Conference Papers

This presentation is a case study on the use of artificial intelligence (AI) to detect issues and operational problems in sewer systems more quickly and reliably by identifying anomalies in sewer flow data, including: -Sudden shifts indicative of issues such as sensor failure or blockages in the system -More gradual changes in the data over longer periods of time, which point to issues like sensor drift or other less immediately obvious operational or environmental problems -Differences between calibrated hydraulic models and sensor data The authors will discuss the application of our AI based approach for detecting anomalies in a large water/wastewater agency located in the midwestern united states. We have found that our AI based approach is highly effective at detecting subtle anomalies such as sensor drift, as well as sudden shifts in the flow data. In some instances, it has detected issues weeks before they would have otherwise been detected during manual review or inspection. Not only is the solution effective, but we also chose an approach that is relatively easy to understand and explain, which makes it more actionable and less challenging to maintain. Through our presentation, the authors intend to highlight the critical role that engineering subject matter expertise played in the formulation of our successful approach, and demonstrate the success that can be achieved through the nexus of data science and engineering.

Introduction In order to better understand the fate and transport of PFAS compounds in wastewater sludge undergoing thermal combustion, a bench-scale system, that mimics a full-scale biosolids incineration, was constructed. Two batch studies were conducted with the system in 2020 and 2021 to investigate: 1) the decomposition and removal efficiency (DRE) of the PFAS in biosolids; 2) the residual PFAS concentrations in the ash, 3) the PFAS concentrations in the flue gas, and 4) the effect of operating parameters on the performance. The operating parameters evaluated were combustion temperature and gas residence time. Methods The two studies are identified as the 2020 and 2021 studies in this abstract. The feed sludge samples used for the two studies were obtained from two municipal wastewater treatment plants. While the same schematic of the bench scale testing system (e.g., a silica quartz tube) was used for the two studies, the sizes of the tubes were different. For example, the length and outside diameter (OD) were 54.0 inches and 2.0 inches for the 2020 study. In order to increase the gas residence time (GRT), the 2021 study used a larger tube with a length of 78.0 inches and an OD of 4.0 inches. As shown in Figure 1, the sludge sample was fed into the reactor tube by a sample boat. The average sludge sample residence time (SRT) in the reactor tube was approximately 5 min for both studies. The reactor tube consisted of three zones. The temperatures over Zone 2 (effective length) of the reactor tube were set 1,150 °C (+1 °C, -51 °C) for the 2020 study and 1,150 °C (+21 °C, -58 °C) for the 2021 study. Air was injected into the reactor tube from the feed end at a constant rate to achieve a pre-set GRT. The GRT was defined based on the volume over the effective length and the actual airflow at 1,150oC. The combustion flue gas was collected by the impingers on the exhaust end of the reactor tube. The impinger setup consisted of four impingers in series. Each 2-L impinger was filled with DI water at a volume of 1.5 L. The above operating conditions are summarized in Table 1. The 2020 study consisted of three triplet tests. Each triplet test generated one impinger sample. The residual ash from the triplet tests was combined to have one ash sample because the amount of ash generated from one triplet test was not enough for performing lab analysis. The 2021 study also consisted of three triplet tests. Due to the large amount of samples used, each triplet from the 2021 study generated one impinger sample and one ash sample. All samples were analyzed for PFAS by Eurofins/TestAmerica with 12-targeted analytes for the 2020 study and 36-targeted analytes for the 2021 study. Fluoride was analyzed for all the impinger samples by the same laboratory for PFAS. Test Results For conciseness, only the major PFAS compounds such as PFOA, PFOS and Total PFAS are presented in Table 2. As shown, total PFAS concentration in the raw sludge was 118 ng/g and 59 ng/g for the 2020 and 2021 studies, respectively. Due to the low total PFAS concentrations, the raw samples were spiked with PFOA, PFHxS, PFHps and PFOS to a total PPFAS level at 31,292 ng/g and 5,893,800 ng/g for the 2020 and 2021 studies, respectively. The purpose of the spiking was to study the PFAS decomposition capability by the combustion process. As shown in Table 2, the spiking level for the 2021 study was 188 times higher than that of the 2020 study. The PFAS level in the ash sample from the 2020 study was much lower than that of the 2021 study (0.32 ng/g versus 11.0 ng/g). This may be due to the large difference in the feed PFAS spiking. The total PFAS level in the ash sample from the 2021 study was 26.1 ng/g at maximum (i.e., 26.1 µg/kg) and 11.0 ng/g on average under the total spiked PFAS concentration of about 5,900,000 ng/g. These residual PFAS results indicate that the ash after the thermal combustion process can meet the PFOS permit level (50 µg/kg) for land application imposed by the State of Michigan. PFAS Decomposition Efficiency - As shown in the schematic (Figure 1), the impinger sampling system was designed to catch the PFAS in the flue gas from the reactor tube, which represents the flue gas coming directly out of the furnace but before the air pollution control (APC) equipment for full scale incineration projects. As mentioned in the experimental setup section, the impinger samples were analyzed for fluoride. The fluorine balance was conducted based on the