Why register for an account?

For many cropping situations, especially in developing countries, quadratic surfaces do not fit the responses of certain crops to fertilizer. Use of the second order designs with standard statistical and economic interpretive techniques may result in costly biases in the estimates of the optimal fertilizer rate. Also, there is a potential pollution problem. A family of linear-plateau models, consisting of intersecting straight lines, is proposed for fitting fertilizer response data which exhibit a plateau effect. The regression coefficients are easily computed using a desk calculator or computer, and the economic interpretations are simple. Techniques for fitting, parameter estimation, and economic interpretation are described. For multi-nutrient experiments, a complete factorial experiment with a number of levels of each nutrient is considered to be the best design for both evaluating the model and then estimating the optimal nutrient levels. Preliminary information may provide a basis for deciding which fertilizer nutrients are apt to produce response. In many soil-crop situations, only NP or N experiments are required, because the other nutrients are already at adequate levels; hence, the amount of experimental material may be redistributed by having fewer factors, but more levels of each factor studied. Two currently used fertilizer response designs, based on preliminary information on optimal nutrient levels, are described; a one-factor-at-a-time design has the disadvantage of providing no estimate of interaction. Several other designs are suggested. We recommend concentrating several treatment levels in the vicinity of the anticipated optimum. Since the sloping phase of the response pattern is more important than the plateau phase, it should receive more attention when distributing treatment levels.