Real-time analysis and operation of drinking water networks (DWN) are key to improving water quality throughout the network, reducing operational cost and improving emergency event response. Network calibration, (i.e., pump head and energy curve calibration, valve calibration, valve closure analysis, etc.) is an essential first step toward the development of a digital twin and real-time operational analysis. A typical drinking water network contains a large number of isolation valves (similar to 10(3)), many of which are often required to be manually closed to restrict or redirect flows (e.g., to perform maintenance operations on parts of the network). Some of these closed valves may remain closed (often unintentionally) without proper records and/or incorporation in a model. Such closed valves can become a major source of energy dissipation in the system and can potentially create water quality issues. Hillsborough County Water Resources Department (HCWRD) is adopting a state-of-the-art digital twin platform capable of real-time monitoring and analysis. As part of the development of the real-time network model, calibration steps were taken to improve model performance, and a valve closure analysis was performed when the real-time data indicated unusual energy loss in the system that could not be explained using a model without closed valves. The valve closure analysis was performed using a greedy search method utilizing real-time pressure data and network state information. Field exploration of the network performed by HCWRD discovered some of the actual closed valves very close to what was indicated by the valve closure analysis.Real-time analysis and operation of drinking water networks (DWN) are key to improving water quality throughout the network, reducing operational cost and improving emergency event response. Network calibration, (i.e., pump head and energy curve calibration, valve calibration, valve closure analysis, etc.) is an essential first step toward the development of a digital twin and real-time operational analysis. A typical drinking water network contains a large number of isolation valves (similar to 10(3)), many of which are often required to be manually closed to restrict or redirect flows (e.g., to perform maintenance operations on parts of the network). Some of these closed valves may remain closed (often unintentionally) without proper records and/or incorporation in a model. Such closed valves can become a major source of energy dissipation in the system and can potentially create water quality issues. Hillsborough County Water Resources Department (HCWRD) is adopting a state-of-the-art digital twin platform capable of real-time monitoring and analysis. As part of the development of the real-time network model, calibration steps were taken to improve model performance, and a valve closure analysis was performed when the real-time data indicated unusual energy loss in the system that could not be explained using a model without closed valves. The valve closure analysis was performed using a greedy search method utilizing real-time pressure data and network state information. Field exploration of the network performed by HCWRD discovered some of the actual closed valves very close to what was indicated by the valve closure analysis.

Conservative chemicals (such as sodium chloride) have been utilized to perform tracer studies within drinking water distribution systems. The resulting signals from a tracer study can provide significant quantitative information to assess the ability of a given network model to represent the underlying hydraulic and transport characteristics of the network. Often, however, the resulting observed water quality time-series data are simply visually inspected to assess the ability of the network model to accurately predict water quality transport. The use of standard quantitative metrics, such as arrival times, sum of squared errors (SSE), and correlation analysis at different time lags to assess the differences between the observed and predicted time-series, can provide some useful information but are not sufficient for paired data signals. In this study, the use of dynamic time warping (DTW)-an approach for estimating the similarity between two time series of data-is presented as a method for quantitative analysis of observed and model-predicted conservative chemical time-series data. DTW uses dynamic programming to match the elements of two time series, in a sequential approach, to minimize the SSE of the two signals. Whereas the SSE provides one goodness-of-fit metric, the resulting length of the warping path also provides additional information as to the degree of the alignment between the two data streams.

Water infrastructure simulation models, such as EPANET and SWMM, have played an important role in the development and dissemination of simulation technologies for water distribution, stormwater and sanitary sewer systems. These software packages have performed varying functions for different portions of the water infrastructure community. Municipalities, water utilities, and consultants use these packages to ensure adequate water

Adsorption to pipe wall materials significantly affects the fate and transport of certain contaminants in water distribution systems. For example, arsenate has a strong affinity for iron oxide, a substance common in water distribution pipes. In this paper a mathematical model for arsenate adsorption to iron oxide pipe wall materials is developed. The effects of two common assumptions on modeled arsenate transport are explored: a theoretical smooth pipe mass transfer coefficient and an assumption of rapid equilibrium of adsorption at the pipe wall surface. The effects of these assumptions are explored in a single pipe sensitivity analysis and found to yield significantly different results than parameters estimated from experimental data. In simulations of a hypothetical arsenate contamination event in a model water distribution system, the two assumptions result in different predictions of system contamination and contaminant exposure to consumers. These results indicate that even though water quality modeling plays an essential role in planning for distribution system decontamination, modeling assumptions must be carefully chosen.

