Journal of Water Resources Planning and Management

The SPANS (SPatial ANalysis System) Geographic Information System (GIS) and the Georgia Landcover Database were used to identify land use within the 333 km² Little River Research Watershed of Georgia and to perform a variety of spatial analyses with the purpose of determining the water quality impact of riparian forest management within the watershed. The watershed contains a dense dendritic stream network, with approximately 54% of a 30 m riparian buffer on either side of all streams of second order or larger currently in forest cover. Approximately 60% of the upland is in some type of agricultural production. GIS analyses were used to extrapolate field-scale measurements on riparian forest nonpoint-source-pollution nutrient buffering capacity to the watershed-scale, and to evaluate the water quality impact of three increasing deforestation and three increasing reforestation riparian forest management scenarios. Nitrogen (N) and phosphorus (P) loadings at the watershed outlet could potentially increase by 16.7 and 9.5%, respectively, under a 50% riparian buffer deforestation scenario. Decreases in potential N and P loadings of 13.2 and 7.4%, respectively, could potentially occur under a 50% reforestation scenario. Fragmentation analysis of the riparian buffer indicated that only six blocks of riparian forest at least 2 km long remained in the entire watershed, confirming the fragmented state of the riparian forest.

In 1946, Massé discussed the optimal operation of a single reservoir for hydroelectric energy production. He obtained his results by both economic reasoning and rigorous mathematical derivations using a generalized form of the calculus of variations. In this article, the procedure of Massé is generalized to cases where: (1) the benefits from release are not related directly to the release but rather to the discharge at a downstream point; (2) there may be several points downstream where an objective is to be achieved; and (3) there are several reservoirs. Massé and Varlet found that the optimal strategy is the one that maintains the marginal value of the release constant in time whenever the reservoir is neither full nor empty. It is shown rigorously here that, in the general case, for a strategy to be optimal, the “memory-integrated future marginal value” of the release must be constant. This result is generalized to a system of several reservoirs and illustrated on a simplified description of the Seine river basin upstream of Paris, France.

Deterministic optimization can produce suboptimal reservoir-control policies by failing to incorporate adequately the impact of low-probability events. Resulting operating policies may not efficiently balance the costs of rationing, minor flooding, or other short-term impacts with the severe impacts of extreme flood or drought. This occurs when deterministic optimization is applied to systems that are not “certainty equivalent.” This paper demonstrates this by contrasting control policies developed using deterministic optimization of inflow forecasts with control policies using stochastic optimization of probabilistic inflows. For a range of hypothetical reservoir-control problems, it is observed that deterministic optimization results in costs that are greater on average and that are much greater for extreme events, particularly for reservoir systems with limited storage capacity and for objectives described by nonquadratic functions. This is true even when forecasts are relatively accurate, such as when streamflows are highly autocorrelated.

This paper reviews a variety of derived single-purpose operating policies for reservoirs in series and in parallel for water supply, flood control, hydropower, water quality, and recreation. Such rules are useful for real-time operations, conducting reservoir simulation studies for real-time, seasonal, and long-term operations, and for understanding the workings of multireservoir systems. For reservoirs in series, several additional new policies are derived for special cases of optimal short-term operation for hydropower production and energy storage. For reservoirs in parallel, additional new special-case rules are derived for water quality, water supply, and hydropower production. New operating policies also are derived for reservoir recreation.

Many important advances have been made in the development of mathematical models for conjunctive water management and reservoir management. However, the conjunctive use models that have detailed stream/aquifer system interactions have not considered reservoir management rules. Conversely, reservoir management models have not simulated stream/aquifer interactions in detail. There is a need for enhanced linking of reservoir and stream/aquifer systems in conjunctive water management models. This paper presents a simulation/optimization model that integrates linear reservoir decision rules, detailed simulations of stream/aquifer system flows, conjunctive use of surface and ground water, and delivery via branching canals to water users. The linear decision rule is an example of a rule that has been widely discussed in reservoir operation literature and is simple to program. State variables, including aquifer hydraulic head, streamflow, and surface water/aquifer interflow, are represented through discretized convolution integrals and influence coefficients. Reservoir storage and branching canal flows and interflows are represented using embedded continuity equations. Results of model application to a hypothetical study area under several scenarios indicate that the more detail used to represent the physical system, the better the conjunctive management. The most detailed representation provides 13% more water than the least detailed system.

Theoretical models for hydraulic network analysis often assume idealized pipe networks. Practical network packages, however, allow for a variety of necessary controlling elements. Sometimes these are introduced as generalized pipe elements, thus hiding the inherent discontinuous operation of nonreturn valves and pressure reduction valves. It is well known that the inclusion of such elements may cause numerical difficulties because of singularities. This paper presents an algorithm based on loop equations. It features a dynamic loop structure that is updated automatically according to the state of the controlling elements. It is shown that this provides a logical solution to the discontinuous operation of the controlling elements and also results in a highly efficient network simulation algorithm.

Petra is located in the southwest region of Jordan about 200 km south of Amman, between the Dead Sea and the Gulf of Aqaba. Petra was carved in sandstone canyons by the Nabatean over 2,000 years ago. Today the city is a major tourist attraction, its monuments being considered the jewels of Jordan. Floods pose a serious threat to the tourist activities in Petra as well as to the monuments themselves. In this paper, a flood analysis model developed and calibrated for the Petra catchment is described. Using the model, flood flows and volumes are estimated for storm events of various return periods. To alleviate the impact of floods on tourism in Petra, several flood mitigation measures are proposed. The impact of these measures on flood peakflow and volume is evaluated. These include afforestation, terracing, construction of check and storage dams, and various combinations of these measures. The flood simulation model predicts that the measures can reduce flood peakflows and volumes by up to 70%.