Foreword: This is the final entry in a four part series by Fedinand Engelbeen. While the narrative is contrary to the views of many of our readers, it is within the framework of WUWT’s goal of providing discussion on the issues. You won’t find guest posts like this on RC, Climate Progress, Open Mind (Tamino), or Skeptical Science where a guest narrative contrary to the blog owner(s) view is not allowed, much less encouraged in a four part series.
That said, I expect this final entry to be quite contentious for two reasons. 1) The content itself. 2) The references to the work of Ernst Georg Beck, recently deceased.
As Engelbeen mentions below, this part was written weeks before, and readers should not get the impression that this is some sort of “hit piece” on him. Unfortunately, it simply worked out that the appearance of part 4 happens after his death, since I had been running each part about once a week. I had considered not running it, but I’m sure he would invite the discussion, and we’d have a lively debate. It is our loss that he will not be able to. For that reason, I’d appreciate readers maintaining a civil tone in comments. Moderators, don’t be shy about enforcing this. My thanks to Ferdinand Englebeen for his hard work in producing this four part series. – Anthony
About background levels, historical measurements and stomata proxies…
1. Where to measure? The concept of “background” CO2 levels.
Although there were already some hints of a “global” background CO2 level of around 300 ppmv in previous years, the concept was launched by C.D. Keeling in the fifties of last century, when he made several series of measurements in the USA. He found widely varying CO2 levels, sometimes in samples taken as short as 15 minutes from each other. He also noticed that values in widely different places, far away from each other, but taken in the afternoon, were much lower and much closer resembling each other. He thought that this was because in the afternoon, there was more turbulence and the production of CO2 by decaying vegetation and/or emissions was more readily mixed with the overlying air. Fortunately, from the first series on, he also measured 13C/12C ratios of the same samples, which did prove that the diurnal variation was from vegetation decay at night, while during the day photosynthesis at one side and turbulence at the other side increased the 13C/12C ratio back to maximum values.
Keeling’s first series of samples, taken at Big Sur State Park, showing the diurnal CO2 and d13C cycle, was published in http://www.icsu-scope.org/downloadpubs/scope13/chapter03.html , original data (of other series too) can be found in http://www.biokurs.de/treibhaus/literatur/keeling/Keeling_1955.doc :
Figure 1. Diurnal variation in the concentration and carbon isotopic ratio of atmospheric
CO2 in a coastal redwood forest of California, 18-19 May 1955, Big Sur St. Pk.
(Keeling, 1958)
Several others measured CO2 levels/d13C ratios of their own samples too. This happened at several places in Germany (Heidelberg, Schauinsland, Nord Rhine Westphalia). This confirmed that local production was the origin of the high CO2 levels. The smallest CO2/d13C variations were found in mountain ranges, deserts and on or near the oceans. The largest in forests, crop fields, urban neighborhoods and non-urban, but heavily industrialized neighborhoods. When the reciprocal of CO2 levels were plotted against d13C ratios, this showed a clear relationship between the two. Again from http://www.icsu-scope.org/downloadpubs/scope13/chapter03.html :
Figure 2. Relation between carbon isotope ratio and concentration of atmospheric CO2 in different air types from measurements summarized in Table 3.4
(Keeling, 1958, 1961: full squares; Esser, 1975: open circles; Freyer and Wiesberg, 1975,
Freyer, 1978c: open squares). All 13C measurements have not been corrected
for N2O contamination (Craig and Keeling, 1963), which is at the most in the area of + 0.6‰
The search for background places.
Keeling then sought for places on earth not (or not much) influenced by local production/uptake, thus far from forests, agriculture and/or urbanization. He had the opportunity to launch two continuous measurements: at Mauna Loa and at the South Pole. Later, other “baseline” stations were added, all together 10 from near the North Pole (Alert, NWT, Canada) to the South Pole, all of them working continuous nowadays under supervision of NOAA (previously under Scripps Institute), some 60 other places working under other organizations and many more working with regular flask sampling.
We are interested in CO2 levels in a certain year all over the globe and the trends of the CO2 levels over the years. So, here we are at the definition of the “background” level:
Yearly average data taken from places minimal influenced by vegetation and other natural and human sources are deemed “background”.
For convenience, the yearly average data from Mauna Loa are used as reference. One could use any baseline station as reference or the average of the stations, but as all base stations (and a lot of other stations, even Schauinsland, at 1,000 m altitude, midst the Black Forest, Germany) are within 5 ppmv of Mauna Loa, with near identical trends, and that station has the longest near-continuous CO2 record, Mauna Loa is used as “the” reference.
As the oceans represent about 70% of the earth’s surface, and all oceanic stations show near the same yearly averages and trends, already 70% of the atmosphere shows background behavior. This can be extended to near the total earth for the part above the inversion layer.
Measurements above the inversion layer.
Above land, diurnal variations are only seen up to 150 m (according to http://www.icsu-scope.org/downloadpubs/scope13/chapter03.html ).
Seasonal changes reduce with altitude. This is based on years of flights (1963-1979) in Scandinavia (see the previous reference) and between Scandinavia and California (http://dge.stanford.edu/SCOPE/SCOPE_16/SCOPE_16_1.4.1_Bishoff_113-116.pdf ), further confirmed by old and modern https://wiki.ucar.edu/display/acme/ACME flights in the USA and Australia (Tasmania). In the SH, the seasonal variation is much smaller and there is a high-altitude to lower altitude gradient, where the high altitude is 1 ppmv richer in CO2 than the lower altitude. This may be caused by the supply of extra CO2 from the NH via the southern branch of the Hadley cell to the upper troposphere in the SH.
From the previous references:
Figure 3. Amplitude and phase shift of seasonal variations in atmospheric CO2
at different altitudes, calculated from direct observations by harmonic analysis
(Bolin and Bischof, 1970)
From https://wiki.ucar.edu/display/acme/ACME :
Figure 4. Modern flight measurements in Colorado, CO2 levels below the inversion layerin forested valleys and above the inversion layer at different altitudes
As one can see, again the values above the inversion layer are near straight and agree within a few ppmv with the Mauna Loa data of the same date. Below the inversion layer, the morning values are 15-35 ppmv higher. In the afternoon, these may sink to background again.
