CERES-N MODEL PREDICTIONS OF NITROGEN MINERALIZED FROM COVER CROP RESIDUES

Winter annual cover crops, widely used in no-till systems, can be an i mportant source of N for the subsequent crop. Because many factors aff ect net N mineralization from cover crop residues, computer models can be powerful tools to predict it. The CERES models, which are some of the most widespread models for simulating the whole crop-soil system, have a common submodel (CERES-N) that describes N transformations. The objectives of this study were to determine decay rate constants under nonlimiting conditions for the carbohydrates and cellulose pools (CAR E and CELL) of CERES-N for residues that decompose on the soil surface , and to test if two modifications to CERES-N could improve the simula tion of N mineralization. The two modifications were to: (i) allow the user to vary the relative size of the residue pools (CARE, CELL, and lignin), and (ii) allow stems and leaves to decompose separately, havi ng a common point of interaction through the inorganic N pool. Results of a 6-mo laboratory incubation experiment with four cover crop resid ues were used to adjust rate constants and test the effect of model mo difications. The decay rates obtained were 0.14 and 0.0034 d(-1) for C ARE and CELL, respectively. Allowing the user to vary the relative siz e of residue pools greatly improved the simulation of net N mineralize d (root mean square error of the model decreased from 1.0 to 0.28 g m( -2)), whereas modeling the separate decomposition of leaves and stems only caused a slight improvement in the prediction of net N mineralize d.