ALBERT

Description: SUMMARYOptimum applications of N fertilizer, Nopt have been related successfully to the amount of mineral N in the soil, Nmin in some parts of Europe but not always in the UK. If there is a body of mineral N, QN, that ultimately lessens the need for N fertilizer, it will not remain constant in its amount or its position. Mineralization will add to QN, while the nitrate component of QN will be leached downwards.Also, part of QN will be taken up into the crop where it will continue to lessen the need for fertilizer N but will be safe from leaching. A computer model was used to simulate these processes for 23 experiments, covering five sites and five years, in which N opt had been estimated. From these simulations we derived trial values of QN that took account of mineral N to a series of depths on a series of dates. For each date we used the trial values to find the depth for which Nopt was best correlated with QN andassumed that this was the depth, dL, of the lower boundary of QN on that date. Thus dL was a collective value for all 23 experiments. The value of dLincreased throughout the winter and the spring and was very closely related to the cumulative average drainage through 0·5 m soil at Rothamsted. By 15 April, dL, was 1·66 m, a depth that was compatible with observations by others that winter wheat can remove mineral N to a depth of at least 1·5 m. We inferred two likely reasons why Nmin may fail as a predictor of Nopt in the UK: insufficient depth of sampling, and too wide a spread of sampling dates. The values of Nopt were shown to be related satisfactorily to the values of QN computed, without any measurements of mineral N, for appropriate depths on single dates.