Introduction

Loss of atmospheric methane is dominated by the reaction with the hydroxyl radical (OH) in the troposphere:

In addition there is minor methane removal through other sinks. Loss rates that have been suggested are: Absorption in the soil, transport to the stratosphere where methane reacts with OH, Cl and O(¹D) and reaction with Cl in the atmospheric boundary layer. Together these minor sinks account for approximately 10 - 15 % of the total loss of atmospheric methane. The methane loss rate is therefore first of all determined by the total atmospheric burden of OH and its changes in time.

The total lifetime of atmospheric methane is given by the equation:

1/t CH4= 1/t OH+1/t additional

Eq. 1

The methane lifetime due to the reaction with OH is t_OH= 1/(k₁¥[OH]), where k₁ is the reaction coefficient for R1 and [OH] the concentration of OH. t_additional denotes the methane lifetime due to the combined minor sinks given above. t_OH = 9.6 years has been obtained by scaling from a methyl chloroform (CH₃CCl₃) lifetime of 5.7 years (Spivakovsky et al., 2000). The total methane lifetime was estimated to be t_CH4=8.4 years. Using this method to estimate methane lifetime implies that any corrections of the methyl chloroform lifetime would lead to changes in the methane lifetime. In fact, from a recent analysis of methyl chloroform observations, Montzka et al. (2000) deduced a reduced atmospheric lifetime of methyl chloroform of 5.2 (+0.2-0.3) years. This implies a methane lifetime that is shorter than the one given above.

The atmospheric lifetime of methane using 3-D tropospheric CTMs is estimated to be similar to what was obtained from the methyl chloroform analysis. For instance, the IPCC Assessment of aircraft impact on the atmosphere (Isaksen and Jackman, 1999) reports atmospheric lifetimes of methane in the range 6.6 to 10.5 years for 1992.

The estimated lifetime of 8.4 years gives a total sink for atmospheric methane of 576 Tg/yr. Estimates of the global source strength is typical of the order 580 to 600 Tg/year. This is in agreement with the estimated sinks and increase in atmospheric burden (Hein et al., 1997; Lelieveld et al., 1998; IPCC, 1996). It should be noted, however, that estimates of sources and sinks are connected with large uncertainties. Hein et al. (1997) estimate the total methane emission to be in the range 520 to 625 Tg/yr, while Lelieveld et al. (1998) estimate the total sink to be 600 ± 80 Tg/yr. There are equal or larger uncertainties connected to the individual sources; Lelieveld et al. (1998) estimate total natural methane emission to be 190 ± 70 Tg/yr, with the largest contribution from wetlands (145 ± 30 Tg/yr). They estimate fossil fuel related emissions to be 120 ± 40 Tg/yr, and agricultural sources to be 230 ± 115 Tg/yr.

Atmospheric abundance of methane has increased approximately a factor 2.5 since pre-industrial time, and it continues to increase. The concentrations have increased from 1.52 ppm in 1978 to 1.76 ppm in 1998. There is, however, a significant declining trend in the methane increase over the last two decades (Dlugokencky et al., 1998), with large year to year variations. The methane growth rate since 1992 has been 4.9 ppb/ yr.

Changes in the emission of methane from major sources like wetlands, fossil fuel related or biomass burning are likely causes for the observed short term, year to year growth in atmospheric methane. An atmospheric lifetime of methane of 10 years or less is sufficiently short that changes in the global OH concentration that last a few years will have significant impact on the methane growth rate.

Chemical processes determining the OH distribution and the methane lifetime

The local abundance of OH is determined by the local abundance of source and sink gases: CO, CH₄, O₃, nonmethane hydrocarbons (NMHC), NO and water vapour, as well as the intensity of solar UV-B radiation. Based on the methyl chloroform loss rate in the atmosphere Prinn et al., (1995) and Kroll et al. (1998) estimated an average global OH concentration of approximately 1x10⁶ molecules/cm³. However, the concentration varies significantly over the day, with season, and with geographical location.

The main production of OH (and odd hydrogen) in the troposphere occurs through the photo dissociation of ozone by UV-B radiation followed by the reaction with of exited state oxygen with water vapor:

Changes in tropospheric ozone, UV-B radiation and water vapour will therefore affect the OH distribution. Tropospheric ozone will increase as a result of increased emissions of NO_X, CO NMHC and CH₄, UV-B radiation will change as a result of changes in stratospheric ozone and changes in tropospheric water vapor could be caused by climate changes.

