Greenhouse effect of CO2 vs CH4

There is considerable interest in the idea that on the Early Earth methane may have taken over much of the role of CO2 in offsetting the Faint Young Sun. In part this interest is due to rather sketchy geochemical evidence that at some times in the Archaean CO2 concentrations may not have been high enough to do the trick, but regardless of whether the evidence actually demands a relatively low-CO2 atmosphere, possibilities abound that in an anoxic atmosphere methane could build up to high concentrations. Even on an abiotic planet, there are possibilities for direct volcanic outgassing of methane, at a rate dependent on the state of oxygen in the planet's interior. Once biology comes on the scene, methanogens can convert volcanic CO2 and H2 to CH4, or can make CH4 by decomposing organic matter produced by anoxygenic photosynthesis.

Methane cannot build up to very high concentrations in a well-oxygenated atmosphere, but the relatively small amounts of methane in the atmosphere today (about 1.7 ppmv) nevertheless contribute significantly to global warming. There is, however, a widespread misconception that methane is in some sense an intrinsically better greenhouse gas than CO2. A few simple calculations will serve to clarify the true state of affairs.

In order to compare the relative effects of CH4 and CO2 on a planet's radiation budget, we calculate the OLR for each case in what we have been calling the canonical atmosphere - a mixture of each gas into a dry atmosphere consisting of 1 bar of Earth air with temperature profile on the dry air adiabat, carried out with Earth's surface gravity. Results for a fixed surface temperature of 280K, computed using the homebrew radiation model employing a constant temperature scaling coefficient T* = 900K, are shown in Fig. 4.35. This graph in essence gives the amount of greenhouse gas needed to sustain a surface temperature of 280K, given any specified amount of solar

Figure 4.35: OLR vs. CO2 and CH4 concentration for each gas individually mixed with Earth air on the dry adiabat. The surface temperature is fixed at 280K.

absorption. For example, with an absorbed solar radiation of 300W/m2 the surface temperature can be sustained with either 464ppmv of CO2 or 35,600 ppmv of CH4 (3.56% of the atmosphere by mole fraction). These results somewhat overestimate the effect of each gas as compared to an actual moist atmosphere, since a moist atmosphere would be on the less steep moist adiabat and in a moist atmosphere the water vapor absorption would compete to some extent with the CO2 and CH4 absorption. Still, as an estimate of the relative effect of the two gases, the story is pretty clear. Methane is, intrinsically speaking, a considerably worse greenhouse gas than CO2. The OLR curve for for methane is everywhere well above the curve for CO2, so that it takes more methane than CO2 to achieve a given reduction of OLR.

The common statement that methane is, molecule for molecule, a better greenhouse gas than CO2 is true only for situations like the present where methane is present in far lower concentrations than CO2. In this situation, the greater power of a molecule of CH4 to reduce the OLR results simply from the fact that the greenhouse effect of both CH4 and CO2 are approximately logarithmic in concentration. Reading from Fig. 4.35, we see that for methane concentrations of around 1ppmv, each doubling of methane reduces OLR by about 2W/m2. On the other hand, for CO2 concentrations near 300 ppmv, each doubling of CO2 reduces the OLR by about 6 W/m2. Hence, to achieve the same OLR reduction as a doubling of CO2 one needs three doublings of methane, but since methane starts from a concentration of only 1ppmv, this only takes the concentration to 8ppmv, and requires only as many molecules to bring about as was needed to achieve the same reduction using a doubling of CO2. Equivalently, we can say that adding 1ppmv of methane yields as much reduction of OLR as adding 75ppmv of CO2. The logarithmic slopes in this example are exaggerated compared to the appropriate values for Earth's actual atmosphere, because of the use of the dry adiabat and because of inaccuracies in the simple temperature scaling used in the homebrew radiation code. Using the ccm radiation code on the moist adiabat, with water vapor at 50% relative humidity, we find instead that each doubling of methane near 1ppmv reduces OLR by 0.77 W/m2, while each doubling of CO2 near 300ppmv reduces OLR by 4.3 W/m2; in this case adding 1ppmv of methane reduces the OLR by as much as adding 38ppmv of CO2. Nonetheless, the principle remains the same: If methane were the most abundant long-lived greenhouse gas in our atmosphere, and CO2 were present only in very small concentrations, we would say instead that CO2 is, molecule for molecule, the better greenhouse gas.

Methane to Total Ratio

Figure 4.36: Total OLR reduction for the canonical atmosphere with a mixture of CH4 and CO2. Each curve gives the OLR reduction relative to a transparent atmophere for a fixed sum of CH4 and CO2 molar concentrations, indicated on the curve in units of ppmv. The results are plotted as a function of the CH4 molar concentration to the total molar concentration for the two gases. The ratio is equal to the ratio of atmospheric carbon in the form of CH4 to total atmospheric carbon. Larger values of the OLR reduction correspond to a stronger greenhouse effect.

