Improved global estimation of N2O emission from nitrification using a process-based model and meta-analysis

Abstract:

Nitrous oxide (N₂O) is recognized as one of the important anthropogenic greenhouse gases and also stratospheric ozone-depleting substances. Natural soil and fertilizer in croplands are large sources of N₂O to the atmosphere, and so their global estimation is of great importance for prediction and mitigation. Two major processes of N₂O production, i.e. nitrification and denitrification, were identified and parameterized using the leaky-pipe scheme. However, their biogeochemical mechanisms are very complicated and it is difficult to fully include them into process-based models, leading to substantial estimation uncertainty. In croplands, estimation is even more difficult because it depends on the amount, composition, and timing of nitrogenous fertilizer input. In this study, we focused on N₂O emission from nitrification, which can play a dominant role in aerobic soils. In the present terrestrial nitrogen cycle models, fractionation of nitrogen in gross nitrification is parameterized in very simplified manner. First, we conducted a meta-analysis of the fraction of N₂O emission from gross nitrification in relation to temperature, moisture (water-filled pore space), and pH, for a wide variety of soils including arable lands. We obtained typical values of N₂O emission ratio from gross nitrification for different soil textures. Second, we assessed sensitivity of N₂O emission to the assumption of nitrification fractionation, using a process-based model VISIT. We changed the value of N₂O fraction in gross nitrification in different manners: default constant ratio, meta-analysis-based ratio, and parameterizations used in other models (DNDC and DLEM). We investigate how much the assumption of nitrification N2O emission affects global estimation of N₂O budget and other parts of nitrogen cycle.