Journal of Water Resources Planning and Management

Sectorization of a water distribution network (WDN) into district metered areas (DMAs) is a proven solution for proactive leakage control. Traditionally, WDN sectorization is conducted by local experts using a trial-and-error approach, often resulting in the identification of arbitrary solutions. Some recently published methods try to improve WDN sectorization by automating the process, especially by using optimization. Various sectorization criteria, constraints, and limitations are introduced, which often fail to consider the issues faced by poorly managed WDNs such as limited funds and shortage of water balance data. These methods also have poor computational efficiency imposed by optimization methods used. This paper presents a new distribution network sectorization method (DeNSE), that overcomes these deficiencies. This method is based on a heuristic procedure in which WDN sectorization is driven by efficient tracking of water balance data and determining the lowest cost investment needed to maintain the same level of operational performance. The above-mentioned set of criteria is particularly well suited for initial sectorization of WDNs when major uncertainties in water balance data often lead to poor management decisions. The DeNSE method is validated and benchmarked against other sectorization methodologies in a case study of a large, real-world WDN. The results show that DeNSE can identify sound, realistic sectorization solutions that are in some respects better than corresponding solutions reported in the literature. DeNSE also enables high computational efficiency, ensuring its applicability to real-world WDNs.

To date, the hydraulics of water distribution networks are calculated using deterministic models. Because many of the parameters in these models are not known exactly, it is important to evaluate the effects of their uncertainties on the results through uncertainty analysis. For the propagation of uncertain parameters, this article for the first time applies the polynomial chaos expansion to a hydraulic model and compares the results with those from classical approaches like the first-order second-moment method and Monte Carlo simulations. Results presented in this article show that the accuracy of the polynomial chaos expansion is on the same level as the Monte Carlo simulation. Further, it is concluded that due to its computational efficiency, polynomial chaos expansion is superior to the Monte Carlo simulation.

The Everglades is a unique ecosystem in Florida that offers a variety of ecosystem services (ES), including water supply and flood risk reduction, water purification, habitats for several endemic species, and recreational opportunities. Therefore, the ramifications of climate change, an imminent threat to the Everglades ecosystem, are going to affect these ecosystem services drastically. The climate-induced changes to the ecosystem services are going to affect people’s lives and livelihoods in the region. Thus, understanding public perception and preferences is an important step in addressing the impending risk. In this study, we used an online survey to assess residents’ perceptions about the risk of climate change and their views on mitigating potential consequences. We find that people’s perceptions and preferences are shaped by their level of education, age, participation in outdoor recreational activities, and elevation and distance from the shoreline of their residential location from the mean sea-level. In general, people who are concerned about the flooding risk and people who tend to prioritize environmental conservation are more likely to agree with the potential impacts of climate change and sea-level rise and more likely to support proposed measures for adaptation. We discuss the policy implications for developing socially acceptable plans based on the findings regarding people’s perception and preferences for adaptation in this region.

Urban water governance is a framework for addressing critical issues in cities’ water resources by creating a more inclusive policy for sustainable transitions. Beyond the physical or direct water resources available to cities, there are water resources embedded within the imported food, energy, electricity, and other materials required for urban life and growth. Combined, these two types of water resources create the water footprint of a city. Expanding the framework of urban water governance to promote a sustainable water footprint is necessary for advancing urban environments in the face of climate change and resource constraints. We discuss the need for open-access data to transition toward a sustainable urban water governance system, considering both direct and indirect resources. This discussion builds on existing literature and asserts a broader goal of achieving a blue city. To this end, we identify a few attributes of the Boston, Massachusetts Metropolitan Area to highlight data and governance opportunities for both its direct and indirect water resources.

Sustainable water management requires understanding the factors that influence water use across multiple time scales, spatial scales, and types of use. However, existing empirical research on water use largely consists of studies at either the municipal or national scale, leaving a sizeable gap at intermediate scales important for planning (e.g., watershed, state, and basin level). This work addresses this gap by using a mixed-effect panel regression of monthly water withdrawals in Virginia to evaluate how well statistical modeling approaches can characterize and explain multisectoral withdrawals. Model fit is high across all sectors, suggesting that statistical models can be effective at these scales as long as they are formulated in a manner that accounts for significant variance in withdrawal volumes and temporal trends. Multiple climatic and economic variables are found to be significantly associated with withdrawals in all sectors evaluated. These relationships suggest that withdrawals in humid regions exhibit similar sensitivities to arid regions that have been the focus of more research and that incorporating economic factors is particularly important for estimating energy and industrial water withdrawals.

