Climate Change 2001:
Working Group I: The Scientific Basis
Other reports in this collection

4.2 Trace Gases: Current Observations, Trends, and Budgets

4.2.1 Non-CO2 Kyoto Gases

4.2.1.1 Methane (CH4)

Methane's globally averaged atmospheric surface abundance in 1998 was 1,745 ppb (see Figure 4.1), corresponding to a total burden of about 4,850 Tg(CH4). The uncertainty in the burden is small (±5%) because the spatial and temporal distributions of tropospheric and stratospheric CH4 have been determined by extensive high-precision measurements and the tropospheric variability is relatively small. For example, the Northern Hemisphere CH4 abundances average about 5% higher than those in the Southern Hemisphere. Seasonal variations, with a minimum in late summer, are observed with peak-to-peak amplitudes of about 2% at mid-latitudes. The average vertical gradient in the troposphere is negligible, but CH4 abundances in the stratosphere decrease rapidly with altitude, e.g., to 1,400 ppb at 30 km altitude in the tropics and to 500 ppb at 30 km in high latitude northern winter.

Figure 4.1: (a) Change in CH4 abundance (mole fraction, in ppb = 10-9) determined from ice cores, firn, and whole air samples plotted for the last 1,000 years. Data sets are as follows: Grip, Blunier et al. (1995) and Chappellaz et al. (1997); Eurocore, Blunier et al. (1993); D47, Chappellaz et al. (1997); Siple, Stauffer et al. (1985); Global (inferred from Antarctic and Greenland ice cores, firn air, and modern measurements), Etheridge et al. (1998) and Dlugokencky et al. (1998). Radiative forcing, approximated by a linear scale since the pre-industrial era, is plotted on the right axis.
(b) Globally averaged CH4 (monthly varying) and deseasonalised CH4 (smooth line) abundance plotted for 1983 to 1999 (Dlugokencky et al., 1998). (c) Instantaneous annual growth rate (ppb/yr) in global atmospheric CH4 abundance from 1983 through 1999 calculated as the derivative of the deseasonalised trend curve above (Dlugokencky et al., 1998). Uncertainties (dotted lines) are ±1 standard deviation. (d) Comparison of Greenland (GRIP) and Antarctic (D47 and Byrd) CH4 abundances for the past 11.5 kyr (Chappellaz et al., 1997). The shaded area is the pole-to-pole difference where Antarctic data exist. (e) Atmospheric CH4 abundances (black triangles) and temperature anomalies with respect to mean recent temperature (grey diamonds) determined for the past 420 kyr from an ice core drilled at Vostok Station in East Antarctica (Petit et al., 1999).

The most important known sources of atmospheric methane are listed in Table 4.2. Although the major source terms of atmospheric CH4 have probably been identified, many of the source strengths are still uncertain due to the difficulty in assessing the global emission rates of the biospheric sources, whose strengths are highly variable in space and time: e.g., local emissions from most types of natural wetland can vary by a few orders of magnitude over a few metres. Nevertheless, new approaches have led to improved estimates of the global emissions rates from some source types. For instance, intensive studies on emissions from rice agriculture have substantially improved these emissions estimates (Ding and Wang, 1996; Wang and Shangguan, 1996). Further, integration of emissions over a whole growth period (rather than looking at the emissions on individual days with different ambient temperatures) has lowered the estimates of CH4 emissions from rice agriculture from about 80 Tg/yr to about 40 Tg/yr (Neue and Sass, 1998; Sass et al., 1999). There have also been attempts to deduce emission rates from observed spatial and temporal distributions of atmospheric CH4 through inverse modelling (e.g., Hein et al., 1997; Houweling et al., 1999). The emissions so derived depend on the precise knowledge of the mean global loss rate and represent a relative attribution into aggregated sources of similar properties. The results of some of these studies have been included in Table 4.2. The global CH4 budget can also be constrained by measurements of stable isotopes (13C and D) and radiocarbon (14CH4) in atmospheric CH4 and in CH4 from the major sources (e.g., Stevens and Engelkemeir, 1988; Wahlen et al., 1989; Quay et al., 1991, 1999; Lassey et al., 1993; Lowe et al., 1994). So far the measurements of isotopic composition of CH4 have served mainly to constrain the contribution from fossil fuel related sources. The emissions from the various sources sum up to a global total of about 600 Tg/yr, of which about 60% are related to human activities such as agriculture, fossil fuel use and waste disposal. This is consistent with the SRES estimate of 347 Tg/yr for anthropogenic CH4 emissions in the year 2000.

