Modifying the LEACHM model for process-based prediction of nitrate leaching from cropped Andosols

Nitrogen (N) management strategies for reducing the risk of groundwater contamination around agricultural fields require precise prediction of N leaching using a process-based model. We modified LEACHM model for use in Andosols, which are characterized by slow soil organic carbon (SOC) mineralization and nitrate adsorption. The modification was made with regard to the SOC mineralization of incoming plant-residue/manure and humus following the RothC model, as well as for nitrate adsorption. Empirical equations were employed to determine the parameters of the modified model. The ability of the modified LEACHM to predict N leaching was tested against existing data from a 4-year lysimeter study for cropped Andosol and sandy soils and compared with that of the original model. The modified model improved the prediction of leached N concentrations and the loss of N from Andosol with relative improvements of 63.5 and 76.5 %, respectively, over the original model, while retaining model applicability in sandy soil. This effective modeling was achieved by using precise predictions of N mineralization in the humus pool along with SOC mineralization processes that were based on the RothC model. The modification extended the applicability of LEACHM and may provide better N management strategies for reducing leaching from cultivated Andosols.