Design storms are an essential element in the design of urban drainage systems. While it is accepted that systems optimized with respect to design storms should perform well under similar rainfall conditions, there is a lack of knowledge on when and why design storms may fail to produce robust solutions—namely, drainage systems that perform consistently under a broad range of rainfall events. To address this question, we contribute a computational framework that evaluates the robustness of drainage systems optimized for a design storm. The framework consists of four building blocks. First, we use sensitivity analysis to identify the most important decision variables [e.g., pipe expansions and low impact development (LID)], thereby reducing the complexity of the design problem. Second, we solve the problem using a multiobjective simulation-optimization scheme, which yields a set of Pareto-efficient solutions optimizing various measures of performance. Third, we simulate each solution under stochastic rainfall events characterized by different duration, intensity, and profile, and finally evaluate their robustness (i.e., the capability of attaining the performance they are designed for). The application of the proposed framework to the Nhieu Loc-Thi Nghe basin (Ho Chi Minh City, Vietnam) indicates that none of the Pareto-efficient solutions are robust across all rainfall events. In particular, we find that the optimized solutions underperform during (1) rainfall events with smaller intensity, but greater depth, than the design storm, and (2) small, but frequent, rainfall events. Results for our case study also suggest that the expansion of the drainage network is more likely to give a robust design than the implementation of LID technologies. Overall, our work elucidates the need to include stochastic rainfall events throughout the design process so as to obtain robust drainage solutions.

Water accounting—the measurement, processing, and communication of information on water availability and use—is essential for effective and sustainable water resource management. Such information becomes especially important during droughts. It facilitates transparent, flexible, and efficient water allocations that benefit the economy and environment, and reduces conflicts among users. This is especially crucial in dry regions such as the western United States, where water demands often exceed available supplies. A web-based review of water accounting practices, supplemented by interviews with more than 50 experts, informed a comparative analysis of how 12 western states and two other countries—Australia and Spain—gather, process, and share information to manage water. These regions were selected because of their similarities in water management challenges, but also because they have comparable levels of economic development and institutional capabilities. We focus on how water availability is assessed; how varying institutional and legal frameworks define water claims; how water use is quantified; how water allocation decisions are made based on regulatory and physical constraints and available information; and finally how the information is shared to improve decision making. We found a huge variety of accounting practices to support different water accounting frameworks in developed regions. This variability is a legacy of the historical evolution of water management practices and challenges, and is usually driven by cultural practices, institutions, and legal frameworks. Drawing on the comparative review, we select management practices that strengthen water accounting. Although these lessons are based on places that share resource-management challenges of advanced economies in dry regions, they can be useful for developing regions to reduce common conflicts in managing groundwater sustainability, dedicating adequate water to support the natural environment, or allocating surface water during times of scarcity.

The reliability of water distribution network (WDN) designs for satisfying water demands under normal and abnormal operating conditions, including during pipe bursts and as systems age, is typically evaluated with surrogate measures of system resistance to potential hydraulic and mechanical stressors. The resilience index and its modifications have been applied to assess the reliability of a WDN in the face of hydraulic variances, including alterations in network friction characteristics, inputs, and demands. A limitation of these indices is that they may only be used as an indirect measure of network reliability in the face of mechanical failures. A uniformity index is proposed to address this limitation. While the definition of the index is simple, the maximization of it is expected to lead to network designs with an increasingly uniform flow, at each node, in which the burden of carrying inflows (or outflows) at each node is shared evenly among all input (or output) pipes. Using a single-loop network example, this paper demonstrates that increasing this index leads to increased mechanical reliability. The incorporation of the uniformity index, along with the goals of maximizing the resilience index and minimizing design costs, in a multiobjective optimization modelling framework, is likely to result in WDN designs with increased mechanical reliability, relative to those based solely on resilience and cost. The efficacy of the multiobjective approach that includes these goals, for improving overall reliability in response to mechanical and hydraulic disturbances, is demonstrated for two WDNs, the two-loop benchmark, and the municipality of Farhadgerd, Iran.

