Current concerns about future climate change are driven in large part by the observational evidence that several long-lived greenhouse gases are increasing at significant rates. However, the detailed biogeochemical and physical knowledge of individual sources and sinks needed to explain quantitatively the greenhouse gas trends, and to project them accurately into the future, is lacking.
For this reason, the sources and sinks of the greenhouse gases, and the potential feedbacks involving changes in these sources and sinks in response to changes in climate, are either ignored or oversimplified in current general circulation models used for climate prediction. Similarly, the complexity of the physics and chemistry of atmospheric aerosols, and the lack of observations of them over the globe, has meant that aerosols, which can cool the Earth by reflection of sunlight, are also poorly simulated in climate models.
Biogeochemistry of Greenhouse Gases and Reflective Aerosols
Although there are now quantitative constraints on the global magnitude of the emissions of the greenhouse gases, uncertainties remain regarding the relative importance of their individual sources: enteric (cattle, etc.) and soil (rice, etc.) methanogens, biomass burning, fossil fuel mining, clathrates, etc. for methane (CH4); soil and oceanic denitrifiers, biomass and fossil fuel combustion, etc. for nitrous oxide (N2O). Knowledge of the nature and magnitude of individual sources, and the control mechanisms operating within each source, is essential for reliable prediction of future emissions and for inclusion of potential feedbacks in climate models. Examples of potential feedbacks are the known sensitivity of N2O emissions to rainfall, and of wetland CH4 emissions, CH4 clathrate decomposition, and net peatland carbon storage to temperature increases. Similar questions arise about the sinks and associated feedbacks for these gases.
Because of the spatial scale and heterogeneity of greenhouse gas sources over the globe, it is impractical if not impossible to deduce accurately the relevant regional or global greenhouse gas source strengths by direct measurements of emissions from each source all over the globe. For this reason selected in situ measurements are used to place constraints on the magnitude and geographical extent of individual sources, and then the deduction of present emissions (or sinks) of the greenhouse gases is approached as an inverse problem.
This methodology has been applied to the trace gases measured in an MIT-led global network (ALE/GAGE/AGAGE), and has produced, among other things, the constraints on global CH4 and N2O emissions mentioned above. However, to gain the detailed knowledge of present emissions necessary to include the sources and sinks of trace gases in climate models, more information is necessary.
In addition to determination of emissions, the processes controlling trace gas fluxes from ecosystems, and hence the magnitude and direction of potential feedback mechanisms, need to be understood. For this purpose, detailed process-oriented studies within ecosystems are needed. Gas fluxes must be understood in terms of microbial trace gas production (or consumption) and the physics of transport in soils and plants. Net gas fluxes from ecosystems depend on the interaction of biological and physical processes, each of which can be strongly influenced by climate change. CGCS researchers are working to improve quantitative knowledge of the magnitude and global distribution of the major sources of the greenhouse gases for realistic incorporation into climate models.
The development, now completed, of a realistic three-dimensional tracer transport model based on observed winds enables a major improvement in the inverse methodology that is used to determine the magnitude and geographical distribution of greenhouse gas (e.g., CH4 and N2O) sources. This will allow full utilization of the available global measurements of concentrations and isotopic composition, current knowledge of source location and isotopic signatures, and current knowledge of atmospheric chemistry. An additional hurdle, apart from the construction and validation of the model, is to devise methods that relate measurements at one locality to emissions from another locality, which can be computed without resort to prohibitively long computer runs.
CGCS researchers have developed global biogeochemistry models for natural surface sources and sinks of several trace gases, which are necessary for handling non-anthropogenic emissions. These natural emissions models are incorporated into a combined chemistry-climate model, and driven by climate variables (e.g., rainfall and temperature from observations or a climate model) and by inputs (e.g., soil properties) from a terrestrial ecosystems model or from observations. This latter work contributes to the MIT Joint Program on the Science and Policy of Global Change.
Ecosystem, Oceanic and Urban Emission Measurements
CGCS researchers are investigating the mechanisms of trace gas release from wetland and peatland ecosystems, which may be the largest non-managed current methane source to the atmosphere. This investigation requires the use of improved in situ techniques to quantify trace gas distributions within peatlands without creating artifacts through disturbance. A promising tool for this is a self-contained battery-operated portable mass spectrometer, a prototype of which was developed at MIT. Complementing this is an effort to determine soil to atmosphere transport mechanisms and rates (particularly the ebullition process) and relate them to climate.
Similarly, the development and testing of portable gas chromatograph systems, which are set up for a period of a year or two in the vicinity of major trace gas source regions, provides measurements sensitive specifically to the emissions from these regions. Locating the station such that some of the time it receives “polluted” air from the source region and at other times it receives clean “background” air should lead to a considerable decrease in the uncertainty of deduced regional emissions. Work is underway using this method for hydrocarbon emissions from natural ecosystems and hydrocarbon and halocarbon emissions from urban areas. CGCS researchers have also worked for several years using similar gas chromatograph systems to measure hydrocarbon fluxes from the world’s oceans.
Local Air Pollution Linked to Global
Failure to consider the interaction between global and local aspects of anthropogenic emissions may prove to be a serious shortcoming in analyses of global climate change because these aspects of environmental policy are linked in important ways: through the chemistry of the atmosphere, through the technology of emissions control, and through the interaction of local pollution policies with greenhouse gas control measures. A key difficulty in utilizing the understanding of physical and chemical transformation processes responsible for the formation of urban, regional and global scale air pollution is that the spatial extent of even major cities is much smaller than the resolution of the current generation of global climate models. In fact many cities may fall within a single grid cell. Since the emissions from cities undergo chemical transformations, from the time they are released until they have dispersed to scales comparable to those used as inputs to the climate models, the effective “source” term needs to reflect these subgrid scale processes.
To address this issue, CGCS researchers are studying the photochemical reactions taking place during long range transport to develop the time scales for chemical oxidation, and the form of the reaction products likely to be of climatic importance. As an illustration, most of the emissions of nitrogen oxides in urban areas are in the form of nitric oxide (NO) while at regional to global scales much of the NO will be oxidized to other products that in turn are used as inputs to the global chemical model. The operational goal is to develop a simplified representation of how to embed the effects of chemical transformations. By using urban-regional models, spatially averaged source terms are being developed for climate models to take into account the effects of pollutant transport from urban areas.