Previous PageTable Of ContentsNext Page

Chapter 6

Nitrous oxide and nitric oxide emissions

The major factors controlling N2O emissions include those that regulate N availability, i.e. N application rate, climate and soil C content. The N application rate determines directly N availability. Soil C content is a proxy for soil fertility, while climate influences the speed of soil processes and plant / microbe competition through temperature and moisture. It is apparent from studies in the tropics that plant uptake can exceed microbe competition, therefore denitrification rates can be relatively low even in warm, moist conditions.

Factors influencing the relative rate of N2O compared to total denitrification loss are soil texture, drainage and pH. Fine soil texture, poor drainage and neutral to slightly acidic soil reaction are conditions that favour N2O emission. Management factors that influence N2O emission are crop type (with N2O emission increasing in the order: grass, non-leguminous annual crops, leguminous crops) and type of fertilizer (with the highest values being for anhydrous ammonia and mixtures of mineral and organic fertilizers).

The major factors regulating NO emission are: N application rate, fertilizer type, soil organic C content and soil drainage. Similar to N2O, NO emissions increase with N application rate and soil C content. However, contrary to N2O, NO emissions from well drained exceed those from poorly drained soils. The factors relating to climate and length of measurement period do not have a significant influence on NO emissions.

Measurement techniques are very important in assessing gas flux measurements. Closed and open chamber techniques yield about equal emission estimates for N2O and NO. However, other techniques yield markedly different emission estimates, particularly for N2O. The data show that N2O emission estimates in agricultural fields should be based on measurements with a frequency of at least one measurement per day in periods with high emission rates (e.g., periods following fertilizer application or rainfall events) and with lower frequencies in periods with low emission rates, extended over periods of at least one year. For NO shorter measurement periods may yield most of the effect of N fertilization because the factor relating to the length of measurement period does not have a significant influence on NO emissions. Hence, N fertilization may have a more long-lasting effect on N2O than on NO emissions.

The estimated global annual N2O-N and NO-N emissions from crop and grassland amount to 3.5 million t of N2O-N emission and 2.0 million t of NO-N emission. The fertilizer induced emissions for N2O and NO amount to 0.8 and 0.5 percent, respectively. The global estimates presented in Chapter 4 account for the major controls of N2O and NO emissions at the landscape-scale from arable land and grassland.

Ammonia volatilization

Although it includes a large number of measurements from the literature, the data set is probably not a fair representation of environmental and management conditions found in the real world. This is because in many cases the measurements were carried out under conditions or in places prone to high N gas losses. Therefore, the report adopted a statistical approach in an attempt to avoid this problem of lack of representativeness.

The comparison of straightforward means and the balanced medians resulted in the selection of the most important regulators of NH3 volatilization. The results are in good agreement with European and global inventories of NH3 volatilization from fertilizers, except for ammonium sulphate, ABC and the different compound NP fertilizers. The data in this report indicate that these fertilizers are more prone to NH3 volatilization than previously thought.

Although the nitrogen fertilizer statistics used may be reliable, the analysis presented in this report involves many uncertainties (Chapter 3). Moreover, the timing and mode of fertilizer application have a strong influence on NH3 volatilization loss. The greatest uncertainty stems from underrepresentation in the data set of measurements of NH3 volatilization rates in tropical cropping systems.

Finally, the regression model developed in the report is a representation of the literature data available. Many factors that are known to be crucial controls of NH3 volatilization, e.g. wind speed, and floodwater pH in flooded systems, could not be included in the regression model. This is because such data are simply not available on the global scale. Therefore, it is difficult to indicate the uncertainty in the prediction of NH3 losses.

Perhaps the major uncertainty is in the omission of the effect of rainfall on NH3 volatilization. In practice, farmers apply fertilizers in periods just after rainfall events, or in periods during which they expect rainfall. The omission of rainfall events may lead to a systematic overestimation of NH3 volatilization losses. However, it is difficult to estimate the uncertainty caused by this omission.

The uncertainty of the results of the summary model is about ±30 percent. The uncertainty for individual fertilizer types depends on the number of measurements available. On a grid cell basis the uncertainty is probably greater because the allocation of arable lands and grasslands is based on statistics combined with land cover maps and land suitability. In particular the allocation of the intensively used grasslands and fertilized grasslands was based on very simple assumptions.

The outcome of the model shows that the potential impact of fertilizer use regulations would be modest from a global emission perspective. The quantities involved, however, constitute a valuable plant nutrient source. Farmers' ability to curtail such losses will primarily relate to economic incentives, in particular in South and South East Asia.

Previous PageTable Of ContentsNext Page