Reservoirs change the environment upstream of dams from lotic to lentic and alter hydrodynamic migration cues, thereby forming barriers to upstream fish migration. Increasing reservoir capacity improves its ability to store water and provide benefits for human water uses but may simultaneously degrade fish migration environment in the reservoir and increase the economic cost for restoring fish migration. Such a tradeoff has been rarely explored. This study develops an integrated hydrodynamic, hydrologic, and economic framework for planning reservoir capacity that accounts for fish migration; water use benefits; and reservoir construction, maintenance, and operation costs. Based on the hydrodynamic simulation of flow velocities along a potential fish migration route in a reservoir, we determine the minimum flow release required to restore migration for different inflow and storage conditions and fish hydrodynamic requirements. The minimum release is then expressed as a constraint in an optimization model that determines the optimal reservoir capacity and a new operating rule curve to maximize reservoir net benefit. Using China’s Danjiangkou Reservoir as a case study, we confirm that increasing reservoir capacity increases the quantity of water abandoned for restoring fish migration. Accounting for fish migration decreases optimal reservoir capacity by 39%, decreases annual costs by 13%, and decreases annual net benefits by 8%–21%, depending on fish hydrodynamic requirements. Large reservoirs may need to sacrifice considerable storage both before and during the migration period, and thus the optimal capacity for a particular reservoir should be decided based on a balance between human water demand and fish migration requirements.

This paper puts forward an improved risk assessment model for a real-time reservoir flood control operation according to the stochastic differential equation (SDE). The model is composed of three components: (1) uncertainties and risk definition, (2) SDE and reservoir flood routing, (3) and risk calculation. The first component presents the considered uncertainty factors and risks. We consider the following six kinds of uncertainties: forecast errors of reservoir inflows (IFE), reservoir outflow errors (OE), observation errors of a reservoir storage capacity curve (SOE), correlations between IFE, correlations between OE, and correlations between SOE. We consider two kinds of risks: the risk of upstream reservoir flooding at every time moment and the total risk of the flood process. The second component establishes the SDE and then conducts reservoir flood routing and derives stochastic water level errors. The third component derives calculation formulas for the proposed risks. The established model was applied to the Foziling reservoir of China. The results reveal IFE is the critical uncertainty factor, while OE and SOE are both nonnegligible in real-time operation. The risks of the reservoir increase with the uncertainty level of uncertainty factors. However, the correlation coefficient between the stochastic water level errors decreases with an uncertainty level of IFE while it increases with an uncertainty level of OE and SOE. The proposed improved risk assessment model is considered as the risk-informed tool for real-time reservoir operation.

Short-term water demand forecasting models address the case of a real-time optimal water pumping schedule. This study focuses on developing artificial neural network (ANN) models to forecast water demand 24 h and 1 week ahead. A number of studies have shown that the relationship between water demand and the driving variables is nonlinear. Two ANN time-series models were developed, a nonlinear autoregressive with exogenous (NARX) model with historical demand and weather data as an exogenous input, and a nonlinear autoregressive (NAR) model with only historical demand as an input. This investigation examines how model structure, length of historical data span, and improvement of an exogenous input can influence model performance. The results show that on average, using a nonlinear ANN model can improve water demand prediction by 18% and 25% when forecasting 24 h and 1 week ahead, respectively. The results also show that training the model (i.e., NARX) with correlated exogenous parameters dropped the error by 30% on average compared with a single-input model. In addition, using historical data for only 4 months compared with 5 years and 1 year decreased the error by 76% and 68% for NARX models and 35% and 33% for NAR models, forecasting 24 h and 1 week ahead, respectively.

Urban areas are facing challenges for the provision of public services, with water scarcity arising as one of the main problems. A twin track approach of supply and demand management is essential, and water-loss management contributes to reducing water demand. However, small municipalities from developing countries have technical, information, and financial limitations to locate and monitor water losses. This paper presents the estimation of real and apparent losses in a small municipality from a developing country in a data-scarce situation. For this, several tools were used, allowing data integration that resulted in a water balance, from which water losses were estimated at 46%, and four alternatives for water-loss reduction were developed. A cost-benefit analysis and financial indicators were estimated for the proposed alternatives, resulting in a water savings of 19%, a payback period of 3 years and an internal rate of return of 39%. The proposed strategies have potential to improve water quantity and quality, the technical stability of the system, utility performance, and water security.

Fluid pressure influences leakage flow rate in water distribution pipe networks. Significant progress has been made in the use of pressure management techniques to control leakage. An empirical power equation (known as the $N1$ power equation) is widely used to model the pressure–leakage relationship in practice. However, recent research has shown that this equation is not able to accurately model leakage under various conditions. When estimating the N1 leakage exponent for a system, the power equation assumes that leakage is located at a point in the network that is representative of the average system pressure, which is rarely true. In addition, parameters of the power equation are not constant but vary with pressure, which is a problem especially for water distribution pipe networks where pressure varies with time, for example, due to diurnal demand variation. These factors result in significant errors when the power equation is used to model leakage in real water distribution networks. This technical note analyzes the factors that introduce errors in the power equation and proposes methods to improve its accuracy. The performance of the methods is demonstrated using a case study.