Journal of Water Resources Planning and Management

Trihalomethanes (THMs), a group of drinking water disinfection by‐products, are suspected carcinogens and are currently regulated by the U.S. Environmental Protection Agency. More stringent regulation of individual and total THM species is anticipated to meet the requirements of the 1986 Safe Drinking Water Act amendments. Because of this new focus on individual compounds, a set of statistical relationships called the bromine distribution factors were developed to characterize the impact of bromide ion on the distribution of THM formation potential (THMFP) species. A new derivation of these relationships is presented in this paper. Careful data analysis in combination with sound reasoning leads to a set of simultaneous equations that, when solved analytically, result in model coefficients that give both statistically sound and physically meaningful predictions for the distribution of THM compounds. Two independent data sets, including simulated distribution‐system THM data as well as THMFP data, were used to demonstrate the validity of the new formulation.

A methodology for determining optimal pump operation schedules for water‐distribution systems is presented. In addition to minimizing the energy‐consumption cost, the model includes a constraint to limit the number of pumps that are switched on during the planning period. A two‐level approach is adopted whereby the system hydraulics are analyzed in an off‐line mode to generate simplified hydraulic and cost functions for an on‐line model. These functions developed for each pump combination allow for rapid evaluation within a dynamic programming optimization algorithm. Constraints can also be included for tank levels, rate of change of tank level, pump switches during each period, and maximum energy consumption. Solution times for small to medium‐sized systems show that the model can be used in an on‐line mode to determine pump operation schedules; however, the applicability of the model is limited by the number of pumps in the system. An extensive sensitivity that highlights operational problems and the utility of the approach is presented.

Flood diversion during dam construction has a major influence on the construction period, and its cost represents a significant part of the total project cost. In general, selection of diversion system capacity and the dimension of the system component are socioeconomic and technical decisions. Quantifying uncertainties is an important problem in risk‐based design of hydraulic structures. This paper presents an optimization model of river diversion that incorporates the uncertainty of the flood magnitude estimator and the hydraulic uncertainties. Confidence interval for the design flood considers the uncertainty associated with an estimated skew coefficient. For hydraulic uncertainties a first‐order second‐moment analysis is carried out. The diversion design flood and the dimensions of the system components are explicitly treated as decision variables. To study its usefulness, the model is applied to the Karoon Dam site, which is now under the design phase in southern Iran. Results show that the model is a promising tool that could be used efficiently as a design aid.

Methods to estimate and generate free‐surface evaporation $E_p$ and potential evapotranspiration $E_t$ monthly time series are presented. Results are applied to the Quabbin and Wachusett water‐supply reservoirs, which serve the Boston metropolitan area, and the Connecticut and Ware River basin watersheds, using historic (1950–80) climate data from several NOAA stations located in New England. The $E_p$ and $E_t$ time series are perturbed using climate‐change scenarios of several general circulation models (GCMs) to investigate the sensitivity of both time series to doubled atmospheric ${\mathrm{CO}}_2$; $E_p$ and $E_t$ are also perturbed by +2°C and +4°C monthly temperature rises. The sensitivity of $E_t$ to a potential climate‐change‐induced vegetation‐canopy resistance is also examined.

A pump‐and‐treat system of extraction wells is modeled to create a hydraulic capture zone to immobilize a plume of contaminants in a confined aquifer with uncertain hydraulic transmissivity. The stochastic approach is used to model the transmissivity as a random field and to generate random realizations of the field. The linear‐response‐matrix method is used to address a joint chance‐constrained optimization problem in which constraints on velocities must be simultaneously satisfied with high probability. In the literature, the multiple‐realization approach provides a robust solution but does not allow for prespecification of a desired reliability. We develop a partial‐infeasibility method, in which the pumping strategy is obtained through heuristic methods that require the designed reliability level to be satisfied for a training set consisting of multiple realizations of the uncertain transmissivity. Empirical Monte Carlo testing shows that as the size of the training set increases, the actual reliability approaches the design criterion. Thus, the method combines the intuitive appeal of the multiple‐realization method with the ability to prescribe a desired reliability level.

To assess the effectiveness of Nassau County's water conservation program, the effects of population changes and weather patterns on water use must first be screened out. Using multiple linear regression, an equation was developed to predict average monthly pumping based upon the number of households in the county and two weather‐related indexes. The first index represents degrees over 65°F of the maximum daily temperature summed for each month. The second index represents consecutive days without rain, summed for the month. Using data for the period 1951–1986, a regression equation was developed describing total monthly pumping as a function of the number of households and the logarithm of the product of the two weather indexes. A Pearson correlation coefficient of 0.95 was attained. The equation was used to “project” pumping between 1987 and 1992, during which time a water conservation program was established. By comparing projected pumping with actual pumping over the period, the effect of water conservation could be estimated. Results indicate that the conservation program achieved a maximum one‐year decrease in water use of 11%.

In this study a demand driven stochastic dynamic programming (DDSP) model is developed that allows the use of actual variable monthly demand in generating the operating policies. In DDSP, the uncertainties of the streamflow process, and the forecasts are captured using Bayesian decision theory (BDT)—probabilities are continuously updated for each month. Furthermore, monthly demand along with inflow, storage, and flow forecast are included as hydrologic state variables in the algorithm. In this model the associated penalty for each operating policy is a function of the release and the expected storage. The operating policies are compared and tested in a hydrologic real‐time simulation model and in a real‐life operation model. The objectives of this paper are: to evaluate the usefulness and the hydrologic reliability of the generated operating policies by DDSP model; and to demonstrate how the assumption of fixed demand in optimization is deficient when the demand is actually variable or uncertain. The reliability of the operating policies is measured in terms of meeting the required demand when the operating policies are applied in simulation/operation models.

This paper presents a technique to efficiently process data to supply the necessary in‐puts to a large‐scale water‐distribution optimization model. The overall goal of this project is to develop an on‐line optimal control strategy for a large municipal and industrial (M&I) water distribution system. That is, which pumps should be turned on to meet the demand and minimize cost? One of the problems in handling a large water‐distribution system is managing the large database required for the optimization model. This paper presents an efficient algorithm to generate, store, and access the necessary data. Some of the characteristics of the method are illustrated through an example of a water‐distribution system.

Responses to willingness‐to‐pay questions in contingent valuation (CV) surveys are being used with increasing frequency to obtain estimates of the value of public goods, such as improved water quality, for use in benefit‐cost analyses. Analysts must decide whether the appropriate conceptual model for any particular application is a private or a political market. When estimating willingness‐to‐pay and total benefits, analysts must decide whether to censor outliers and certain zero responses, called protest bids, from data sets. Issues associated with identification of protest bids are reviewed, and alternate methods for the treatment of outliers are discussed. The significance of these methodological decisions is illustrated with an example involving willingness to pay for new storm water programs in Baltimore County, Md. to help achieve Chesapeake Bay nutrient‐reduction objectives. It is argued that decisions concerning analysis of troublesome responses should be consistent with the choice of market model for the application.