Journal of Water Resources Planning and Management

Forum papers are thought-provoking opinion pieces or essays founded in fact, sometimes containing speculation, on a civil engineering topic of general interest and relevance to the readership of the journal. The views expressed in this Forum article do not necessarily reflect the views of ASCE or the Editorial Board of the journal.

Water distribution systems are vulnerable to hazards that threaten water delivery, water quality, and physical and cybernetic infrastructure. Water utilities and managers are responsible for assessing and preparing for these hazards, and researchers have developed a range of computational frameworks to explore and identify strategies for what-if scenarios. This manuscript conducts a review of the literature to report on the state of the art in modeling methodologies that have been developed to support the security of water distribution systems. First, the major activities outlined in the emergency management framework are reviewed; the activities include risk assessment, mitigation, emergency preparedness, response, and recovery. Simulation approaches and prototype software tools are reviewed that have been developed by government agencies and researchers for assessing and mitigating four threat modes, including contamination events, physical destruction, interconnected infrastructure cascading failures, and cybernetic attacks. Modeling tools are mapped to emergency management activities, and an analysis of the research is conducted to group studies based on methodologies that are used and developed to support emergency management activities. Recommendations are made for research needs that will contribute to the enhancement of the security of water distribution systems.

This study investigates the role of parameter uncertainty on evaluation of stormwater control measures. Specifically, several formal and informal Bayesian uncertainty analysis techniques are linked with the modified Fair and Geyer model to predict effluent total suspended solids (TSS) concentrations from extended detention basins (EDBs). Moreover, a global sensitivity analysis is performed to identify the most important parameters of the model. The results indicate that TSS removal in EDBs is most sensitive to the particle size distribution and particle density of solids in the runoff entering EDBs. Although formal Bayesian techniques estimated narrower prediction intervals, the inclusion rate of observed data was substantially lower than those estimated using informal methods. These results suggest that formal Bayesian methods may overconstrain the posterior parameter space and hence culminate in underestimation of the uncertainty in pollution removal effects of EDBs. The study reveals that selection of uncertainty analysis methods should be carefully conducted to ensure statistically rigorous and appropriate probabilistic characterization of the performance of stormwater control measures and the accompanying uncertainty in that performance.

Deriving optimal reservoir operation rules in a selective withdrawal framework (SWF) considering quality and quantity objectives is a challenging problem due to its computational burdens. To overcome the computational bottleneck, two surrogate models have been developed and coupled with the evolutionary algorithm in an adaptive-recursive framework to form the surrogate-based multiobjective optimization technique (SBMOOT). SBMOOT is used to derive the optimal reservoir operating strategies and the set of nondominated optimal solutions to enhance reservoir outflow water quality and maximize water supply and hydropower energy generation. The most desirable scenarios of the Pareto front have been identified to derive monthly operating rules in the SWF. The operating rules focusing on water supply, hydropower energy, and water quality objectives are estimated using polynomial regression technique. The performances of the operating rules, with and without regarding the uncertainty of inflows, have been compared with the historical operating strategy in Karkheh Reservoir, Khuzestan, Iran. The results show that water quality measure may be enhanced while maintaining desirable water supply and/or hydropower energy compared with the historical operating strategy of Karkheh Reservoir.

Behavioral research on natural resource management has revealed a number of variables that can impact collective action. This research builds upon an interactive decision game using experimental economics methods with a focus on production decisions and the corresponding impact they have on ambient water quality. Using hierarchical clustering algorithms, four primary types of behavior are identified: competitive, hypercompetitive, cooperative, and hypercooperative. The results from the experiment are used to test the following three hypotheses: (1) financial incentives increase cooperative behavior, (2) increasing the number and frequency of water quality sensors increases cooperative behavior, and (3) the spatial location of the agents and sensors affect cooperative behavior. Mixed-effect multinomial logistic models reveal that policy incentives, sensor location, and frequency of sensing alter the behavioral strategies of decision makers in the experiment and that outcomes vary by spatial location. From a watershed planning perspective, minimal investments in advanced environmental monitoring/sensing systems can potentially have large effects in improving water quality; however, there is also some evidence of marginal diminishing returns associated with such investments.