measured PFAS level in the spiked sludge samples and the measured fluoride concentration in the impinger samples. A fluorine balance ranging from 78% to 88% was achieved in the combustion tests. The fluorine balance above 80% indicates a reasonably good mass balance on the combustion system, thus using total PFAS balance around the combustion reactor to calculate the decomposition efficiency is valid. The mass balance results and the decomposition efficiencies from the two studies are presented in Table 3. As shown, the PFAS mass in the feed was about 3,000 times more in the 2021 study than the 2020 study due to that the higher spiking concentration and larger sample mass were used for the 2021 study. However, the total recovered PFAS mass based on the impinger and ash samples were not significantly different between the two studies. The average total PFAS recovery by the impinger and the ash samples is 207.9 ng and 243.8 ng for the 2020 and 2021 studies, respectively. Therefore, the PFAS mass destroyed in the 2021 study was about 3,000 times that of the 2020 study, which was reflected by the PFAS decomposition efficiency. As shown, the PFAS decomposition efficiency based on total PFAS (sum of the 36 analysts) was at a level of 99.9999% (6 logs) for the 2021 study and 99.67% (close to 3 logs) for the 2020 study, on average. The major difference in operating conditions between the two studies was the gas residence times - 2 seconds for the 2020 study and 4 seconds for the 2021 study. The two tests were run at the same combustion temperature of 1,150oC. Therefore, it was believed that doubling the gas residence time from 2 s to 4 s improved the PFAS decomposition efficiency. Estimated PFAS Level in Flue Gas - The PFOA, PFOS, and 6:2 FTS levels in the flue gas were estimated based on their mass caught by the impinger divided by the total air volume (in m3 at STP) supplied during each triplet batch. The estimates represent the flue gas PFAS levels at the point before the air pollution control (APC) equipment for full-scale systems. The estimated flue gas PFOA, PFOS, and 6:2 FTS concentrations are summarized in Table 4. Statistical analysis indicates that even though the variability of concentrations within each triplet batch is large, the average concentrations of PFOA and PFOS in the flue gas obtained from each study are significantly different. The average flue gas PFOA and PFOS levels for the 2020 study were about 132.3 ng/m3 and 152.7 ng/m3, respectively. However, the average levels of PFOA and PFOS for the 2021 study were about 60.8 ng/m3 and 59.4 ng/m3, respectively, which are below the permit levels of 70 ng/m3 imposed by the State of Michigan for the air emission. It must be noted that the PFOA and PFOS levels in the flue gas estimated by the two studies represent the concentration before the APC equipment. The results from the 2021 study suggest that the air combustion operated at 1150oC and 4 seconds GRT can meet the emission requirements. Conclusions This bench scale study demonstrates that thermal combustion can be effectively and efficiently utilized to achieve near complete PFAS decomposition in the biosolids. The ash residuals generated from the two studies conducted at a combustion temperature of 1150oC and GRT of 2 and 4 seconds meet the PFAS requirements for biosolids land application. The estimated PFAS level in the flue gas indicates the potential of the combustion technology to meet the air emission requirements at 70 ng/m3 for PFOA and PFOS imposed by the State of Michigan. The gas residence time of 4 seconds at a combustion temperature of 1150oC improved the total PFAS decomposition and lowered the air emission for PFOA and PFOS, compared to the 2 seconds GRT.

Everyone has to do it. Few talk about it. It's one of those things you do the same way you did it last year, the year before, and the year before. Open last year's CCR, update the figures, send to your subject matter experts for a quality check and it's off to the printer and mail house. Your job is done, and you've complied with EPA Regulations. A couple weeks later, a customer opens their mailbox to find an oftentimes visually bland and unappealing letter from their water provider. If U.S. Postal Service statistics are correct, it's competing with nearly 500 other marketing pieces and catalogs that customer will get that year. If they even open it — and that's a big if — they'll notice from the first sentence that the language is overly technical, uses confusing industry jargon, and reads like a legal document. If you're lucky, they might scan the dense tables of figures, still without absorbing a single concept, and then give up. The next stop is the recycling bin. Utilities can do better. Many years ago, Charlotte Water's Public Affairs Division recognized that their CCR provided a rare opportunity to communicate with customers, enhance their brand, build relationships, and make human connections. Their annual Water Quality Report (WQR) was born. In 2021, Charlotte Water partnered with Raftelis to create a WQR that stands out among its competitors in the mailbox. Their piece focuses on two key factors that influence whether the report gets read or quickly recycled: design and accessibility. It's these same two factors that Raftelis focused on to create the winning entry to the Environmental Policy Innovation Center's Water Data Prize Competition last year. Both Charlotte Water and the Raftelis team saw the CCR not for what it is, but for the opportunity it delivers — a rare and critical moment to connect customers to the one thing they can't go without. This session will provide participants with ideas, insights, and guidance that can be scaled for use by the smallest utility with minimal resources to the largest utilities with maximum resources. Attendees will come away with a fresh eye on the old CCR, and the inspiration to turn this mandated report into a masterpiece.

