/% CO2 with the greenhouse gas as the independent variable. The 1073 year offset slightly changes the operating point on the solubility curve. The product of the two slopes, r^2, is 0.7609, and r is thus increased from 0.860 to 0.872. (Computation of correlation by the straight line fit method does not involved data wrapping.)
For several reasons, the catastrophic fit can be put to rest. Carbon dioxide is dependent on temperature, and not the reverse. The reason is not just the fact that concentration lags temperature changes, but because it is a physical consequence of the ocean temperature distribution.
G. FINDING THE OPERATING POINT FOR THE VOSTOK CO2 RECORD ON THE SOLUBILITY CURVE
The slope of the solubility curve is 3.49%/ºC at 8.26ºC. This is where the straight line fit to the lag–adjusted Vostok CO2 concentration is tangent to the solubility curve. It occurs at the solubility level of 0.247 g/100g water, as shown in Figure 16.
Locating the first order operating point on the original solubility data is made difficult by the granularity of the solubility data. The final point comes from analysis of the slope of the solubility curve in various polynomial representations, as shown in Figure 17.
The Vostok CO2 data occur over a relative temperature region, which mathematicians call the domain, of 14ºC. The best fit of the solubility curve to the Vostok data occurs in the region of 0ºC to 14ºC, the segment of the solubility curve shown in Figure 18.
H. THE CO2 CONCENTRATION IN THE VOSTOK ICE CORE DATA IS IMPRINTED BY THE PHYSICS OF THE SOLUBILITY OF CO2 IN WATER
The operating region from the solubility curve transforms into a curve representing the Vostok CO2 concentration, as shown in Figure 19.
This segment of the solubility curve fit to the Vostok CO2 data accounts for all the Vostok CO2 data. That is, there is no additional concentration of CO2 in the Vostok record which is not imprinted with the solubility data. Additional, long term CO2 not involved in the solubility process would reduce the percentage variations, moving the operating point to hotter and physically meaningless temperatures, or even off the solubility curve altogether.
I. ERROR ANALYSIS SHOWS THE PHYSICS OF CO2 SOLUBILITY IN WATER REPRESENTS VOSTOK DATA BETTER THAN CAN ANY POLYNOMIAL
What remains is assessment of the goodness of the solubility fit and the consequences of the analysis.
First, the solubility curve lies comfortably within the one standard deviation bands of the best linear fit. That fit is shown in Figure 20.
The CO2 solubility model even fits well within the catastrophe trend, as shown in Figure 21.
In fact, the CO2 solubility representation of the relationship between the CO2 concentration data and temperature records at Vostok is superior to any reasonable polynomial fit, as shown by Figure 22.
Superimposed in Figure 21 are every polynomial fit to the Vostok data, from the first to the tenth degree, with temperature the independent variable. Unlike the polynomials, the solubility fit has well behaved end effects. At high orders, the polynomials chase measurement errors, including transient effects like volcano eruptions or forest fires, a weakness that worsens as the order increases. The solubility curve chases neither measurement errors nor transients.
The solubility fit is accurate to within a fraction of a percent of the least error, that of the highest order polynomial. The polynomials are slightly superior at error reduction because they have the effect of reducing measurement errors along with representing the physical process. Polynomials are malleable, mathematically guaranteed to fit the data of the underlying process along with the errors and disturbances, but physically meaningless. The solubility model shape is fixed by the underlying physics, and fits according to whether those physics are applicable. Lastly, the solubility model is insensitive to measurement errors or transient events.
Figure 20
The solubility physics represents
the Vostok CO2 data within
one standard deviation
of the trend line.
Figure 21
The solubility reaction fits
well both linear trend lines
for the Vostok CO2
concentration data.
Figure 22
The physics of CO2 solubility
in water is better suited
to represent the Vostok records
than is any polynomial.
IV. CONCLUSIONS
A. A NEW MODEL FOR ATMOSPHERIC CARBON DIOXIDE
Science is about models of the real world that, first of all, fit all the data. This analysis is a first step in postulating a scientific model for the CO2 observations. The short term objective here is to characterize the observed concentration that science demands future models reproduce, and to assess the consequences.
Looking beyond that characterizing of the Vostok data, the pattern in the data suggests a model for CO2 such as shown in the sketch of Figure 23.
Figure 23
Several processes are simultaneously underway in the Carbon Dioxide Stream of Figure 23. Superimposed on a latitude–temperature graph is the solubility curve (shown without its ordinate axis). Solubility gets a shaded thickness to suggest the temperature dependent potential to absorb or release CO2 everywhere.
The atmosphere is a cloud to portray the global mixing of atmospheric gases by the winds. The CO2 exchange should occur to some extent distributed over the surface of the ocean. It should also occur focused by the ocean’s meridional overturning circulation, also known as the thermohaline circulation, and popularly called a conveyor belt. The circulation descends at the poles and rises to touch the surface dominantly in the Indian Ocean and the Eastern Pacific. When the belt rises to the surface, the current is saturated with CO2 because of the rising temperature and falling pressure. It is ripe to release the gas.
Insofar as the thermohaline circulation governs the rate at which deep waters are exposed to the surface, it may also play an important role in determining the concentration of carbon dioxide in the atmosphere.
Wikipedia, Thermohaline Circulation. The Wikipedia entry also gives 1200 years as the period of the circulation, which is quite close to the observed lag, supplying additional corroboration for the model. See Figure 11, above. This source supplies no hint of the accuracy of the period, or of the probable geographic locations for the release of the CO2. See also http://www.grida.no/climate/vital/32.htm for a nice diagram of the circulation. For a recent revelation that integration of the ocean patterns into the GCMs was still a decade away, see IPCC [2001], Ch. 14 Advancing Our Understanding, ¶14.2.3.2 Thermohaline circulation. http://www.grida.no/climate/ipcc_tar/wg1/508.htm.
The distribution of evaporation and precipitation over the ocean (its hydrologic cycle) is one of the least understood elements of the climate system. However, it is now considered one of the most important, especially for ocean circulation changes on decadal to millennial time-scales.
The Ocean Component of the Global Water Cycle, Raymond W. Schmitt, Department of Physical Oceanography, Woods Hole Oceanographic Institution, [2002]. http://www.earthscape.org/t1/scr01/scr01a.html.
The atmosphere only holds a few centimeters of liquid water, or 0.001% of the total.
heating one part in 100,000 of the water, he seems to attribute to the Man Behind the Curtain that
[i]n a stronger CO2 greenhouse climate it is hypothesized that the hydrologic cycle will intensify.
Id. The cause and effect perversely get reversed. Intensification of the hydrological cycle through heating of the ocean should increase the concentration of CO2 in the atmosphere, where it will have a minute effect on atmospheric temperature.