The current emissions from CH4 hydrate deposits appear small, about 10 Tg/yr. However, these deposits are enormous, about 107 TgC (Suess et al., 1999), and there is an indication of a catastrophic release of a gaseous carbon compound about 55 million years ago, which has been attributed to a large-scale perturbation of CH4 hydrate deposits (Dickens, 1999; Norris and Röhl, 1999). Recent research points to regional releases of CH4 from clathrates in ocean sediments during the last 60,000 years (Kennett et al., 2000), but much of this CH4 is likely to be oxidised by bacteria before reaching the atmosphere (Dickens, 2001). This evidence adds to the concern that the expected global warming may lead to an increase in these emissions and thus to another positive feedback in the climate system. So far, the size of that feedback has not been quantified. On the other hand, the historic record of atmospheric CH4 derived from ice cores (Petit et al., 1999), which spans several large temperature swings plus glaciations, constrains the possible past releases from methane hydrates to the atmosphere. Indeed, Brook et al. (2000) find little evidence for rapid, massive CH4 excursions that might be associated with large-scale decomposition of methane hydrates in sediments during the past 50,000 years.

Table 4.2: Estimates of the global methane budget (in Tg(CH4)/yr) from different sources compared with the values adopted for this report (TAR).
Reference: Fung et al. Hein et al. Lelieveld et al. Houweling et al. Mosier et al. Olivier et al. Cao et al. SAR TARa
  (1991) (1997) (1998) (1999) (1998a) (1999) (1998)    
Base year: 1980s - 1992 - 1994 1990 - 1980s 1998
Natural sources
 
 
 
 
 
 
 
 
 
Wetlands
115
237
225c
145
 
 
92
 
 
Termites
20
-
20
20
 
 
 
 
 
Ocean
10
-
15
15
 
 
 
 
 
Hydrates
5
-
10
-
 
 
 
 
 
Anthropogenic sources
 
 
 
 
 
 
 
 
 
Energy
75
97
110
89
 
109
 
 
 
Landfills
40
35
40
73
 
36
 
 
 
Ruminants
80
90b
115
93
80
93b
 
 
 
Waste treatment
-
b
25
-
14
b
 
 
 
Rice agriculture
100
88
c
-
25-54
60
53
 
 
Biomass burning
55
40
40
40
34
23
 
 
 
Other
-
-
-
20
15
 
 
 
 
Total source
500
587
600
 
 
 
 
597
598
Imbalance (trend)
 
 
 
 
 
 
 
+37
+22
Sinks
 
 
 
 
 
 
 
 
 
Soils
10
-
30
30
44
 
 
30
30
Tropospheric OH
450
489
510
 
 
 
 
490
506
Stratospheric loss
-
46
40
 
 
 
 
40
40
Total sink
460
535
580
 
 
 
 
560
576
a TAR budget based on 1,745 ppb, 2.78 Tg/ppb, lifetime of 8.4 yr, and an imbalance of +8 ppb/yr.
b Waste treatment included under ruminants.
c Rice included under wetlands.

The mean global loss rate of atmospheric CH4 is dominated by its reaction with OH in the troposphere.

OH + CH4 CH3 + H2O

This loss term can be quantified with relatively good accuracy based on the mean global OH concentration derived from the methyl chloroform (CH3CCL3) budget described in Section 4.2.6 on OH. In that way we obtain a mean global loss rate of 507 Tg(CH4)/yr for the current tropospheric removal of CH4 by OH. In addition there are other minor removal processes for atmospheric CH4. Reaction with Cl atoms in the marine boundary layer probably constitutes less than 2% of the total sink (Singh et al., 1996). A recent process model study (Ridgwell et al., 1999) suggested a soil sink of 38 Tg/yr, and this can be compared to SAR estimates of 30 Tg/yr. Minor amounts of CH4 are also destroyed in the stratosphere by reactions with OH, Cl, and O(1D), resulting in a combined loss rate of 40 Tg/yr. Summing these, our best estimate of the current global loss rate of atmospheric CH4 totals 576 Tg/yr (see Table 4.2), which agrees reasonably with the total sources derived from process models. The atmospheric lifetime of CH4 derived from this loss rate and the global burden is 8.4 years. Attributing individual lifetimes to the different components of CH4 loss results in 9.6 years for loss due to tropospheric OH, 120 years for stratospheric loss, and 160 years for the soil sink (i.e., 1/8.4 yr = 1/9.6 yr + 1/120 yr + 1/160 yr).

Continues on next page



Other reports in this collection