Journal of Water Resources Planning and Management

An investigation of water‐quality sampling strategies, applicable to real‐time sampling programs, is presented. The sampling strategies are designed to be a component of the data‐collection process for the determination of the annual mass load of pollutants, or nutrients, in a river. As such, the sampling schemes are intended for use in situations where stream‐flow data are collected regularly, but concentration data are collected during only a limited number of time periods. The sampling strategies are evaluated through the use of a Monte Carlo experiment that created synthetic stream‐flow and associated nutrient concentration data. The performance of each sampling strategy is evaluated and compared in terms of an accuracy measure and a precision measure to establish the relative merits of the proposed schemes. A sampling scheme incorporating stratified sampling with realtime updating of the sampling characteristics is found to give the best estimate of the annual mass load.

Evaluation of the effects of land‐use change in a natural catchment and what can be done to mitigate these effects requires accurate quantitative representation of the mechanisms by which precipitation contributes to run off in the natural and changed states. It is necessary to predict the locations of flow production zones, how these zones change with time, and where flow from each zone enters the channel system. A model is presented that simulates flux rates and spatial distributions of evapotranspiration, Horton and saturated overland flow, subsurface storm flow, and dry‐weather base flow. The model is operated continuously in time and has parameters that are readily measured or estimated, allowing land use changes (forest cover removal, regrading, litter zone removal) to be explicitly represented. The model is applied to a 39‐ha (97‐acre) ungauged catchment in King County, Washington, and an approximate calibration is achieved by comparing the location of simulated runoff production zones against field‐mapped locations. The model is used to explore the relative temporal and spatial effects of land use change on storm hydrographs, flow duration, and long‐term mass balance. The utility of a model structure that permits comparison of directly observable field states with comparable model predicted states is demonstrated.

An optimal experimental design algorithm is developed to facilitate the planning and the optimal configuration and scheduling of a ground‐water‐tracer test whose data are to be used to estimate model parameters. A maximal information criterion is used to select among competing designs. The proposed criterion is equivalent to a weighted sum of squared sensitivities, employing the observation that parameters are most accurately estimated at points with high sensitivity to the parameter but that the relative magnitudes of sensitivities to different parameters are different. The fundamental advantage of this criterion is that it is comparatively simple. The influence‐coefficient method is used to compute the sensitivity coefficient. A zero‐one integer hueristic is used to solve a simplified example for experiment configurations under a given experimental duration. The design considers the installation cost, which is a function of location and depth of the observation well and of the samples themselves. The resulting designs are intuitively reasonable. It was found that a dramatic increase in information can be obtained with an experimental budget increase in a heterogeneous example case.

A least‐cost methodology is presented for use in upgrading existing single‐source water‐distribution networks in order to sustain any single component failure (link or node). The methodology is developed by casting the network‐reliability problem in terms of an explicit level of system redundancy. As a result, minimum cut set computations are avoided as well as the need to select an arbitrary level of system reliability. The design engineer can thus select a design based on the total cost and the desired level of service. The proposed methodology results in a network that provides two different levels of system redundancy: topologic and hydraulic. Topologic redundancy is satisfied through the application of methods from graph theory. Hydraulic redundancy is satisfied through the application of linear programming. Application results are presented for the Federally Owned Water Main System in Washington, D.C.

A procedure for calculating the spatial and temporal distribution of a direct runoff hydrograph from an ungauged watershed is presented. Nodal values of hydraulic roughness and slope are used to avoid kinematic shock. This formulation allows the simulation of overland flow over spatially variable surfaces typical of natural watersheds. The procedure is compared to an analytic solution for a simple case of a watershed composed of two planes. Broader application to watersheds with spatially variable parameters is possible because of the flexibility of the finite element. The Galerkin formulation of the finite element method is used to solve the kinematic wave equations. The method of characteristic and the finite element method were compared on a two‐plane system with an abrupt change in slope. The numerical and analytical considerations are presented to enable practicing engineers to adopt this method for the calculation of flow depth and discharge rate in a distributed manner throughout a watershed dominated by overland flow. A procedure using two‐dimensional elements is being developed by the writers.

The problem of optimal reservoir operations consists of obtaining optimal releases, reservoir storages, and downstream reach routed flows based on forecasted inflows and precipitation. Five goal‐programming schemes that minimize deviations from a set of preferred target storage and flow values are considered: (1) Preemptive goal programming; (2) weighted goal programming; (3) min‐max goal programming, (4) fuzzy goal programming; and (5) interval goal programming. The reservoir operation problem is also formulated as a multiobjective linear program (MOLP). The optimal solutions of the goal programs are contained among the efficient points of the MOLP. It is also shown that the min‐max and fuzzy goal programs can yield inefficient points as optima; however, there exist efficient alternative optima. Because goal programs concentrate on achieving predetermined subjective target values, the potential optima for objectives may not be explored. An examination of the extreme points of the feasible region without the goal constraints overcomes this difficulty and provides a way to set realistic targets. These techniques are applied to the Green River basin in Kentucky data reported previously by other researchers.