Journal of Water Resources Planning and Management

There is interest in the possibility of purchasing Indians' reserved water rights to use either on or off the reservation for various purposes. Those wishing to purchase such reserved rights should examine the law closely, since it appears that the courts will not allow those reserved rights to be transferred from the reservation lands.

Many empirical, mass balance phosphorus models have been developed and tested on northern temperate lakes. Sixteen published models were tested on New Mexico reservoirs to determine their usefulness in simulating reservoir behavior in the Rio Grande basin. The models overpredicted measured, seasonal phosphorus concentrations in the reservoirs. Models with settling velocities that varied with selected hydrologic variables performed better than models with constant settling velocities. Model predictions were improved by establishing a relationship between removal rates of phosphorus and certain hydrologic variables. The relationship between settling velocity of phosphorus and reservoir flushing rate was the most significant. Flushing rate also was significantly correlated with the loading rate of total suspended solids. This helps to explain the removal rates observed in the reservoirs. High flushing rates bring in high loads of suspended solids, which may combine with phosphorus and remove it from the water column as the solids settle.

In many cases, mitigation of the impact of localized groundwater contamination can be achieved, or enhanced, by using groundwater withdrawal and/or injection to control plume migration. However, for complicated flow domains, where multiple wells are required, the determination of the pumping rates necessary to achieve a specific plume manipulation objective is nontrivial. A relatively simple and flexible methodology is presented for initial screening estimation of pumping rates that will achieve plume manipulation objectives. Plume manipulation objectives are formulated as unconstrained nonlinear minimization problems using penalty function methods. For solution, a nonlinear mathematical programming algorithm is loosely coupled to a finite difference simulator of the hydraulic response of the flow domain to pumping and a numerical integration routine for plume tracking.

Among a large number of random variables related to water storage for given sample size n, the most studied are the range $R_n\text{,}$ deficit $D_n\text{,}$ and negative run‐sum $V_n\text{.}$ A simulation of 5,000 samples for various n was performed for three assumed population processes: normal independent, normal dependent, and gamma independent. It was found by simulation, and verified by an analytical derivation, that the range and deficit are equal $(R_n=D_n)$ in 50% of generated samples, while in the other 50% the range is greater that the deficit $\left(R_n>D_n\right)\text{.}$ Probabilities of $D_n=V_n$ range from zero to more than 0.60; they decrease as n and skewness $C_s$ increase, but increase as the lag‐1 autocorrelation coefficient increases. The case of $R_n=D_n=V_n$ has similar patterns. The ratios of sample means for these three storage‐related variables, either for 5,000 samples or only the samples for which the values of these variables are not equal, exhibit these general patterns: (1) Ratios $D_n/R_n$ increase with n converging to an asymptotic value of 0.7854 in case of 5,000 samples; (2) for $R_n>D_n$ samples, the ratios $D_n/R_n$ are much smaller than for the entire set of 5,000 samples with $R_n⩾D_n;$ (3) ratios $V_n/D_n$ and $V_n/R_n$ are much smaller than $D_n/R_n\text{.}$ The pairwise lag‐0 correlation coefficients of $R_n\text{,}{D}_n\text{,}$ and $V_n$ for 5,000 sample estimates show that corr $\left(R_n\text{,}{D}_n\right)$ changes around 0.45 as n increases in the case $C_s=0;$ this correlation is small $(0-0.25)\text{,}$ but increase with n for $C_s=3\text{.}$ For $R_n>D_n$ samples, corr $\left(R_n\text{,}{D}_n\right)$ becomes negative, oscillating around $-0.07\text{.}$ corr $\left(D_n\text{,}{V}_n\right)$ shows a decrease with an increase of n for $C_s=3\text{,}$ but increases with an increase of autocorrelation, corr $\left(R_n\text{,}{V}_n\right)$ decreases with an increase of n, changes slightly with autocorrelation and with skewness.

Although the engineering basis for managing depleted aquifer systems has been theoretically established, governmental management of such systems has been slow to develop, due to various economic, legal and institutional constraints. New Jersey, one of a number of states facing severe groundwater problems, has been successful in developing an approach which will reduce user withdrawals from depleted aquifers, while requiring all users to contribute to the cost of the necessary alternative surface supplies. Conjunctive use arrangements, water conservation, and recharge will be facilitated by this approach.

This paper presents a linear programming formulation of the resource allocation problem applied to water distribution system rehabilitation and reconstruction. The decision variables are fractions of large bundles of water mains to be rehabilitated or replaced throughout the planning horizon. The model can be used by municipal planners for scheduling capital expenditures related to water distribution system improvement and maintenance and is therefore a major component of a maintenance management system.

Three regional sources of freshwater around Saudi Arabia are identifed: the Indus River in Pakistan, the Nile River in Sudan and Egypt, and the Euphrates and Tigris Rivers in Iraq. Transportation costs are calculated using tanker shuttle service, in which case tankers are used exclusively for the transportation of freshwater between loading and unloading ports. The average water transportation costs from Karachi to Dammam, Port Sudan to Jeddah, and Suez to Jeddah are $0.79, $0.29, and $\$0.61m^3\text{,}$ respectively, in a 300,000‐dwt tanker. Reported transportation costs through backhaul service (return cargo) are $1 and $\$2/m^3$ at worldscales of 30 and 60, respectively. These costs compare favorably with desalination costs of $\$1.54-\$4.5/m^3$ in Saudi Arabia. Using these unit costs, four scenarios are discussed for transportation of $910\text{,}000m^3$ of water/day through different schemes. Water transportation from Pakistan appears to be the most cost‐effective of the alternatives and can be made operational at short notice with a small capital investment.

Three groundwater modeling studies of a water rights dispute in the San Juan basin of New Mexico are compared. Although these studies all use the same computer model and rely on the same data base, the conclusions they reach are very different. The discrepancies appear to be due primarily to different interpretations of the pump test data used to estimate aquifer parameters in the vicinity of a proposed well field. Since none of the modeling studies attempts to estimate the effect of ambiguous data interpretations on predicted drawdowns, there is no way to judge the relative value of the conflicting results. Although this outcome is disturbing it is not unusual. The credibility of such modeling studies could be greatly improved if they explicitly addressed the factors that control and limit model accuracy.

Indexed sequential modeling (ISM) has been proposed by the Western Area Power Administration as an alternative approach to developing firm marketable capacity, i.e., project‐dependable hydropower capacity (PDC), in contrast with the usual approach of basing PDC on the most adverse period of record. ISM allows a probabilistic analysis of hydropower capacity by extracting a series of overlapping short‐term (say, 10 year) inflow sequences directly from the historical record, which includes the most adverse period, and then simulating reservoir operations over this interval for each sequence. As a means of evaluating ISM, the New Melones Reservoir system in the federal Central Valley Project of California was selected as a case study for comparing hydropower output generated from ISM input with use of a multivariate stochastic inflow generation model. A comparison of monthly power and energy output at 95% confidence limits, 10% risk level, and most adverse, showed reasonably good correspondence between the two methods, except for a few months of energy production in the final year of simulation.