If we take the 1000 m as the average upper level for the influence of local disturbances, that represents about 10% of the atmospheric mass. Thus the “background” level can be found at 70% of the earth’s air mass (oceans) + 90% of the remaining land surface (27%). That is in 97% of the global air mass. Only 3% of the global air mass contains not-well mixed amounts of CO2, which is only over land. These measured values show variations caused by seasonal changes (mainly in the NH) and a NH-SH lag. Yearly averages are within 5 ppmv:
Figure 5. Yearly average CO2 levels at different baseline stations plus a non-baseline station (Schauinsland, Germany, only values taken when above the inversion layer and with sufficient wind speed).
General conclusion:
Background CO2 levels can be found everywhere over the oceans and over land at 1000 m and higher altitudes (in high mountain ranges, this may be higher).
2. The historical data
2.1. The compilation by Ernst Beck.
Note: this comment was written weeks before we heard of the untimely death of Ernst Beck. While I feel very uncomfortable that this is published now, as he can’t react anymore on this comment, I think that one need to know the different viewpoints about the historical data, which is a matter of difference in opinion, and has nothing to do with what one may think about Ernst Beck as person.
What about the historical data? While I only can admire the tremendous amount of work that Ernst Beck has done, I don’t agree with his interpretation of the results. Not in light of the above findings of what one can see as “background” CO2 levels.
The historical measurements show huge differences from place to place, sometimes within one year, and extreme differences within a day or day to day or over the seasons for the same place. That there are huge differences between different places shows that one or more or all of these places are not measuring background CO2, but local CO2 levels, influenced by local and/or regional sources and sinks. This is clear, if one looks at the range of the results, often many hundreds of ppmv’s between the lowest and highest values. Modern measurements, sometimes interestingly done at the same places as the historical one’s, either don’t show such a wide range, and then can be deemed background for the modern ones and therefore the historical one’s must be inaccurate as method or there were problems with the handling or with the sampling. Others show huge variations also today, which means that neither the modern, nor the historical data are background.
But let us have a look at the compilation of historical CO2 measurements by Ernst Beck:
Figure 6. Compilation of historical data by Ernst Beck.
From: http://www.biomind.de/realCO2/realCO2-1.htm
Beck only gives the yearly and smoothed averages and the instrument error. That doesn’t say anything about the quality of the places where was measured, thus which of these measurements were “background” and which were not. One may be pretty sure that measuring midst of London, even in 1935, would give much higher (and fluctuating) CO2 levels than near the coast with seaside wind. Moreover, a peak of some 80 ppmv around 1942 is hardly possible, but removing such a peak in less than 10 years is physically impossible. The total amount of CO2 involved is comparable to burning down one third of all living vegetation on land and growing back in a few years time. The oceans are capable of having a burst of CO2 with a sudden decrease of pH, but simply can’t absorb that amount back in such a short time span, even if the pH would go up again (and what should cause such a massive change in pH?). Therefore I decided to look into more detail at the peak period in question, the years 1930-1950.
2.2. The minima, maxima and averages
Here is a plot of all available data for the period 1930-1950, as used by Ernst Beck (plus a few extra I did find in the literature). These can be found at his page of historical literature:
http://www.biomind.de/realCO2/historical.htm
Figure 7. Minima, maxima and averages of historical measurements in the period 1930-1950
Not all measurements were published in detail. Several authors did provide only an average, without any indication of number of samples, range or standard deviation. But for those where the range was given, the results are widely varying. What is obvious, is that where the range is small, in most cases the average of the measurements is around the ice core values (Law Dome in this case, the values of three cores, two of them with a resolution of 8 years and an accuracy of 1.2 ppmv, 1 sigma). That is especially the case for the period 1930-1935 where several measurements were performed during trips over the oceans. And even most of the worst performers show minima below the ice core values.
And as one can see, the “peak” around 1940-1942 is completely based on measurements at places which were heavily influenced by local/regional sources and sinks. That doesn’t say anything about the real background CO2 level of that period. Moreover, the fact that the average of measurements at one part of the world is 600 ppmv and at the other side of the globe it is 300 ppmv within the same year, shows that at least one of them must be at the wrong place.
2.3. The accuracy of some apparatus
Some of the measurements were done at interesting places: Point Barrow and Antarctica, where currently baseline stations are established. Unfortunately, for these measurements, the portable apparatus was as inaccurate as could be:
Barrow (1948) used the micro-Schollander apparatus, which was intended for measuring CO2 in exhaled air (some 20,000 ppmv!). Accuracy +/- 150 ppmv, accurate enough for exhaled air, but not really accurate to measure values of around 300 ppmv.
The same problem for Antarctica (1940-1941): Accuracy +/- 300 ppmv, moreover oxygen levels which were too low at high CO2 (1700 ppmv), which points to huge local contamination.
2.4. What caused the 1941 peak?
The 1941 peak is heavily influenced by two data series: Poona (India) and Giessen (Germany). With a few exceptions, the results of Poona should be discarded, as these were mostly performed within and below growing vegetation, which may be of interest for those who want to know the influence of CO2 on growth figures, heavily influenced by CO2 production from soil bacteria, but not really suitable to know the background CO2 levels of that time.
Giessen is a more interesting place, as the measurements were over a very long period (1.5 years), three samples a day over 4 heights were taken. And we have a modern CO2 measuring station now, only a few km from the original place, taking samples every 30 minutes. Thus let us see what the historical and modern CO2 levels at Giessen are, compared to baseline places:
Figure 8. Historical data of Giessen, during a few days of extra sampling to measure diurnal changes.
Figure 9. A few days in the modern summer life of CO2 at Linden-Giessen compared to the raw data from a few baseline stations for the same days.
Data for Linden-Giessen are from http://www.hlug.de
Baseline stations hourly average CO2 levels, derived from 10-second raw voltage samples, are from ftp://ftp.cmdl.noaa.gov/ccg/co2/in-situ/
These are all raw data, including all local outliers at Barrow, Mauna Loa, the South Pole and Giessen. It seems to me that it is rather problematic to figure out anything background-like from the data of Giessen, modern and historical alike. And I have the impression that Keeling made not such a bad choice by starting measurements at the South Pole and Mauna Loa, even if the latter is on an active volcano.