B. CARBON DIOXIDE SHOULD NO LONGER DRIVE PUBLIC POLICY
Over those 420,000 years, warm ocean water has regulated the concentration of CO2 by release of this gas into the atmosphere. Because there is no trace of build–up of CO2 from forest fires, volcanoes, or the oceans themselves, cold waters must be scrubbing CO2 out of the air. Since there is no difference between manmade and natural CO2, anthropogenic CO2 is sure to meet the same fate.
To the extent that the analyst’s Vostok temperature trace represents a global atmosphere temperature, so does the concentration of CO2. Thus, CO2 is a proxy for global temperature, and attempting to control global temperatures by regulating anthropogenic CO2 is unfounded, futile, and wasteful.
C. GREENHOUSE CATASTROPHE MODELS (GCMs)
Since the industrial revolution, man has been dumping CO2 into the atmosphere at an accelerating rate. However the measured increase in the atmosphere amounts to only about half of that manmade CO2. This is what National Geographic called, “The Case of the Missing Carbon”. Appenzeller [2004].
Climatologists claim that the increases in CO2 are manmade, notwithstanding the accounting problems. Relying on their greenhouse gas theory, they convinced themselves, and the vulnerable public, that the CO2 causes global warming. What they did next was revise their own embryonic global climate models, previously called GCMs, converting them into greenhouse gas, catastrophe models. The revised GCMs were less able to replicate global climate, but by manual adjustments could show manmade CO2 causing global warming within a few degrees and a fraction!
The history of this commandeering is documented in scores of peer-reviewed journal articles and numerous press releases by the sanctified authors. Three documents are sufficient for the observations here, though reading them is rocket science. (An extensive bibliography on climate, complete with downloadable documents, covering the peer-reviewed literature and companion articles by peer-published authors is available on line from NASA at http://pubs.giss.nasa.gov/.) The three are Hansen, et al., [1997], Hansen, et al., [2002], and Hansen, et al., [2005]. Among Hansen’s many co-authors is NASA’s Gavin Schmidt, above. He is a frequent contributor to the peer–reviewed literature, and he is responsible for a readable and revealing blog unabashedly promoting AGW. http://www.realclimate.org/.
The three peer-reviewed articles show that the Global Climate Models weren’t able to predict climate in 1997. They show that in the next five years, the operators decoupled their models from the ocean and the sun, and converted them into models to support the greenhouse gas catastrophe. They have since restored some solar and ocean effects, but it is a token and a concession to their critics. The GCMs still can’t account for even the little ice age, much less the interglacial warming.
All by themselves, the titles of the documents are revealing. The domain of the models has been changed from the climate in general to the “interannual and decadal climate”. In this way Hansen et al. placed the little ice age anomaly outside the domain of their GCMs. Thus the little ice age anomaly was no longer a counterexample, a disproof. The word “forcing” appears in each document title. This is a reference to an external condition Hansen et al. impose on the GCMs, and to which the GCMs must respond. The key forcing is a steadily growing and historically unprecedented increase in atmospheric CO2. “Efficacy” is a word coined by the authors to indicate how well the GCMs reproduce the greenhouse effect they want.
In the articles, Hansen et al. show the recent name change from Global Climate Models to Global Circulation Models, a revision appropriate to their abandonment of the goal to predict global climate. The climatologists are still engaged in the daunting and heroic task of making the GCMs replicate just one reasonable, static climate condition, a condition they can then perturb with a load of manmade CO2. The accuracy and sensitivity of their models is no longer how well the models fit earth’s climate, but how well the dozens of GCM versions track one another to reproduce a certain, preconceived level of Anthropogenic Global Warming. This suggests that the models may still be called GCMs, but now standing for Greenhouse Catastrophe Models.
In these GCMs, the CO2 concentration is not just a forcing, a boundary condition to which the GCM reacts, but exclusively so. In the GCMs, no part of the CO2 concentration is a “feedback”, a consequence of other variables. The GCMs appear to have no provision for the respiration of CO2 by the oceans. They neither account for the uptake of CO2 in the cold waters, nor the exhaust of CO2 from the warmed and CO2–saturated waters, nor the circulation by which the oceans scrub CO2 from the air. Because the GCMs have been split into loosely–coupled atmospheric models and primitive ocean models, they have no mechanism by which to reproduce the temperature dependency of CO2 on water temperature evident in the Vostok data.
GCMs have a long history. They contain solid, well-developed sub-models from physics. These are the bricks in the GCM structure. Unfortunately, the mortar won’t set. The operators have adjusted and tuned many of the physical relationships to reproduce a preconceived, desired climate scenario. There is no mechanism left in the models by which to change CO2 from a forcing to a feedback.
Just as the presence of measurable global warming does not prove anthropogenic global warming, the inclusion of some good physics does not validate the GCMs. They are no better than the underlying conjecture, and may not be used responsibly to demonstrate runaway greenhouse effects. Science and ethics demand validation before prediction. That criterion was not met before the climatologists used their models to influence public opinion and public policy.
The conversion of the climate models into greenhouse catastrophe models was exceptionally poor science. It is also evidence of the failure of the vaunted peer review process to protect the scientific process.
D. WHAT CLIMATOLOGISTS NEED TO DO
The GCMs need to be revamped. They need to have the primary thermodynamic loop restored. This is the chain of dynamic events from solar radiation, through the shading and reflection of clouds responding to temperature changes, absorption primarily in the ocean, and the transport and exchanges of heat and gases by which the oceans create and regulate the earth’s climate and atmosphere. The models need to reflect the mechanisms which make the earth’s climate not vulnerable, but stable.
The CO2 concentration is a response to the proxy temperature in the Vostok ice core data, not a cause. This does not contradict that CO2 is a greenhouse gas, but it does contradict the conjecture that the presence of a greenhouse gas has any destabilizing effect on global climate. Other forces overwhelm the conjecture of a runaway greenhouse effect. The concentration of CO2 is dynamic, controlled by the solubility pump. Global temperature is controlled first by the primary thermodynamic loop.
The Vostok data support an entirely new model. Atmospheric CO2 is absorbed by the oceans. Fires, volcanoes, and now man deposit CO2 into the atmosphere, but those effects are transient. What exists in steady state is CO2 perpetually pumped into the atmosphere by the oceans. Atmospheric CO2 is a dynamic stream, from the warm ocean and back into the cool ocean.
Public policy represented by the Kyoto Accords and the efforts to reduce CO2 emissions should be scrapped as wasteful, unjustified, and futile.
BIBLIOGRAPHY
Appenzeller, Tim, National Geographic Magazine, Feb. 2004, The Case of the Missing Carbon. http://magma.nationalgeographic.com/ngm/0402/feature5/.
