How Much of Atmospheric Carbon Dioxide Accumulation Is Anthropogenic?

By Gary W. Harding

Global warming has become a contentious issue. This debate is primarily a clash between science and economics. The natural laws of gas chemistry and the greenhouse effect are scientifically understood. The dominant greenhouse gas is carbon dioxide. Regardless of the source, it is clear from direct measurements and proxy data that the atmospheric concentration of carbon dioxide has increased by about 80 parts per million over the last 200 years. The greenhouse effect from this added carbon dioxide has produced a positive forcing of about one degree centigrade upon mean global surface temperature. This temperature increase, in turn, is expected to produce climate change.

The argument, however, is about the source of the carbon dioxide. If human carbon emissions are the primary source, then attempts to reduce these emissions would have a significant economic impact. A consensus from many scientists has concluded from the data that human activities are at least partly responsible [#1]; many say for at least half of it. Climate-change skeptics have argued, however, that anthropogenic carbon emissions are not precipitating global warming. They say that the human contribution to carbon emissions from all sources is too small to be significant and that the observed changes in atmospheric carbon dioxide concentration are just natural variation. This position, of course, is in response to a perceived threat to the economic status quo.

To shed some light upon this issue, let's take a look at the scientific data. Human emissions of carbon are associated primarily with carbon dioxide from the combustion of fossil fuels and some from other combustion (e.g., wood), and cement production. But, carbon emissions also include minor amounts of other carbon compounds. Scientists measure carbon dioxide emissions by the weight of carbon in it rather than the weight of carbon dioxide itself. The weight of carbon emissions differs from that of carbon dioxide because carbon dioxide is 3.67 times heavier than carbon alone.

Anthropogenic carbon emissions and atmospheric carbon dioxide concentration from 1900 to present are shown in Figure 1. Present emissions are almost 7 billion metric tons (Gton) per year. Current carbon dioxide concentration is about 360 parts per million (ppm). Although these two curves follow a similar exponentially growing course, they are not in close agreement except during the last 2 decades.


Fig. 1: Anthropogenic carbon emissions [green circles; 0 - 7 billion metric tons (Gton); scale to left] and atmospheric carbon dioxide concentration [red squares; 295 - 365 parts per million (ppm); scale to right] from 1900 to present.


Anthropogenic carbon emissions per year has reached a troublesome magnitude. Today, the atmosphere contains about 720 Gtons of carbon. The concentration of carbon dioxide is about 360 ppm. Regardless of its source, one billion tons of carbon released into the atmosphere as carbon dioxide would increase its concentration by 0.5 ppm (360 / 720) if all of it stayed there. However, scientists estimate that an amount eaual to about half of annual human carbon emissions are absorbed by the environment each year. Of the half absorbed, scientists have accounted for where half of that goes. Where the other half goes is the "mystery of the missing carbon" (about 1.8 Gton per year).

Since about half of human carbon emissions are not absorbed by the environment, this fraction accumulates in the atmosphere from year to year. A better way to look at the relation between carbon emissions and atmospheric carbon dioxide concentration is to examine the cumulative total of emissions. These data are shown in Figure 2. They have been fitted with a linear regression (i.e., best linear fit) from the year 1900 to present. The slope of the relation is 0.266 ppm per year per Gton of emissions. If exactly half of human carbon emissions have been absorbed by the environment, the slope would be 0.25 (0.5 / 2). The expected intercept is 298 ppm (in the year 1900). That of the regression is 293. The r-square goodness of fit is 0.996. This means that 99.6% of the variance (i.e., variability) in atmospheric carbon dioxide concentration is completely accounted for by anthropogenic carbon emissions. The correlation coefficient between cumulative anthropogenic carbon emissions and atmospheric carbon dioxide concentration is r = 0.998 (e.g., r = 1.0 represents an absolutely perfect match).


Fig. 2: Atmospheric carbon dioxide concentration (280 - 380 ppm) as a function of cumulative anthropogenic carbon emissions (0 - 300 Gton) from 1800 to present (red circles). Black line is a linear regression fit to these data from 1900 to present (CO2 = 0.266 x CE + 293).


From these data, the 270 Gtons of human carbon emissions over the past 200 years would have increased carbon dioxide concentration from 280 to 415 ppm, if it had all stayed there. Since present carbon dioxide concentration is about 360 ppm, 41% of the cumulative emissions have been absorbed by the environment; assuming, of course, that the carbon source is entirely human. This differs from the present estimate of 50%, due to lower absorption rates during the first three quarters of the last two centuries.

Is the source of carbon dioxide accumulation in the atmosphere entirely human? The data presented above would suggest that it is. However, these data are circumstantial, representing the net appearance. The concentration of carbon dioxide in the atmosphere is a consequence of several factors. These factors are the elements in the ecosystem's carbon cycle. The net effect is the sum of all of the carbon sources and sinks.

