0 avis
Long-term modelling of soil N mineralization and N fate using STICS in a 34-year crop rotation experiment
Archive ouverte : Article de revue
International audience. Net soil N mineralization is a driver for N uptake and N losses at an annual scale, but is itself dependent on long-term N surplus and C-N storage in agricultural systems. The accurate modelling of N mineralization remains challenging. Thus, the STICS research version V1610 that includes modified soil organic nitrogen (SON) mineralization and root biomass turnover modules was assessed in this study regarding its predictions of net N mineralization and long-term N fate in a 34-year experiment comparing crop rotations with or without catch crops (CC) and bare soil. The in situ gross balance method was used as a reference to estimate net N mineralization based on measured N variables (i.e. N uptake, exported N and N leaching). The Index of Agreement (IA) of STICS predictions concerning crop biomass, crop yield, N uptake and exported N ranged between 0.61 and 0.76, 0.79-0.89, 0.49-0.64 and 0.47-0.58, respectively, depending on the crop rotations. STICS also enabled a good simulation of annual drainage and N leaching with IA ranges of 0.92-0.96 and 0.78-0.93, but high leaching values were not captured by the model. The STICS research version simulates the decay of deep roots (below a depth of 25 cm) but it neglects their decomposition. This simplification could cause an underestimation of N leaching. The observed N surplus ranged from 27 to 51 kg N ha(-1) yr(-1) in the cropped rotations depending on the crop rotations, and the N surplus was accurately simulated with an IA of 0.75-0.84. STICS produced a good prediction of changes in SON stocks under cropped rotations and bare soil, with both the rRMSE and rMBE being lower than 10%. Estimated mean annual N mineralization was 115 kg N ha(-1) under cropped rotations and 42 kg N ha(-1) under the bare soil treatment. STICS relatively well predicted net N mineralization regarding both differences between crop rotations and over time. Moreover, STICS correctly simulated the long-term effects of CC on drainage, N leaching, SON accumulation and net N mineralization. To conclude, STICS is a useful model to predict net N mineralization and N fate in long-term crop rotations. Moreover, this work raised new questions concerning the long-term fate of N stored in deep dead roots. Further improvements to describe the fate of these residues should enhance the prediction of N leaching by the STICS model and enable the optimization of N management in cropping systems.
Consulter en ligne
Suggestions
Du même auteur
Du même sujet