Hansen, J., et al., 1997. Forcings and chaos in interannual to decadal climate change. J. Geophys. Res. 102, 25679-25720, doi:10.1029/97JD01495.
Hansen, J., et al., 2002. Climate forcings in Goddard Institute for Space Studies SI2000 simulations. J. Geophys. Res. 107, no. D18, 4347, doi:10.1029/2001JD001143.
Hansen, J., et al., 2005. Efficacy of climate forcings. J. Geophys. Res. 110, D18104, doi:10.1029/2005JD005776.
International Panel on Climate Change (IPCC), Climate Change 2001: Working Group I: The Scientific Basis.
Schoen, Deborah, Learning from Polar Ice Core Research, Environmental Science & Technology, April 1, 1999 / Volume 33, Issue 7 / pp. 160 A-163 A. http://pubs.acs.org/hotartcl/est/99/apr/learn.html.
Schmitt, R. W., Department of Physical Oceanography, Woods Hole Oceanographic Institution, [2002], Columbia Earthscape, “an online resource on the global environment”, The Ocean Component of the Global Water Cycle. http://www.earthscape.org/t1/scr01/scr01a.html.
Dr. Glassman has a BS, MS, and PhD from the UCLA Engineering Department of Systems Science, specializing in electronics, applied mathematics, applied physics, communication and information theory. For more than half of three decades at Hughes Aircraft Company he was Division Chief Scientist for Missile Development and Microelectronics Systems Divisions, responsible for engineering, product line planning, and IR&D. Since retiring from Hughes, he has consulted in various high tech fields, including expert witness on communication satellite anomalies for the defense in Astrium v. TRW, et al, and CDMA instructor at Qualcomm. Lecturer, Math and Science Institutes, UCI. Member, Science Education Advisory Board. Author of Evolution in Science, Hollowbrook, New Hampshire, 1992, ISDN 0-89341-707-6. He is an expert modeler of diverse physical phenomena, including microwave and millimeter wave propagation in the atmosphere and in solids, ballistic reentry trajectories, missile guidance, solar radiation, thermal energy in avionics and in microcircuit devices, infrared communication, analog and digital signals, large scale fire control systems, diffusion, and electroencephalography. Inventor of a radar on-target detection device, and a stereo digital signal processor. Published A Generalization of the Fast Fourier Transform, IEEE Transactions on Computers, 1972. Previously taught detection and estimation theory, probability theory, digital signal processing.
© 2006 JAGlassman. All rights reserved.
TrackBack
TrackBack URL for this entry:http://www.rocketscientistsjournal.com/cgi-bin/mt/mt-tb.cgi/46
Comments (45)
Jeff:
I do not understand the thermocoline of deep ocean currents. Where can I read more, or perhaps you can simply it for me?
al
[RSJ: Try thermohaline.]
Posted by Alvin Clavin | October 24, 2006 11:44 AM
For completeness, what does the correlation plot for temperature lagging CO2 look like (Figure 11 with negative tau)? Are there any peaks worth noting?
[RSJ: A graph showing the negative axis was easy to compute, but for the moment too difficult to post here in a comment. It shows a major peak around -800 years, and otherwise is roughly similar to the correlation along the positive axis. The two-sided graph supports well the conclusion not to use zero lag, but it leaves rather arbitrary which of many peaks one might use. The two-sided graph is roughly an even function, but quite noisy. It shows that one should not place too much reliance on cross-correlation. It provides a good clue how one might model the relationship between parameters, and may provide good rejection criteria.]
Posted by Stephen | October 30, 2006 9:53 PM
FYI, Gavin Schmidt has replied to this (though briefly).
[RSJ:See Gavin Schmidt's Response to the Acquittal of CO2 Should Sound the Death Knell for AGW on this blog.]
Posted by Crust | November 1, 2006 7:40 AM
Here is a graph that clearly shows the CO2 lagging the temperature changes;
http://www.daviesand.com/Choices/Precautionary_Planning/New_Data/IceCores1.gif
[RSJ: I recommend against drawing any such conclusions by eyeball assessment of parametric data plots. It's a numeric problem, suited to more objective computer calculation.]
Posted by Jesse | November 7, 2006 2:59 PM
Oceans and Global Warming
November 7th, 2006 by globalwarming2000
Number of science reviews in this field have linked solar activity to the climate change. Rise in global temperature is always accompanied by the rise in CO2 concentration. Human contribution may be significant but it is not critical. By far the greatest amount of CO2 is released by the world's oceans; they are also the largest absorbers. The release of CO2 is not, but its absorption is affected by the Sun. The culprits are UV and gamma radiations reaching the oceans' surface during periods of high sunspot activity.
Some 2 years ago I wrote:
Increased solar activity results in an increase of the harmful radiation, reducing bio-mass of the oceans' surface plankton trough process of sterilisation by irradiation. Result of this is reduced uptake of CO2 from the atmosphere and rising in the 'green-house' effect. Reverse process takes place during reductions in the solar activity
[RSJ: The consensus among climatologists seems to be that CO2 uptake by the oceans is affected by the solubility curve, wind, and land area, and that it runs between 92 and 107 PgC/yr. But where is their computation? And where is the physics of the additional radiation effects?]
Posted by C2j+Cjs | November 8, 2006 3:43 AM
Interesting. What peer reviewed journal will it appear in and about when?
[RSJ: Rev. 10/13/07. A well-publicized study by Naomi Oreskes started with 928 abstracts from refereed scientific journals published between 1993 and 2003 and containing the key phrase "climate change", or some say, "global climate change". Among these, she found that 75%, or 696, articles discussed what she considered the Consensus proposition: global warming is occurring because of manmade greenhouse gas. Of those 696, 100% agreed!
[The results prove not Oreskes' conclusion about the existence of a consensus, but instead that with a high probability, the refereed journals in her survey have, for whatever reasons, published no papers disputing the anthropogenic climate change conjecture.
[Consequently, submitting The Acquittal of Carbon Dioxide to one of these refereed journals is a major waste of time.
[A journal dedicated to science and not a cause would seek out and encourage articles challenging the models of the day. That's how science must progress. Instead, too many journals screen against opposing views.
[Journals should adopt and publish standards for acceptance of its papers. They should, of course, require stylistic standards, clarity, and relevance to the field of the journal, but most importantly, compliance with the strict scientific method. The recommendation is that obedience to the scientific method be a prerequisite for peer review and publication, never a consequence of it. What Oreskes' study shows is the degree to which climatology has sunk to astrology, phrenology, sociology, and paranormal-ology, ostensibly peer-reviewed, published fields.
[Of course, any journal that wishes to publish The Acquittal of Carbon Dioxide is free to do so at the cost of nothing but attribution. It has been reproduced on other blogs.