Facility Overview: South Platte Renew holds the distinction of being the third-largest Water Resource Recovery Facility (WRRF) in the state of Colorado. Serving as a vital lifeline, it caters to the needs of over 300,000 residents and consistently processes an average of 20 million gallons per day (MGD) of wastewater. SPR's asset registry is a testament to its complexity and scale, comprising nearly 6,000 assets across multiple disciplines, encompassing mechanical, electrical, instrumentation and controls, structural, and fleet components. The diversity within this asset portfolio is further underscored by the wide-ranging age of these components, stemming from the original plant construction in the 1970s, complemented by significant upgrades undertaken during the 1990s and the early 2000s. This intricate blend of assets has introduced a formidable challenge for SPR, as it endeavors to strike a balance between the renewal needs of variable aging infrastructure with other process and regulatory capital improvement needs. Asset Management Background: While SPR has been performing certain asset management activities since its original construction, SPR's formal programmatic asset management journey started in 2008 with the acquisition of an Enterprise Asset Management (EAM) system. Since then, SPR has advanced its asset management program by launching a series of initiatives designed to enhance overall asset lifecycle health. These initiatives encompass recurring asset-focused programs, such as equipment rebuilds, tank cleaning, HVAC maintenance, valve exercises, and concrete repair and rehabilitation projects. In addition, SPR has integrated asset management principles into its capital projects through standard development and data utilization. This ongoing commitment to enhancing asset management practices reflects SPR's dedication to continually evolve and improve its approach to ensure the sustained health and efficiency of its vital infrastructure. The Challenge: SPR has diligently worked across various departments to develop data management and maintenance strategies to support the ongoing advancement of its asset management program. However, one thing that became evident to SPR was the lack of a cohesive and organization-wide strategy amid the ongoing tactical endeavors. The lack of understanding and inconsistencies in asset management practices among SPR staff posed a notable challenge. It was not well understood among SPR staff what asset management meant and what the overall goals and objectives of the program were. It was also evident that there were inconsistencies in how asset management practices were being performed. The Solution: To address the challenges faced, SPR embarked on the development of a Strategic Asset Management Plan (SAMP) which had a dual purpose. Firstly, it formalized asset management strategies and policies. Secondly, it actively engaged key stakeholders and nurtured an asset management culture within the organization. The SAMP encompassed critical components such as: Identifying and recruiting key asset management stakeholders across operations, maintenance, engineering, data analysis, and senior leadership staff Developing a clear vision, goals, objectives, and policies around SPR's asset management program Establishing programmatic governance, roles, and responsibilities Defining levels of service and key performance indicators Implementing risk-based renewal planning methodologies in consideration of an asset's condition, criticality, and redundancy Optimizing maintenance strategies Benefits Realized: SPR has already realized several benefits from the development and implementation of the SAMP. There is overall a greater sense of understanding of asset management among staff, there is excitement and the realization of value from asset management activities, and there is consistency and clarification on many procedures and activities. Through the implementation of a SAMP, SPR will also gain an increased resiliency in financial planning through transparent and accurate reporting, as well as operations and maintenance practices, allowing SPR to better maintain high water quality and maximize value to customers. Lessons Learned: SPR's asset management journey offered valuable lessons, including the importance of starting with simplicity and gradually building complexity, eliminating silos, and the foundational importance of thoughtful and strategic data management. The SAMP emerged as a concise, repeatable document that can guide any utility in creating an asset management framework. Key Value Points for UMC Attendees: This presentation by SPR staff and its external consultant will provide attendees with insights into SPR's asset management journey, including: Step-by-step guidance for creating a SAMP and its associated benefits Strategies to avoid single points of failure by sharing institutional knowledge The significance of fostering an asset management culture across an organization Transitioning from fragmented and inconsistent initiatives to a formalized organizational strategy The advantages of self-performing SAMP development in collaboration with external consultants The importance of building upon past successes and discarding outdated approaches to meet current and future programmatic needs. SPR's asset management story serves as a valuable case study, offering practical guidance and inspiration for utilities seeking to enhance their asset management practices.