This work introduces two novel methods to the solution of water distribution networks equipped with pressure-control valves based on the global-gradient algorithm (GGA). One method, which can be viewed as an extension of that used to estimate the variable speed coefficient of variable speed pumps, leads to a nonsymmetric system to be solved at each iteration by introducing a new unknown (the head loss at the valve) and imposing the head at the controlled node. The second method maintains the symmetry of the system to be solved by imposing the head at the controlled node, removing the pipe equipped with the valve, and adjusting the mass-balance equation in the other node of the removed pipe. The performances of the proposed methods were analyzed on four case studies, and the results were also compared to those of EPANET 2 in terms of hydraulic accuracy and efficiency.

Low/negative pressure events that increase the risk of contaminant intrusion may take place in distribution systems and may become more common in ageing infrastructure. Guidance on whether to issue an advisory after loss of pressure is based on the duration and extent of pressure loss and is accompanied by Escherichia coli monitoring obligation. In this paper, the limitations of E. coli monitoring to detect intrusion is demonstrated through hydraulic and water quality modeling using a conservative 5-h pressure loss and considering intrusion of raw sewage. Ingress of contaminated water and fate and transport of E. coli throughout a 30,077-node network are simulated using a realistic pressure-driven hydraulic model coupled to a multispecies water quality model (EPANET-MSX). For a chlorinated distribution system, the limited positive nodes show the challenge of any contamination confirmation even for the studied conservative scenario. In a chloraminated system, a larger number of nodes (2,905 nodes) experienced E. coli over the simulation duration compared to the chlorinated system (166 nodes), increasing the likelihood of detecting contamination. Overall, numerical predictions can guide utilities for redefining more reliable sampling strategies for both confirmation and clearance sampling. Large-volume sampling at at-risk nodes identified by advanced numerical models provide greater credence in negative results to manage boiling advisories.

Unreported partially/fully closed valves or other types of pipe blockages in water distribution networks result in unexpected energy losses within the systems, which we also refer to as faults. We investigate the problem of detection and localization of such faults. We propose a novel optimization-based method, which relies on the solution of a nonlinear inverse problem with ${\ell }_1$ regularization. We develop a sequential convex optimization algorithm to solve the resulting nonsmooth nonconvex optimization problem. The proposed algorithm enables the use of nonsmooth terms within the problem formulation, and exploits the sparse structure inherent in water network models. The performance of the developed method is numerically evaluated to detect and localize blockages in a large water distribution network using both simulated and experimental data. In all experiments, the sequential convex optimization algorithm converged in less than 3 s, suggesting that the proposed fault detection and localization method is suitable for near real-time implementation. Furthermore, we experimentally validate the developed method for near real-time fault diagnosis in a large operational water network from the United Kingdom. The method is shown to successfully detect and localize blockages, with real system modeling uncertainties.

We demonstrate that a nonhomogeneous hidden Markov model (NHMM) can be useful for simulating future daily rainfall at 19 stations in South Florida. Using upper atmosphere circulation variables that are typically better represented than precipitation in general circulation models (GCMs), a NHMM conditioned on GCM circulation variables is shown to provide credible stochastic simulations of daily precipitation for future conditions. Seasonality changes as well as changes in seasonal extreme precipitation quantiles, total seasonal rainfall, and number of wet days are assessed. The Coupled Model Intercomparison Project phase 5 simulation of the coupled ocean-atmosphere Euro-Mediterranean Center on Climate Change Climate Model CMCC-CMS for 1948–2100 is used for the demonstration. Seasonality changes emerge naturally from the driving variables, and each season is not modeled separately. The future projections for CMCC-CMS indicate that South Florida may have drier conditions for most of the year. The number of wet days reduces, while extreme rainfall frequency increases. These findings are consistent with recent rainfall trends. A modest reduction in total rainfall in the February–May period and a slight increase in the September–October projected rainfall is noted. Changes in the expression of the North Atlantic subtropical high in the CMCC-CMS simulations appear to influence the new seasonality and patterns of rainfall.