2.5. Estimation of the historical background CO2 levels.
Francis Massen and Ernst Beck used a method to estimate the background CO2 levels from noisy data, based on the fact that at high wind speeds, a better mixing of ground level CO2 with higher air masses is obtained (see http://www.biokurs.de/treibhaus/CO2_versus_windspeed-review-1-FM.pdf ). This works quite well, if you have a lot of data points with wind speeds above 4 m/s and a relative narrow range at high wind speeds. Here the “fingerlike” data range at high wind speed measured at Diekirch (small town in a shielded valley of Luxemburg):
Figure 10. CO2 levels vs. wind speed at Diekirch, Luxemburg.
Compare that to a similar plot of the historical data from Giessen:
Figure 11. Historical CO2 levels at Giessen vs. wind speed.
There are only 22 data points above 4 m/s, still a wide range (300 ppmv!) and no “finger” in the data at high wind speeds.
Further, the historical three samples of Giessen, taken in the morning, afternoon and evening already give a bias of some 40 ppmv (even the continuous modern sampling at Giessen shows a huge bias in averages). The afternoon measurements have a higher average than the morning and evening samples, which is contrary to almost all other measurements made in that period (and today): during daylight hours, photosynthesis lowers the CO2 levels, while at night under an inversion level, CO2 from soil respiration builds up to very high levels. And at the other end of the world (Iowa, USA) in 1940, CO2 levels of 265 ppmv were found over a maize field. Unfortunately, there are no measurements performed at “background” places in that period, except at Antarctica, which were far too inaccurate.
My impression is that the data of Giessen show too much variation and are too irregular, either by the (modified Pettenkofer) method, the sampling or the handling of the samples.
2.6. Comparing the historical peak around 1941 with other methods:
The ice core data of Law Dome show a small deviation around 1940, within the error estimate of the measurements. Any peak of 80 ppmv during years should be visible in the fastest accumulation cores (8 years averaging) as a peak of at least 10 ppmv around 1940, which is not the case (see Figure 7.).
Stomata data don’t show anything abnormal around 1940 (that is around 305 ppmv):
Figure 12. CO2 levels vs. stomata data calibration in the period 1900-1990.
From: http://igitur-archive.library.uu.nl/dissertations/2004-1214-121238/index.htm
And there is nothing special to see in the d13C levels of coralline sponges around 1940. Coralline sponges follow the 13C/12C ratios of CO2 in the upper ocean waters. Any burst and fall of CO2 in the atmosphere would show up in the d13 levels of the ocean mixed layer: either with a big drop if the extra CO2 was from vegetation, or with a small increase, if the extra CO2 was from the deep oceans. But that is not the case:
Figure 13. d13C levels of coralline sponges growing in the upper ocean layer.
2.7. Conclusion
Besides the quality of the measurements themselves, the biggest problem is that most of the data which show a peak around 1941 are taken at places which were completely unsuitable for background measurements. In that way these data are worthless for historical (and current) global background estimates. This is confirmed by other methods which indicate no peak values around 1941. As the minima may approach the real background CO2 level of that time, the fact that the ice core CO2 levels are above the minima is an indication that the ice core data are not far off reality.
3. About stomata data.
Stomata index (SI) is the ratio between the number of stomata openings to the total number of cells on leaves. This is a function of CO2 levels during the previous growing season (Tom van Hoof, personal communication). Thus that gives an impression of CO2 levels over time. As that is an indirect proxy of CO2 levels, one need calibration, which is done by comparing the SI of certain species over the past century with ice core and atmospheric CO2 measurements. So far, so good.
The main problem of the SI is the same as for many historical measurements: the vegetation of interest grows by definition on land, where average CO2 levels may vary within certain limits for one period of time, but there is no guarantee that these limits didn’t change over time: the MWP-LIA change might have been caused in part by changes in the Gulf Stream away from NW Europe, this bringing less warm wet air over land, even changing the main wind direction from SW to E. That may have introduced profound changes in type of vegetation, soil erosion, etc., including changes in average CO2 levels near ground over land.
Further, land use changes around several of the main places of sampling might have been enormous: from wetlands and water to polders and agriculture, deforestation and reforestation, all in the main wind direction, as all happened in The Netherlands over a full millennium.
Conclusion:
Stomata index data may be useful as a first approximation, but one shouldn’t take the historical levels as very reliable, because of a lack of knowledge of several basic circumstances which may have influenced the local/regional historical CO2 levels and thus the SI data.
@Engelbeen
Using Modtran (default settings except altitude and CO2) looking down from 1km and 1ppm CO2 examining the IR spectrum output there is no absorption at 15um which is what one would expect. Changing the CO2 to 999,999ppm there is STILL no dent at 15um where in reality this would easily result in complete extinction of 15um.
The indisputable result is that Modtran is completely broken at least when it comes to looking down from 1km. It’s accuracy elsewhere must therefore be questioned.
Try again.
Dave Springer says:
September 30, 2010 at 6:08 am
Modtran with default settings (375ppm CO2) except for altitude changed to 1km gives me 406.316 w/m^2 (not the number you claimed – you must have screwed up the settings somehow).
You probably didn’t change the Locality to “1976 U.S. Standard Atmosphere”, as the initial choice is for the tropics only.
I changed the CO2 setting to 999999 (pure CO2) just to see what would happen and got a reading of 401.606 w/m.
Then I changed it to 1ppm and got a reading of 406.944 w/m
Then I changed the altitude to 20km and 999999ppm CO2 and got a reading of 201.243.
The interesting thing about the last reading is that the shoulders broadened to extinguish 400 wavenumbers.
Going from 1ppm to 999,000ppm CO2 in the first kilometer of air changed the output only 5 w/m^2!
I don’t trust the physics and/or the programming behind the Modtran model farther than I can throw it. It gives ridiculous results. In engineering (I understand you’re an engineer) we call what I did with the model a “sanity test”. It didn’t pass.