Background Portland Water District (PWD) has four wastewater treatment facilities (WWTFs) with a capacity of more than 25 million gallons per day. Two of the four facilities - East End and Westbrook process most of the solids in the system and were the focus of PWD's comprehensive approach to assess the current solids system and develop a strategic plan that outlines the necessary upgrades, process modifications, and state-of-good-repair projects that PWD should undertake during the next 20-year planning period. PWD began contemplating this planning effort not only due to aging infrastructure, but rather due to increasing concern over lack of biosolids management options and rising public concern around per- and polyfluoroalkyl substances (PFAS) in biosolids. Throughout New England, decreased biosolids management capacity has been a rising concern. Much of the region relies upon landfills and regional incinerators, but beneficial use (land application) has served as a reliable, sustainable option for a number of utilities. In Maine, land application had been an accepted practice, with good market understanding of the benefits of biosolids, including biosolids based compost and thermally dried products. In 2019, however, PFAS contamination was discovered at a Maine dairy farm that had historically land applied both biosolids and short paper fibers. This led the Maine Department of Environmental Protection (DEP) to promulgate soil screening standards for two PFAS - perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS). These screening standards were used to limit biosolids land application rates based on the concentrations of these two compounds, and the understanding that PWD might be able to land apply in the future served as an initial assumption of the study. Shortly after study initiation, the Maine legislature passed LD 1911, which prohibits the land application or sale of biosolids and biosolids-derived products in the state of Maine. While options were already limited before this new law, the only option for biosolids management in Maine is now landfill disposal. The passage of LD 1911 thus focused the biosolids master plan on technologies that significantly reduce volume and/or employ thermal destruction. Methodology The goal of this project is to create a 20-year system-wide implementable plan, including near- and long-term improvements that improve the reliability of PWD's solids handling assets while meeting the project goals of mitigating emerging contaminants risks, enhancing reliability and regulatory resiliency, and increasing the public's confidence in PWD's environmental stewardship. PWD worked with its consultant to create a master plan that has both near- and long-term improvements to improve the reliability of PWD's solids handling. Alternatives were evaluated based on regulatory, financial, social, and environmental goals and objectives. An overview of the master planning process is shown in Figure 1. Custom solids and energy models were created to reflect the operation of East End and Westbrook, which were used to determine alternatives for solids handling. The consultant team also provided PWD with estimated operational and capital costs, solids production, and energy requirements based on actual PWD unit costs. Throughout this process, the PWD and consultant team developed and evaluated seven alternatives at six workshops. The outcome of these workshops was a roadmap with several triggers rather than a final solution due to the uncertainty and expected changes in Maine's regulatory framework. Alternatives were evaluated using a mass and energy balance model that supported evaluation of both technical and operational performance. When combined with the capital investment required for each alternative, the model produces a net present cost (NPC) lifecycle cost of each alternative that, when compared to the baseline process condition, determines its financial viability. For this study, it was assumed that the planning horizon for the project would be 20 years. The alternatives evaluated are: -Alternative 0: Status quo. No operational or capital improvements. -Alternative 1: Modified baseline. Baseline operation with new dewatering and sludge storage. This alternative considers the possibility of sending dewatered solids to a merchant facility or to a landfill. -Alternative 2: Thin film Dryer. Westbrook dewatered solids hauled to East End, and the combined solids are dried in a thin film dryer and then disposed of at a landfill. -Alternative 3: PS AD + thermal dryer. Westbrook dewatered solids hauled to East End. East End primary sludge (PS) only mesophilic anaerobic digestion (AD) with thermal drying of all East End and Westbrook sludge. Landfill disposal with possible beneficial end use options in the future. -Alternative 4: Regional Pyrolysis. Westbrook and East End dewatered solids hauled to a PWD-owned regional pyrolysis facility, with either biodrying or thermal drying. Biochar is beneficially used. -Alternative 5: Regional Solar Drying. Westbrook and East End dewatered solids hauled to a PWD-owned regional solar drying facility. Landfill disposal with possible beneficial end use options in the future. Table 1 presents the results of the NPC analysis for these alternatives at a range of solids management costs, which are the largest contributors to the annual solids handling and processing costs. Note: Pyrolysis alternatives do not increase with rising tip fees, as it was assumed PWD had an agreement with the vendor to take the biochar at no cost. Sludge pyrolysis has not been demonstrated at PWD's size, so PWD decided the uncertainty with this technology was too significant at this high capital cost; therefore, PWD decided to continue to monitor the technology before making a final decision. The remaining alternatives were then evaluated based on non-cost factors including de-risking PWD on end use, sustainability, cost predictability, and improved reliability. Table 2 shows a summary of the goal factors for each of the alternatives. The modified baseline, which centers around improved dewatering, was found to be advantageous under current tip fees, and helps de-risk PWD as tip fee volatility increases. This was considered a 'no regrets' project as it both replaces aging infrastructure, and saves PWD money over the long term. With respect to the other alternatives, external factors, particularly the manner in which Maine DEP does or does not permit certain technologies, will affect which alternative should be implemented. As a result, a roadmap was made with key regulatory, market, and other implementation triggers to help guide PWD's actions in the next five years. The roadmap is presented as Figure 2, with the 'no regrets' project serving as the starting point (e.g. will be implemented). Key triggers will cause PWD to go on a certain path of the map, while additional triggers will further refine the likely or viable options. The final two branches (red and orange) of the roadmap are highly unlikely but could still be an outcome. The three key triggers are tip fees continue to rise, landfills reject biosolids, and tip fees stabilize or drop. This adaptable model allows PWD to make immediate improvements that will result in cost savings (dewatering upgrades and thickened sludge storage), while allowing for planning in the face of continuing regulatory flux. It is anticipated that over the next few years, Maine's regulatory approach will stabilize, clarifying the appropriate path forward. Conclusions While state and federal biosolids regulations are still uncertain, PWD can take the following actions recommended by the master plan: -Engage with regional partners now. Should support exist (both from Maine DEP and peer utilities) for regionalization, PWD and its potential partners would have time to determine challenging aspects of partnership such as governance. -Continue to engage with service providers as they explore merchant regional solutions. The master plan established a range of tip fees at which it could be advantageous for PWD to secure long-term management capacity. -Conduct a preliminary design project for the no-regrets projects at East End and Westbrook. This project would provide PWD with a basis of design report (BODR) in which dewatering technology and location would be selected as well as selection of size and location for sludge storage tanks at East End. The BODR would also include evaluation of conveyance of cake, automation of the screw press, and polymer changes at Westbrook. -Continue to engage Maine DEP on technology updates and data updates on PFAS mitigation. -Explore funding mechanisms and continue discussions with vendors on viable solutions. -Visit dryer and other vendor installations.

