Journal of Water Resources Planning and Management

A climate-driven water resource model of California’s Sacramento River Basin (SACB) is presented, based on the Water Evaluation and Planning model Version 21 (WEAP21). The model’s configuration, calibration, testing, and limitations are presented. The major contribution includes an integration of the watershed’s surface and subsurface hydrology, consumptive, and nonconsumptive use, and the water management infrastructure and controls that determine how water naturally flows and is managed. The SACB was subdivided into numerous catchments; groundwater basins; irrigated areas; urban/export use; environmental requirements; and canals, diversions, and reservoirs in an attempt to characterize the forces that act on water throughout the basin. A monthly climate time series forced an embedded hydrologic model that simulates runoff, groundwater-surface water interactions, and consumptive agriculture and urban water uses. Results show that the model can reproduce both local and regional water balances, including managed and unmanaged streamflow, reservoir storage, agriculture and urban water uses, and the allocation of ground water and surface water supplies, which should be useful for various California water planning processes.

A linked simulation-optimization model using artificial neural networks (ANNs) and genetic algorithms (GAs) is developed for deriving multiple objective management strategies for coastal aquifers. The GA-based optimization approach is especially suitable for externally linking a numerical simulation model within the optimization model. However, the solution of a linked simulation-optimization model is computationally intensive, as a very large number of iterations between the optimization and the simulation models are necessary to arrive at an optimal management strategy. Computational efficiency and feasibility for such linked models can be enhanced by simplifying the simulation process by an approximation. A possible approach for such approximation is the use of an ANN model. In this paper, an ANN model is developed initially as an approximate simulator of the three-dimensional density dependent flow and transport processes in a coastal aquifer. A simulation-optimization model is then developed by linking the ANN model with a GA-based optimization model for solving multiple objective saltwater management problems. The performance of the optimization models is evaluated using an illustrative study area. For comparison of the solution results, a multiple objective management model is also solved using embedded formulation and classical nonlinear optimization technique. The comparison of results shows potential feasibility of the proposed methodology in solving multiple objective management model for coastal aquifers.

A stochastic mixed integer programming decision support system (DSS) to aid farmers in the Sahel-Sudan region of West Africa is formulated. It provides optimal crop planting practices that minimize the caloric deficit of a household and maximize limited household income given a seasonal probabilistic rainfall forecast and the farmer’s risk level. The recommendations of the DSS are subjected to sensitivity analyses related to labor availability, prices, and risk level. Furthermore, survey data gathered from a case study of Bonam Village are presented for comparison with the output of the DSS. It is found that the DSS captures many of the complexities of agricultural management practices such as the strategy of diversification. An optimization model like this may have future applications in West Africa to improve agricultural decision making and also test hypotheses related to climate forecasts and improvements in household food security.

The area of systematic identification of contamination sources in water distribution systems is in its infancy and is rapidly growing. The real water distribution network problem poses many challenges that current methods usually assume away to facilitate manageable method development and testing. Current methods may not readily and efficiently address issues, such as multiple sources, unknown contamination types with different reaction kinetics, use of different types of sensors with varying degree of resolution, dynamically varying demand and sensor information, and uncertainty and errors in the data and measurements. With the aim of addressing these imminent challenges, this paper reports the findings of an ongoing research investigation that develops and tests an evolutionary algorithm-based flexible and generic procedure, which is structured within a simulation-optimization paradigm. This paper describes the specific implementation of the method using evolution strategies (ESs), a population-based heuristic global search algorithm. A key component of designing this source characterization method is to define a compact, but comprehensive, solution encoding structure. The new method is constructed using a tree-based encoding design to enable the representation of variable-length decision vectors and a set of associated genetic operators that enable an efficient search. This algorithm is successfully tested and demonstrated to have consistently good performance for several instances of an illustrative water distribution contamination case study. As the ES-based algorithm conducts a probabilistic search, its robustness is tested using multiple random trials, and the method is shown to exhibit a robust behavior.