The use of pipes no smaller than 150 mm (6 in.) in diameter is often recommended for fire protection in North America. This work examines the some of the costs and benefits of this restriction by looking at a single pipe in isolation. First, we argue that North American fire flow requirements are quite conservative by international standards, with European requirements approximately 25% of those in North America. It is shown that smoother 100-mm PVC in place of older, rougher 150-mm cast iron can produce 60% of the available fire flow, in principle still exceeding the European requirement. Furthermore, the estimated capital cost is reduced by 30%, and water age by 56%. No differences in energy use were observed, owing to very low demands in normal service. A simple model of biological growth showed some potential for increased biological growth in smaller pipes, however. Smaller pipes likely have more dynamic shear stresses, which can mitigate discoloration. Overall, there may be many benefits if smaller-diameter pipes are permitted in low-density suburban service. Fundamentally, the amount of water needed to fight modern fires in North America is largely unknown, suggesting a need for additional research.

Water distribution systems (WDSs) are critical infrastructures that supply drinking water from water sources to end-users. Smart WDSs could be designed by integrating physical components (e.g., valve and pumps) with computation and networking devices. As such, in smart WDSs, pumps and valves are automatically controlled together with continuous monitoring of important systems’ parameters. However, despite its advantage of improved efficacy, automated control and operation through a cyber-layer can expose the system to cyber-physical attacks. The one-class classification technique is proposed to detect such attacks by analyzing collected sensors’ readings from the system components. One-class classifiers have been found suitable for classifying normal and abnormal conditions with unbalanced datasets, which are expected in the cyber-attack detection problem. In the cyber-attack detection problem, typically, most of the data samples are under the normal state, while only a small fraction of the samples can be suspected as under attack (i.e., abnormal state). The results of this study demonstrate that one-class classification algorithms can be suitable for the cyber-attack detection problem and can compete with existing approaches. More specifically, this study examines the support vector data description (SVDD) method together with a tailored features selection methodology, which is based on the physical understanding of the WDS topology. The developed algorithm is examined on the Battle of the Attack Detection Algorithms (BATADAL) datasets that demonstrate a quasi-realistic case study and on a new case study of a large-scale WDS.

Using the water source as a factor, a typology of residential irrigation water users was developed by segmenting 3,310 Floridians who used irrigation by their irrigation water source (i.e., well water, reclaimed water, and city water). Based on three years of survey data, there was a moderate association between living in a homeowners’ association and using city or reclaimed water for irrigation. There was an association between the water source and engagement in nine conservation behaviors. Well water users were less likely to use recycled water, use a rain sensor, calibrate their sprinklers, and use smart irrigation controls. Reclaimed water users were more likely to use recycled water and use a rain sensor and also less likely to have retrofitted a portion of the landscape so that it is not irrigated and have turned off zones or capped irrigation heads for established plants. City water users had the strongest personal and social norms surrounding water conservation, although this did not translate into conservation as the theory indicates it should have. The findings reveal that identifying the irrigation water source can provide meaningful insights into outdoor water use and should be integrated into residential water conservation interventions.

Water distribution systems reliability has typically been studied using single-period simulations of the hydraulic behavior of the system, due to high computational requirements. The aim of this work was to develop and test a framework for the evaluation of mechanical reliability, hydraulic reliability, and firefighting reliability under extended-period simulations. Four functionality functions were proposed to quantify the serviceability of the system under the perturbation scenarios. The functions were tested in five case studies based on real-world networks and compared with the previously developed mechanical reliability estimator (MRE) and hydraulic reliability estimator (HRE), as well as with similar definitions for single-period simulation. Results showed that both MRE and HRE consistently correlate with the supply/demand ratio functionality, and are easy to compute even in optimization routines. Comparison results with single-period definitions of functionality showed that, on average, evaluating hydraulic conditions at the peak hour works is a good estimation of the extended-period behavior, but in some cases, discrepancies up to 50% can be found.

