Evaluation of LEACHMN for simulating seasonal changes in plant available nitrogen across a variable landscape

A model that accurately simulates the seasonal variation in nitrogen (N) dynamics may represent an important additional tool in N management. The objectives of this study were to evaluate whether seasonal changes in the amount of N that becomes available to a maize crop in a variable landscape can be simulated with LEACHMN and, in particular, to assess the potential value of LEACHMN in estimating N available at the time of the presidedress soil nitrate test (PSNT). Soil mineral N (SMN) and shoot N were measured biweekly over seven growing seasons in corn (Zea Mays L.) grown under conventional tillage in a variable landscape in Southern Ontario, Canada. The model was calibrated using data from 2002 and 2003 from each of five positions in the landscape and then evaluated using data from the 1997–2001 growing seasons. Although the model under-estimated SMN and over-estimated shoot N, the model was more successful in simulating the sum, defined as plant available N (PAN). Simulations of PAN were best at the summit and shoulder positions. The agreement between measured and simulated PAN were poorest early in the season. Although the accuracy of PAN simulations late in the growing season indicates this model has potential value in N management decisions, the errors in simulating SMN early in the season suggest adjustments are required before it can be used, along with other tools, as a substitute for the PSNT in cool humid environments. Key words: Soil mineral N, Plant available N, variable landscapes, LEACHMN