Journal of Water Resources Planning and Management

Dealing with significantly large design problems for water-resource systems a mixed optimization procedure based on network linear programming and the subgradient method will be described. Using a linear problem formulation, the procedure uses network linear programming as a subproblem that assumes the knowledge of design variables. Since inside its domain, the global objective functions is a convex piecewise linear function, a subgradient method is used to obtain the direction of the improvement of design variables at each iteration using the solutions of the network subproblem. The mixed technique permits an efficient evaluation of the design variables in order to reach a good approximation of the global objective function optimum. The solution technique performs well in the purely linear case and, moreover, allows some kinds of nonlinearities in the cost functions of design variables.

Tank irrigation has been in existence for centuries. However, optimization of the tank irrigation system is of recent origin. Even this has been attempted in piecemeal, optimizing every segment/aspect of the irrigation system individually. Thus, although each component may be optimal, the system as a whole may not be optimal. This paper presents a methodology to optimize the design of the tank irrigation system by taking into account the constraints, objectives, options of the location, conveyance and application methods, crops and their irrigation levels, productivity, demand of the population dependent upon that area in terms of amount of foodgrains, pulses, oilseeds, nutrients, and the availability of different inputs to achieve an optimal design of all components of the system.

A multicriterion decision-making (MCDM) methodology is developed to design a long-term water supply system involving multiple actors. A 1989 regional development plan for augmenting water resources in the Adour-Garonne Basin (southwestern France) is used to illustrate the approaches. The specific problem of the choice of a site for building a large reservoir is investigated. The problem elements consist of four actors, 13 criteria, and 38 alternatives. The first phase of the methodology, the ELECTRE III technique, is used to screen alternatives and reduce the choice set to eight “efficient” alternatives, which are then examined in some detail. In the second phase, the multiactor multicriterion decision process is modeled by means of an extension of ELECTRE III. The solutions thus obtained are compared to the outcomes of the actual decision process. The similarities and differences between the model and reality are used to assess the suitability of that approach.

The writers present a framework for increasing the effectiveness of remedial design decision-making at ground-water contamination sites where there is uncertainty in many parameters that affect remediation design. It is specifically designed for broad, “big picture” analyses, such as in the preliminary stages of remedial design. The presented framework is used to (1) select the best remedial design from a suite of possible ones; (2) estimate if additional data collection is cost-effective; and (3) determine the most important parameters to be sampled. The framework is developed by combining elements from Latin-Hypercube simulation of contaminant transport, economic risk-cost analysis, and regional sensitivity analysis (RSA). The framework is demonstrated using a hypothetical contamination problem where radionuclide strontium ( ⁹⁰ Sr) is leaching from a trench into the ground water. Three remediation design alternatives are considered: monitoring only, isolating the source trench, and installing a plume containment and treatment system. Uncertainty in remediation design performance is due to uncertainty in 13 flow and transport parameters including hydraulic conductivity and source strength. The methodology can be applied to a variety of remediation problems.

A Virtual Runoff Hydrograph System (VROHS) based on artificial neural network technology was designed and developed to generate runoff hydrograph. Data from 45 lab experiment sets were used to develop the VROHS. A recurrent back-propagation neural network was trained to generate runoff hydrograph. Twenty-nine of the 45 lab experiment sets were randomly selected to train the network, while 16 experiment sets were selected to test the VROHS. It was found that the VROHS could predict the runoff hydrograph system very accurately for sets of input data (experimental conditions) that it had never seen before. The values of the correlation coefficients and coefficient of determination for the testing sets ranged from 0.96 to 0.99 and 0.92 to 0.99, respectively. These high coefficient values demonstrated the good correlation between the observed data and the predicted data and also the high performance of the VROHS.

A heuristic optimization methodology is developed for obtaining the optimal allocation of storage capacity and release rate for the design of storm-water detention facilities. The nonlinear optimization procedure is formulated within the framework of production function theory. The model is set up in terms of a nonlinear cost-objective function with a technological production constraint, which is the storage/release isoquant defining runoff volume or pollutant load trap efficiencies. The process constraint is expressed as a function of the design variables of storage capacity and release rate, in terms of runoff process statistical parameters and desired trap efficiency. The analytical structure of the model allows the use of marginal analysis in obtaining the least cost solutions of storage capacity and release rate for any level of pollutant trap efficiency. It also allows prompt assessment of sensitivity, resiliency, and variability of pollutant loading rates. The methodology is applied to a catchment in Minneapolis.

Lag-1 correlation for inflow between consecutive periods is considered into reservoir model in many publications. This consideration greatly improves the reservoir models to incorporate the uncertainty of inflow. However, no publication was found to incorporate a correlation relationship for releases between consecutive periods to improve the operating policy for reservoir operations. This paper presents a methodology of autoregressive decision rule for an aggregated reservoir operation as a surrogate of a multireservoir system of the Upper Colorado River Basin. The method incorporates a lag-1 correlation for the releases between consecutive periods with the optimal operating policy solved by stochastic dynamic program. The decision rules with and without incorporation of the autoregressive correlation for the releases were then used in simulated operation of the reservoir with historical inflow records to evaluate their relative effectiveness. The results showed that the autoregressive decision rule yields more stable and higher reliability of annual water supply for the aggregated reservoir operations.