Journal of Water Resources Planning and Management

Escalating salinity levels in the Bonriki aquifer due to unplanned groundwater extraction is a major concern for the people of Kiribati. A multiobjective management model capable of providing sustainable optimal groundwater pumping strategies and simultaneously confining salinity concentrations in the aquifer within specified limits is needed for the Bonriki aquifer system. This study applies a regional-scale linked simulation-optimization (S/O) methodology with a Pareto front clustering technique to prescribe optimal groundwater withdrawal patterns from the Bonriki aquifer. A numerical simulation model is calibrated and validated using available field data. For computational feasibility, support vector regression (SVR) surrogate models are trained and tested utilizing input-output data sets generated using the numerical flow and transport simulation model. The developed surrogate models were externally coupled with the multiobjective genetic algorithm optimization (MOGA) model as a substitute for the numerical model. The study area consisted of freshwater pumping wells for extracting groundwater for beneficial use. Pumping from barrier wells installed along the coastlines were also considered as a management option to hydraulically control saltwater intrusion. The executed multiobjective linked S/O model generated 700 Pareto-optimal solutions. Analyzing a large set of Pareto-optimal solutions is a challenging task for the decision maker. Hence, the $k$-means clustering technique is utilized to reduce the larger original Pareto-optimal solution set for solving the large-scale saltwater intrusion management problem in Bonriki.

The Yaqui River Basin (YRB) is in the semiarid state of Sonora in northwest Mexico. The flow in this watershed is controlled by three reservoirs: Angostura, Novillo, and Oviáchic. A daily reservoir operation model is developed for this system to assist local reservoir operators in the decision-making process. This model is composed of a semidistributed daily watershed simulator combined with a nonlinear optimization model. The hydrologic simulator was developed using the Hydrologic Engineering Center–Hydrologic Modeling System (HEC–HMS) developed by the USACE Hydrologic Engineering Center and the optimization model was developed using the General Algebraic Modeling System (GAMS) methodology. Two climatology forcings (precipitation, temperature, and solar radiation) were applied to the HEC–HMS simulation: (1) a network of daily ground observations from Mexican water agencies during the historical period of October 2001 to September 2010; and (2) gridded fields from the North American Land Data Assimilation System (NLDAS) at a 12-km resolution during the same simulation period. The NLDAS forcing improved the spatial representation of the simulator, compared to ground observation forcing, in areas draining into Angostura and Novillo. However, ground observation forcing resulted in the best performance of the hydrologic model in the contributing area of Oviáchic. The simulated reservoir releases were compared to historic releases at the three reservoirs to generate confidence in the simulation tools. Findings show that existing water allocation strategies are unable to satisfy demands from most users. The simulator, in combination with the optimization model, was useful to generate water allocation scenarios that would meet these demands; therefore, the proposed model is a useful tool for reservoir operations in the YRB.

Traditional urban drainage systems rely heavily on centralized network-based infrastructures. Recently, the idea of centralized urban drainage networks has increasingly been questioned. The latest investigations suggest a transition from centralized to decentralized or hybrid schemes. Therefore, there is a need for tools and methodologies to evaluate and optimize drainage networks with arbitrary degrees of centralization (DC). For this purpose, this paper introduces an algorithm called the hanging gardens algorithm to generate all possible sewer layouts and to explore different degrees of decentralization. The proposed algorithm starts with generating a centralized layout and introducing a list of outlet candidates. Next, it adds arbitrary outlets from candidates to the generated layout and uses a graph theory–based approach to assign parts of the layout to different outlets. This procedure is iterated until all (combinations) outlet candidates have been included. To demonstrate the performance of the proposed algorithm in enumerating all different DC and generating realistic layouts, the algorithm is coupled with an optimization engine in order to optimize the stormwater collection network of a section of the city of Ahvaz, Iran. The number and location of outlets, the layout configuration of each part and the size of pipes are used as optimization variables to minimize costs subject to hydraulic and feasibility constraints. The proposed algorithm performs well in exploring different DC and finding near-optimum solutions.

