IGAC | Where Does Nitrous Oxide Come From?

Contributed by M. Keller, U.S.D.A. Forest Service, Puerto Rico
Reprinted from IGACtivities Newsletter No. 2, September, 1995.

[Editor's Note: BATGE is IGAC's "Biosphere-Atmosphere Trace Gas Exchange in the Tropics: Influence of Land Use Change" Activity.]

In many ways, nitrous oxide is a forgotten greenhouse gas. Its rate of increase, 0.3% per year during the 1980's, does not sound terribly shocking. Its sources are poorly known and the prospects for "instant gratification" following possible controls of nitrous oxide emissions are small because of its 150 year atmospheric lifetime. From another vantage, the long atmospheric lifetime of nitrous oxide gives each additional molecule emitted to the atmosphere a large global warming potential. The effects of increasing concentrations of atmospheric nitrous oxide will be with us for a long time.

Our ignorance of nitrous oxide sources and sinks is troubling. The current budget is seriously unbalanced. Sources (14 Tg-N/yr; 1 Tg = 1 teragram, or 1 million metric tons) exceed sinks (10 Tg-N/yr) by 40%. Data from polar ice cores show that nitrous oxide concentrations in the pre-industrial atmosphere hovered around 280 ppb. Today the concentration exceeds 310 ppb. Where does the excess contribution to the nitrous oxide budget come from? Based on numerous measurements, we know that fossil fuel combustion cannot be a major source. Other industrial processes such as the manufacture of adipic acid (for nylon) and nitric acid contribute perhaps one eighth of the imbalance.

Looking to the natural biological sources of nitrous oxide gives us strong hints about where to find the perturbed sources. Nitrous oxide is formed by microbial processes. For soil processes which produce about two thirds of the natural nitrous oxide, the "Hole-In-the-Pipe" conceptual model (Figure 1) relates the total amount of nitrous oxide released through "leaks" in the pipe directly to the overall "flow" of nitrogen (N) through the pipe. The size of the leaks may be controlled by a number of soil properties, chief among them soil moisture content. From our understanding of terrestrial ecology, we know that most temperate ecosystems are nitrogen poor. In contrast many natural tropical ecosystems are nitrogen rich. And recent budgets suggest that while unmanaged temperate and boreal ecosystems produce about 1.2 Tg nitrous oxide-N annually, unmanaged tropical ecosystems produce 4 times as much: 4.8 Tg annually.

Figure 1. The "Hole in the Pipe" conceptual model indicates the flows of inorganic nitrogen through the microbial processes of nitrification and denitrification. Nitrogen oxides escape through "leaks" in the pipe. (Adapted from Firestone and Davidson, 1989).

What happens when these ecosystems are disturbed? Disturbance of temperate ecosystems generally does not lead to large releases of nitrous oxide because nitrogen is in short supply. In contrast, large nitrous oxide emissions have been observed following natural (e.g., hurricanes) and anthropogenic disturbances of tropical forest. Figure 2 shows findings from a study of nitrous oxide release from a sequence of pastures representing various times following deforestation. Because forest to pasture conversion is a dominant land-use in tropical America, this effect may account for a tenth of the annual global imbalance of nitrous oxide.

Figure 2. Emissions of nitrous oxide from soils along a chronosequence following the conversion of forest (0 years) to pasture through secondary succession in the Atlantic Lowlands of Costa Rica. (Adapted from Keller and Reiners, 1994).

While forest disturbance is important, we need to look for higher flows in our plumbing if we are going to find a big enough leak. Big flows of nitrogen can be found wherever farmers are applying nitrogen fertilizer to fields and also where domestic animals are kindly returning much of what they take in. The annual use of nitrogen fertilizer, about 80 Tg-N, is now greater than natural biological nitrogen fixation. It has been known for two decades that fertilizer use increases nitrous oxide emissions from farm soils. Fertilizer use in the developed world appears to have reached a plateau. In contrast, in the developing world (a near synonym for the tropical world), fertilizer use nearly doubled during the 1980's. Moreover, we have indications that the same amount of nitrogen fertilizer yields more nitrous oxide under tropical conditions than under temperate conditions.

Is the world faced with agonizing decisions pitting food production against atmospheric composition change? Not necessarily. Through improved fertilizer management, we may be able to readjust the nitrogen plumbing with the help of plants. We need to design agronomic systems for the tropics that reduce the nitrogen flow through the pipe by directing more fertilizer to the crops and less to the microbes that make nitrous oxide. Results from studies of sugar cane in Hawaii and wheat in Mexico suggest that careful fertilizer management can significantly limit emissions of nitrous oxide while at the same time cutting fertilizer costs and increasing crop yields. Reduced nitrous oxide emissions to the atmosphere may someday be the result of richer harvests by farmers in the tropics.

Selected References:

Bouwman, A.F., K.W. Van der Hoeck, J.G.J. Olivier. 1995. Uncertainties in the global source distribution of nitrous oxide. J. Geophys. Res., 100, 2785-2800.
Firestone, M.K. and E.A. Davidson. 1989. Microbiological basis of NO and nitrous oxide production and consumption in soil, in Exchange of Trace Gases between Terrestrial Ecosystems and the Atmosphere, edited by M.O. Andreae and D.S. Schimel, pp. 7-21, John Wiley & Sons, New York.
M. Keller and P.A. Matson. 1994. Evaluating the effects of tropical land use changes on atmospheric composition, pp. 103-118, in R.G. Prinn (ed.) Global Atmospheric-Biospheric Chemistry, Plenum Publishing Company, New York.
M. Keller and W.A. Reiners. 1994 Soil-atmosphere exchange of nitrous oxide, nitric oxide, and methane under secondary succession of pasture to forest in the Atlantic lowlands of Costa Rica. Global Biogeochem. Cycles, 8, 399-409.