[Peer review in American science is severely compromised. It is undergoing a treasured rebirth on the Internet. Anyone, peer or not, is free to criticize by commenting on this blog. Peer silence indicates the absence of error or of the need for further exposition.]
Posted by Jeff Stewart | March 22, 2007 3:48 PM
Please comment on the linked plot. It explores some of the same concepts you have addressed.
I cross-plotted the Vostok data as you have, but used CO2 concentration as the independent variable as AGW advocates assume. I then used the logarithmic best-fit trend to project temperatures expected if CO2 is indeed "forcing" T.
The CO2 values come from the US South Pole Station, 1958-2004. These values are similar to those from other sampling sites, including Mauna Loa, suggesting good atmospheric mixing of CO2, and should be applicable to Vostok.
Temperature variation at Vostok is projected at 6.23d C in 2004.
Actual dT at Vostok over this period ranges from -1.8d C in 1960 to +2.3d C in 1980, reaching +0.8d C in 2004. Net change over the period is negligible.
This suggests to me that CO2 definitely is not forcing T.
[RSJ: The AGW advocates do assume that CO2 is the independent variable. However, they contradict that claim in the Third Assessment Report trying to account for the discovery that CO2 lags (the surrogate for) Temperature. To accommodate this inconvenient fact, the IPCC (aka the Consensus) conjectured that while some unknown process triggers a temperature rise, CO2 amplifies it. This conjecture is unsupported by measurements, and results in a model that fails to shape the CO2 concentration according to the complement of the solubility curve.
[Your plotting CO2 as the independent variable, that is, along the abscissa, unfortunately proves nothing. The fact that you found a functional fit likewise proves nothing about dependence or independence. You merely found a way to characterize the shape of the data. Almost any function convex down in the region would have produced a similar result, and looking at the data, a function convex up might have worked even better. As the Acquittal of CO2 shows, one investigator found a nice fit with a quadratic polynomial. Such mathematical regression is the way to "parameterize" (find a parametric equation) relating the variables, but it doesn't provide a cause and effect.
[What you need to do is postulate a cause and effect relationship, and then see if the physics of that C&E relationship fit the data. The complement of the solubility curve for CO2 in water does the trick, and it provides as good a fit as quite high order polynomials.
[I assume you did your curve fitting at zero lag. At what lags (or leads) do you get the best fit? That should give you a clue about which is the dependent variable. You need to explore the cross-correlation function between these data traces, along with a cause and effect model, to support the dependency analysis.
[Your data called "Predicted 1958-2004 Vostok Trend" appear to be the extrapolation of your logarithmic fit. Just to test your method, you might find the logarithmic fit with Temperature as your independent variable. Is the fit about as good? Now find the optimum lags and see which curve is best.
[The Vostok data are biased well above your logarithmic fit at CO2 concentrations from about 270 ppmv to 300 ppmv. Your logarithmic fit fails to predict above 270 ppmv.
[The reverse log fit might do much better, and your temperature extrapolation will be much hotter, maybe up to 10 degrees C in 2004. The point is that one method has as much validity as the other - none. You cannot draw a valid cause and effect (dependent/independent) conclusion from your method.
[Turn now to the well-mixed issue. The IPCC (Consensus) needs that assumption. Some of the graphs in the literature indicate that South Pole readings appear to be from the same population as readings from other parts of the globe, including in particular Mauna Loa in the TAR. This assumption is important to AGW claims that CO2 levels are at unprecedented levels and that man's CO2 pollution has a residence time in the atmosphere between multiple decades and centuries. The data contradict these conclusions, and draw into question the calibration methods used in the various readings.
[The residence time of CO2 is easily calculated from IPCC (Consensus) data. It is about 1.5 years to 2.0 years, depending on whether you include the leaf water uptake reported by the IPCC (Consensus).
[As to being well-mixed, the IPCC (Consensus) reports that the CO2 north-south gradient is ten times greater than the east-west gradient. This implies first that the east-west gradient is discernable, and second that the north-south gradient is at a minimum substantial, at least 10 times what is discernable. This directly contradicts the well-mixed assumption.
[The western Pacific Ocean perpetually emits a huge quantity of CO2. That gas rises at the Equator and splits toward the poles. It rises into Hadley cells which bring the gas down and feed it into the trade winds. This circulation puts Mauna Loa directly in the chimney of the great efflux of CO2 from the ocean. A little decadal shift in climate patterns could move this CO2 plume across Mauna Loa to cause some or all of the observed increases. On the other hand, the cold waters at the poles create a massive sink for CO2. The Vostok data are drawn from the interior of this sink.
[Charles Keeling, the father of the Mauna Loa measurements, warned not to mix such data. However, he was known to merge data from different locations by calibration techniques and adjustments. (Rev. 8/27/07.)
[Except for its well-mixed assumption, the IPCC (Consensus) offers no explanation for matching data from the sink to data from the source, nor how the gradient bias might have been removed by their data calibration.
[As a footnote to the unprecedented CO2 levels in the last 400 to 600 millennia, that is known with a 3% confidence. The present record has exceeded the Vostok maximum for about 50 years. The Vostok data are about 1,500 years between samples. The chances a similar epoch, if it existed, would have been caught by a Vostok sample is about 50/1500.
[The Consensus and the IPCC are wrong. CO2 does not persist for multiple decades or longer, but only for a couple of years. CO2 in the atmosphere is not well mixed, but has a substantial, circuitous gradient from the Equatorial effluxes to the polar uptakes. Present day CO2 is not known to any acceptable degree of confidence to be at unprecedented concentrations relative to the past 400 millennia. CO2 is not well-modeled as a slug of gas inserted as a forcing, but instead is overwhelmingly a temperature related feedback from the ocean.]
Posted by Ursus | April 15, 2007 5:07 PM
A little clarification- in doing this exercise I was not attempting to determine the cause of the shape of the Vostok data crossplot. I was attempting only to test the assumptions adopted by the AGW camp to see if these data support the hypothesis that CO2 forces T. Thus I assumed CO2 as the independent variable, a logarithmic relationship between CO2 and T, and no adjustment for lag.
[RSJ: I wasn't clear enough last time. (a) You didn't accomplish anything by making CO2 the independent variable. (b) Your logarithmic fit proves nothing. These methods cannot solve the Cause & Effect riddle of science. Lack of correlation can disprove C&E. The existence of correlation only suggests where a C&E relationship might exist to be modeled by physics (or chemistry even).]
Linear actually is a better fit and predicts a higher temperature, but again the "accepted" relationship is logarithmic.
[RSJ: You went half-way toward a function convex up. Try making an exponential fit, which is equivalent to making T your independent variable.
[I don't know what you mean by "the 'accepted' relationship is logarithmic." Do you have a reference for this claim? Did you happen to reject the linear fit because the T forecast was too big? You need to set an objective standard first.
