In many parts of the world, emissions of carbon monoxide (CO) are regulated because high concentrations in cities are a threat to human health and add to air pollution. CO is present at such low concentrations in less populated areas that it has no known health or direct environmental effect.
by M. A. K. Khalil, Department of Physics, Portland State University, Portland, Oregon
The concentrations and trends of carbon monoxide in the remote atmosphere are extremely important to understanding global atmospheric chemistry and environmental changes caused by human activities. If, for instance, the concentration of carbon monoxide increases, it may reduce the levels of hydroxyl radicals (OH), with which it reacts rapidly. Hydroxyl radicals can be thought of as the chemical scavengers of the atmosphere. These radicals react with many man-made and natural atmospheric gases to form intermediate compounds that are soon removed from the atmosphere by rain or deposition. Lower concentrations of OH could therefore lead to increases of trace gases that could in turn, cause global warming or destroy the ozone layer. A naturally high level of OH is also required to remove many hydrochlorofluorocarbons (HCFCs) and hydrofluorocarbons (HFCs) proposed as substitutes for the ozone-depleting chlorofluorocarbons. The fact that changes of carbon monoxide can affect the concentration of hydroxyl radicals, and therefore cause an indirect influence on the global environment is an important practical reason for studying its long- and short-term trends.
CO concentrations have been measured worldwide for the last 15 years. Reconstructions of solar spectra taken at astronomical observatories have been used to measure concentrations of CO since the 1950s. For earlier times there are no reliable data.
The concentration of CO changes with the seasons and over longer periods of several years (Figure 1). In the early 1980s the concentration of CO was increasing, according to our data. Spectroscopic data from the 1950s to the 1980s suggests the increase may have been occurring for many years. This trend, in turn, implied that OH concentrations may be decreasing. The increasing trends did not persist, however. Concentrations of CO started decreasing rapidly and most unexpectedly sometime in the late 1980s (Figure 1).
Fig. 1.
a) Seasonal average surface concentration of CO at various latitudes (polar, middle, and tropical latitudes of each hemisphere). b) Rate of change of CO over successive 7-year overlapping periods for the data shown in a). Values above zero represent increasing concentrations of CO, and values below zero represent decreasing concentrations of CO.
A gas can increase in the atmosphere if the emissions increase or the removal processes slow down. Increases or decreases in CO affect only the part of the world where they occur. The atmospheric lifetime of CO is short, an average of 2 months, so what happens in one part of the atmosphere does not greatly affect concentrations at distant locations. For instance emissions from automobiles in the northern middle latitudes have a minor effect on the CO concentrations in the Southern Hemisphere because CO does not survive the 1 to 2 year period it takes for air to mix between hemispheres. Also, because of the short lifetime, changes in sources or sinks are rapidly reflected in the atmospheric concentrations. Because of these properties, the trends observed at each location represent the imbalances of sources, which produce CO, and sinks, which use up CO, in regions surrounding the site but not necessarily global-scale processes. Consequently, there is probably no single explanation for the trends observed worldwide.
Various lines of reasoning suggest that the rising concentrations during the early 1980s were caused mostly by increases in worldwide emissions. The recent rapid decreases in the concentration of carbon monoxide have come as a surprise. One idea is that, as the stratospheric ozone layer thins due to chlorofluorocarbons, more UV can reach the lower atmosphere where it can stimulate the production of hydroxyl radicals. The increased OH would then consume more CO, causing decreases in the concentrations. This explanation is feasible in some regions over which significant ozone depletion has occurred, namely the high latitudes of both hemispheres. There has been no significant ozone depletion over the tropics. Wherever a significant fraction of CO comes from methane oxidation, more rapid oxidation of methane will increase the production rate of CO. The faster loss of CO from increased OH would be canceled by the greater production of CO from methane oxidation, thus resulting in a steady CO concentration.
In the middle and higher Northern Hemisphere, it is possible that the changes of hydroxyl radicals are contributing to the trends of CO. Here also it is probable that emissions are an important factor since anthropogenic sources have been decreasing continuously as more stringent controls are applied to automobiles and other urban sources.
In the Southern Hemisphere several factors suggest that the possible increase of OH is not the main reason for the rapid decreases being observed. One likely possibility is the reduction of biomass burning in the tropics during recent years. This could happen if the area of biomass burned annually has stopped increasing, or has declined. But even if the area burned has not declined, the amount burned may have, because of climatic variations, such as reduced rainfall that may reduce biomass productivity. Recently, several researchers have come to the same conclusions about decreases of tropical biomass burning using completely independent measurements of 13CH4.
At present, atmospheric concentrations of CO are declining. Whatever the specific causes may be, this trend is an indicator of important global changes probably caused by human activities and may reflect a slowdown in tropical biomass burning.
Acknowledgments: Samples for data in Figure 1 were collected by the National Oceanic and Atmospheric Administration/Climate Monitoring and Diagnostics Laboratory, Cape Grim Station, and CSIRO-Australia. Samples were analyzed at R. A. Rasmussen's laboratory at the Oregon Graduate Institute.
Return to Science and Society
Return to Starting Point