As large electricity consumers, water distribution system (WDS) pumping stations have the potential to become meaningful participants in demand response (DR) programs. The authors propose an optimization framework for assessing the DR capacity of a WDS and identifying the optimal bidding strategy for maximizing WDS revenue in the DR spot market. The proposed mixed integer linear programming (MILP) model overcomes computational constraints of previous DR optimization models by adopting a preprocessing procedure to minimize the number of binary variables and implementing a convex relaxation technique to linearize the hydraulic equations. The proposed MILP model also explicitly accounts for varying levels of risk tolerance of WDS operators by varying the recovery period over which pumping returns to business-as-usual operation. The optimization framework is implemented on a skeletonized 48-node WDS model that includes 7 pumps, 6 tanks, and 39 pipes. Using a simulated DR event and water consumption profile, the authors derive the optimal DR supply curves (i.e., compensation price versus load curtailment quantity) and revenue potential of the WDS under six scenarios for DR participation.

The value of water use quantification assessments is hindered by the use of inconsistent terminology and reporting standards. Challenges associated with data collection and maintenance are made unnecessarily worse by the community’s lack of agreement on definitions and reporting standards. Three major problems stand out: terminology conflicts, imprecise units, and data integrity. This work illustrates the impact of these problems using recent work on water use in the US energy system as a case study. Relatively minor changes to the definition of water consumption can change reported water consumption by $-50\%$ to $+270\%$, with no change to underlying data. Quantitative impacts of imprecise units and data integrity are more difficult to estimate, but this work demonstrates that minor changes to reporting standards in these realms can substantially improve certainty. This article identifies major terminology conflicts and recommends a mass flow–based approach to definitions, with the goal of clearly separating conversations about water quantity versus quality. Regardless of chosen approach, standardizing terminology and reporting within the research community can improve data quality at no to low cost.

End-uses at water supply systems typically follow a random pulse behavior, which blurs as consumptions are aggregated upstream, affecting flow rate variability along the spatial scale. Instantaneous variability has impact on the capacity of a hydraulic model to represent rapidly changing flow network scenarios, but traditional models only simulate average conditions. This paper analyzes the spatial scale effect in instantaneous flow variability by making use of a novel analytical approach to SIMDEUM (standing for SIMulation of water Demand, an End-Use Model) microcomponent-based stochastic demand model. Analytical results show good correspondence to previous results at the Benthuizen case study and demonstrate the potential use of the approach to assess the effect of network size in a realistic system. Results prove that demand coefficients of variation increase in the periphery of water systems according to power laws, highlighting the necessity of considering real variability rather than average conditions in these areas where real water flows never correspond to average flows. This is of utmost importance when dealing with real measurements and water quality applications.

This paper proposes the modulation of nodal outflows in water distribution networks (WDNs) to solve the problem of low disinfectant concentrations at critical dead-end nodes, in which low flow velocities and long residence times cause excessive disinfectant decay. The slight increase in nodal outflows at these sites, which can be obtained through the opening of a blowoff at the hydrant site, can help to address this problem with no need to increase disinfectant doses at the source(s) or of install additional disinfectant booster stations. The methodology is based on the combined use of optimization and flow routing/water quality modeling of WDNs. The concentration of disinfectant at the source(s) and the values of nodal emitter coefficients at the critical dead-end nodes are the decisional variables to be optimized. Two objective functions are considered in the optimization: the total volume of water delivered in the network (inclusive of supply, leakage, and additional nodal outflow considered for fixing disinfectant residuals); and the total mass of disinfectant injected into the network. The effectiveness of the methodology was proven on a real WDN, yielding insight into the economic feasibility of the solution.

The paper presents the results of the Battle of Postdisaster Response and Restoration (BPDRR) presented in a special session at the first International water distribution systems analysis & computing and control in the water industry (WDSA/CCWI) Joint Conference, held in Kingston, Ontario, Canada, in July 2018. The BPDRR problem focused on how to respond and restore water service after the occurrence of five earthquake scenarios that cause structural damage in a water distribution system. Participants were required to propose a prioritization schedule to fix the damages of each scenario while following restrictions on visibility/nonvisibility of damages. Each team/approach was evaluated against six performance criteria: (1) time without supply for hospital/firefighting, (2) rapidity of recovery, (3) resilience loss, (4) average time of no user service, (5) number of users without service for eight consecutive hours, and (6) water loss. Three main types of approaches were identified from the submissions: (1) general-purpose metaheuristic algorithms, (2) greedy algorithms, and (3) ranking-based prioritizations. All three approaches showed potential to solve the challenge efficiently. The results of the participants showed that for this network, the impact of a large-diameter pipe failure on the network is more significant than several smaller pipes failures. The location of isolation valves and the size of hydraulic segments influenced the resilience of the system during emergencies. On average, the interruptions to water supply (hospitals and firefighting) varied considerably among solutions and emergency scenarios, highlighting the importance of private water storage for emergencies. The effects of damages and repair work were more noticeable during the peak demand periods (morning and noontime) than during the low-flow periods; and tank storage helped to preserve functionality of the network in the first few hours after a simulated event.