[You don't want to fit data including a part that might constitute your validation specimens. Reserve the hypothetical future data as a test case to validate your model.]
I am a little puzzled about the CO2 data. I obtained both Mauna Loa and South Pole data from the CDIAC website. Both data sets apparently were processed by Keeling, and on comparison, show maximum divergence in 2004, the last year of record, of 2.8ppmv. This accounts for my assumption that atmospheric CO2 is well mixed. In fact, on cursory examination, the South Pole records show the lowest 2004 concentration of the eight stations reported worldwide, but varies from the highest only by 4 ppmv. But again, I am using the "accepted" data.
[RSJ: Puzzled is right. Physics indicates a CO2 gradient should follow the wind circulation from the tropical oceanic outgassing to the polar uptakes. The IPCC admits as much, but contradicts it. Keeling warned not to link data from sinks or sources. Yet Mauna Loa sits right in an outgassing plume, and the South Pole data come from inside a sink. The Mauna Loa data should not fit the polar data. Some climatologists linked the Mauna Loa data to the Sipple ice core data by arbitrarily shifting the few Sipple data 83 years! Was something like that done again? The burden is on the Consensus to justify the method Keeling or others employed. Otherwise, we must reject the results. The Consensus may not adjust data from different sources on the grounds that the gas is well-mixed, and then claim it is well-mixed because once adjusted it fit together. That is a bootstrap and abysmal science.]
The conclusion I draw from this plot is that if CO2 were forcing T in a logarithmic manner, we should see a temperature trend, based on 1958-2004 CO2 concentrations, scattered about the plotted projection, which obviously is not the case.
[RSJ: Why? You can fit a logarithmic function to any old scatter of data. You're lacking both an objective standard and a physical model.]
As an aside, I am most impressed by the lags in the Vostok data where T falls dramatically while CO2 remains high for thousands of years. I cannot see how this possibly could occur if the data are correct and if CO2 is forcing T, unless they represent some catastrophic events like massive and sustained volcanic eruption.
[RSJ: Your impressions and the drama of the traces are all subjective, and not science. You need to measure the lead-lag relationship, and that requires calculating the cross-correlation function. As the Acquittal of CO2 paper shows, this particular function exhibits lots of local peaks, including a few strong ones around 1 millennium. That suggests we should look at the 1 millennium thermohaline half circulation. We should find mechanisms for all the peaks.]
Posted by Ursus | April 19, 2007 8:49 AM
It seems you really don't understand what I was trying to accomplish. I wasn't looking for the same outcome you were, I was trying to hang the "consensus" with their own rope. That's why I used the tenets they accept, although I don't agree with them.
[RSJ response: Thanks for being persistent. I get your point now by reading your latest comments with reference to your chart at
[http://bp2.blogger.com/_DVW_My87o0E/ReTQzIHwdeI/AAAAAAAAADM/0jlif89LghQ/s1600-h/Temp+v+CO2b.jpg
[My apologies for not including the link with your first post.]
These are, (1), the Vostok data show a relationship between CO2 and dT, that relationship being that CO2 "forces" T; (2) a rise in CO2 produces a logarithmic rise in T (http://adsabs.harvard.edu/abs/2005AGUFM.V41H..03R); (3), CO2 is well enough mixed in the atmosphere that CO2 levels reported for the South Pole station should approximate those at Vostok.
[RSJ: (1) The correlation at Vostok between CO2 and Temperature supported the Consensus' greenhouse gas conjecture. The lag discovered some time later disproved it. But data traces cannot establish cause and effect; something like the greenhouse effect or outgassing was necessary.
[(2) Your reference to the abstract was helpful to your claim where it said, "the effect of CO2 on temperature is logarithmic." However, I could find no way to retrieve the full text of the Ruddiman paper, even if I wanted to pay a fee (which I refuse to do). I had never run across a claim by the Consensus that CO2 and Temperature had a logarithmic relationship, and the abstract is for a 12/05 paper, well after the Consensus locked on to its conjecture. Regardless of the authority, your chart shows the logarithmic fit to be poor. And my paper, the Acquittal of CO2, posted on this cite shows that the relationship is the complement of the solubility curve, something quite different than logarithmic. The Ruddiman claim is dubious.
[If I might restate your point, it is that even with as poor a fit as logarithmic, the history of the CO2 and T relationship doesn't predict recent Vostok readings (which I take at face value). Much more downward convexity is needed, but the data are actually convex up. Your argument is valid without making any claims for the goodness of the logarithmic fit.
[(3) Vostok and the South Pole ice cores should be quite similar, well-mixed CO2 or not.]
Therefore, if the supposed "forcing" trend seen in the Vostok ice core data is applied to modern CO2 levels, Vostok should be much warmer than it is today. Should be even warmer assuming a linear fit, far warmer with a convex upward fit. But I stuck with the "consensus" tenets, it's off enough even when using them.
[RSJ: you're correct that with the better fits of linear or convex up, the ice core history makes CO2 a much worse predictor of Temperature. Whatever the mechanism is that the Consensus contends causes CO2 to force temperature, it remains hidden in GCM code. It seems to remain optimistic that the GCMs will eventually show that CO2 drives T.
[The GCMs as presently configured will never reveal the curvature between natural CO2 and Temperature. This is because the Consensus makes CO2 a forcing, inserting a slug and watching what happens to temperature. It needs to be a feedback for the overwhelming natural portion, and that alone should demolish the CO2 theory.
[But if the data don't fit, you must acquit. Nice discovery.]
I did not deal with the lag because it was unnecessary to demonstrate my point. In stating I was "impressed" by the lag, I meant that this seems to be the most compelling indication that CO2 does not force T, though I have not attempted to quantify or account for it.
Posted by Ursus | April 20, 2007 11:10 AM
I plotted up CO2 vs. temp records for Law Dome, Dome-C and the rest of EPICA, as you have done here (just for giggles, I was genuinely curious). In Dome-C, CO2 leads temperature.
[RSJ: This is remarkably good news for the IPCC folks (the Consensus on AGW). For the Third Assessment Report, they had invented the naked theory that while CO2 may not have actually CAUSED global warming, it somehow amplified it! This was because some accepted investigators reported that CO2 lagged the Vostok temperature trace, making a shambles of the CO2-cause temperature-effect conjecture.
[Now by lagged, a scientist means that the sample cross-correlation function has a significant lagging peak. Assumedly, you meant that you calculated the cross-correlation between Dome-C temperature and CO2 concentration and found a significant lead component for the CO2 - didn't you? Quantify for the readers, if you would, how the function looked. You didn't just eyeball the graphs, did you?]
Further, the error in age control of the Vostok cores does not permit your analysis, particularly in the early portion of the record.