The City of Austin has regularly tracked odor complaints within its major interceptor systems and has sought ways to reduce complaints. In order to assess the feasibility and effectiveness of head space air extraction for odor control within four of its large diameter interceptors, the City initiated a comprehensive fan test in early 2021. Large-scale tests of the impact of air extraction were performed in areas served by the City's Walnut Creek Interceptor (WCI), Little Walnut Creek Interceptor (LWCI), and Williamson Creek Interceptors (WICI) and the Crosstown Tunnel / (CTI), and Onion Creek Interceptor (OCI) systems. All are large diameter pipeline systems (diameters up to 96-inch) with headspace restrictions at downstream ends. Fan testing and monitoring was performed on five different combinations of extraction points and associated monitoring locations. For each extraction point, The City of Austin and its consultants selected a combination of upstream and downstream monitoring manholes to provide an indication of the zone of influence of the air extraction. The testing area included 34 separate monitoring locations along approximately 50 miles of interceptor. The purpose of the fan test monitoring was to provide data on headspace pressurization and hydrogen sulfide concentrations, and the variability of each over time, both under existing conditions and with air extracted from the sewer headspace. Perkins Engineering Consultants, Inc. (now Mead & Hunt) served as the prime contractor for the testing events, placing and maintaining equipment within the manholes for monitoring hydrogen sulfide and differential pressure, monitoring results during testing, and transmitting results to the project team on a daily basis. V&A Consulting Engineers, Inc. providing equipment, ducting, and personnel for fan operation during the tests. Differential pressure and hydrogen sulfide loggers were installed two days prior to starting the fan test at each location to establish baseline sewer pressure and hydrogen sulfide conditions. The loggers continued to monitor differential pressure and hydrogen sulfide during the two-day fan test, with most loggers transmitting data continuously during the test. Loggers were left in place at least six hours after each test was completed to ensure the system had reverted to normal conditions. During the two-day fan test, the fans were run over a range of airflows to provide a more accurate picture of the zone of influence potentially available within each interceptor. Two fan test locations were tested on the WCI: one at Balcones Park, and the other at Loyola Lane, located upstream of Balcones Park. The Loyola Lane test showed a zone of influence of at least 0.05 in-w.c. 16,000 ft upstream and 8,500 ft downstream of the extraction point. The Balcones Park test had a zone of influence 24,000 ft upstream. The LWCI test showed a zone of influence of 15,000 ft upstream. Testing on the CTI siphon Produced a zone of influence 75,000 ft upstream in certain areas. Testing on the WICI showed a zone of influence of 20,000 ft downstream on the OCI and only 10,000 ft upstream. This paper presentation the test methodology and results, and presents lessons learned and considerations that should be incorporated when choosing fan test extraction and monitoring points for testing. Extreme weather conditions were presented during part of the testing by Winter Storm Uri. The experience of running the test and having instruments in place through the extreme weather event will also be discussed.

In modern water management, integrating system performance needs with stakeholder input is essential for effective capital planning. Parker Water & Sanitation District (PWSD) is completing a long-term water supply planning process. This presentation outlines the technical aspects of PWSD's approach, which integrated stakeholder input, water supply models, and a decision-support tool for capital planning. The process began with stakeholder workshops to establish objectives, assess system performance gaps, and define alternatives. Multi-criteria decision analysis was employed, allowing input from multiple stakeholders on evaluation criteria and capital plan alternatives that supported increasing water supply or decreasing demands through water efficiency measures. A non-proprietary tool, developed using the Microsoft Power Platform, enabled stakeholders to view how each capital plan variation met system performance metrics, organizational objectives, and budget constraints. Through a consensus-building process, commonalities across the alternative plans were identified, leading to a capital improvement plan that met system performance needs, diverse goals, and budget constraints. Unlike traditional methods that debate a single plan's merits, this approach allowed each stakeholder's perspective to be considered, resulting in improved buy-in, and a plan that is robust and defensible. The decision-support tool clearly demonstrated how each plan addressed needs and goals through embedded visualizations about system performance metrics. This data-driven approach has broad applicability and can serve as a model for other utilities undertaking similar capital planning efforts. PWSD's process represents a significant advancement in capital planning, emphasizing the importance of stakeholder involvement, technological integration, and a methodical approach. It provides a practical roadmap for other utilities, highlighting the role of digital technologies in modern capital planning for water management.