A planning strategy is presented that aims at establishing the optimal replacement schedule for a water distribution network. Two performance indicators are defined. The first accounts for structural state (total cost defined as the sum of pipe replacement cost and the expected cost of pipe break repairs) and the second for hydraulic performance (minimization of the pressure deficit). A multiobjective objective function is defined based on these two indicators and a genetic algorithm optimization technique is used to identify optimal solutions (Pareto front). Three management strategies are considered to choose a replacement schedule among those making up the Pareto front: (1) a prostructural strategy that only considers the structural indicator; (2) a prohydraulic strategy that integrates both structural and hydraulic indicators; and (3) a budgetary constraint strategy which assumes a predefined budget for replacement expenditures. The proposed planning strategy was tested on two hypothetical networks. Synthetic pipe break records were generated using a statistical pipe break model. A Bayesian inference approach was then used to estimate parameters values from these pipe break series. A comparison of the different management strategies is provided as advantages of using Bayesian inference are discussed.

Water rationing in reservoir operation is essential for mitigating the damage of water deficits from prolonged and severe droughts. This problem is addressed in this study using hedging rules characterized by three hedging parameters: starting water availability, ending water availability, and a hedging factor. The effects of hedging are evaluated using two conflicting objectives: (1) a total shortage ratio and (2) a maximum 1-month shortage ratio, which represent the long- and short-term water-shortage characteristics for water supply. A multiobjective genetic algorithm is employed to solve this multiobjective reservoir operation optimization problem. The Pareto-optimal solutions of various hedging rules, including time-varying parameters and multiple periods ahead hedging, are searched using the nondominated sorting genetic algorithm II. The proposed optimization framework is applied to the Nanhua Reservoir in southern Taiwan. The results show that time-varying hedging promotes both objectives. The merit of multiperiod ahead hedging can be superimposed on the hedging rules incorporating time-varying hedging parameters. Significant improvement on the Pareto front is obtained for all time-varying hedging when 2 months ahead hedging is implemented. Further increasing length of periods in hedging leads to an improved Pareto front associated only with higher time-varying frequency hedging parameters but not the constant hedging scheme.

The purpose of irrigation scheduling is to coordinate the timing and amount of water used to irrigate crops in a way that maximizes profits. Using a case study on corn production in the Havana Lowlands region, Illinois, this paper examines the extent to which the incorporation of different types of weather forecasts into irrigation scheduling can increase the profitability of irrigated agriculture. A soil water atmosphere plant (SWAP) model is employed to simulate crop yields using weather forecasts from the growing seasons during a period running from 2002 to 2006. The net profits from the modeling analysis based on the modeled soil moisture and weather forecasts are compared to that from an irrigation schedule that the farmers roughly followed in 2002. The results show that if farmers just use the real-time soil moisture information and the empirical rules set with the SWAP model, crop profits may increase by 16%; the inclusion of 7-day forecasts in a rule-based simulation model boosts the net profit by 21%. Over the 5 testing years, it is found that the proper use of the 7-day forecast can save irrigation water and increase crop yield in dry years. Next, these results are also compared to ones derived using deterministic optimization techniques that assume perfect 2-week and seasonal forecasts to assess the extent to which net profits could increase with further weather forecasting improvements. On average over the 5 testing years, the perfect 2-week and seasonal forecasts yield profits of 42% and 48%, respectively, greater than the scenario based on the 7-day weather forecasts. Furthermore we conduct a stochastic optimization analysis using the monthly probabilistic climate prediction provided by the Climate Prediction Center of the National Oceanic and Atmospheric Administration. A review of the retrieved irrigation schedule shows that the farmers at the study site sometimes applied more water than necessary and, at other times, less than the model based on a 7-day weather forecast and soil moisture simulation suggested.

Few regions in North America have directly incorporated the implications of climate change in water resources planning initiatives because the relevant information is not readily accessible, and methods for adjusting policy and operations are not obvious. To help one community and to provide an example for managers of other watersheds, we engaged stakeholders in a group model building process to explore plausible water resources futures for the Okanagan Basin, British Columbia, Canada. The process was conducted by a team of academic and federal government researchers and comprised of a sequence of five one-day participatory workshops held within the basin over a 12-month period. Primary workshop objectives included: creating a shared learning experience, developing a customized exploration tool, and fostering trust in the model among the participants. Survey results indicate that the exercise helped participants to expand their thinking to the basin scale and to appreciate the complexity of water management. Participants found the resulting model to be suitable for futures exploration and outreach; however, the high turnover rate in participation limited the sense of ownership in the model by the completion of the final session.