[RSJ: What do you mean by "age control of the Vostok cores"? What is the error in the handling or analysis of the Vostok cores, and in what way do you claim it affects the analysis in the Acquittal of CO2? What of the Vostok record must be disregarded? This will truly disappoint the Consensus.
[But not to worry. The analysis in the Acquittal of CO2 assumes that each CO2 concentration reading is at the temperature reading at the greatest ice age less than the given gas age. In theory, a large mean error between the two reported ages could disrupt the correlation and the pattern evident between CO2 and Temperature. This did not happen, however. With whatever errors were manifest in the data, the CO2 and Temperature traces were still highly correlated (86%). The pattern matched the complement of the solubility curve as well as any mathematical polynomial up to 10th order. Natural CO2 comes from the ocean. It is a product of global warming, not a cause.]
I am no doubter or promoter of 'global warming' etc. but you need to do a little more research (and cite some people for cripe's sake) before you conduct such an analysis.
[RSJ: When you became not a doubter or promoter of global warming, but a fan and promoter of Anthropogenic Global Warming, how did you judge the science and reasoning? Was there something specific you appreciated, or was it just a matter of the alleged voting? In science, someone always has to go first. How do you measure the adequacy of his research? If the research in the Acquittal of CO2 was inadequate, it must have overlooked something relevant. You must have discovered a paper that refutes the Acquittal of CO2. Please share it.]
It is typical of the arrogance of physics to think such a complex problem can be distilled into an equation (if it could, our climate models would be a little more precise, eh?). This is not 'consider a spherical cow' and I find the science and reasoning of the prominent geologists and climate scientists to be much more sound than this analysis. Moreover, they have a far greater understanding of the system.
[RSJ: Physics has arrogance!? Climate models are filled with equations, and they still can't predict. Climatologists, not climatology, urge public policy based on models that neither predict nor match the historical record. That is not arrogant; it is unethical. A few more equations are certain to be added to the GCMs. But remember, climate is a statistic; don't expect precision. ]
The stoichiometric control over CO2 equilibrium between atmospheric and oceanic reservoirs is a compelling argument, but you have neglected to mention the C sinking effects of oceanic biomass (our friends the foraminifera and nano-plankton) -- changing calcite compensation depth and the net sinking of carbon via CaCO3 in the deep ocean account for a great deal of the CO2 variability observed in the long-term record (check out some of Wally Broecker's work). For instance, what is your mechanism for the early to mid-Holocene CO2 anomaly? Your proposed mechanism requires a fundamental change in the ocean conveyor over very short time-scales to induce such events; from what we know about thermohaline circulation and the ocean conveyor, these changes would require great volumes of freshwater dumped into the North Atlantic, or some other such mechanism.
[RSJ: As to "stoichiometric control over CO2 equilibrium", I give you Caveman's answer: "Yeah, I have a response. What?" Besides, how could the Acquittal of CO2 be negligent in not mentioning an irrelevancy?
[The Acquittal of CO2 shows that in the Vostok record, the atmospheric CO2 concentration is curved according to the complement of the solubility of CO2 in water. A sound physical reason supports this result. That curvature of atmospheric CO2 is not modeled in the GCMs. Instead, the operators insert a massive slug of CO2 as a forcing, allowing it to decay slowly over several decades to a century plus. That is now known to be false from the Vostok record, and it is contradicted by the climatologists calculations in the TAR showing 90 GT or more of carbon uptake by the oceans per year. The Consensus position is supported by their well-mixed CO2 conjecture, but that, too, is contrary to the IPCC admission of large gradients (at least ten times the minimum detectable) in CO2 atmospheric concentration. It is also contrary to the physics of the CO2 exchange with the ocean. These findings demolish the CO2 forcing model. As a minimum, the great majority of atmospheric CO2 should be represented as a climatology feedback.
[Not only did the Acquittal of CO2 not mention anything like a "stoichiometric control of CO2 equilibrium", it relied on no stoichiometry at all. Even the TAR refers to stoichiometry only three times, and then with no real consequence. Nor does the TAR make mention of any "Holocene anomaly". Climatologists recognize Holocene anomalies, as in the El Niño Southern Oscillation (ENSO), and in insolation. But the TAR says the CO2 concentration was flat, between 260 and 280 ppmv, during the Holocene, making it actually anomalous. Whatever you mean by "mid-Holocene CO2 anomaly", the paper needs no mechanism for it. It is irrelevant to the inconvenient Vostok data.
[The TAR says, "The total amount of carbon in the ocean is about 50 times greater than the amount in the atmosphere, and is exchanged with the atmosphere on a time-scale of several hundred years. Dissolution in the oceans provides a large sink for anthropogenic CO2, due in part to its high solubility, but above all because of its dissociation into ions and interactions with sea water constituents (see Box 3.3)." Then it says, "This process depletes surface CO3(2-), reduces alkalinity, and tends to increase pCO2 and drive more outgassing of CO2 (see Box 3.3 and Figure 3.1)." Thus the chemical processes are alleged to crate an excess of the carbonate ion, CO3, which is supposed to be a bottleneck to the absorption of CO2!
[This incorporates much of the following opinion of David Archer, a contributing author to the TAR, and while not W. S. Broecker, at least one of his co-authors. Elsewhere Archer has claimed,
["When you release a slug of new CO2 into the atmosphere, dissolution in the ocean gets rid of about three quarters of it, more or less, depending on how much is released. The rest has to await neutralization by reaction with CaCO3 or igneous rocks on land and in the ocean. These rock reactions also restore the pH of the ocean from the CO2 acid spike. My model indicates that about 7% of carbon released today will still be in the atmosphere in 100,000 years. I calculate a mean lifetime, from the sum of all the processes, of about 30,000 years. That's a deceptive number, because it is so strongly influenced by the immense longevity of that long tail. If one is forced to simplify reality into a single number for popular discussion, several hundred years is a sensible number to choose, because it tells three-quarters of the story, and the part of the story which applies to our own lifetimes."
[This is patent nonsense on two separate grounds. First according to the TAR, about 65% of the 730 GT of atmospheric CO2 is removed every year. That gives a mean residence time of 1.52 years, and a half-life of 0.65 years. Second, the dissolution of atmospheric CO2 in water is a physical process, not a chemical reaction. The ratio of carbonate, bicarbonate, carbon dioxide and carbonic acid, and hence the pH, will adjust thermodynamically until the CO2 concentration is on the solubility curve for CO2 in water. Archer and the TAR make the solubility depend not on just temperature and pressure, but also on the carbonate concentration or the pH. This would be a major new result in physics!
