Evaluation of denitrification and decomposition from three biogeochemical models using laboratory measurements of N<sub>2</sub>, N<sub>2</sub>O and CO<sub>2</sub>

Abstract. Biogeochemical models are essential for the prediction and management of nitrogen (N) cycling in agroecosystems, but accuracy denitrification decomposition sub-modules is critical. Current were developed before suitable soil N2 flux data available, which may have led to inaccuracies how was described. New measurement techniques, using gas chromatography isotope-ratio mass spectrometry (IRMS), enabled collection more robust N2, N2O CO2 data. We incubated two arable soils – a silt-loam sand 34 58 d, respectively, with small field-relevant changes made control factors during this period. For soil, seven treatments varying moisture, bulk density NO3- contents included, temperature changing incubation. The sandy without incorporation litter (ryegrass), temperature, water content DeNi, Coup DNDC tested No systematic calibration model parameters conducted since our intention evaluate general structure or “default” runs. Measured fluxes generally responded as expected factors. assessed direction modeled responses three categories: no response, response same measurements opposite measurements. 14 %, 52 % respectively. 47 19 DeNi 0 67 33 magnitudes underestimated by overestimated while there trend soil. None able determine litter-induced correctly. To conclude, currently used not consistently simulate processes. better evaluation development, we need design experiments, take frequent measurements, use new updated address complexity, add missing processes models, calibrate denitrifier microbial dynamics, anaerobic volume concept.