Predicting nitrogen leaching with the modified LEACHM model: validation in soils receiving long-term application of animal manure composts

Abstract

A variety of process-based models have been developed for predicting nitrogen (N) dynamics in agro-ecosystem; however, no reliable models have been validated for N leaching from soils receiving a long-term application of different types of animal manure composts. The Leaching Estimation and Chemistry Model (LEACHM) was recently modified by incorporating the basic structure of Rothamsted Carbon Model for extending its ability to describe soil organic matter decomposition and subsequent N leaching in soils rich in organic matter. We evaluate the applicability of the modified LEACHM in cropped Yellow soils receiving 10-year application of cattle or swine manure compost in addition to chemical fertilizers, where high-frequency field monitoring data of soil water contents, soil N contents and leachate N concentrations were available for the last 3 years. Particular attention was paid to determine all input parameters from independent measurements, parameterization from known soil properties or databases without optimisation to fit the measured field data. The model reasonably predicted temporal changes in the soil NH4-N and NO3-N contents, and inorganic N concentrations in the leachate as well as their differences due to different manure compost/chemical fertilizer applications. The simulations of leached N concentration yielded a Willmott index of agreement (IA) of 0.62–0.68, with those for soil moisture, soil nitrate content and crop N uptake all within an acceptable IA range. In view of the good performance without site-specific calibrations, the modified LEACHM appears to be a valuable tool for predicting N leaching from cropped soils receiving long-term manure compost applications.