Design of Agriculture Drainage Under Uncertainty: 2. A Stochastic Programming Approach

The effect of soil property uncertainty on drainage system design was presented in the first of a series of papers on methods for optimal design of agricultural drains. A First Order--Second Moment (FOSM) approach was developed for the Hooghoudt steady-state drainage design ,equation to provide an estimate of the of the uncertainty of the dewatering zone between the drains as a function of the design variables and the uncertainty in the soil properties. In this paper, a Stochastic Programming Model for optimal design of drains under uncertainty, based upon the FOSM approach, is developed. The Stochastic Programming Model incorporates uncertainty in the objective function of the model as the expected loss in crop production as a function of uncertainty in the dewatering zone. The Stochastic Programming model is extended to included a multiple cropping situation and finally, the Chance Constraint approach, presented in the first paper, is compared with the Stochastic Programming Approach to drainage design and advantages of each are presented.