The calculation includes water vapour and a lot of other GHGs in the “standard” atmosphere, which doesn’t exist in a Venusian atmosphere. But the calculation still may be right if the extra CO2 absorption is only for the 15 micron band and only for the first km…
And it seems that there is something wrong with the plotting. If you look at the detailed calculations (“Save output for later retrieval”, after submissions ask for “view the whole output file”), there are a lot of zero’s in the transmissions around 15 micron, but the plot only shows them from 10 km height, but then even with only 1 ppmv CO2.
@Engelbeen
Modtran breaks when CO2 gets too high. When confined to commonly accepted CO2 maximums and temperature in the geologic column (such as the Eocene Optimum) it matches up pretty good. I withdraw any qualms I had about it.
At 2800ppm (10x pre-industrial, Eocene typical) with all other gases including water vapor set to zero, 76 US Std Atmos., no clouds/rain, 70K looking down, surface temp rise is 3.8K which is just about how much warmer it was in the Eocene Optimum.
Everything else I tried (within reasonable bounds) was exactly what I expected.
I never really had any substantial objections to observed CO2 in recent history being due to human emission nor to the greenhouse effect of CO2 in isolation. I just don’t think it can be proven any more than it can be proven that all swans are white. It can only be disproven. I don’t think I’d agree it’s ready to be a theory of CO2 quite yet but I’d call it the best explanation and best explanations are what science is all about.
The bone of contention for me are twofold:
1) The only way to match up the geologic column CO2/temperature relationship to what Modtran indicates is with zero feedback.
2) CO2 of 10x pre-industrial level is a great net benefit to the biosphere. The earth was green from pole to pole in the Eocene Optimum.
3) In the current climate with CO2 at geologically low concentrations we are at extreme risk of runaway cooling and at no risk of runaway warming. The former happens frequently (every 120,000 years currently) while the latter never happened.
What we should be doing, if we should be doing anything, is figuring out how to get Eocene-level CO2 concentration restored and hope that it ends the ice age. I’m not sure there’s enough carbon stored in economically recoverable fossil fuels to accomplish that. There certainly isn’t much more than enough.
@Engelbeen
Plus I’ve never had any objections to CO2 being well mixed in the atmosphere. Countless measurements confirm it. Those measurements include my own as I happen to have an infrared CO2 meter I got for a song at a surplus sale. It’s a Honeywell duct-mount control made to turn on ventilators in occupied buildings when too many people have been breathing the same air too long. No display on it but any old digital voltmeter can get a precise measure from the scaled voltage or current outputs. Nowhere outdoors that I personally took a precise CO2 reading differed substantially from Mauna Loa – it was well mixed even at waist level.
Hi Dave Springer. I’ve followed your fascinating exchange with Dr. Engelbeen about CO2 modelling. You say
At 2800ppm (10x pre-industrial, Eocene typical) surface temp rise is 3.8K which is just about how much warmer it was in the Eocene Optimum…
…The bone of contention for me are twofold:
1) The only way to match up the geologic column CO2/temperature relationship to what Modtran indicates is with zero feedback.
2) CO2 of 10x pre-industrial level is a great net benefit to the biosphere. The earth was green from pole to pole in the Eocene Optimum.
3) In the current climate with CO2 at geologically low concentrations we are at extreme risk of runaway cooling and at no risk of runaway warming. The former happens frequently (every 120,000 years currently) while the latter never happened.
What we should be doing, if we should be doing anything, is figuring out how to get Eocene-level CO2 concentration restored and hope that it ends the ice age. I’m not sure there’s enough carbon stored in economically recoverable fossil fuels to accomplish that. There certainly isn’t much more than enough.
First, that’s three-, not two-fold 🙂
Second, while I agree with the thrust of everything you say here up to that point, I don’t think that it would be healthy to return to Eocene levels of CO2. Everything has a toxicity level. You can even die from ‘water poisoning’ (by drinking too much water). So we must determine what the “optimum” level of CO2 is from a biosphere and human interest perspective before saying what level we’d like to return to.
Plants love these megadose quantities of CO2. Animals don’t. The mildest toxic effects of CO2 in animals occurs somewhere between 1500 and 2000 ppmv. Where, exactly, is a bit of a judgement call. For humans, mild headache and dizziness begin somewhere in this range. 10x pre-industrial levels would be a bit higher, in a range in which toxicity effects are not controversial, though not fatal. It would probably put stress on the fauna, however.
I guess it depends on what one means by “optimal”, but my money is on about 4x pre-industrial levels, somewhere around 1000 to 1200 ppm. I think if instantly exposed to that level one might find it a bit “stuffy” (it would be comparable to sitting in a crowded theatre). However, I also believe that over the long term one simply adapts; we are pretty plastic with regard to lung capacity, efficiency of oxygen transport and alveoli activity. Given a very slow transition to these levels one would not even notice, and the biosphere would adjust. I confess that I have no evidence for my statements above, they just “make sense” from experience and general knowledge about biology. I’d be interested in what is known from (if any) studies of long-term exposure to 1000 ppm-range levels of CO2 does for various animal species.
In any case, I don’t think we could burn fossil fuels fast enough to exceed 1000 ppmv. The response we’re already seeing in the vegetative world is enough to indicate that any elevation of CO2 is going to be met with massive increase in photosynthesis across the biosphere. We’d have to accelerate far beyond what even the alarmists are projecting to outrun the world’s vegetation.
But if we start to approach 1000 ppm with little slowing in the rate of increase, I confess I will begin to feel a little nervous. Up til then I’m happy as a lark to see the CO2 increase, and anyone who loves the starving millions, and the biosphere, should be happy too. I don’t think there’s anything radical about these thoughts whatsoever.
Dave Springer says:
September 30, 2010 at 12:42 pm
I think we do agree on all three of your points…
As I wrote before:
Thus in my opinion (and of many others), while there may be an influence of CO2 on temperature, it is very modest and (far) below what the climate models (and the IPCC) “project”…
But I think that there is little doubt left that humans are responsible for the recent increase in CO2. After many (years) discussions about this topic, I am still waiting for an alternative explanation that fits all observations…
Dave Springer says:
September 30, 2010 at 1:04 pm
Nowhere outdoors that I personally took a precise CO2 reading differed substantially from Mauna Loa – it was well mixed even at waist level.