Our presentation will summarize the development and implementation of artificial intelligence and machine learning (AI/ML) tools Citizens Energy Group has developed to provide predictive operational decision support for combined sewer overflow (CSO) facilities. Citizens developed these tools by efficiently leveraging its prior investments in collection system modeling and metering. Our AI/ML tools consist of artificial neural networks fit to long-term datasets to predict inflows to CSO facilities using the 72-hour NOAA radar rainfall forecast. Both the rainfall and CSO facility inflow are embedded in a Power BI dashboard for efficient communication to Citizens staff. We developed this dashboard without incurring any additional software cost and believe any wastewater utility operating wet weather facilities can implement a similar AI/ML tool to ours. Citizens Energy Group is the combined water, wastewater, natural gas, steam, and chilled water utility in Indianapolis. Citizens is presently implementing a 20-year CSO consent decree with required completion in 2025. Consent decree compliance is primarily achieved through the DigIndy Tunnel system, a 28-mile long, 18-foot diameter tunnel network. Citizens is also implementing a pair of storage facilities in the Upper White River and Upper Pogues Run watersheds. Citizens, like all utilities implementing a consent decree, has made significant investments in modeling and metering its collection system, and has multiple years of meter data and model results on the shelf from recent regulatory and design support activity. While Citizens has confidence in its InfoWorks ICM model of the collection system, computational limitations prevent the model from being applied in real time or for immediate prediction of future rainfall events. In order to transform the collection system model from a design tool into a predictive operational tool, Citizens developed neural networks for the DigIndy Tunnel system and the Upper Pogues Run storage tank. These neural networks effectively serve as an extension of Citizens' collection system model to simulate inflow to the tunnel or tank in a matter of seconds. Since both facilities will begin operation at the beginning of 2022, Citizens utilized several years of collection system model input and output data to develop the neural networks. Figure 1 presents the architecture diagram for the AI/ML tools. Within Power BI, embedded Python scripts using the MetPy library (Unidata, 2021) connect to the NOAA server and obtain the 72-hour National Digital Forecast Database (NDFD) radar forecast for rainfall and temperature. The forecasted rainfall data is compiled for the CSO facility service area and temperature is utilized to estimate evapotranspiration (Hargreaves and Samani, 1982) for use in the neural networks. Citizens developed standard feed forward neural networks to predict inflow volume to the tunnel and storage tank similar to the collection system model. Figure 2 presents a scatterplot comparing each modeled inflow event to the tunnel system for precipitation years 2016 through 2019 to the neural network prediction. The inputs to the neural network are rainfall, peak intensity, antecedent dry days, the previous day's rainfall, and evapotranspiration. The total inflow volume from the neural network was within one percent of the InfoWorks collection system model simulation, with a R2 of 93%. Citizens evaluated input parameter sensitivity of the neural network and determined that for the storage tunnel, rainfall is the nearly an order of magnitude more important than any of the other input parameters. However, for the Upper Pogues Run storage tank that serves a much smaller watershed area, peak intensity is nearly as important as rainfall. Citizens initiated a desktop test deployment in early 2021 and progressed to an automated test deployment that has been completed at the time of the abstract. In the automated deployment, the rainfall forecast and subsequent inflow volume forecast is generated hourly, with results posted on Citizens' internal SharePoint. The rainfall forecast and forecasted inflow volume are presented though an online Power BI dashboard. We will present our method of deployment of this tool as well as other options for automated deployment that wastewater utilities may consider. Table 1 presents an example forecast from the neural network for July 15 – July 17, 2021. The forecasted inflow volume is based on the NDFD on July 14, 2021. In other words, should the tunnel system have been operational on July 16th, operational staff would have had an estimate of the magnitude of inflow more than a day before the event occurred. Over the course of the deployment, the BI dashboards were expanded to also collect NWS river stage forecasts and actual rainfall data through the application programming interface (API) provided by ADS as part of the PRISM interface to Citizens' rain gauge and flow meter network. Using the API key, Citizens developed a process to compare the forecasted rainfall from NOAA to the actual rainfall from the gauge network. From March 2021 through July 2021, the total forecasted rainfall was 23.5 inches, compared to 22.3 inches of actual gauge rainfall. An encouraging finding is that seventy-six percent (76%) of the rain events were forecasted within two-tenths of an inch of the actual gauge rainfall. We believe our presentation will benefit any wastewater utility that is operating or soon to be operating a wet-weather storage, treatment, or pumping facility to address combined sewer or sanitary sewer peak flows. We will provide a roadmap for any utility to implement a similar AI/ML solution for predictive operational decision support that can be implemented without any commercial software cost.