Several chemical reactions are involved in the ozone loss in the troposphere. The main loss occurs through the reaction with CO converting OH to HO₂:

Approximately 70% of the global OH loss is estimated to go through this reaction (Karlsdottir and Isaksen, 2000), and approximately 15–20% through the reaction with CH₄ (R1). Additional loss occurs via reactions with CH₂O, NMHC and ozone. The lifetime of OH is very short (less than 1 second).

Conversion from HO₂ through gas phase reactions is important for the tropospheric abundance of OH. The two main reactions are:

One important result of the OH forming and loss reactions given above is that increased emission of carbon monoxide and hydrocarbons lead to reduced OH concentrations, and longer methane lifetime. Enhanced emission of NO has the opposite effect. The NO/CO ratio in emissions, is therefore more important for the abundance of OH than the absolute emission of the gases.

Since reaction R1 affects the atmospheric lifetime of both methane and OH, increases in methane emission will lead to reduced ozone levels and thereby to a longer atmospheric lifetime of methane (Eq. 1) which further increases atmospheric concentrations of methane (positive feedback) (Isaksen, 1988; Berntsen et al., 1992). The methane feedback factor F is given by:

where D[CH₄] is the change in global methane equilibrium concentration for a change in global emission DEM_CH4. Extensive studies of the feedback factor have been performed over the last 10 years. IPCC (1995) reported values in the range 1.2–1.7 from 2-D and 3-D model studies. In a recent intercomparison between 5 different 3-D CTMs (Karlsdottir, 2000), values for F in the range 1.34 to 1.63 for atmospheric oxidation by OH was obtained. These studies demonstrate that methane oxidation of OH play a significant role for atmospheric increase of methane

Changes in methane lifetime

It is likely that the global OH distribution has changed since pre-industrial time due to increases in precursor emissions (CO, CH₄, NO_x) and ozone. How large the change is is, however, uncertain due to the counteracting effects of CO and CH₄ (reduces OH through reactions R1 and R4 respectively), and NO_x and O₃ (increases OH increases through reaction R6 and R7 respectively). Lelieveld et al. (1998) calculate a significant reduction in OH and increase in methane lifetime from pre-industrial time (Eq.1) (6.2 yr) to 1992 (7.9 yr), while Berntsen et al. (1997) calculate an increase in global OH of approximately 6% (and decrease in methane lifetime). Future changes in methane lifetime, particularly toward the end of the 21st century, are highly uncertain due to the large uncertainties in precursor emissions (IPCC, 2001).

Figure 1. The evolution of the fossil fuel emission of CO (a), NOX (b), for North America, Western Europe, Eastern Europe, and Australia; and fossil fuel consumption (c) for South America, Africa, Middle East, South East Asia (China and India included) and Japan compared to 1985 values.

We have used the 3-D Oslo CTM1 (Berntsen and Isaksen, 1997) to calculate changes in global OH distribution and methane lifetime for the time period 1980 to 1996 (Karlsdottir and Isaksen, 2000). We further estimated the implication the changes in lifetime have had for the global methane emissions. Methane changes over the time period were constrained by observations (WMO, 1992; IPCC; 1996: Dlugokencky et al., 1998). Nine different regions were selected where changes in NO_X, CO and NMHC emissions were given. Figure 1 shows the emission changes for the time period 1980 to 1996 adopted for different region for CO and NO_X, and for oil consumption. The latter numbers were used to characterize the emissions for developing countries and Japan. References to the adopted emissions are given in Karlsdottir and Isaksen, 2000. There are large differences in the adopted emission trends between the different regions, and in the ratio of the OH forming to the OH loss precursors (e.g. NO_X to CO ratio). Eastern Europe had a marked reduction in NO_X emissions during the 1990s, while the rest of the world experienced either small changes in NO_X emissions or significant increases. CO emissions on the other hand, decrease significantly during the 1990s in the industrial regions. This different emission pattern is estimated to have led to significantly different developments in the OH abundance in the different regions.

Figure 2. The change in global average OH concentration (—), and tCH4 (– –) from 1980 to 1996 due to changes in emissions of CO, NOx, and NMHCs, and observed changes in CH4. Calculations of tCH4; with the global NOx emissions held at 1985 level (), with the South East Asia (Japan included) NOx emissions held at the 1985 level (D), with the global CO emissions held at the 1985 level (x), with the South East Asia (Japan included) CO emissions held at the 1985 level (o).

