Science Briefs

Understanding Carbon Monoxide as Pollutant and as Agent of Climate Change

Carbon monoxide (CO) is a pollutant that affects methane, carbon dioxide, and tropospheric (lower atmospheric) ozone. It thus plays a role in both air pollution and climate change, and is therefore regulated in many parts of the world. CO is unique among pollutants in the lower atmosphere in that it lasts for roughly a month, long enough for it to be transported long distances but not so long that it becomes distributed nearly uniformly. Further, it has been observed from satellites for several years. These features make comparison of modeled and measured CO a very useful way to evaluate our current understanding of and ability to model air pollution.

As part of an effort to study future air quality and climate change, a team of researchers from ten countries recently used 26 state-of-the-art atmospheric chemistry models to simulate present-day and projected near-future CO. We compared the present-day simulations with near-global observations from the Measurements of Pollution in the Troposphere (MOPITT) instrument flown on NASA's Terra satellite and with local surface measurements. The models show large underestimates of CO at middle and high latitudes in the Northern Hemisphere, while typically performing reasonably well elsewhere (see figure).

Four maps of carbon monoxide as observed and simulated

Figure above: Maps of carbon monoxide in the middle troposphere observed from MOPITT and the multi-model average, both in units of parts per billion by volume (ppbv, i.e., molecules of CO per billion molecules of air). The top row shows MOPITT observations from 2000, while the bottom row shows the equivalent for the average of the 26 models analyzed. Values in the upper right corner are the global mean area-weighted CO (ppbv). Grey indicates areas where satellite observations were not available. Click for large GIF or PDF.

Detailed analysis of the results suggests that year-round surface emissions of CO, probably from fossil fuel burning in East Asia, and seasonal biomass burning emissions in South-central Africa, are greatly underestimated in current inventories used to drive models. Models appear to be able to reproduce many aspects of the transport and chemical transformation of CO in the atmosphere relatively well. However, variability among models is large. This is primarily due to model-to-model differences in simulating the hydrologic cycle and in representations and emissions of non-methane hydrocarbons, which can be chemically oxidized into CO. Better information on surface emissions and simultaneous observations of additional gases would clearly help resolve the limitations of current air pollution modeling revealed in this study.

Projections of the global average CO response to emissions changes in the year 2030 for three different scenarios are quite consistent among the models. Global average CO around the middle of the lower atmosphere increases by 16±4% for a high-emissions scenario, decreases by 11±3% for a low-emissions scenario, and changes by 2±2% for a mid-range scenario. Projected 2030 climate changes play a smaller role, decreasing the global average CO by 2±2%. Local changes can be much larger, however. These range from decreases of 10-15% over much of the industrialized Northern Hemisphere for the mid-range scenario to CO increases worldwide under the high-emission projection, with the largest changes over central Africa (20-30%), southern Brazil (25-40%) and South and East Asia (20-50%). The trajectory of future emissions thus has the potential to greatly influence air quality over most of the world's highly populated areas.


Shindell, D.T., G. Faluvegi, D.S. Stevenson, M.C. Krol, L.K. Emmons, J.-F. Lamarque, G. Pétron, F.J. Dentener, K. Ellingsen, M.G. Schultz, O. Wild, M. Amann, C.S. Atherton, D.J. Bergmann, I. Bey, T. Butler, J. Cofala, W.J. Collins, R.G. Derwent, R.M. Doherty, J. Drevet, H.J. Eskes, A.M. Fiore, M. Gauss, D.A. Hauglustaine, L.W. Horowitz, I.S.A. Isaksen, M.G. Lawrence, V. Montanaro, J.-F. Müller, G. Pitari, M.J. Prather, J.A. Pyle, S. Rast, J. M. Rodriguez, M.G. Sanderson, N.H. Savage, S.E. Strahan, K. Sudo, S. Szopa, N. Unger, T.P.C. van Noije, and G. Zeng 2006. Multi-model simulations of carbon monoxide: Comparison with observations and projected near-future changes. J. Geophys. Res. 111, D19306, doi:10.1029/2006JD007100.

Note: PDF documents require the free Adobe Reader or compatible viewing software to be viewed.