Ecological bias introduced by spatial data aggregation causes significant variation in correlation statistics between pathogen exposures and illness rates. Modifiable areal unit problem sensitivity analysis is introduced to investigate the impact of spatial aggregation on ecological bias. Simulation produces numerical estimates for the relative magnitudes of components that effect ecological bias: (1) spatial autocorrelation of exposure concentrations; (2) scaling; (3) zoning; (4) network-clustered structure of illness events; (5) clustering of exposure measurements; and (6) the statistical distribution of exposure concentrations. These six components are mixed and used to compare random illness patterns to patterns determined from a dose–response model. Of the six, spatial autocorrelation of exposure data has the greatest influence on ecological bias. Spatial aggregation can cause high correlations in random illness patterns. More importantly, if pathogen concentrations are randomly distributed in space, then there is a greater likelihood that data aggregation might obscure a strong association.

Raw sewage discharges from Combined Sewer Overflows (CSOs) during rainfall can severely impact microbial drinking water quality. These pathogen loading events are episodic and short lasting. Traditional risk assessment that uses mean exposure values will therefore average out these short term duration high risk events, and result in underestimates of risk. A more accurate approach requires a characterization of the exposure dynamics that result from multiple upstream CSO communities, creating a challenge for more computational and data intensive urban watershed models. To this end, we developed a simple dynamic model of CSOs to estimate overflow discharges from combined sewer networks for river basin scale exposure assessments. The impervious subcatchment and sewer system are modeled as linear reservoirs in series. The overflow volume estimates of this CSO model were found to be in good agreement with a sophisticated hydrodynamic model (SWMM), and with real overflow data with R-2 values of 0.96 and 0.91, respectively. Pathogen loadings from CSO's were estimated by superimposing simulation of overflow discharges on raw sewage enteric pathogen concentration. We apply this simplistic approach to estimate pathogen concentration due to multiple upstream CSO's in a hypothetical river basin, demonstrating that this simplified model is suitable for representing the dynamics of CSO induced pathogen loadings into receiving waters. The model serves a framework to estimate the dynamics of pathogen loadings that are central to river basin microbial risk assessments.

In this paper, an overview of a strategy for automatic meter reading (AMR) data interpretation and aggregation is presented along with the proposed stochastic models adequate for representing the intrinsic characteristics of the data. Water demand measurements from single user accounts are obtained from an AMR system that continuously monitors consumption in different zones of Cincinnati, Ohio. The data represent volumetric measurements characterized by fixed increments, which depend on the sensitivity of the instruments used and occur at irregular times due to the polling method of the AMR system. Given the nature of the data, a nonhomogeneous Poisson process is proposed to model the arrivals of the increments within a selected time interval of 350days. An exponential-polynomial-trigonometric rate function with multiple periodicities (EPTMP) is assumed to describe both trends and periodicities in the observed data. A specific methodology for estimating the parameters of the EPTMP rate function is presented, based on the method of maximum likelihood. In order to evaluate the estimation technique, a performance evaluation is carried out on synthetic data generated in simulation. Finally, the estimation method is applied and tested on samples of the complete AMR data set, which is obtained from aggregating randomly selected subsets of different magnitude. The results provide significant evidence of the numerical stability and accuracy of the modeling procedure and encourage the use in simulation and prediction of water demands at network nodes from available AMR data.

