Assessment of economically optimum application rates of fertilizer N on the basis of response curves

Abstract

N-response curves of numerous N-fertilizer trials with sugar beet and potato are described by a quadratic and a modified exponential equation. For both sugar beet and potato the modified exponential equation was much better than the quadratic equation when the residual sum of squares (RSS) was taken as a measure of the degree of fit. In order to take into account the few occasions when the quadratic model was superior, it is suggested that both models should be used for the data of each individual trial. The economically optimum application rate of fertilizer N is calculated on the basis of the best-fitting model. This procedure yielded optima which covered entire ranges of fertilizer-N levels tested: 0–250 kg ha−1 for sugar beet and 0–400 kg ha−1 for potato. The magnitude of the confidence intervals (p > 95%) of the optimum N-fertilizer application rate frequently was very high. In 46% of the sugar beet trials and even in 60% of the potato trials it was higher than 300 kg ha−1 N. It is suggested that N-fertilizer recommendations be drawn up only with reliable optima.