Journal of Water Resources Planning and Management

The determination of the design capacity and operation policy of water delivery systems for farming areas is complicated by the temporal variability of demands and the corresponding use of delivery capacity, the probabilistic occurrence of supplies, the spatial distribution of users, and losses in the distribution systems. This paper develops a general method to integrate the choices of delivery system capacity and of conjunctively operated local reservoir capacities (the decisions of delivery system operators and farmers) to make more efficient use of water resources. The method consists of using the results from local economic optimization models of farm water requirements and benefits in a discrete dynamic programming model of the delivery system to determine the most efficient method of distributing limited water supplies. In previous papers, the writers have worked with intrayearly probabilistic supplies and transportation losses. The main improvments here are that the method: (1) Allows variable peak‐delivery proportions; (2) includes local (on‐farm) intrayear storage; (3) is compatible with many other existing optimization models for farms; (4) addresses irrigation scheduling problems; and (5) addresses efficiency‐equity conflicts.

A statistical technique for delineating groups of stations to be considered a region for regional flood frequency analysis is presented. The technique, which utilizes cluster analysis, allows the inclusion of a diversity of factors which might be considered to be of relevance when seeking stations in differing basins having similar extreme flow characteristics. The method incorporates both a basin similarity measure, imbedded in the clustering algorithm, and a regional homogeneity measure used to evaluate station partitionings obtained from the clustering algorithm. The result is that groups of stations are identified that can be considered sufficiently homogeneous to effect an efficient spatial data transfer. An application of the methodology to rivers in southern Manitoba, Canada, is presented to illustrate pertinent aspects of the procedure.

Quantification of the relationship between stream discharge and the availability of suitable aquatic habitat requires detailed information on the values of local depths and velocities as they vary through riffles and pools. Conventional flow models are inadequate to supply the needed information. A basin model is presented which provides a new approach to basinwide flow modeling for stream habitat evaluation. Streams within a basin exhibit similar patterns of development, varying in scale with the drainage area. Average width, depth, and velocity increase in a predictable way with drainage area and discharge. Hydraulic geometry relations define the average values of width, depth, and velocity for any stream in the basin. Local variations of depth and velocity about the average are evaluated from probability distribution models developed from field measurements made throughout riffle and pool sequences. The basin flow model provides the needed hydraulic data to apply the Instream Flow Incremental Methodology (a state‐of‐the‐art methodology for stream habitat evaluation) of the U.S. Fish and Wildlife Service.

A conjunctive groundwater‐surface water planning model is presented. The optimization model maximizes the net revenue generated from the production of wheat, corn, and cotton over a one‐year planning horizon. The costs of production include distribution costs for Yellow River water, fertilizer, and nonlinear groundwater pumping costs. The decision variables of the model are the groundwater pumping schedules and the allocation of surface water, fertilizer, and land to the four drainage basins of the study area. Agricultural production functions are developed for the Yucheng County region of the North China Plain. The production functions relate yield per ha of wheat, corn, and cotton and the applied water and fertilizer. The groundwater hydraulic response equations are developed using the finite element method and the matrix exponential. The groundwater equations are imbedded as constraints in the planning model. Optimization results are presented for a range of hydrologic conditions. The results are compared with present allocation patterns in Yucheng County.

The use of graphical‐display training simulators is presented as a means of enhancing efficiency in operating multi‐reservoir water delivery systems. Water delivery systems of this type include municipal water supply systems and irrigation water delivery systems (ditch companies). As a stand‐alone tool, the training simulator can improve human operational performance. However, a coupling of human operational capabilities and mathematical programming techniques may also be beneficial. The mathematical programming technique called aggregate state dynamic programming (ASDP) explicitly draws upon the knowledge base of the system operator. Augmentation of that knowledge base is possible through the use of graphical‐display training simulators to obtain improved operating rules for spatial allocation of water. The dynamic programming component of ASDP handles the temporal allocation. A case study is presented to display the application of the procedure and to provide direct comparisons against human operational performance.