Drinking water quality sampling in distribution systems has typically been approached in an ad-hoc manner, based on simple metrics such as geographic coverage that may, or may not, relate to sampling goals. Sampling goals will depend on the envisioned uses for the data: compliance monitoring, operation and control, or research (including model calibration). This paper develops a methodology for quantifying the worth of alternative sampling plans for compliance monitoring, where samples are collected to efficiently represent some feature of the true underlying distribution of space- and time-varying water quality. Conceptually, the evaluation of sampling plans could proceed through prior collection of an exhaustive historical data set. One could then retrospectively examine, for example, the frequency with which critical water quality conditions were identified by a particular sparse sampling of the data. Of course this procedure is impractical due to time and economic constraints, but such a procedure could be implemented using simulated water quality data, so long as the simulated data were thought to represent the true field conditions. The sampling design evaluation methodology developed and applied here is based on Monte Carlo simulation and advanced methods for simulating chemical reaction dynamics in water distribution systems. We have used a general Monte Carlo simulation tool built on an extension of the Epanet network model that allows for interactions between multiple species or components. This core capability allows for generation of alternative water quality scenarios that may represent the true variability and uncertainty in water quality, and reflect the best available knowledge about water quality models and processes for any given application. Significant work is needed, however, on probabilistic models that mimic variation in system boundary conditions, water demands, and network operation. The Monte Carlo framework will be applied to two illustrative situations: sampling of chlorine residuals in the distribution system to ensure adequate residual maintenance, and sampling of chlorination by-products to ensure consumer protection from these potential carcinogens. Both these applications can involve uncertainty and variability in kinetic model parameters, source water quality, water demands, and system operation. The latter application is timely in the USA as it relates to promulgation of the Initial Distribution System Evaluation (IDSE) rule, which requires utilities to develop sampling plans for disinfection by-products that are more likely to capture critical water quality conditions. These results will be presented at the WDSA2006 conference. This paper was presented at the 8th Annual Water Distribution Systems Analysis Symposium which was held with the generous support of Awwa Research Foundation (AwwaRF).

Calibration is a process of comparing model results with field data and making the appropriate adjustments so that both results agree. Calibration methods can involve formal optimization methods or manual methods in which the modeler informally examines alternative model parameters. The development of a calibration framework typically involves the following: (1) definition of the model variables, coefficients, and equations; (2) selection of an objective function to measure the quality of the calibration; (3) selection of the set of data to be used for the calibration process; and (4) selection of an optimization/manual scheme for altering the coefficient values in the direction of reducing the objective function. Hydraulic calibration usually involves the modification of system demands, fine-tuning the roughness values of pipes, altering pump operation characteristics, and adjusting other model attributes that affect simulation results, in particular those that have significant uncertainty associated with their values. From the previous steps, it is clear that model calibration is neither unique nor a straightforward technical task. The success of a calibration process depends on the modeler's experience and intuition, as well as on the mathematical model and procedures adopted for the calibration process. This paper provides a summary of the Battle of the Water Calibration Networks (BWCN), the goal of which was to objectively compare the solutions of different approaches to the calibration of water distribution systems through application to a real water distribution system. Fourteen teams from academia, water utilities, and private consultants participated. The BWCN outcomes were presented and assessed at the 12th Water Distribution Systems Analysis conference in Tucson, Arizona, in September 2010. This manuscript summarizes the BWCN exercise and suggests future research directions for the calibration of water distribution systems. DOI: 10.1061/(ASCE)WR.1943-5452.0000191. (C) 2012 American Society of Civil Engineers.

A water quality sampling model is developed that provides a quantitative basis for determining sampling locations and schedules. The deterioration of water quality in distribution systems is known to correlate strongly with water age. The sampling model developed here uses water age as the basis for determining sample "representativeness." A mixed integer linear programming formulation is described and an example application is developed. The application illustrates how sampling objectives can be parameterized within the model and how unmodeled issues effect the implementation of the sampling plans produced.

