Figure 52 Scheme of the revised Daycent trace gas module

Note: D1, reduction of nitrate to NO; D2, reduction of NO to N2O; D3, reduction of N2O to N2; N2den, production of N2 from denitrification; N2Oden, production of N2O from denitrification; N2Onit, production of N2O from nitrification; NOden, production of NO from denitrification; NOnit, production of NO from nitrification. Source: Stehfest (2005)

models for simulating N2O emissions on a field to regional scale, other models have included algorithms to estimate N2O emissions. Examples of these models are EPIC, LPJ (Xu-Ri and Prentice, 2008), DRAINMOD-N (Youssef et al, 2005), NLOSS (Riley and Matson, 2000), WNMM (Li et al, 2005, 2007), FASSET (Chatskikh et al, 2005), and CERES-NOE (Gabrielle et al, 2006). A comprehensive review of field- to regional-scale N2O models has been carried out by Chen et al (2008).

Better performance of mechanistic models for N2O emissions at these scales is not only hampered by insufficient understanding of the various processes controlling N2O emissions, but also by lack of data on hydraulic conductivity and porosity, and their spatial heterogeneity, among other variables. As a result, the performance of the above mechanistic models remains rather poor, and uncertainty ranges are large, both for process-based models and for statistical approaches (Figure 5.3). The modelling efficiency or R2 correlations with measurement data are often not reported or average 0.5 or less when data cover periods up to the crop growing season; agreement tends to be poorer when models are evaluated with daily measurements (Del Grosso et al, 2000, 2009; Li et al, 2005). For the EF approaches, the uncertainty range remains large, for example from —85 per cent to +250 per cent (Stehfest and Bouwman, 2006).

In addition to the direct N2O emissions that occur in an agricultural field, all N leaching, NH3 volatilization and NOx emission from the field can eventually be converted into N2O in groundwater and surface water, leading to indirect emission of N2O. So far, there is no process-based modelling of indirect N2O emissions, and all estimates are based on the emission factor approach as outlined by IPCC (2006) (Table 5.1). Starting from the amount of N added to the system as synthetic fertilizer input, as organic N input from manure spreading, dung and urine excretion on the field and crop residues, or from mineralization of soil organic matter following a land-use change, the fractions that are volatilized and leached and the subsequent N2O production are given as generic factors. The N2O emissions from groundwater and surface drainage are not based on N input, but rather on the relative concentrations of N2O and dissolved inorganic N in the water. The EFs for indirect N2O emissions are based on a rather small number of measurements with an uncertainty range that often exceeds one order of magnitude (IPCC, 2006).

0 0

Post a comment