Introduction and Objectives Our scientific understanding of per- and poly-fluoroalkyl substances (PFAS) in our environment has evolved rapidly in the last several years and PFAS have increasingly gained attention from the public, scientific, and regulatory sectors due to their carcinogenic and reproductive and endocrine disruptive nature. Previous research has illustrated that water resources recovery facilities (WRRF) are notable PFAS emission routes to the environment, both liquid and solid discharge routes. Despite industrial phasing-out of some PFAS and increasing source management of industrial discharges, PFAS are expected to be present in wastewater solids for the foreseeable future. According to Northeast Biosolid and Residual Association (NEBRA), more than seven million dry tons of wastewater solids were produced at WRRFs in 2004 and solids management practices have resulted in the release of approximately 3000 kg PFAS per year to agricultural lands and landfills (Venkatesan and Halden, 2013). Much work has been done to develop and demonstrate wastewater treatment technologies to destroy PFAS compounds (mineralize to non-fluorinated end-products) in wastewater solids, but few studies have documented the practical application of PFAS destruction technologies for wastewater solids or estimated the cost of implementing those technologies. Our team conducted an evaluation for the Minnesota Pollution Control Agency (MPCA) to document the technology options that could be implemented today to remove and destroy PFAS within municipal wastewater effluent, compost contact water, landfill leachate, and municipal wastewater solids. The objectives of this project were to: -establish a framework for determining which technologies meet the definition of commercially available and effective for PFAS destruction -conceptualize upgrades to destroy PFAS at existing WRRFs -estimate the cost of completing these upgrades for a range of WRRF capacities. Systems were conceptualized for facilities ranging from 0.1-10 MGD. For capacity greater than 10 MGD, a regionalized option was considered. This paper will highlight the findings from wastewater solids portion of the study. This information will be valuable for utilities looking to implement PFAS destruction technologies as regulations around PFAS evolve rapidly. Target PFAS Initial work for the study involved selecting a set of PFAS to represent the range of PFAS known or predicted to be in wastewater solids. Selected PFAS are well-described in PFAS treatment literature, frequently detected in wastewater solids, and diverse in chain length and functional groups. Table 1 lists the selected compounds, assumed typical and high concentrations in wastewater biosolids, and associated target post-treatment. The selected PFAS are generally long-chain perfluoroalkyl acids (PFAAs) or are precursors to PFAA formation, which research has shown to partition into solids. Representative concentrations of PFAS were selected from available literature. Technology Screening PFAS destruction technologies were screened for their ability to consistently remove at least one identified PFAS to below the currently accepted laboratory detection limit, as outlined in Table 1. Technologies were further screened by their readiness for full-scale implementation. Wastewater solids PFAS destruction technologies passing the screening included pyrolysis with thermal oxidation of the process gas, gasification with thermal oxidation, and supercritical water oxidation (SCWO). Pyrolysis and gasification systems were considered to use similar enough equipment to be evaluated as a single alternative for conceptual design. Pyrolysis/gasification systems included a thermal dryer as part of the equipment assembly since both technologies require high solids content. Conceptual Design and Cost Estimating A conceptual design was created for both short-listed PFAS destruction alternatives and for solids production rates of 1, 5, and 10 dry tons per day at a given WRRF, corresponding roughly with the plant influent flow rates used for the liquids portion of the study. The feed solids to the pyrolysis/gasification process were assumed to be dewatered to 25 percent total solids. For SCWO, the feed solids were assumed to be screened and dewatered to 15 percent total solids. Construction cost estimates (AACE Class 5) were prepared from recent project estimates and input from vendors. The construction costs were plotted against the system feed rate to create cost curves for each alternative. One example of a cost curve for pyrolysis/gasification is provided as Figure 1. The construction cost curves for pyrolysis/gasification and SCWO, and parameters used to develop each, will be included in the paper. Additionally, a conceptual design for a regional pyrolysis/gasification facility for wastewater solids was developed to represent another approach for utilities. Our findings showed that due to high costs associated with PFAS treatment, regional approach might become viable for smaller facilities (facilities less than 10 MGD). A regional facility would receive dewatered solids from two or more individual WRRFs within a reasonable hauling distance. Fifty dry tons per day was assigned as a representative capacity for a regional PFAS destruction facility. Operation and maintenance costs estimated for recent pyrolysis/gasification projects were assembled to select representative 2022 O&M costs for a 50 dry ton per day facility. A 20-year lifecycle cost analysis was prepared and tested for sensitivity to variations of the tipping fees and interest rates. This paper will include the results and conclusions from the lifecycle cost analysis of the regional facility concept. Conclusion This study documents the cost of implementing currently available technologies to destroy PFAS in wastewater solids. For pyrolysis/gasification, the O&M costs per dry ton trended lower as the system capacity increased, indicating that smaller utilities could reduce costs if they formed a partnership and developed a regional PFAS destruction facility. Demonstration testing of SCWO systems is expected to provide evidence to clarify if similar results can be expected for SCWO systems.

Near the end of 2021, New Castle County's Public Works initiated an extensive project for gravity sewer system O&M planning, implementation and support. As a complement to physical infrastructure improvements, an interactive asset management dashboard was developed. By visualizing and forecasting past and present data, leaders are enabled to make critical decisions and improve operational efficiency. This presentation will detail the process of developing a comprehensive asset management dashboard and its impacts on utility maintenance and operations. A unified dashboard was developed in Microsoft Power BI to support the creation and implementation of data management, analysis, and visualization solutions for the Sewer O&M Program. The dashboard connects directly with New Castle County's existing Cityworks maintenance management and permitting software. This integration eliminates the cumbersome need to compile figures and tables manually, streamlining data analysis and ensuring that decision-makers have access to the most current and relevant information. A specific goal of New Castle County was to provide detailed information on the efficacy of the new SL Rat pipe cleaning program. A pipe is first tested and assigned a condition score based on blockages. For the most severely affected pipes, a cleaning team is sent out. A secondary inspection then occurs where a condition score is once again determined. Prior to this project, there was little information available to track the effectiveness of this process. The dashboard now provides analysis of cleaning performance by individual workers, correlation between pipe dimensions and condition score, details on specific pipes that could not be accessed by crews, and other visual aids to better plan and monitor the SL Rat cleaning program. The dashboard also provides other data visualizations and insights. One area is the real-time tracking of performance against American Water Works Association (AWWA) metrics, providing an invaluable reference point for gauging operational success. Furthermore, it offers a window into work order progress and unveils critical patterns among sewer assets and regions. The versatility of this dashboard equips decision-makers with a comprehensive understanding of the program's performance metrics. The dashboard is in its final stages of development and will be in frequent use by New Castle County at the time of the conference. The presentation will include lessons learned in implementation and use that can be transferred to other utilities.