[Lastly, it is nonsense because the ocean is not stagnant as the ionic build-up conjecture suggests. The solubility pump loads CO2 in the cold waters, transports it deep and undersaturated at high pressure, returning it to the surface, heated and at atmospheric pressure, to be unloaded from saturated water. The water then circulates on the surface to cool and reabsorb CO2 from the air. The solubility pump depends on the solubility curve, and needs no modification. It accounts well for the Vostok record.]
Finally, you entirely ignore the atmospheric physics at play here. For the sake of entertainment, if the post-IR CO2 increases are anthropogenic, that does this change the fact that CO2 is indeed a very effective greenhouse gas?
[RSJ: If we assume the AGW conjecture is valid, then the AGW conjecture is valid. But, it is not. The adjectives "indeed" and "very effective" don't help. Anthropogenic CO2 is quite as effective as natural CO2, but not as effective as H2O, each of which overwhelm the ACO2 in quantity and effectiveness.]
We've pumped a great deal of it into the atmosphere in a very short amount of time, can you deny the physics of surface warming due to this increase?
[RSJ. Yes. It is not zero, just negligibly small. Watch this blog for additional explanations.]
I've met very few physicists who dared deny it. Your arguments are indeed well thought out, but poorly researched. I would be slightly more convinced if you took the time to do your homework on the subject and publish it.
[RSJ. It is published!]
Nevertheless, I'm circulating this with my colleagues just to see what they think. Thanks for entertaining this long-winded post, I'm neither a nay-sayer nor dooms-dayer, but I want to see all the data considered; something you've failed to do.
-Morgan
[RSJ: Where along the AGW route did Morgan fall off the AGW bandwagon? Does he deny that he bought into yet another doomsday scenario?]
Posted by morgan | May 8, 2007 9:15 PM
You posit that CO2 does NOT accumulate in the atmosphere. How do you explain atmospheric concentrations of CO2 increasing over the last 100 years?
[RSJ: A full response appears as a separate entry: On Why Co2 Is Known Not To Have Accumulated in the Atmosphere & What Is Happening with Co2 in the Modern Era, 6/11/07. ]
Posted by Myles Goodman | June 6, 2007 12:03 AM
You make the perceptive comment that the Mauna Loa Observatory results may be significantly impacted by the MLO's relative proximity to the equatorial ocean regions where the significant amount of the CO2 release is taking place. Since the amount of the release of CO2 will be controlled by both the CO2 content and the surface temperature of the ocean, the theory could be tested by looking for the impact of El Nino events on the MLO data. If MLO data is impacted by its proximity, the ocean surface temperature change during El Nino should be sufficient to show up in the MLO data.
[RSJ: Many articles are available on line that identify El Niño events in the Mauna Loa data. Keeling himself along with his fellow researchers, especially Bacastow, investigated the El Niño relationship to the Mauna Loa measurements.
[http://scrippsco2.ucsd.edu/program_history/charles_david_keeling_biography.html
[Note the References at
[http://cdiac.ornl.gov/trends/co2/sio-mlo.htm
[(2) Keeling's model of the Southern Oscillation Index included upwelling and releasing of about 0.6 GTons per year of CO2 from depths of 50 to 150 meters during periods when the SOI was positive, that is, during La Niña states. In the opposing, El Niño states, he claimed there was no appreciable CO2 flux. Keeling, C.D. and R. Revelle, Effects of El Nino/Southern Oscillation on the Atmospheric Content of Carbon Dioxide, Meteoritics, Vol. 20, No.2, Part 2, June 30, 1985. P. 437.
[(3) The great bulk of the 90 to 100 GTons of ocean CO2 outgassing appears to be associated with the thermohaline circulation, known under various names, including the ocean conveyor belt. By comparison, SOI effects may be minor eddy currents. Most of the outgassing occurs south and east of Hawaii, where it rises into the Hadley cells, then north and down into the trade winds feeding the Islands.
[Maps of air-sea CO2 flux, along with charts of the THC, strongly support the deep current theory. The maps, frequently attributed to Taro Takahashi, were prepared from pCO2 measurements, the surface wind, and the gas transfer velocity. IPCC, Fourth Assessment Report, p. 523, Figure 7.8. See also
[http://www.ldeo.columbia.edu/res/pi/CO2/carbondioxide/pages/air_sea_flux_rev1.html
[The following site provides a map of the Mean Annual Air-Sea Flux for 1995 for non-El Niño conditions! http://www.pmel.noaa.gov/pubs/outstand/feel2331/images/fig06.jpg
[linked from
[http://www.pmel.noaa.gov/pubs/outstand/feel2331/feel2331.shtml
[Somewhere on line you might find an animated, monthly Takahashi map.]
Posted by Tom Klein | June 24, 2007 2:13 PM
Thanks for the references. Keeling & Whorf paper indicates that they recorded the largest increase in CO2 content in the year 1998 - a very strong El Nino year - . The increase in that year, 2.87 ppmv, was more than double than the 46 year average of 1.4 ppmv. Monthly data also show that the largest year to year increases were in the months of August 1998 to about February 1999, the period of the strongest El Nino activity.
Tom Klein
RSJ: Keeling and Revelle discussed the effects of El Niño, including on the carbon cycle, in Effects of El Nino/Southern Oscillation on the Atmospheric Content of Carbon Dioxide, Meteoritics, Vol. 20, No.2, Part 2, June 30, 1985. P. 437. The IPCC Third Assessment Report restates much of what K&R said, but in the summary omitting the part about the CO2. See Climate Change 2001, Technical Summary of the Working Group I Report, p. 52. Climate Change 2001 discusses the El Niño/CO2 link in the context of CO2 variability, and notes the conincident "reduced upwelling of CO2-rich waters". Id., pp. 208-209. Then it concludes with the following observation:
[In any case, the slowdown (of the early 1990s) proved to be temporary, and the El Niño of 1998 was marked by the highest rate of CO2 increase on record, 6.0 PgC/yr. Id., p. 210.
[This is your citation, and it contradicts Keeling's earlier model.
[The correlation between El Niño events and atmospheric CO2 concentration was shattered. It might be repaired by examining the correlation between SOI and the CO2 concentration data a bit less processed than the Keeling curve. For more, see RSJ, Solar Wind, El Niño/Southern Oscillation, & Global Temperature: Events & Correlations, 7/6/07.
[Outgassing from the thermohaline circulation is a better model than the damaged El Niño model because it, too, accounts for the shape of CO2 concentration with temperature (the complement of the solubility curve) and it accounts for the one millennium lag. Moreover, the THC-CO2 model fits the Takahashi maps. ]
Posted by Tom Klein | June 25, 2007 3:59 PM
Dear Dr Glassman,
I was delighted to see your analysis here. As an electronics engineer I had been surprised by the EPICA Dome C data which also shows the same phenomena. CO2 lags behind Temperature. Naturally I visited Real Climate to understand why the ice cores are considered crucial to demonstrating the validity of MMGW theory. On their site they mention that CO2 forces temp in a positive feedback mechanism. I then went back to the EPICA Dome C data and found that the increase in temperature at the end of an interglacial is linear, not exponential as you would expect for a positive feedback mechanism. Furthermore, there is no point of inflection in the temperature curve at the time when the CO2 starts to rise. Basically this claim is untrue, as you have said.