Methane to Total Ratio

Figure 4.36: Total OLR reduction for the canonical atmosphere with a mixture of CH4 and CO2. Each curve gives the OLR reduction relative to a transparent atmophere for a fixed sum of CH4 and CO2 molar concentrations, indicated on the curve in units of ppmv. The results are plotted as a function of the CH4 molar concentration to the total molar concentration for the two gases. The ratio is equal to the ratio of atmospheric carbon in the form of CH4 to total atmospheric carbon. Larger values of the OLR reduction correspond to a stronger greenhouse effect.

Kirschvink and others have proposed that the Makganyene Snowball came about through a methane catastrophe, in which oxygenation converts methane to CO2 and reduces the greenhouse effect sufficiently to precipitate a snowball. A methane crash, due to a reduction in methanogenic activity of some other mechanism, has sometimes been proposed as a trigger for the Neoproterozoic Snowballs as well. It is by no means easy to make these scenarios play out as they are supposed to, since methane contributes much less to the greenhouse effect than CO2 when the two gases have similar abundances in the atmosphere. Conversion of methane to CO2 will only reduce the greenhouse effect if methane is initially present in sufficiently small concentrations - but if there is too little methane present, the contribution of methane to the total greenhouse effect is too small to make much difference. Let's use the data in Fig. 4.35 to illustrate how the conversion of methane to CO2 would affect climate in a few illustrative cases. The detailed numbers would change with a more accurate radiation model, or if the effect of water vapor were brought in, but the basic conclusions would remain much the same.

For example, suppose we started out with an atmosphere that contained 36,650 ppmv of methane, which would be sufficient to maintain a 280K surface temperature given 300 W/m2 of absorbed solar radiation. If this were all converted to CO2 by oxidation, then, according to Fig. 4.35 the OLR would plunge to 255 W/m2. In order to re-establish radiation balance, the planet would have to warm up to a temperature well in excess of the the initial 280K. Far from causing a Snowball, in this case the oxidation of methane would cause a hot pulse, followed by gradual recovery to the original temperature as the CO2 is drawn down by silicate weathering.

Next let's consider a more general situation, and identify the conditions necessary for a conversion of CH4 to CO2 to substantially reduce the greenhouse effect. Because the absorption features of CH4 and CO2 do not overlap significantly for the range of concentration under consideration, the combined effects of the two gases can be obtained by summing the OLR reduction AOLR for each of the gases taken in isolation. We wish to ask the question: if we have a given number of carbon atoms to use in supplying the atmosphere with greenhouse gases, how does the net greenhouse effect depend on the way we divvy up those atoms between CH4 and CO2? Since each molecule has a single carbon, this question can be addresed by varying the molar concentration of CH4 while keeping the sum of the molar concentrations of CH4 and CO2 fixed. Results of a calculation of this type are shown in Fig. 4.36. For any fixed total atmospheric carbon content, the OLR reduction has a broad maximum when plotted as a function of the methane ratio, and varies little except near the extremes of an all-methane or all-CO2 atmosphere. The only case in which one can get a substantial reduction in greenhouse effect by oxidizing methane into CO2 is when the initial CO2 concentration is very high, the initial CH4 fraction is between about 10% and 90% of the total, and the CH4 is almost entirely converted to CO2. For example, with a total carbon concentration of 10000ppmv, reducing the CH4 concentration from 1000ppmv to 1ppmv reduces the greenhouse effect from 104W/m2 to 80 W/m2. Because the curve is so flat, starting from an atmosphere which is 80% methane works almost as well: in that case the greenhouse effect is reduced from 101 W/m2 to 80 W/m2. If we have a total of 100,000 ppmv of carbon in the atmosphere, then the maximum greenhouse effect occurs for an atmosphere which is about 25% methane, and has a value of 151 W/m2. Reducing the methane to 1 ppmv brings down the greenhouse effect 36 W/m2, to 115 W/m2. In the Paleoproterozoic or Archaean, when the net greenhoue effect needed to be high to offset the Faint Young Sun, it is possible that a methane crash could have reduced the greenhouse effect enough to initiate a snowball, but it is essential that in a methane crash, the methane concentration be brought almost all the way down to zero; a reduction of methane from 50% of the atmosphere to 10% of the atmosphere would not do much to the greenhouse effect. By the time of the Neoproterozoic, when the solar luminosity is higher and less total greenhouse effect is needed to maintain open water conditions, it is far less likely that a methane catastrophe could have initiated glaciation. Some further remarks on atmospheric transitions that could initiate a Snowball will be given in Chapter 8.

Was this article helpful?

0 0

Post a comment