If you can lend it to TonyB, he wants to measure the CO2 levels personally at Mauna Loa…
Ferdinand Engelbeen wrote:
The historical and current CO2 measurements over the oceans, coastal and at high altitude (mountains) or latitude (South Pole) all show the same CO2 levels and trends, plus a recurrent seasonal cycle and a NH-SH lag. The degassing/absorption of CO2 by the oceans is huge, but spread over a year and with sufficient wind speed fast mixed in, so that the change of levels within a day is even unmeasurable in the trend. Even the sporadic volcanic outgassing with downslope wind at Mauna Loa only disturbs the data with not more than 4 ppmv, this is included in the raw data of figure 9…
I would like to rebut this with this quote from Timothy Casey B.Sc. (Hons.) Consulting Geologist, this is a quote from a paper he uploaded to the net.
Uploaded ISO:2009-Oct-25
Revised ISO:2010-May-17
1.1 The Importance of CO2 in Volcanic Emissions
The importance of juvenile (erupted and passively emitted) volcanic CO2 is due to the fact that carbon, and particularly carbon dioxide has a strong presence in mantle fluids, so much so that it is a more abundant volcanic gas than SO2 (Wilson, p. 181; Perfit et al., 1980). According to Symonds et al. (1994) CO2 is the second most abundantly emitted volcanic gas next to steam. Although you might imagine that there is no air in the mantle, the chemical conditions favour oxidation, and shortages of oxygen ions are rare enough to ensure a strong presence of CO2 (Schneider & Eggler, 1986). Oxidation of subducted carbon sources such as kerogen, coal, petroleum, oil shales, carbonaceous shales, carbonates, etc. into CO2 and H2O makes volcanic CO2 quite variable in back arc and continental margin volcanoes, where these volatile gases can be surprisingly abundant (eg. Vulcano & Mount Etna). Subduction isn’t the only way CO2 enters magma. At continental rift zones, where an entire continent is being pulled apart by divergent mantle convection, magma rising to fill the rift is enriched in CO2 from deep mantle sources (Wilson, 1989, p. 333). Oldoinyo Lengai is an example of a continental rift zone volcano, which has above average CO2 outgassing at 2.64 megatons of CO2 or 720 KtC per annum (Koepenick et al., 1996).
If volcanoes produce more CO2 than industry when they are not erupting, then variations in volcanic activity may go a long way towards explaining the present rise in CO2.
1.2 The Location of Co2 Monitoring Station in regions enriched by volcanic CO2
Volcanic CO2 emission raises some serious doubts concerning the anthropogenic origins of the rising atmospheric CO2 trend. In fact, the location of key CO2 measuring stations (Keeling et al., 2005; Monroe, 2007) in the vicinity of volcanoes and other CO2 sources may well result in the measurement of magmatic CO2 rather than a representative sample of the Troposphere. For example, Cape Kumukahi is located in a volcanically active province in Eastern Hawaii, while Mauna Loa Observatory is on Mauna Loa, an active volcano – both observatories within 50km of the highly active Kilauea and its permanent 3.2 MtCO2pa plume. Samoa is within 50 km of the active volcanoes Savai’i and/or Upolo, while Kermandec Island observatory is located within 10 km of the active Raoul Island volcano.
Observatories located within active volcanic provinces are not the only problem. There is also the problem of pressure systems carrying volcanic plumes several hundred kilometers to station locations. For example, the observatory in New Zealand, located somewhere along the 41st parallel, is within 250 km of Tanaki and the entire North Island active volcanic province. Low pressure system centres approaching and high pressure system centres departing the Cook Strait will displace volcanic plumes from the North island to the South Island.
Another class of problem for monitoring stations plagues “Christmas Island”, which is actually Kiribati Island (02º00’N, 157º20’W) where the Clipperton Fracture Zone (Taylor, 2006) crosses the Christmas Ridge and is nowhere near Christmas Island (10º29’S, 105º38’E; located on the other side of Australia, 10,000 km due west of Kiribati). Christmas Ridge is formed in a concentration of Pacific Seamounts. Extraordinary numbers of seamounts are volcanically active (Hillier & Watts, 2007). Moreover, active fracture zones also offer a preferred escape route for magmatic CO2, as this CO2 also finds its way into aquifers (eg. Giggenbach et al., 1991), which can be cut by fracture zones that consequently provide a path to the surface (Morner & Etiope, 2002). This may raise doubts concerning measurements taken at the La Jolla observatory, which is located near the focal point of a radial fault zone extending seaward from the San Andreas Fault (see imagery sourced to SIO, NOAA, USN, NGA, & GEBCO by Europa Technologies & Inegi, for Google Earth).
Amundsen Scott South Pole Station appears to be well separated by 1300 km from the volcanic lineation extending along Antarctica’s Pacific Coast (From the Ross Shelf to the Antarctic Peninsula), However, Antarctic volcanoes are not nearly as well mapped as those in more populated regions, such as Japan. In any case, the strong circumpolar winds that delay mixing will inevitably concentrate Antarctica’s volcanic CO2 emissions over the Antarctic continent, including Amundsen Station. The same potential problem exists with the observatory at Alert in Northern Canada, because it is located inside the circumpolar wind zone along with the Arctic Rift and thousands of venting seamounts along key parts of the Northwest Passage.
That leaves us with Point Barrow, arguably the only CO2 monitoring station whose CO2 measurements are unlikely to be influenced by magmatic gas plumes. However, the Canada Basin, extending seaward from Point Barrow, is also referred to as “the Hidden Ocean” because of poor access, which consequently leaves us with very little information about the sea floor in this region. The high probability of active seamounts in the vicinity of Point Barrow has not been ruled out, and in view of the fact that the other observatories probably experience significant skew due to magmatic CO2, it would not be unreasonable to remain skeptical until this possibility has been ruled out.
This question of volcanic skew in CO2 measurements has been raised a number of times, in addition to other more serious allegations (Bacastow, 1981; Jaworowski et al., 1992; Segalstad, 1996).