The development of NetSafe —a network simulation software for predictive simulation of the fate and transport of chemical and biological agents in drinking water distribution systems (DWDS)-is described in this paper. Using real-time hydraulic and water quality measurements and run-time models, NetSafe is capable of identifying the presence of contaminants and predicting their sources and destinations in DWDSfor vulnerability analysis and optimal emergency response. NetSafe uses a model-based anomaly detection approach that can be used with any type of sensor. The contaminant source identification leverages the Contaminant Status Algorithm (CSA), and the contaminant spread is determined using the EPANET MSX engine. NetSafe has been developed as .NET application written in C#. It is based on a robust design pattern called "Model-View-ViewModel (MVVM)" that permits scalability as well as flexibility. The user interface can be completely defined using an XML like syntax called the eXtensible Application Markup Language (XAML). NetSafe has the ability to connect to relational databases to obtain SCADA data, and it has been tested with both Microsoft SQL Server 2005 and Microsoft Access 2003. The results displayed by NetSafe were independently verified using an Excel-based workbench. The architecture evolved through the creation of use case diagram, class diagram, communication diagram, sequence diagram, state activity diagram. These capabilities of NetSafe were tested using "Network 3" from the EPANET sample data and multi-species water quality kinetics, and the initial results appear to be promising. The next phase of research will compare NetSafe with existing approaches using real-life networks.

Multi-species water quality models can be used to predict the fate and transport of contaminants such as arsenic in water distribution networks. In recent work, water quality models have been used to simulate hypothetical contamination events, estimate potential human health effects, and characterize the ability of sensors to detect contamination. Little work has been done to calibrate water quality models and validate them against experimental data generated in Distribution System Simulators (DSSs). In this paper, results are reported from bench scale and pilot scale experiments performed with a DSS at U. S. EPA’s Test and Evaluation Facility in Cincinnati, Ohio. The parameters for a reversible adsorption model were estimated from bench scale data generated over two days. The model was used with the EPANET-MSX software package to simulate the pilot scale experiment in the DSS. Model results match the pilot scale data very well for the first two days after the arsenate injection, however pilot scale data after this time deviates from model predictions. This deviation may be due to limitations in the time scale or sample size of the bench scale experiment. Additional modeling, simulation, and experimental work is planned to develop a fate and transport model that can be used in practical settings to design decontamination strategies following intentional arsenic contamination of water distribution systems.

Upon determination of a possible contamination threat in a water distribution network, a variety of response actions (e.g., public notification and operational changes) can be pursued in order to minimize public health and economic impacts and ultimately return the utility to normal operations. Flushing is a relatively common operational response option employed by utilities to address water quality concerns. Previously, an optimal hydraulic response tool was developed to help identify the best hydrant locations to flush. However, in order to apply this tool the contaminant injection location needs to be known. In previous research efforts, either the injection location was assumed to be known, or a sensor coverage map, which displays all contamination incidents potentially detected by a sensor, was employed to identify all possible injection locations. While the flushing locations selected for a known source location were effective in reducing impacts, the locations selected based on sensor coverage maps were not as effective. Therefore, in this study, a source location algorithm based on an event backtracking analysis is used to identify the most likely source locations. An example network model and multiple injection locations are used to evaluate the effectiveness of this approach. In addition, the reduction in impacts between the three different source identification approaches (i.e., known, sensor coverage map, backtracking) were compared. Overall, knowing the contaminant injection location greatly influences the effectiveness of the flushing response. For this study, the smaller amount of possible source locations, the greater the reduction in impacts. If only one source location is identified, the impact reduction could be as high as 98%. However, when 18 possible sources were identified from the sensor coverage map approach, only a reduction of 2% was achieved.

This research aims to explore how Human Computation can be used to aid economic development in communities experiencing extreme poverty throughout the world. Work is ongoing with a community in rural Kenya to connect them to employment opportunities through a Human Computation system. A feasibility study has been conducted in the community using the 3D protein folding game Foldit and Amazon's Mechanical Turk. Feasibility has been confirmed and obstacles identified. Current work includes a pilot study doing image analysis for two research projects and developing a GUI that is usable by workers with little computer literacy. Future work includes developing effective incentive systems that operate both at the individual level and the group level and integrating worker accuracy evaluation, worker compensation, and result-credibility evaluation.