The calculated trends in the global OH distribution, and in the methane lifetime, are shown in Figure 2. The changes in emissions from fossil fuel usage over the time period considered is estimated to have had a significant impact on the OH distribution, and on the methane lifetime. Of particular importance is the shift in emission pattern from northern industrial regions to developing countries (e.g., South East Asia (SEA) as pointed out by Gupta et al., 1998). We estimate the global average concentrations of OH to have increased from 0.95 x 10⁶ in 1980 to 1.01 molecules/cm² in 1996. Different sensitivity studies were performed to estimate the impact of CO and NO_X emissions globally, as well as from SEA sources. In each of the model runs, which were performed for the last year of the period, either the CO or the NO_X emissions were kept constant at the 1985 levels. The results of these sensitivity studies, which are included in Figure 2, reveal that the main contribution to the reduced atmospheric lifetime of methane after 1985, was mainly caused by the rapidly growing NO_X emission in SEA. The increase in CO emission in the same region had some effect on the methane lifetime. NO_X emission increase in SEA since 1985 reduced the methane lifetime by approximately 0.5 years, while CO emission in the same region increased the lifetime by approximately 0.2 years between 1985 and 1996. Emission changes in other regions of the world have had smaller impact on the methane lifetime. There are two reasons for this: Firstly, the rapid growth in anthropogenic emissions and , secondly, methane lifetime is basically determined by oxidation in the lower troposphere at low latitudes.

 Implications for methane emission changes

The global source strength of CH₄ needed to give the observed global CH₄ concentration can be determined by the equation:

C(t) is the concentration of CH₄, P(t) is the global methane emission, and L(t) is the calculated global average loss rate = 1/t _CH4. t _CH4 is given by Eq. 1.The calculations give a yearly mean methane emission between 1980 and 1996 of 542.7 Tg/yr, which is in good agreement with other estimates (Fung et al., 1991; Dlugokencky et al., 1998). The emission increases from 515 Tg/yr in 1980 to 573 Tg/yr in 1996. In Table 1 we summarize our calculated yearly trends for the time period 1980 to 1996 in the OH distribution, in t _CH4, and in the estimated global methane emission. The yearly average increase in methane emission is calculated from the observed changes in methane concentrations and the calculated changes in the t _CH4. Included in the table are the estimated changes in OH and t _CH4 from Krol et al (1998), and the estimated increase in methane emissions from Dlugokencky et al. (1998); the latter is based observed methane concentrations, and on a constant methane lifetime. The calculated change in methane lifetime, which is similar to the change calculated by Krol et al. (1998), implies a significant larger increase in emission than what is obtained from the observed methane changes assuming a constant methane lifetime (Dlugokencky et al., 1998).

It is of interest to estimate what increase in methane concentrations we can expect from the calculated increase in global methane emission if methane´s lifetime was affected only by the increasing emission of methane through reaction R1 (no changes in the emissions of CO, NO_X and NMHC). In this case the mixing ratio of methane would be raised to 1.86 ppm in 1986 from a value of 1.57 in 1980 as compared to the observed value in 1996 of 1.76 ppm. This clearly demonstrate how important the emissions of precursors like NO_X and CO, particularly the rapid increase in fossil fuel SEA is for the methane lifetime.

Observations have shown that there have been large variations in the atmospheric abundance in CO during the late 1980s and early 1990s (Novelli et al., 1998). Part of the observed reductions can probably be explained by the decrease in fossil fuel related CO emissions given in Figure 1. However the large variation in emission from biomass burning over the time period studied has probably also contributed to the large CO variations. In order to study the impact of the changing emissions from biomass burning we have made several model runs with the 3-D Oslo CTM where the individual global biomass sources (forest fire, savannah burning, waste, biofuel) are removed one at a time (Isaksen et al., 2000). The resulting impact on [OH] and on t _CH4 is given in Table 2.

The large CO/NO_X ratio in most of the biomass burning emissions leads to a decrease in OH levels when the emission increases (reaction R4 dominates over reaction R6, except for savannah burning which have a smaller ratio) and leads to an increase in t _CH4. This is in striking contrast to the impact of fossil fuel emission where enhanced emission leads to enhanced OH and reduced t _CH4. This is due to a much smaller CO/NO_X ratio in fossil fuel emissions. In the situations with general increase in fossil fuel use the OH chemistry is dominated by reaction R6. The overall effect of biomass burning, even with the large observed variation in the emissions, has been much less than the effect of increasing fossil fuel use over the time period 1980 to 1996.