2.0 Calculated Estimates: Glorified Guesswork
The estimation of worldwide volcanic CO2 emission is undermined by a severe shortage of data. To make matters worse, the reported output of any individual volcano is itself an estimate based on limited rather than complete measurement. One may reasonably assume that in each case, such estimates are based on a representative and statistically significant quantity of empirical measurements. Then we read statements, such as this one courtesy of the USGS (2010):
Scientists have calculated that volcanoes emit between about 130-230 million tonnes (145-255 million tons) of CO2 into the atmosphere every year (Gerlach, 1991). This estimate includes both subaerial and submarine volcanoes, about in equal amounts.
In point of fact, the total worldwide estimate of roughly 55 MtCpa is by one researcher, rather than “scientists” in general. More importantly, this estimate by Gerlach (1991) is based on emission measurements taken from only seven subaerial volcanoes and three hydrothermal vent sites. Yet the USGS glibly claims that Gerlach’s estimate includes both subaerial and submarine volcanoes in roughly equal amounts. Given the more than 3 million volcanoes worldwide indicated by the work of Hillier & Watts (2007), one might be prone to wonder about the statistical significance of Gerlach’s seven subaerial volcanoes and three hydrothermal vent sites. If the statement of the USGS concerning volcanic CO2 is any indication of the reliability of expert consensus, it would seem that verifiable facts are eminently more trustworthy than professional opinion.
This is not an isolated case. Kerrick (2001) takes a grand total of 19 subaerial volcanoes, which on p. 568 is described as only 10% of “more than 100 subaerial volcanoes”. It is interesting to observe that Kerrick (2001) leaves out some of the more notable volcanoes (eg. Tambora, Krakatoa, Mauna Loa, Pinatubo, El Chichon, Katmai, Vesuvius, Agung, Toba, etc.). Nevertheless, despite these omissions Kerrick calculates 2.0-2.5 x 1012 mol of annual CO2 emissions from all subaerial volcanoes, which is understated on the assumption that the sample is from the most active volcanic demographic. This is in spite of the fact that eight of the world’s ten most active volcanoes are omitted from Kerrick’s study (Klyuchevskoy Karymsky, Shishaldin, Colima, Soufriere Hills, Pacaya, Santa Maria, Guagua Pichincha, & Mount Mayon). At 44.01g/mol, 2.0-2.5 x 1012 mol of CO2 amounts to a total of 24-30 MtCpa – less than 0.05% of total industrial emissions (7.8 GtCpa according to IPCC, 2007). My main criticism of Kerrick’s guess is that it putatively covers only 10% of a highly variable phenomenon on land, and with the cursory dismissal of mid oceanic ridge emissions, ignores all other forms of submarine volcanism altogether. If we take the Smithsonian Institute’s list of more than 1000 potentially active subaerial volcanoes worldwide, Kerrick’s 10% is reduced to 1-3%.
According to Batiza (1982), Pacific mid-plate seamounts number between 22,000 and 55,000, of which 2,000 are active volcanoes. However, none of the more than 2,000 active submarine volcanoes have been discussed in Kerrick (2001). Furthermore, Kerrick (2001) justifies the omission of mid oceanic ridge emissions by claiming that mid oceanic ridges discharge less CO2 than is consumed by mid oceanic ridge hydrothermal carbonate systems. In point of fact, CO2 escapes carbonate formation in these hydrothermal vent systems in such quantities that, under special conditions, it accumulates in submarine lakes of liquid CO2 (Sakai, 1990; Lupton et al., 2006; Inagaki et al., 2006). Although these lakes are prevented from escaping directly to the surface or into solution in the ocean, there is nothing to prevent superheated CO2 that fails to condense from dissolving into the seawater or otherwise making its way to the surface. It is a fact that a significant amount of mid oceanic ridge emissions are not sequestered by hydrothermal processes; a fact which is neglected by Kerrick (2001), who contends that mid oceanic ridges may be a net sink for CO2. This may well sound reasonable except for the rather small detail that seawater in the vicinity of hydrothermal vent systems is saturated with CO2 (Sakai, 1990) and as seawater elsewhere is not saturated with CO2, it stands to reason that this saturation is sourced to the hydrothermal vent system. If the vent system consumed more CO2 than it emitted, the seawater in the vicinity of hydrothermal vent systems would be CO2 depleted.
Morner & Etiope (2002) published a somewhat more representative estimate of subaerial volcanogenic CO2 output based on a more comprehensive selection and found as a bare minimum that subaerial volcanogenic CO2 emission is on the order of 163MtCpa. Morner & Etiope (2002) also provide a much better explanation of how CO2 is cycled through the mantle and the lithosphere. However, this still does not account for active volcanic emissions and remains vulnerable to eruptive variability. Based on data reproduced in Shinohara (2008), there were on average about five subaerial volcanic eruptions every year producing an average of 300KtSpa (kilotons of sulphur per year) from 1979-1989. Shinohara (2008) also presents molar ratios of CO2, SO2, & H2S from which, via the same academic daring as Gerlach (1991) and Kerrick (2001), we might derive an average ratio of 3.673 mol carbon for every mol of sulphur in gaseous volcanic emissions. That would loosely translate to 1.376KtC for every 1.000KtS. This gives us a figure of around 2MtCpa for minor volcanic activity based on SO2 emission events reported in Shinohara (2008). However, applying the same statistical assumption to some of the more notable eruptions of recent history, contrasted with one or two slightly older examples, gives us the following estimates:
Year Volcano Mean Sulphurous Output Source Est. Carbon output during year(s) of eruption
1883AD Krakatoa 38 MtSO2pa Shinohara (2008) 26.14 MtCpa
1815AD Tambora 70 MtSO2pa Shinohara (2008) 48.16 MtCpa
1783AD Laki 130 MtSO2pa Shinohara (2008) 89.44 MtCpa
1600AD Huaynaputina 48 MtSO2pa Shinohara (2008) 33.02 MtCpa
1452AD Kuwae 150 MtH2SO4pa Witter & Self (2007) 67.40 MtCpa
934AD Eldja 110 MtSO2 Shinohara (2008) 75.68 MtCpa
1645BC Minoa 125 MtSO2pa Shinohara (2008) 86.00 MtCpa
circa 71,000BP Toba 1100 MtH2SO4pa Zielenski et al. (1996) 494.24 MtCpa
Notice how all but one of the individual annual volcanogenic carbon outputs, estimated above, dwarf the global subaerial volcanogenic carbon outputs estimated by both Gerlach (1991) & Kerrick (2001). Even the Morner & Etiope (2002) subaerial estimate (163 MtCpa) is shaken by most of these figures and dwarfed by one. If this is not enough evidence of just how unreliable volcanic emission estimates can be, let us take a closer look at my 89 MtCpa estimate for the 1783AD Laki eruption. Consider the difference it makes if, instead of using the average ratio by weight for carbon and sulphur emissions I derived from Shinohara (2008), we take the ratio we use for the Laki estimate from more direct observations. Agustsdottir & Brantley (1994) studied emissions from Grimsvotn, from which Laki extends as a fissure, and found that Grimsvotn outgasses 53 KtCpa for 5.3 KtSpa. In other words, the weight of carbon emitted at Grimsvotn is ten times that of the sulphur emitted there. This would extend to Laki, which shares the same source, and is described by Agustsdottir & Brantley (1994) as a fairly stable ratio. By this ratio, Laki’s 130 Mt of sulphur dioxide in 1783AD translates to an emission of 650 MtCpa that year. This demonstrates just how much uncertainty is involved when trying to audit the volcanic contribution to the “carbon budget”.