The City of Vancouver, British Columbia, is working to develop an integrated or 'one water' approach to delivering water-related infrastructure (i.e., potable, sanitary, and stormwater) to the Cambie Corridor, a rapidly developing transit corridor south of downtown. The current water-related infrastructure is beyond capacity; and a recent land-use plan for the 1,000-hectare corridor revealed that the population living and working there is expected to more than double in the coming decades. These land-use pressures along with climate change and ecosystem health concerns necessitate a fresh, holistic look at all forms of water infrastructure serving the corridor. To create a plan for the corridor, the City used a collaborative engagement process to develop a water-servicing vision and a multi-benefit assessment approach for evaluating one-water opportunities (i.e., programs, policies, and projects). Opportunities include various types of distributed green and blue-green stormwater infrastructure, regional green stormwater management facilities, water conservation and efficiency strategies, site and district scale non-potable water reuse (e.g., rainwater, greywater, and foundation drainage), and policies and programs to mandate, incentivize, and/or encourage private investments and partnerships. The assessment approach, which will be the focus of this presentation, was used to identify, value, screen, and ultimately optimize the suite of opportunities needed to meet the City's objectives. The assessment approach included several analytical tools working in tandem, including (1) a custom-built Water Balance Model paired with a calibrated hydrologic and hydraulic model to evaluate opportunity performance, (2) a GIS tool to help site opportunities, (3) a Decision Support Tool to value ecological and community co-benefits and measures of reliability and feasibility, (4) a scenario planning analysis to account for risks and future uncertainty, and (5) an optimization tool that uses advanced statistical algorithms to identify the ideal combination of opportunities that meets water-servicing targets, minimizes costs, and maximizes co-benefits. This assessment approach allowed the City to balance the trade-offs inherent in a multi-objective analysis in a defensible, transparent, and easy-to-understand way. The effort culminated in development of a high-level long-term water servicing strategy for the corridor and a more detailed short-term capital plan for investments in the next 4 to 8 years.

As Buffalo Sewer prepares for its second decade of implementation of the approved Combined Sewer Overflow Long Term Control Plan (LTCP), results of a recalibrated hydraulic model, updated Financial Capability Assessment, a cost-benefit analysis of early projects, a deeper understanding of the existing system's design, and increasing recognition of issues of environmental justice and implications of systemic racism and climate change have all driven a re-evaluation of how to achieve the goals of the LTCP while also building a more equitable, inclusive, and resilient city. Buffalo Sewer's approved 2014 LTCP made use of innovative strategies with a focus on Green Infrastructure and Real Time Controlled Smart Sewers. As we move forward into the next phase of the LTCP, we are learning how to implement Green Infrastructure in a more sustainable and maintainable way that also addresses long-standing disinvestment in the existing increasingly aged collection system. We are also finding new and innovative ways to implement Real Time Controlled Smart Sewer technology to take advantage of opportunities created by the historical design of our collection system and the disinvestment in redlined neighborhoods while also creating visible community benefits and simplifying maintenance. This presentation will review the development of the approved LTCP and its components. It will also review the projects completed under that plan and implementation issues experienced to date. It will also discuss how the recalibrated model has impacted the viability of the 2014 LTCP and its components. Additionally, the historical, economic, and climatic conditions of Buffalo and the resultant opportunities, risks, and restrictions that they create will be discussed. Finally, the updated LTCP will be presented. Currently, the updated Financial Capability Analysis is under review with the United States Environmental Protection Agency and New York State Department of Environmental Conservation with final approval expected in late 2021. The Model Recalibration is also under ongoing review by the regulators with approval expected in mid-Fall 2021. Six traditional Smart Sewer Projects have been completed, three additional traditional projects are in bidding or construction phases, and a final project which implements Smart Sewer technology at an existing pumping station is poised to go into operation by October 2021. Three additional traditional Smart Sewer projects have been identified for design during the 2022 Calendar Year, in addition, two next generation Smart Sewer projects have been identified for design during this same period, and several of the projects currently in design and implementation represent a global implementation strategy rather than a more traditional CSO or even waterbody specific strategy. Similarly, on the Green Infrastructure front, RainCheck 1.0 has been completed and it has become increasingly clear that while the demolition of vacant and abandoned homes has created vast swaths of new green space throughout large tracts of Buffalo at limited cost to Buffalo Sewer, the opportunity for further demolitions is limited and at the same time, bioretention cells within the public right of way are costly to construct and maintain and other green infrastructure techniques are more sustainable and create opportunities for other co-benefits without negatively impacting long-term residents in disinvested neighborhoods.