A chance constrained model is presented for the minimum cost design of water distribution networks. This methodology attempts to account for the uncertainties in required demands, required pressure heads, and pipe roughness coefficients. The optimization problem is formulated as a nonlinear programming model which is solved using a generalized reduced gradient method. Details of the mathematical model formulation are presented along with example applications. Results illustrate that uncertainties in future demands, pressure head requirements, and pipe roughness can have significant effects on the optimal network design and cost.

The first‐order surface pollutant washoff model is examined under noise‐corrupted runoff conditions. Statistics of the pollutant load are derived using the derived probability distribution concept and verified using Monte Carlo simulations. The investigation demonstrates that under noise‐corrupted runoff conditions a bias is introduced in the estimate of the pollutant load. The bias is sensitive to: (1) The time step of the simulation; (2) the variance of the noise in runoff; and (3) the time from the start of the simulation (duration of the simulation). Statistics are derived for both the incremental pollutant washoff load $\left(\Delta P\right)$ and the mass of pollutant remaining on the street surface (P), The results of this study can be incorporated in any rainfall‐runoff model (e.g., SWMM, STORM, etc.) that makes use of the first‐order pollutant washoff model.

This paper presents an application of chance‐constrained goal programming methodology to a system of multipurpose reservoirs. Several objectives which may be conflicting and noncommensurate such as flood protection, municipal and industrial (M&I) water supply, hydroelectric power generation, recreation, etc. are considered in the model. The system goals and constraints are expressed either deterministically or probabilistically depending on whether the random variable term for inflow into a reservoir is present or not. For a constraint, the requirement must be strictly satisfied. For a goal, it is desired to achieve the solution which is as close as possible to the specified target. Stochastic inflows with one period correlation in successive periods are explicitly considered with conditional distribution functions based on normal and lognormal distributions of inflows provided. The application of the methodology to a three‐reservoir system in Oklahoma is reported. The proposed model uses a time period of a month and may be employed for planning rather than real‐time applications.

The U.S. Bureau of Reclamation has the responsibility for developing implementation strategies for salinity control projects in the Colorado River Basin. Determining when and which projects should be built depends primarily upon the planning and construction funds available at various times during the program period and other considerations. The paper describes the development of a methodology that is used by Reclamation for analyzing implementation strategies for salinity control projects in the Colorado River Basin. The methodology uses a variant of the more traditional dynamic programming formulations of the capacity expansion problem which allows total project costs to be used as the state variable. Objective‐space dynamic programming was used to efficiently develop multiple solutions for a range of possible funding levels for the life of the program. The solutions allow the evaluation of the sensitivity and effects of various funding scenarios on the overall salinity control program. Examples of the information developed by the methodology and the types of sensitivity analysis that can be conducted are presented.

A raw water supply Master Plan has been prepared for the City of Boulder, Colorado. The city's portfolio of water rights includes decrees for native water from Boulder Creek and shares in two large transbasin diversion projects. The supply system is operated to meet municipal and industrial demands, provide minimum streamflows for Boulder Creek, and generate revenues from hydroelectric turbines installed in raw‐ and treated‐water transmission lines. The Master Plan was developed with the use of three applications of the Central Resource Allocation Model, a microcomputer‐based network‐optimization tool, which uses the Out‐of‐ Kilter Algorithm. The three applications addressed the yield of the city's native water rights, the yield of its transbasin diversion project holdings, and the economic operation of its water delivery system to meet future demands. The Master Plan recommends modifications to certain operating policies, enlargement of some delivery and treatment facilities, and sale or lease of one of the transbasin water sources.

Spatial data are collected at various locations. Two cases are considered: (1) The observations are independent; and (2) the observations are correlated. If the data are assumed to be normally distributed with constant mean and variance, it is shown that the optimal sample sizes under the independent assumption discussed in this note provide further details correlated case (nó) are related by the expression n*=[(nó²]/(n*+2ΣΣΣS$r_{\mathrm{ij}}$), where $r_{\mathrm{ij}}$ denotes the correlation coefficient between observations at locations i and j. This note derives the preceding expression and provides further details concerning the statistical inference of correlated spatial variables. The importance of the relationship between correlated and uncorrelated sample sizes for hydrologic spatial data collection is emphasized in this work.