As you can see, volcanic systems are diverse and unpredictable. They cannot be statistically second-guessed for the same reason that lottery numbers cannot be statistically second-guessed. This in itself raises serious doubt concerning the reliability of volcanic carbon dioxide emission estimates. This is especially problematic when significant elements of the estimates, such as passive submarine volcanic emission, all active volcanic emission, and at least 96% of passive subaerial emissions, are based on statistical assumptions rather than on any actual measurement.
R. Craigen says:
September 30, 2010 at 2:44 pm
OSHA safe exposure limit is 5000ppm in the workplace. Prolonged exposure to 10,000ppm causes mild drowsiness. Headache begins at 30,000ppm but breathing air at that level has been tolerated for at least as long as one month. The body becomes acclimated to it.
http://en.wikipedia.org/wiki/Carbon_dioxide#Toxicity
Atmospheric CO2 of 2000ppm is on the low side of the geologic norm. It ranges up to about 5000ppm. It is only during ice ages that it drops as low as it is today (the earth is currently in an ice age). Given that the vast majority of time since terrestrial forms of life appeared (air breathing plants and animals have been around for about 500 million years) had atmospheric CO2 at 10x-20x current levels it is reasonable to presume this is what evolution optimized terrestrial life around.
Thanks for the reference Dave, I defer to you. Apparently the EPA hasn’t yet pushed an alarmist slant on Wikipedia; now that’s a refreshing surprise. 🙂 I was working by memory from a technical source; it’s entirely possible that in my head I slipped an order of magnitude.
In any case I seriously doubt that we could reach 1000 ppmv with anyone’s projected use of fossil fuels — partly because of Henry’s law, which tell us that 90 or 95% of it will go into the ocean, and partly because of the known response of plant life.
I’m sure animals can adapt to higher CO2 levels, but I believe there are two kinds of adaptation that are relevant. First, the adaptation of a single individual when exposed to CO2. Over a period of months I think the blood oxygen transport, chest capacity and surface area of the alveoli would adjust to the oxygen content, which would probably slightly extend the upper range of the individual’s tolerance of the gas.
Further, natural selection, over several generations, would help entire species adapt to major changes in CO2 beyond the tolerable range. The key here is that animals need merely to be able to “tolerate” CO2, we do not rely on it directly for life, as plants do. This is why we see that plants, even millions of years after the higher CO2 range, thrive when concentrations are returned to those levels. They evolved in a high-CO2 environment, and while they can adapt to lower levels, they cannot eliminate their CO2 dependence without transforming into something else altogether.
Thus animals can adapt over time to different levels and probably do as well at 5000 ppm as at 100 ppm, but plants cannot because of their fundamental dependence on the nutrient.
Keith D says:
October 1, 2010 at 3:45 pm
I don’t differ in opinion about 1.1, except to point to the d13C level of most of the volcanic outgassing: mostly around zero per mil, positive if the origin is carbonate subduction and slightly negative it is deep core CO2. I have only found one volcanic outgassing which was quite low at -7 per mil. Thus if volcanic outgassing was the cause of (most of) the increase, that would increase the d13C levels of the atmosphere (currently at -8 per mil). But we see a year by year decrease.
Adn I don’t want to discuss chapter 2: The total amounts of volcanic outgassing are indeed only best guesses, but if the 30% increase was the result of more outgassing, there should be an increasing tectonic activity, which isn’t observed. Further the largest eruption of the previous century, the Pinatubo eruption, caused a drop in CO2 increase, as the resulting temperature drop caused by the volcanic ash and sulfates did give more uptake of CO2 than the Pinatubo emitted.
But let us have a look at chapter 1.2:
1.2 The Location of Co2 Monitoring Station in regions enriched by volcanic CO2
Volcanic CO2 emission raises some serious doubts concerning the anthropogenic origins of the rising atmospheric CO2 trend. In fact, the location of key CO2 measuring stations (Keeling et al., 2005; Monroe, 2007) in the vicinity of volcanoes and other CO2 sources may well result in the measurement of magmatic CO2 rather than a representative sample of the Troposphere. For example, Cape Kumukahi is located in a volcanically active province in Eastern Hawaii, while Mauna Loa Observatory is on Mauna Loa, an active volcano – both observatories within 50km of the highly active Kilauea and its permanent 3.2 MtCO2pa plume. Samoa is within 50 km of the active volcanoes Savai’i and/or Upolo, while Kermandec Island observatory is located within 10 km of the active Raoul Island volcano.
Observatories located within active volcanic provinces are not the only problem. There is also the problem of pressure systems carrying volcanic plumes several hundred kilometers to station locations. For example, the observatory in New Zealand, located somewhere along the 41st parallel, is within 250 km of Tanaki and the entire North Island active volcanic province. Low pressure system centres approaching and high pressure system centres departing the Cook Strait will displace volcanic plumes from the North island to the South Island.