Groundwater overdraft is a major problem globally and has been a growing problem for California for decades. This overdraft is predominantly driven by the economic value of water for agricultural production and cities. Spurred by the recent drought, California passed legislation requiring the elimination of groundwater overdraft by 2040. This paper employs a statewide hydroeconomic optimization model to explore potential water supply effects of ending long-term groundwater overdraft in California’s Central Valley for several general water policies with historical and warmer–drier climates. The model minimizes agricultural, urban scarcity, and operating costs over 82 years of historical hydrologic variability, given today’s infrastructure and environmental flow constraints. The model results assess effects of overdraft and Delta policies for different climates on water deliveries, economic costs, environmental flows, water market operations, and the economic value of expanding infrastructure capacities. Prohibiting long-term overdraft leads to reduced agricultural water use and reoperations, and reduced outflows to the sea from the Sacramento-San Joaquin Delta, where water availability policies become important. In combination with a warmer–drier climate, ending overdraft further exacerbates water scarcities, increases environmental and economic costs, and increases the marginal economic value of water exports from the Delta, which are likely to worsen water conflicts and illustrate connections of California’s groundwater and surface water problems. Economically useful adaptation actions include more water transfers involving the Delta, water markets, and trades; conjunctive use of surface water and groundwater; and recycled wastewater supplies for coastal urban users.

Many water distribution system (WDS) solution methods have been developed to perform demand-driven steady-state analysis. These methods are used to solve the nonlinear system of equations that model a WDS. WDS networks have structural properties that can often be exploited to speed up these solution methods. One solution method that exploits these structural properties is the forest-core partitioning algorithm that was proposed as a preprocessing and postprocessing method that can be used to separate the network into a linear forest component and a nonlinear core component. This paper presents a complementary method for preprocessing and postprocessing called the bridge-block partitioning algorithm (BBPA). This method further partitions the core component of the network into a number of linear bridge components and a number of nonlinear block components. The use of BBPA to partition a WDS network provides significant advantages over current solution methods in terms of both speed and solution reliability.

Water distributions networks (WDNs) are exposed to multiple hazards, leading the network to operate under a range of critical conditions. This paper explored the relationship between the impact of anomalous events (AEs) of WDNs and the consequent palliative actions (PAs) to be implemented in the network to minimize such impact. Both AEs and PAs were assessed through a network resilience criticality index adapted to WDNs. The results were compared with those obtained from normal operating conditions with respect to the satisfaction rate of nodal demands. The proposal was evaluated by two case studies. The first corresponded to a small synthetic network and the second to a medium-size utility network. After a pipe burst event analysis, two different isolation actions were scrutinized in each of the two WDNs. The results quantify system resilience and support water utility managers in further decision-making processes. This is done through critical resilience indicators that provide information and support for better crisis preparedness (planning) and management (mitigation).

This research investigated the dynamic effects of iron bacteria–formed biofilms and associated material accumulation on pipeline head loss and roughness for the Salt Interception Scheme (SIS) in South Australia. Two case studies were conducted in high-density polyethylene (HDPE) pipelines pumped with ground water that contained iron bacteria. The friction factor and equivalent pipe roughness were calibrated using the measured head loss and flow. The topological and microstructure of the material accumulation were studied, and deoxyribonucleic acid (DNA) analysis was used to reveal the bacteria that contributed to the material accumulation. The results showed that, for pipelines with various diameters but at similar initial flow velocity (Case study 1), the material accumulated faster and imposed more significant head loss in smaller pipes. For pipelines with the same size (Case study 2), the material accumulated faster and introduced higher head loss in pipes in which the flow velocities were higher. The conventional calibration approach that assumes a constant pipe diameter is not appropriate for pipes with excessive biofilm growth (as the case in this study). The approach that considers a 2-mm reduction in pipe effective diameter with every 1-mm increase in equivalent roughness produced friction factor and roughness results in reasonable physical ranges, and the calibrated equivalent roughness was generally consistent with the observed thickness of the accumulated material. The material accumulation was highly nonuniform, and comprised microbial iron nanowires. The nanowire had a helically coiled structure which was different from that of common brown compounds of oxidized iron. The bacteria Mariprofundus ferrooxydans were the producers of the iron nanowires. Overall, to limit the growth of iron biofilms and material accumulation in pipes used in the SIS, smaller pipes should be avoided, and use of the largest pipe diameter practical to achieve low flow velocity is recommended.

The uncertainty of a model to determine the damping of pressure head oscillations following a sudden valve closure in a simple piping system in a pressurized closed conduit is quantified using Bayesian inference. The joint probability density of the model parameter is estimated based on experimental results published in the literature as well as experiments performed at the University of South Carolina. A Markov chain of the posterior joint distribution of the model parameters is calculated and used to predict the pressure head oscillations. The prediction is performed in a probabilistic fashion, estimating an interval of pressures as a function of time rather than estimating a single point by a traditional regression analysis. The 95% high posterior density of the damping ratio ranges from 1% to 6%. The probabilistic model also correctly predicts the experimental damping ratios when compared with experimental data.