Well done for sticking your neck out on this.
One worrying thing have noticed is that since the meaning of the ice core data has been disputed, it is very difficult to get hold of graphs on the internet with high resolution showing the EPICA Dome C data. They were common a while ago, with a good graph being given on Wikipedia. Now you can only find the graphs with a running averager applied to the data to remove the 800year lag and all the peculiar peaks and troughs ironed out. It is shocking that supposedly reputable scientists are prepared to lie to the public to support this highly dubious science.
Keep up the good work.
Regards, P. Hoy.
[RSJ: Interesting points about the shape of data under positive feedback.
[If there were a positive feedback in effect between a greenhouse gas and temperature, the growth of water vapor caused by hotter surface temperatures could have turned Earth into a suffocating hothouse long ago. Sufficient positive feedback will destroy the system or exhaust its dynamic range. The dynamic range limit for water vapor is the vaporization of the last of the liquid water. (Rev. 8/27/07.)
[The paucity of good data is a huge problem in this field. The global "Temperature anomaly", like all other data series, should be published without temporal smoothing. The Mauna Loa CO2 concentration since 1958 should be accompanied by a record of the local wind vector and time of day, assuming those data were recorded. From time to time, I have suspected that the Mauna Loa data are a composite including other sites, that it has been adjusted for El Niño events, or that it has been subjected to temporal smoothing. The data should be unambiguous about any such processing, and it should be published without it.]]
Posted by Phillip Hoy | July 9, 2007 9:31 AM
Hello Dr. Glassman
I just posted your Solar Wind article on Global Warming Skeptics homepage. It is an excerpt (The introduction section) The rest of your interesting paper is linked to your website blog.
I have already posted your paper: The acquittal of CO2 on the GWS website about 3 months ago and repeated on the homepage recently. It got 62 (the repost) reads meaning they went to your link to read the rest.
I also posted Gavin Schmidt's response too. It got 40 reads meaning they went to your link to read the rest.
I also posted The Acquittal of CO2 at an Atheist forum. A few months ago. There they called you names. I kept trying to get them to make a rebuttal against your paper. They kept resisting as they/ kept calling you unflattering names.
I posted it because there were a number of people who claimed to be a scientist who subscribe to the CO2 warming propaganda. I wanted to see if they with science backgrounds find problems with your paper. After a lot of prodding on my part to get these name callers to make a rebuttal. One person finally did and it was not a good one.
I was not impressed.
I will admit that I struggle to understand what you write because I lack a science background. However I do read a lot on the subject and now participate in posting a mix of articles on a Global Warming Skeptic website.
We accept that there has been some warming over the last 100 years. We just do NOT accept the idea that mankind's CO2 emissions adds a lot of warming effect to the atmosphere.
Thank you for your time and effort to post your papers for us to read and learn.
If you want I can provide the link to the ATHEIST forum. If you care to respond.
Cheers
[RSJ: 10/5/04. The promos are appreciated. Sorry I don't know enough about your sites yet to comment on them.
[At some risk of pedantry, but in the interest of promoting the precision science demands, allow me to reflect on the names "Global Warming Skeptics" and the Atheist forum.
[Global Warming is a fact not to be rationally denied. That climate on any scale is either warming or cooling is a tautology. It's a fractal-like property. At present, Earth is warming from the last Ice Age, from the last Glacial epoch, and from the Little Ice Age. The inevitable turn-around is not in sight. But to the discredit of the Consensus on climate, it shows no sign of accounting for this warming background before it computes what it considers presently to be an anomalous warming. What should be controversial is not Global Warming but Anthropogenic Global Warming.
[Why "Global Warming Skeptics"? To deny global warming is equivalent to denying ice ages.
[But skepticism is a quality of every good scientist. When the Consensus attacks skeptics as contrarians, it exposes a deficit in its science literacy. As a minimum, every member of the Consensus community should be skeptical, especially of its own results, fraught with uncertainty, lack of success, and want of forthrightness. As to the contrarian accusation, it is the rejection of the crowd and a widely shared belief. Science is neither a democracy nor a repository for belief systems.
[I was piqued by the thought of a peculiar interest an atheist forum might have in errors in the AGW conjecture. I do grant you the Consensus is an evangelical group for a new theology. The fruits of conversions in the new temple are the heady stuff of power and money.
[On the other hand, scientific models are, and indeed must be, secular (in the theological not chronological sense). That is not atheistic, and no legitimate model should provide any support for atheism.
[Won't you share the unworthy comment?
[If you have difficulty at some point in reading one of my papers, please post a pointed comment. Perhaps a clarification would get you back on track, and help others.]
Posted by Sunsettommy | July 19, 2007 8:35 PM
Consulting my old textbook in inorganic chemistry, written in 1978, I found some numbers which are quite illuminating.
Atmospheric CO2: 330 ppm by volume corresponding to 480 ppm by weight.
Oceanic CO2: 100 ppm by weight.
The bottom line:
2.4E15 kg CO2 in the atmosphere
1.4E17 kg CO2 in the oceans
5E14 kg CO2 biomass flux
2E13 kg CO2 added yearly from fossil fuels
So the rise from 330 to 380 ppm (+15% in 29 years) would be caused by man adding a fraction of a per cent to the total of the CO2-stream?
If man would burn fossil fuels at the rate of 1978 for 7000 years, the amount of CO2 in the CO2-stream would double. I can live with those numbers.
[RSJ: 10/29/07. Your numbers are part of a set that is a linchpin of the AGW theory: the residence time of anthropogenic CO2 in the atmosphere. For a brief discussion of the topic see On Why CO2 Is Known Not To Have Accumulated in the Atmosphere & What Is Happening with CO2 in the Modern Era in the Journal, and written before the new Fourth Assessment Report could be absorbed. This is a good opportunity to update that report a bit.
[First, a brief facts check is appropriate. Your CO2 concentrations of 330 ppmv and 380 ppmv fit the data in the IPCC Third Assessment Report (TAR) (Figure 3.2a, p. 201) and its Fourth Assessment Report (4AR) (Figure 2.3, p. 138).
[Apparently you did a multiplication by volume or weight in the reservoirs for the two stocks, and something similar for the two fluxes, too. Your set of four numbers in Petagrams (1 Pg = 10^15g) along with the usual government authorities, is in the table below.
Parameter
Ivarsson
TAR
4AR