Climate Change 2001: Working Group I: The Scientific Basis

	Other reports in this collection

4.2.3 Reactive Gases

4.2.3.1 Carbon monoxide (CO) and hydrogen (H₂)

Carbon monoxide (CO) does not absorb terrestrial infrared radiation strongly enough to be counted as a direct greenhouse gas, but its role in determining tropospheric OH indirectly affects the atmospheric burden of CH₄ (Isaksen and Hov, 1987) and can lead to the formation of O₃. More than half of atmospheric CO emissions today are caused by human activities, and as a result the Northern Hemisphere contains about twice as much CO as the Southern Hemisphere. Because of its relatively short lifetime and distinct emission patterns, CO has large gradients in the atmosphere, and its global burden of about 360 Tg is more uncertain than those of CH₄ or N₂O. In the high northern latitudes, CO abundances vary from about 60 ppb during summer to 200 ppb during winter. At the South Pole, CO varies between about 30 ppb in summer and 65 ppb in winter. Observed abundances, supported by column density measurements, suggest that globally, CO was slowly increasing until the late 1980s, but has started to decrease since then (Zander et al. 1989; Khalil and Rasmussen, 1994), possibly due to decreased automobile emissions as a result of catalytic converters (Bakwin et al., 1994). Measurements from a globally distributed network of sampling sites indicate that CO decreased globally by about 2 %/yr from 1991 to 1997 (Novelli et al., 1998) but then increased in 1998. In the Southern Hemisphere, no long-term trend has been detected in CO measurements from Cape Point, South Africa for the period 1978 to 1998 (Labuschagne et al., 1999).

Some recent evaluations of the global CO budget are presented in Table 4.6. The emissions presented by Hauglustaine et al. (1998) were used in a forward, i.e., top-down, modelling study of the CO budget; whereas Bergamasschi et al. (2000) used a model inversion to derive CO sources. These varied approaches do not yet lead to a consistent picture of the CO budget. Anthropogenic sources (deforestation, savanna and waste burning, fossil and domestic fuel use) dominate the direct emissions of CO, emitting 1,350 out of 1,550 Tg(CO)/yr. A source of 1,230 Tg(CO)/yr is estimated from in situ oxidation of CH₄ and other hydrocarbons, and about half of this source can be attributed to anthropogenic emissions. Fossil sources of CO have already been accounted for as release of fossil C in the CO₂ budget, and thus we do not double-count this CO as a source of CO₂.

Table 4.6: Estimates of the global tropospheric carbon monoxide budget (in Tg(CO)/yr) from different sources compared with the values adopted for this report (TAR).
Reference:	Hauglustaine et al.	Bergamasschi et al.	WMO	SAR	TARa
	(1998)	(2000)	(1999)	(1996)
Sources
Oxidation of CH4		795		400 - 1000	800
Oxidation of Isoprene		268		200 - 600b	270
Oxidation of Terpene		136			~0
Oxidation of industrial NMHC		203			110
Oxidation of biomass NMHC		-			30
Oxidation of Acetone		-			20
Sub-total in situ oxidation	881	1402			1230
Vegetation		-	100	60 - 160	150
Oceans		49	50	20 - 200	50
Biomass burningc		768	500	300 - 700	700
Fossil & domestic fuel		641	500	300 - 550	650
Sub-total direct emissions	1219	1458	1150		1550
Total sources	2100	2860		1800 - 2700	2780
Sinks
Surface deposition	190			250 - 640
OH reaction	1920			1500 - 2700
Anthropogenic emissions by continent/region	Y2000	Y2100(A2p)
Africa	80	480
South America	36	233
Southeast Asia	44	203
India	64	282
North America	137	218
Europe	109	217
East Asia	158	424
Australia	8	20
Other	400	407
Sum	1036	2484
a Recommended for OxComp model calculations for year 2000. b Includes all VOC oxidation. c From deforestation, savannah and waste burning.

It has been proposed that CO emissions should have a GWP because of their effects on the lifetimes of other greenhouse gases (Shine et al., 1990; Fuglesvedt et al., 1996; Prather, 1996). Daniel and Solomon (1998) estimate that the cumulative indirect radiative forcing due to anthropogenic CO emissions may be larger than that of N₂O. Combining these early box models with 3-D global CTM studies using models from OxComp (Wild and Prather, 2000; Derwent et al., 2001) suggests that emitting 100 Tg(CO) is equivalent to emitting 5 Tg(CH₄): the resulting CH₄ perturbation appears after a few months and lasts 12 years as would a CH₄ perturbation; and further, the resulting tropospheric O₃ increase is global, the same as for a direct CH₄ perturbation. Effectively the CO emission excites the global 12-year chemical mode that is associated with CH₄ perturbations. This equivalency is not unique as the impact of CO appears to vary by as much as 20% with latitude of emission. Further, this equivalency systematically underestimates the impact of CO on greenhouse gases because it does not include the short-term tropospheric O₃ increase during the early period of very high CO abundances (< 6 months). Such O₃ increases are regional, however, and their magnitude depends on local conditions.

Molecular hydrogen (H₂) is not a direct greenhouse gas. But it can reduce OH and thus indirectly increase CH₄ and HFCs. Its atmospheric abundance is about 500 ppb. Simmonds et al. (2000) report a trend of +1.2 ± 0.8 ppb/yr for background air at Mace Head, Ireland between 1994 and 1998; but, in contrast, Novelli et al. (1999) report a trend of –2.3 ± 0.1 ppb/yr based on a global network of sampling sites. H₂ is produced in many of the same processes that produce CO (e.g., combustion of fossil fuel and atmospheric oxidation of CH₄), and its atmospheric measurements can be used to constrain CO and CH₄ budgets. Ehhalt (1999) estimates global annual emissions of about 70 Tg(H₂)/yr, of which half are anthropogenic. About one third of atmospheric H₂ is removed by reaction with tropospheric OH, and the remainder, by microbial uptake in soils. Due to the larger land area in the Northern Hemisphere than in the Southern Hemisphere, most H₂ is lost in the Northern Hemisphere. As a result, H₂ abundances are on average greater in the Southern Hemisphere despite 70% of emissions being in the Northern Hemisphere (Novelli et al., 1999; Simmonds et al., 2000). Currently the impact of H₂ on tropospheric OH is small, comparable to some of the VOC. No scenarios for changing H₂ emissions are considered here; however, in a possible fuel-cell economy, future emissions may need to be considered as a potential climate perturbation.