The operating environment of water infrastructure is inherently dynamic due to factors such as water-related events, media coverage of such events, or changes in institutional expectations. This dynamic environment impacts public perceptions of infrastructure services and providers. In an effort to understand public views regarding water infrastructure, decision makers may periodically assess perceptions to aid in pursuing successful community-supported management strategies. However, single cross-sectional samples from previous studies do not consider the changes in perceptions over time. This study assessed dynamic perceptions of local water providers using two surveys deployed in 21 US shrinking cities, before (2013) and after (2016) media attention on nationwide water infrastructure challenges arising from events such as the Flint water crisis. Questions of interest pertained to whether respondents trust their water providers and want to partake in participatory processes of local utilities. Results indicated a statistically significant change in perceptions, highlighting the importance of temporal sampling. Statistical modeling revealed that geographic parameters typically have homogeneous impacts, indicating localized factors influencing utility–customer relationships in shaping perceptions, whereas demographic parameters have heterogeneous impacts.

The very geography of Nepal and India is such that these two countries must engage with each other in social, economic, and political spheres; water resources is one of the most important spheres. This study reviews the experience of cooperation in the Kosi basin between India and Nepal, focusing on the perceptions of risks and opportunities by decision makers. The following five categories of risk are examined at four tipping points, 1954, 1966, 1991, and 2015: capacity and knowledge, accountability and voice, sovereignty and autonomy, equity and access, and stability and support. The study provides deeper insights about the perceived risks as far as negotiations on the proposed Kosi high dam are considered. The perceived risks have grown significantly from 1954 to 2015 for Nepal as compared to India. In 2015, Nepal has all five categories of perceived risks. The study also proposes risk reduction strategies to reduce the level of perceived risks for both Nepal and India, which is an area of future research. The study may facilitate the negotiation process for the Kosi High Dam Project leading to mutually agreed win-win solutions for both riparian countries.

Water resource management is a complex process involving multiple stakeholders with different interests and different expertise. Typically, planning and management of water resources incorporate the analysis of social, economic, and environmental impacts as well as the consideration of hydrology and energy aspects. This study demonstrated the applicability of multicriteria decision analysis (MCDA) methods for the planning and management of a system of multipurpose reservoirs given competing development goals. We applied the methodology to the Mahaweli multipurpose project of Sri Lanka. The economic, social, and environmental planning goals were evaluated using multicriteria decision analysis techniques with multiple attributes measured in several units and trade-off weights elicited from multiple decision makers. Results showed that decision makers tend to prioritize objectives according to their institutional affiliations; however, they also demonstrate an understanding of the importance of other, less preferred objectives. The best alternative chosen by the models performed well across all objectives, decision makers, and MCDA methods. Sensitivity analyses confirmed the robustness of the best alternative under varying input parameters.

Boosting chlorine concentrations at intermediate locations within water distribution systems is a common way to maintain a specified minimum level of disinfectant to control microbiological regrowth. At the same time, maximum levels of both chlorine and its disinfection by-products should not be exceeded. Various mathematical procedures are available to find optimal solutions for the locations and dosing schedules of these booster plants. The problem is that most formulations have assumed unrealistic linear kinetics for chlorine decay in bulk water and at pipe walls, resulting in unrealistic solutions. Recent research shows that accurate prediction of chlorine and by-product concentrations resulting from initial and booster dosing requires more complex kinetic representation, in which chlorine reacts with dissolved substances remaining after treatment and with biofilms adhering to pipe walls. Optimization then requires more general nonlinear solution techniques. In addition, minimization of the total costs involved, both operating and capital, is needed to match the decision-making processes of water utilities in this field.

Statistical process control (SPC) identifies any nonrandom patterns in the system output variables of a water distribution system (WDS) by comparing them to their normal historic mean and variance. While each SPC method has different performance characteristics, there has been little effort expended to develop a hybrid method that combines the different characteristics. This paper proposes a hybrid SPC method that combines a modified Western Electric Company (WECO) method and the cumulative sum (CUSUM) method. First, the original WECO method is modified to incorporate a user-defined parameter $c$ that manipulates the tolerance for warning and control limits to fit the specific network of interest. Then, the best parameter set is identified for each of the two individual methods so that coupling them should not increase false alarms. The detection effectiveness and efficiency of the WECO, CUSUM, and hybrid methods were compared by using common data sets obtained from a hydraulic model of the Austin network. The results showed that a simple coupling of individual SPC methods with different detection characteristics can significantly improve pipe burst detection probability while reducing false alarm rates and average detection time.