Another class of problem for monitoring stations plagues “Christmas Island”, which is actually Kiribati Island (02º00′N, 157º20′W) where the Clipperton Fracture Zone (Taylor, 2006) crosses the Christmas Ridge and is nowhere near Christmas Island (10º29′S, 105º38′E; located on the other side of Australia, 10,000 km due west of Kiribati). Christmas Ridge is formed in a concentration of Pacific Seamounts. Extraordinary numbers of seamounts are volcanically active (Hillier & Watts, 2007). Moreover, active fracture zones also offer a preferred escape route for magmatic CO2, as this CO2 also finds its way into aquifers (eg. Giggenbach et al., 1991), which can be cut by fracture zones that consequently provide a path to the surface (Morner & Etiope, 2002). This may raise doubts concerning measurements taken at the La Jolla observatory, which is located near the focal point of a radial fault zone extending seaward from the San Andreas Fault (see imagery sourced to SIO, NOAA, USN, NGA, & GEBCO by Europa Technologies & Inegi, for Google Earth).
Amundsen Scott South Pole Station appears to be well separated by 1300 km from the volcanic lineation extending along Antarctica’s Pacific Coast (From the Ross Shelf to the Antarctic Peninsula), However, Antarctic volcanoes are not nearly as well mapped as those in more populated regions, such as Japan. In any case, the strong circumpolar winds that delay mixing will inevitably concentrate Antarctica’s volcanic CO2 emissions over the Antarctic continent, including Amundsen Station. The same potential problem exists with the observatory at Alert in Northern Canada, because it is located inside the circumpolar wind zone along with the Arctic Rift and thousands of venting seamounts along key parts of the Northwest Passage.
That leaves us with Point Barrow, arguably the only CO2 monitoring station whose CO2 measurements are unlikely to be influenced by magmatic gas plumes. However, the Canada Basin, extending seaward from Point Barrow, is also referred to as “the Hidden Ocean” because of poor access, which consequently leaves us with very little information about the sea floor in this region. The high probability of active seamounts in the vicinity of Point Barrow has not been ruled out, and in view of the fact that the other observatories probably experience significant skew due to magmatic CO2, it would not be unreasonable to remain skeptical until this possibility has been ruled out.
This question of volcanic skew in CO2 measurements has been raised a number of times, in addition to other more serious allegations (Bacastow, 1981; Jaworowski et al., 1992; Segalstad, 1996).
While Mauna Loa is on an active volcano, the influence of the volcanic vents is easely recognised:
– the values are slightly higher (up to 4 ppmv) and very irregular.
– only with downslope wind from certain directions.
The opposite conditions happen sometimes in the afternoon, with upwind conditions: CO2 levels are slightly depleted due to vegetation in the valleys (down with about 4 ppmv)
Both conditions are recognised and the hourly average data are flagged and not used for daily, monthly and yearly averages.
The data used are mainly from trade winds blowing over halve the Pacific Ocean. These show not more variability that a few tenths of a ppmv over a day…
See the hourly average raw data, including all outliers and the selected daily and monthly data from Mauna Loa and the South Pole:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/co2_mlo_spo_raw_select_2008.jpg
The same for data from New Zealand: these are only taken into account if the wind is from the south, as that is only ocean influenced over a very long distance…
Further, there are 10 “baseline” stations today, but some 70 more which measure CO2 in pristine areas, plus regular flight measurements and ship’s measurements. All show values close to the Mauna Loa values if averaged over a year. And all show the same trend over the years, even the 400+ stations which are intented to measure CO2 fluxes over land in/out vegetation and human sources show the same trend.
See the carbon tracker:
http://www.esrl.noaa.gov/gmd/ccgg/iadv/
It would be quite strange that the outgassing of volcanoes and/or faults would be quite identical all over the world… Thus there is a common, worldwide source at work.
Last but not least: the increase in the atmosphere is about 50% of the human emissions over the past 110 years. If volcanoes were the main cause of the increase, that would show over 100% increase, higher than the emissions.
From both sides of this idiotic discussion it is always the same… your words get twisted anyway… I explained how stomata response works to Ferdinand, but anyways the authors think they studied this proxy enough to be able to make their own good scientific judgement… Funny part of the stomata data is that neither the deniers as the pro-global warming activists really like it… this is how SCIENCE works… both camps claim it is purely coincidental that the dutch data match the USA data and that both curves match earlier ice core data. Mind you that fact that the wiggle around 1200AD is found in the Netherlands, in the USA and on Antarctica is already evidence enough that this is not a local signal…. but anyways, I won’t even try anymore to have an unbiased climate discussion with both sides, fruitless anyways
Tom says:”both camps claim it is purely coincidental that the dutch data match the USA data and that both curves match earlier ice core data. Mind you that fact that the wiggle around 1200AD is found in the Netherlands, in the USA and on Antarctica is already evidence enough that this is not a local signal…. ”
I am dutch living in South Africa,
skeptic scientist, e.g.
http://letterdash.com/HenryP/more-carbon-dioxide-is-ok-ok
I am trying to figure out what you mean here Tom?
Tom van Hoof says:
October 4, 2010 at 5:37 am
“From both sides of this idiotic discussion it is always the same… your words get twisted anyway… I explained how stomata response works to Ferdinand, but anyways the authors think they studied this proxy enough to be able to make their own good scientific judgement…”
As our previous discussion was quite clear: stomata data are a good proxy, and quite reliable. One point is that it responds to local/regional CO2 levels which may have a relative stable bias in the current period, compared to “background” CO2 levels, thus can be eliminated by calibration with ice cores and direct measurements.
The unresolved question in our discussion is that one doesn’t know how the bias changed over longer periods. As the landscape in The Netherlands changed tremendously over the centuries, how do we know what effect that had for CO2 levels in St. Odiliënberg? Further, temperature and current changes over the oceans (MWP-LIA) may give quite huge changes in main wind direction over land. How does that change CO2 levels locally more inland?
The SI data from Jay Bath (Western USA) and The Netherlands (and ice core CO2 from Antarctica) are similar around the MWP-LIA transition, but Jay Bath shows much more variability than The Netherlands, thus one of both is less backgound (usually the most variable…).
Note: see the differences in monthly averages between Giessen (Germany), somewhat southeast of St. Odiliënberg, farther inland:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/giessen_mlo_monthly.jpg