Partitioning a water distribution system (WDS) into district metered areas (DMAs) is a difficult task due to the complex WDS structure and simultaneous consideration of multiple constraints. Further, alternative metrics can define DMA goals, and expressing and quantifying those objectives in an efficient algorithm is also challenging. To address the multifaceted set of objectives, a multiphase DMA design method is developed. DMA feed pipes are identified as the primary flow paths in a branched network. In the methodology presented here, the feed pipe network is first laid out by determining node clusters and boundary pipes that maximize the dissimilarity of pressures and modularities between clusters and minimize the number of cuts between clusters while defining the number of DMAs closest to the desired number. In the methodology’s second phase, secondary DMA feed pipes are identified by minimizing the number of secondary feed pipes and maximizing the nodal excess energy while maintaining desired pressure. Finally, a postoptimization analysis compares the performance of the Pareto solutions based on their availability, water quality, and daily leakage. System availability is calculated based on the minimum cut-set method combined with a new pressure-driven analysis method. To accelerate the optimization algorithm, two strategies are applied: step-by-step optimization and reducing the decision variable searching space by considering desirable DMA characteristics. The effectiveness of the methods is examined by applying it to the C-Town and real B-Town water distribution networks. Results demonstrate that the search space reduction method effectively decomposes the full network into DMAs in the face of multiple hydraulic and water quality metrics.

In a world that is ever more focused on energy efficiency and climate change mitigation, minimizing energy consumption associated with pumping groundwater is a growing concern. In this study, a aquifer storage and recovery simulation-optimization model (ASRSOM) is developed to optimize aquifer storage and recovery (ASR) wellfield operations. ASRSOM combines an analytical hydraulic model and a numerical optimization model to optimize wellfield operations. The objective function used to minimize energy consumption is the temporal integral of the products of temporally varying total dynamic head values and pumping rates. Comparison of ASRSOM results with work by others for idealized aquifer operations supports the validity of ASRSOM. Four scenarios were simulated to evaluate the role that optimization of operations and aquifer recharge plays in reducing the energy required to pump groundwater out of the aquifer. The operations of a municipal ASR wellfield located in the Denver Basin, Colorado, were simulated for a 10-year period. Optimization decreased energy consumption by $\mathrm{2,179}\mathrm{MW}\mathrm{h}$ of power ($-19.6\%$ compared with historical scenario) and 1,541 t of atmospheric carbon. For the conditions considered, managed aquifer recharge reduced power consumption by 1%. The limited benefit of recharge is attributed to the small recharge volume in the case study, the short duration of the analysis, and the depth of water levels in the aquifer. Additional opportunities to address economic and environmental impacts associated with groundwater pumping include optimal position of wells and factors controlling total pumping head.