Mathematical simulation models that are used to provide information about the age of water in potable water distribution networks usually represent the age of water as the flow weighted average age value of the merged flows at a particular location. This approach is questioned and it is argued that age values should not be averaged in this way. This paper describes a methodology to improve the description of the age of water by using the outputs from a simulation model to identify the component ages that contribute to the mean age. In this case, the AQUIS software was used as the simulation engine and the model has been enhanced to calculate the volume and age of the different parcels of flow that meet at all nodes in the network. The program has then been adapted to subsequently track these parcels of water throughout the network. The model provides a description of the component age profiles at any point in a distribution network and this information can be translated into operational strategies to manage the flows, and hence residence times, thereby providing the opportunity to improve overall water quality performance by identifying potential water quality problem areas and hence to promote proactive remediation strategies. The paper is concluded by a case study example of the application of the model.

Reservoir operating rules are often based on deterministic models using critical period (CP) analysis, which results in very conservative decisions. CP is defined as the historical hydrologic period that includes the lowest flow or the most severe drought, which represents functionally as the full-empty cycle of reservoir storage under a given firm yield. This paper compares CP with another time scale, the forecast horizon (FH) underlying hedging rule policies for reservoir operation. When the decisions in the initial few periods are not affected by forecast data beyond a certain period, the period is known as FH, and the number of the initial periods is known as the decision horizon. When a reservoir operation policy follows the concept of CP, CP is used as FH and the optimal yield generated within the FH is the firm yield. FH as long as the CP is not realistic because in the real world data for a such a long future period is unavailable; and using CP as FH is not effective either, because CP—the longest FH-involving the most severe drought leads to very conservative release decisions, i.e., the “near-zero-risk” decision and the time preference of utility diminishes the influence of hedging for over long periods.

A contamination event in a water distribution system can result in adverse health impacts to individuals consuming contaminated water from the system. Assessing impacts to such consumers requires accounting for the timing of exposures of individuals to tap-water contaminants that have time-varying concentrations. Here we present a probabilistic model for the timing of ingestion of tap water that we developed for use in the U.S. Environmental Protection Agency’s Threat Ensemble Vulnerability Assessment and Sensor Placement Tool, which is designed to perform consequence assessments for contamination events in water distribution systems. We also present a statistical analysis of the timing of ingestion activity using data collected by the American Time Use Survey. The results of the analysis provide the basis for our model, which accounts for individual variability in ingestion timing and provides a series of potential ingestion times for tap water. It can be combined with a model for ingestion volume to perform exposure assessments and applied in cases for which the use of characteristics typical of the United States is appropriate.

Two erosion control methods are compared for cost effectiveness in reducing sediment loading into the streams; targeted replacement of agricultural crops and placement of vegetative filter strips (VFS) along field drains. The first method implements an optimization approach for reducing sediment load through targeting and replacement of row crops in sensitive areas with switchgrass, such that watershed income is maximized while maintaining sediment load at a desired level. In the second approach, the soil and water assessment tool (SWAT) was used with simplified grass filter model , a filter strip model, to estimate sediment load. The sediment load predicted for each hydrologic response unit in SWAT was input to a filter strip model which calculates the sediment trapped as flow passes through the filter strip. The study shows that placement of VFS along field drains is more cost effective compared to targeted replacement approach in reducing sediment load. However, the targeted replacement approach renders in situ soil conservation, because it reduces the amount of sediment generated on site compared to placement of VFS along field drains to trap sediment that has already been displaced. Further analysis was made to evaluate the relative benefits of placing vegetative filter strips on selected fields based on the sediment yield. The comparison was made between placing a VFS in all fields and placing VFSs on 75, 50, and 25% of the fields with the greatest sediment loads. It was found to be more effective to place small sized filter strips in as many land units as possible, as opposed to placing larger size VFSs on a few selected fields. It was also found to be slightly more beneficial to vary the width of the VFS in proportion to the size of the field contributing to the sediment inflow. In addition, the relative benefits of harvesting the VFS instead of designating the areas as the Conservation Reserve Program lands was evaluated. It was found that producers could obtain more income from harvesting the VFS than they would earn from water quality incentive payments while maintaining the same level of sediment reduction, resulting in a savings of conservation expenditures.