Four source-sink scenarios which explain the observations are presented in Table 1 below. Scenario A reflects the data presented above which indicates that all, or nearly all, of the accumulated carbon dioxide in the atmosphere is anthropogenic. Scenario B allows for a minor amount of natural emissions, about 5%. Scenario C assumes that there are natural sources, as yet unidentified, of the same magnitude as human emissions and that there are, as yet, undetected sinks which can account for the observed net concentration of carbon dioxide. This is the position taken by the Intergovernmental Panel on Climate Cahange (IPCC) [#1]. Scenario D assumes that human contributions to carbon dioxide accumulation are insignificant (about 5%), as argued by the climate-change skeptics [#2].

Cumulative human carbon emissions over the past 200 years (270 Gton) as well as net carbon accumulation in the atmosphere (160 Gton) are known quantities. Scenarios A - D balance these quantities with the required net natural sources and sinks which would determine the specified human contribution.

TABLE 1. Four scenarios of net cumulative carbon sources (+) and net sinks (-) for carbon added to the atmosphere from 1800 to present in billions of metric tons (Gton).

Scn. Human (+)  Natural (+)  Subtotal (+) Natural (-) Net (+&-)
---------------------------------------------------------------
A    270 (100%)    0 ( 0%)    270          110 (41%)     160
B    270 ( 95%)   14 ( 5%)    284          124 (44%)     160
C    270 ( 50%)  270 (50%)    540          380 (70%)     160
D    270 (  5%) 5400 (95%)   5670         5510 (97%)     160

One of the basic principles of science is the simplicity hypothesis. That is, among competing explanations for a phenomenon, the simplest explanation is most likely to be the correct one. Scenario A in Table 1 satisfies the simplicity hypothesis. Scenario B accounts for the possibility that there could be a minor amount of natural emissions. Scenario C, advanced by the IPCC, leaves considerable wiggle room for adaptation to the, as yet, incompletely understood carbon cycle; including the mystery of the missing carbon. Scenario D, put forth by the climate-change skeptics, is preposterous on its face. Here, a 5.4 trillion ton, undetected natural source (emitting 95%) as well as a 5.5 trillion ton natural sink (absorbing 97%) are required to explain the observations. Further, in accordance with the data in Fig. 2, this hypothetical net source and sink must have grown over the past 200 years almost exactly in parallel with human carbon emissions.

TABLE 2. Same four scenarios as in Table 1, but for projected year 2000 net cumulative carbon sources (+) and net sinks (-) for carbon added to the atmosphere per annum in billions of metric tons (Gton).

Scn. Human (+)  Natural (+)  Subtotal (+) Natural (-) Net (+&-)
---------------------------------------------------------------
A    7.0 (100%)    0  ( 0%)    7.0          3.5 (50%)    3.5
B    7.0 ( 95%)   0.4 ( 5%)    7.4          3.9 (53%)    3.5
C    7.0 ( 50%)   7.0 (50%)   14.0         10.5 (75%)    3.5
D    7.0 (  5%) 133.0 (95%)  140.0        136.5 (98%)    3.5

Table 2 lists the data for the four scenarios, but for projected year 2000 annual emissions of carbon (see Fig. 1). Scenarios A - C represent a rationally pessimistic to optimistic range. Although there is the mystery of the missing carbon, the climate-change skeptic's scenario D requires such a huge net natural source and sink that it would be impossible for scientists to have missed them.

References

#1 Intergovernmental Panel on Climate Change, THE SCIENCE OF CLIMATE CHANGE 1995, J. T. Houghton et al., Eds., Cambridge University Press, Cambridge, 1996.

#2 Gelbspan, R., THE HEAT IS ON, Addison Westly, Reading, MA, 1997.

Postscript (2008):

A decade has passed since these data were gathered and this analysis was performed. So, where do we stand now? Fortunately, a much more comprehensive data set from the Earth Policy Institute recently appeared on the internet (Earth Policy Institute). These data cover 1751 through 2007 in 1 year increments. The update for Fig. 1 is shown in Fig. 3 with anthropogenic carbon dioxide emissions by the weight of carbon (left y axis) in green and atmospheric carbon dioxide concentration (right y axis) in red. Both have continued to grow over the last decade at about the same rates as the previous 2 decades.

.

.

.

.

The update for Fig. 2 is shown in Fig. 4 with cumulative carbon emissions (E) in green and the regression line in black. The data from 1880 (when the entry of oil and natural gas turned the curve upward) to 2007 were used. The linear regression is [CO2] = 0.271 x E + 293 with an r-squared of 0.996. The coefficients have changed little. Therefore, the conclusions drawn here a decade ago still stand.

.

.

.

.

.

.

I have calculated per capita carbon emissions (i.e., carbon footprint) by adding population data from Kremer (1993) as reported in Cohen (1995) and the United Nations as reported in Cohen (1995) and the World Almanac and Book of Facts (1995-2008). Fig. 5 shows these results with per capita carbon emissions (left y axis) in green and population (right y axis) in red. Per capita emissions turned upwards in 1850 with a slowdown in the growth rate during the Great Depression. They turned upward again during WWII and thereafter. Curiously, per capita emission growth stopped in about 1975. However, the growth rate picked up again starting in 2000.

.

.

.

© 1998, 2008; Gary W. Harding

Correspondence: Trajcom@aol.com

Last updated: 11 Nov 2008

Back to the Tragedy of the Commons Homepage