This paper aims to show how the total demand signal (i.e., the sum of all user demands) of a district metered area (DMA) can be reconstructed using simple linear models starting from the readings taken at influential user locations with a preassigned temporal resolution (e.g., 1 h). By comparing the total reconstructed demand pattern with the inflow into the DMA, which is usually monitored at DMA boundaries using flowmeters, water utilities have the potential to identify occurrences of anomalous events, such as pipe bursts, unauthorized consumption, and leakage increases. To address this issue, two procedures are proposed that are applicable when smart meters at the end of life were previously installed at all locations and when no smart meter was present, respectively. The first procedure is based on the application of the stepwise regression that measures the demand time series aimed at both the selection of user locations and the model calibration, whereas Fisher’s test or economic considerations is used as stopping criterion. The second methodology consists of the application of criteria to identify the subset of representative locations on the basis of available data and practical considerations (user typology and consumption on the annual bill). In addition, a new method to calibrate the model using billed annual consumptions is provided. Both methodologies enable the construction of linear models that express the total pattern as a function of single-user consumption patterns, allowing the reconstruction of the original signal. Both procedures were applied to a case study in Naples (Italy) with a total of 1,406 users, that is, 1,067 residential users and 339 non-residential users. The results proved that, as expected, the accuracy of the total demand pattern reconstruction of both procedures increases as the sample size of the representative locations grows. However, the results indicating the trade-off between the sample size and the goodness of the fit reveal that the accuracy is good even for low model order (25–50 users selected). Indeed, the coefficient of determination, $R^2$, of the fit increases from 0.66 to 0.83 for 25 and 50 users selected, respectively, and the accuracy becomes almost perfect for a size of 100 (i.e., approximately 7% of the total number of users considered). Overall, this paper demonstrates that an effective strategy for the accurate characterization of the total demand in a DMA consists of distributing the preassigned number of smart meters between the different categories of users present in that DMA (i.e., residential and non-residential) and privileging users with the highest annual consumption.

Modeling intermittent water distribution networks (WDNs) requires effective incorporation of the transient flow conditions, including the transitions from pressurized to partial flow regimes and the inverse associated with periodic filling and emptying of pipes. Most existing hydraulic analysis models fail to account for the partial-flow conditions, and due to this reason, they cannot be directly applied for practical situations. This paper presents an improved pressure-dependent analysis (PDA) model, the PDA-PF model, integrating partial-flow characteristics in the algorithm. The PDA-PF model was developed based on the presumption that the pressure-deficiency triggers partial flows in pipes. The outflows at the demand nodes of highly intermittent WDNs are represented as uncontrolled orifice-based demands. In analyzing the pipe flows, the PDA-PF model performed better than the established models. The suitability of the proposed model for conducting extended period analysis was demonstrated by applying it to a real-world highly intermittent WDN of a municipality of India. Under the circumstances considered, partial-flow conditions existed in the distribution pipes for one-fourth of the total flow duration. The residual flows caused by the emptying of the pipes increased the volumetric water demand satisfaction at the low-elevation nodes of the considered WDN.

Due to the volatility, intermittency and, randomness of photovoltaic (PV) power output, hydropower is utilized as a complementary power source. The largest grid-connected hydro-PV power station, the Longyangxia project, is operational and located in northwestern China. The annual power station operational plan assists in making appropriate annual electricity purchasing and sales contracts and in economically benefiting the enterprise. Focusing on the Longyangxia project, this study proposes using a multiobjective optimization model to make a long-term equilibrium operational plan for a hydro-PV power system while simultaneously considering the benefits, stability, and tolerance of the power grid. The Non-dominated Sorting Genetic Algorithm version II (NSGA-II) is adopted to optimize the model, and the Pareto fronts for different hydrological years are presented. The results demonstrate the competitive relationships between the objectives, but there are distinct points where the relationships are more in line with one another. The inflection points are identified, based on the analysis, as the recommended solutions for operating the hydro-PV power system, and the corresponding reservoir operational schemes are also presented.

The uncertain nature of monsoon and the high temporal and spatial variability of rainfall in India make it imperative to utilize the available water resources optimally and efficiently. Reservoirs are one of the most effective means of matching the temporal and spatial availability of water with demands. In India, the state governments are considered custodians of the water in their respective states. Sometimes, water-sharing disputes arise in large river basins that are shared by multiple states. A mechanism to resolve such disputes is through tribunals, which allocate the basin waters to different states and projects in accordance with their past and projected demands, water availability at different dependability levels, and other constraints. In such basins, the operation of individual projects/reservoirs, so as to follow all the stipulations of the tribunal, is a challenging task. In this paper, a methodology has been framed to develop the operation procedure for individual projects so as to honor the tribunal awards. The Krishna River basin in India has been used as an example case study. The proposed procedure keeps track of the total inflows received, the total water used, and the balance utilizations to be made at each project, each state, and the whole river basin in real time. Using this procedure, the percent utilization achieved by different sharing states of the Krishna basin, especially in deficit years, is quite close to the values specified by the tribunal. The developed methodology is quite general and can be applied to any river basin with suitable modifications.