Progress in source strength estimates of nitrous oxide (N_2O)

摘要

From the above, we drew the following conclusions: First, at NCGG-3 no new sources of N_2O were presented. Second, the IPCC Guidelines may be evaluated based on new information. The emission factors for mobile combustion need to be revisited. For agriculture we recommend that proposals for revisions of the IPCC Guidelines need to be carefully evaluated. We need to be sure that the proposed changes are real improvements and reduce the uncertainties in the national emission estimates. There may be reasons to evaluate emission factors for soils, as well as the method for indirect emissions. Third, we may consider process-based simulation modeling as an alternative to emission factor inventory methods. This requires a careful evaluation and inter-model comparison of model results. Process-based models may be used to evaluate the effect of emission reduction technologies. Fourth, there is a need for more experimental data in terrestrial and aquatic systems. Biogenic emissions of N_2O from terrestrial and aquatic systems are extremely variable in time and space, and influenced by many environmental and management factors. Smith & Dobbie (2002), for instance, observed that in the UK, annual emissions varied over 20-fold over 8 years at a grassland site, and by over 5 fold in the same year at grassland sites in different parts of the UK. This clearly illustrates the need for long term experiments at a number of representative sites. This holds for all world regions. Realizing that up till now most experimental research has been performed in terrestrial systems in temperate zones, there is in particular need for experimental work in tropical regions and aquatic systems. Finally, the most uncertain emissions are likely aquatic (indirect agricultural) emissions. The number of experimental studies in aquatic systems is limited. There is therefore a need for more data on aquatic emissions of N_2O, induced by agricultural activities on the land (the indirect emissions). These studies preferably take into account the relations between human activities in the watersheds, the landscape N flows and N_2O formation in groundwater, riparian zones, lakes, rivers, estuaries and coastal systems. Meanwhile, process-based models may improve our understanding of aquatic systems.