References

1. Berntsen, T.K., J.S. Fuglestvedt, and I.S.A. Isaksen, Chemical-dynamical modelling of the atmosphere with emphasis on the methane oxidation, Ber. Bunsen Ges. Phys. Chem., 96, 241-251, 1992.
2. Berntsen T.K., Isaksen I.S.A., 1997, A global three-dimensional chemical transport model for the troposphere. 1. Model description and CO and ozone results, J. Geophys. Res., Vol 102, No. D17, pp. 21239-21280, 1997.
3. Berntsen T.K., Isaksen I.S.A., Myhre G., Fuglestvedt J.S., Stordal F., Larsen T.S., Freckleton R.S., Shine K.P.: Effects of anthropogenic emissions on tropospheric ozone and its radiatve forcing, J. Geophys. Res., Vol 102, No. D23, pp. 28101-28126, 1997.
4. Dlugokencky E.J., Masaire K.A., Lang P.M., Tans P.P., 1998: Continuing decline in the growth rate of the atmospheric methane burden. Nature Vol. 393 pp. 447 - 450.
5. Fung, I., J. John, J. Lehner, E. Matthews, M. Prather, L.P. Steele, P.J. Fraser, Three-dimensional synthesis of the global methane cycle, J. Geophys. Res., 96, 13033-13065, 1991.
6. Gupta M., Elliott S., Cicerone R., 1998, Perturbation to global tropospheric oxidizing capacity due to latitudinal redistribution of surface sources of CH₄, CO, and NO_x, Geophy. Res. Lett. Vol. 25 No. 21 pp. 3931 - 3934.
7. Hein, R., P.J. Crutzen, and M. Heinmann, An inverse modelling approachto investigate the global atmospheric methane cycle, Global Biogeochem. Cycles, 11, 43-76, 1997.
8. IPCC, 1996, Climate Change 1995, The science of climate change, Intergovernmental Panel on Climate Change, J.T. Houghton et al., eds, Cambridge U. Press, 1996.
9. Isaksen,I.S.A., Is the oxidation capacity of the atmosphere changing? eds. F.S. Rowland and I.S.A. Isaksen, pp 141-159, John Wileys and Sons, 1988.
10. Isaksen, I.S.A. Jackman, S., Modelling the chemical composition of the future atmosphere, Chapter 4: Modelling the chemical composition of the future atmosphere, Special Report on Aviation and the Global Atmosphere, Eds. J. Penner, D.H. Lister, D.J. Griggs, D.J. Dokken, M. McFarland, Intergovernmental Panel of Climate Change, Cambridge University Press, pp 121-163, 1999.
11. Isaksen, I.S.A., S. Dalsøren, J.K. Sundet, The contribution of biomass burning to the global ozone budget and oxidation of chemical compounds in the troposphere, paper presented at the Quadrennial Ozone Symposium, Sapporo, July 3 - 8, 2000.
12. Karlsdottir, S., Model studies of methane in the atmosphere, PhD thesis, University of Oslo, 2000.
13. Krol M., van Leeuwen P.J. Lelieveld J., 1998: Global OH trend inferred from methylchloroform measurements. J. Geophys. Res., Vol. 103 No. D9 pp. 10697 - 10711.
14. Lelieveld J., Crutzen P.J., Dentener F.J., 1998: Changing concentration, lifetime and climate forcing of atmospheric methane. Tellus 50 B, pp. 128 - 150.
15. Montzka, S.A. C.M. Spivakovsky, J.H. Butler, J.W. Elkins, L.T. Lock and D.J. Mondeel, New Observational Constraints for Atmospheric Hydroxyl on Global and Hemispheric Scales. Science, 288, 500-503, 2000.
16. Novelli P.C., Masarie K.A., Lang P.M., 1998, Distributions and recent changes of carbon monoxide in the lower troposphere, J. Geophys. Res., Vol. 103 No. D15 pp. 19015 - 19033.
17. Prinn R.G., Weiss R.F., Miller B.R., Huand J., Alyea F.N., Cunnold D.M., Fraser P.J., Hartley D.E., Simmonds P.G., 1995: Atmospheric trends and lifetime of CH₃CCl₃ and global OH concentrations, Science, Vol. 269 pp. 187 - 192.
18. Spivakovsky, C.M., J.A. Logan, S.A. Montzka, Y.J. Balkanski, M. Foreman-Fowler, D.B.A. Jones, L.W. Horowitz, A.C. Fusco, C.A.M. Brenninkmeijer, M.A. Prather, S.C. Wofsy, and M.B. McElroy, Three- dimensional climatological distribution of tropospheric OH: update and evaluation, J. Geophys. Res., in press, 2000.
19. Wourld Meteorological Organization 1992 [WMO, 1991], Scientific assessment of ozone depletion: 1991, WMO/UNEP, WMO Global Ozone Research and Monitoring Project, Report No 25, Geneva, 1992.