This study is motivated by the need to develop stochastic models of water demand that can be applied to different scales of spatial and temporal aggregation to accurately represent hydraulics and water quality dynamics in water distribution systems. Previous work has produced models that were not used to represent spatial-temporal demands for extensive data sets, mainly due to limitations in collection of data necessary for testing and validation, in the mathematical structure of the models, and in the methods used for parameter estimation. The main goal of this work is to address such limitations by exploiting a unique opportunity to collect large volumes of data at the individual service connection (ISC) level and by introducing proposed enhancements to stochastic point process modeling methods. The methodology contemplates the implementation and testing of temporal models for single-site or ISC water demands, based on stochastic point processes. The Poisson Rectangular Pulses (PRP) and the Neyman-Scott Rectangular Pulses (NSRP) are the two point process models selected for representing the temporal variation of water demand. The model parameters account for the physical processes involved in water usage, namely, time dependent arrival rate, intensity, and duration of individual water usage events. A Bayesian parameter estimation methodology is proposed and is implemented using a Markov-Chain Monte Carlo (MCMC) method based on the Metropolis-Hastings algorithm. The MCMC approach produces samples from the posterior distribution of the model parameters providing more information than single point estimates.

The self-organization of social networks and the emergence of ideas have both been studied extensively in recent years, but seldom in a single framework. In this paper, we describe a distributed multi-agent model for the self-organization of social networks from encounters between agents with specific ideas, which are seen as combinations of words. Each agent maintains a semantic network of the words it knows, which implicitly defines the ideas in its repertoire. Agents exchange their ideas over their social networks, and incorporate the received ideas in their semantic networks. Social bonds are made and broken based on the agents' social and semantic preferences (i.e., shared ideas), leading to the emergence of social communities. Thus, the model embodies a circular interaction between the formation of social networks and new ideas. We mine the resulting communities for novel ideas that are generated by their members, and look at the effect of interaction choices on their formation.

New network modeling software such as EPANET-MSX allows researchers and, eventually, practitioners to develop and use water quality models that account for multiple reactive species in the distribution system. This capability allows for a more complete analysis of network water quality, including processes such as adsorption and biological inactivation by a disinfectant, and attachment and detachment of pathogens to and from pipe biofilm. In the author's opinion, such models offer the first true ability to model water quality in distribution pipeline networks, as single species models - which have been maintained as the standard for more than 20 years - can only consider rough surrogates of water quality (e.g., water age), and can not in principle consider chemical or biochemical processes and reactions that impact water quality (e.g., free chlorine is modeled as a single specie decay process, rather than as an interaction of free chlorine with natural inorganic and organic matter). Multi-species models require, however, that adequate process models be developed and investigated, and field-verified, before they can be used in practice; our field is only beginning to contemplate such work, which will require reaching out to environmental engineers, physical scientists, and microbiologists. The benefits of doing so should be worth the effort, as we may be able to finally develop water quality models of sufficient accuracy for quantitative assessment of public health risks, and development and evaluation of water quality sensing and management strategies.

As urbanization continues to increase and climate changes continue to impact our water cycle, the importance for understanding the complex interactions occurring between and within the urban and natural environments continues to be more important. When considering an urban center, the intake of water — primarily for potable water production — is impacted by the interplay of the hydrological cycle with both the natural and engineered systems located upstream of that center. Within an urban center, the production of potable water is utilized for both potable and non-potable uses and must satisfy a complex, sometimes conflicting set of regulatory constraints. These systems also must handle changes in water use through conservation efforts, development of "grey water" infrastructure, and continuing challenges to protect public health. The return of this water back to the environment through the sewer and storm water system is challenged by an increasingly aged infrastructure that must avoid direct discharges of untreated water into the receiving waters; and, factoring the potential benefit of low impact development, on improving and maintaining these systems as the hydrologic cycle changes. As the water then leaves the urban center, the remaining water quantity and quality now impacts the downstream natural and urban environments, where the urban water cycle repeats itself. Understanding the complex interactions within one of these natural or urban systems, let along the interactions between them, is a challenging but necessary objective if we are to continue developing sustainable water resource strategies into the future. The role of the developing "observatories" will be an important part of collecting the data necessary to understand these complex, spatially diverse systems. The additional role of "cyber-infrastructures" as data repositories and simulation frameworks will provide the platforms to amplify research efforts by making this information available to research groups throughout the world. The following sections describe out on-going efforts in developing the cyber-infrastructure for implementing an urban observatory within drinking water distribution systems.