Modelling nitrous oxide emission from water-logged soils of a spruce forest ecosystem using the biogeochemical model Wetland-DNDC

During the last decades, decision makers and policy have increasingly demanded for regional and national inventories of greenhouse gas emission, such as nitrous oxide (N2O), to develop appropriate strategies and mitigation options. A potential way to derive large-scale estimates of N2O emission is the use of process-based models, such as PnET-N-DNDC or Wetland-DNDC. While PnET-N-DNDC has been effectively applied for various upland forest ecosystems, the Wetland-DNDC model has not yet been validated with regard to N2O emission. We calibrated and validated the Wetland-DNDC model on the basis of a 4-year field data set of two water-logged soils (Humic Gleysol and Histic Gleysol) of a spruce forest ecosystem. Model calibration by means of the Levenberg-Marquardt algorithm considerably improved the model performance for the period of calibration (2001-2002). The error variance was reduced by up to a factor of two and the modelling efficiency was increased from -1.24 to -0.15 (Humic Gleysol) and from -0.42 to 0.1 (Histic Gleysol). However, the model performance for the period of validation (2003-2004) and particularly for the extreme dry period in summer 2003 was not fully satisfying, notably with regard to the temporal pattern of the N2O emission.