A Web-Based Model of N Mineralization from Crop Residues.

Cover crops can provide substantial quantities of N for subsequent crops, but estimating the amount of N that will be mineralized from residues is challenging because of the complexity of the processes and the variety of factors involved, including residue quality, temperature, water content, and soil characteristics. Simulation models of N mineralization have been shown to provide reasonable estimates of N mineralized as crop residues decompose. To improve the reliability of simulations further, model inputs could be improved by linking models to more specific soil water and soil temperature data from weather stations, and by residue testing for better input for model simulations. Our objectives were to describe a web-based N mineralization model and its operation, calibrate the model with results from published laboratory studies, and validate it with field studies conducted with surface-applied or incorporated crimson clover (Trifolium incarnatum L.) or rye (Secale cereale L.) during three years.  Modifications to rate constants and some parameters led to good fit between estimated and measured values in the laboratory studies used for calibration. In the field studies used for model validation, net N mineralized in 120 days ranged from 42 to 83 kg N ha^-1 during the three years. Simulated values of net N mineralized were within the 95% confidence intervals for observed values for incorporated residues, but tended to overestimate observed values for surface-applied residues. Thus, the model performed well for incorporated residues, but additional work is needed to improve its performance for surface-applied residues.