Sewer pipelines are deteriorating due to aging and sub optimal material selections, leading to the infiltration of clean ground and rainfall water into the pipes. It is estimated that a significant portion (up to 40-50%) of the water entering wastewater treatment plants is actually clean infiltrated water. This infiltration not only contributes to unnecessary energy consumption but also poses the risk of flooding the sewer network and treatment plants. Finding these broken pipes is utmost importance but is not straight forward due to the pipes being located a few meters below ground. There exist methods of pinpointing where these leaks occur, but they are often time consuming and expensive. This thesis seeks to address the following question; Can the estimation of infiltration be accomplished solely through the temperature data obtained from discrete pump stations, or is the inclusion of precipitation data essential for achieving accurate results? Two machine learning algorithms are investigated to solve the regression problem of estimating the amount of rainfall derived infiltration. The first model is a classical linear regression model. The second model is a Convolutional neural network (CNN). Both of these models are trained on the same data set. The temperatures recorded at the stations are reliable and can be trusted. However, the data labeling process involves utilizing calculated flows to the stations during both dry and wet weather periods. This means that the labels of the data cannot be trusted to be the actual ground truth, and there exists an uncertainty in the data set. Both models manage to capture large temperature drops which indicates infiltration has occurred. The linear regression model seems to be too sensitive towards small temperature drops and predicts infiltration when there is none. The CNN model on the other hand seems to be able to capture only large temperature drops when infiltration occurs. However, both models are trained with data from only one station, this means that the models are biased towards the average temperature of that particular station, other stations may have a higher or lower average temperature. When testing the models on a different station with lower average temperature the models predict infiltration when there is none.

A rural-urban divide in health status and healthcare infrastructure has been well-documented in the U.S., where populations residing in census regions classified as rural often exhibit more negative health outcomes, adverse health behaviors, and have reduced access to affordable and proximal health services, compared to their urban and peri-urban counterparts. However, it is important to note that such disparities vary based on specific rural regions and individual circumstances. Rural areas may face elevated risk factors for infectious diseases such as increased proximity to wildlife and livestock and disproportionately high reliance on private, non-federally regulated, primary drinking water sources. Chronic conditions prevalent in rural communities such as diabetes and hypertension are frequently linked with longer duration and higher severity of symptoms than in urban areas; this association suggests that the risk of exposure to infectious diseases and the likelihood of progression to serious illness and hospitalization may be elevated, although this is not universally the case across all rural settings. Alongside documented urban-rural health disparities, there also exist disparities in the nature and quality of data on health-related behaviors, outcomes, and service provision in rural areas compared to urban and peri-urban regions. In this dissertation, two key environmental matrices –drinking water and wastewater– were highlighted as vectors of information to better estimate levels of contaminant exposures and health outcomes in rural communities. First, baseline data on drinking water contaminant levels and associated health outcome data were highlighted as crucial for refining holistic exposure estimates as well as understanding drinking water related health burdens in rural communities where a larger proportion of households use private drinking water sources, such as well water, that are not federally regulated. Second, systematic sampling and testing of pathogen biomarkers in wastewater to non-invasively measure population-level health status, also known as wastewater based surveillance (WBS) and, depending on the context, wastewater based epidemiology (WBE) is not constrained by disadvantages of clinical testing, e.g., limited health-care access, long travel times to testing facilities, delay between symptom-onset and testing. Thus, expanded implementation of WBS in rural communities is proposed here as a strategy to address data disparities in clinical testing for infectious diseases. Collectively, this dissertation advances knowledge on estimated drinking water contaminant levels, exposures, and associated public health outcomes and corresponding research gaps in rural Appalachian U.S., and elucidates pathways toward best practices and considerations for public-health focused wastewater testing adoption in rural communities. For the latter, the question of whether WBS challenges unique to rural wastewater systems hinder application of WBS in small, rural communities was explored, as well as methods to advance best-practices for rural WBS. To summarize existing publicly available peer-reviewed literature on drinking water contaminants in rural Appalachian U.S., in Chapter 2, a systematic review and meta-analysis of microbial and chemical drinking water contaminants was performed. Key contaminants were identified as being elevated beyond regulatory, health-based, maximum contaminant levels in our meta-analyses from rural drinking water sources in Appalachia, including E coli, lead, arsenic, uranium. Overall, we found data on drinking water source quality under baseline conditions (i.e., rather than post anomalous contamination events such as chemical spills) in rural Appalachian U.S. was sparse relative to widespread media coverage on the issue. Epidemiologic-based research studies that collected both drinking water exposure data and paired health outcome data were also limited. As a result, although some instances of anomalously high levels of drinking water contaminants were identified in rural Appalachia from the published literature, we could not obtain a clear picture of baseline exposures to drinking water contaminants in most rural Appalachian communities, highlight need to address these knowledge gaps. In Chapter 3, to evaluate whether wastewater could serve as a reliable metric for estimating community circulation of viruses and antimicrobial resistance (AMR) markers, even when sourced from aging and low-resource sewer collection networks, a 12-month wastewater monitoring study was conducted in a small, rural sewer conveyance system with pronounced infrastructural challenges. Specifically, the field site under study was compromised with heavy inflow and infiltration (IandI). Detection rates and concentrations of viral, AMR, and human fecal markers were grouped by levels of IandI impact across the sewershed, and location-, date-, and sample- specific variables were assessed for their relative influence on viral, AMR, and human fecal marker signal using generalized linear models (GLMs). We found that while IandI likely adversely impacted the magnitude of wastewater biomarker signal to some extent throughout the sewershed, especially up-sewer at sites with more pronounced IandI, substantial diminishment of wastewater signal at WWTP influent was not observed in response to precipitation events. Thus, our data indicated that WWTP influent sampling alone can still be used to assess and track community circulation of pathogens in heavily IandI impacted systems, particularly for ubiquitously circulating viruses less prone to dilution induced decay. Delineations were also made for what circumstances up-sewer sampling may be necessary to better inform population shedding of pathogens, especially where IandI is prevalent. Various normalization strategies have been proposed to account for sources of variability for deriving population-level pathogen shedding from wastewater, including those introduced by IandI-driven dilution. Thus, in Chapter 4, we evaluated the temporal and spatial variability of viral and AMR marker signal in wastewater at different levels of IandI, both unnormalized and with the adoption of several normalization strategies. We found that normalization using physicochemical-based wastewater strength metrics (chemical oxygen demand, total suspended solids, phosphate, and ammonia) resulted in higher temporal and site-specific variability of SARS-CoV-2 and human fecal biomarker signal compared to unnormalized data, especially for viral and AMR marker signal measured in wastewater from sites with pronounced IandI. Viral wastewater signal normalized to physicochemical wastewater strength metrics and flow data also closely mirrored precipitation trends, suggesting such normalization approaches may more closely scale wastewater trends towards precipitation patterns rather than per capita signal in an IandI compromised system. We also found that in most cases, normalization did not significantly alter the relationship between wastewater trends and clinical infection trends. These findings suggest a degree of caution is warranted for some normalization approaches, especially where precipitation driven IandI is heightened. However, data and findings largely supported the utility of using human fecal markers such as crAssphage for normalizing wastewater signal to address site-specific differences in dilution levels, since viral signal scaled to this metric did not result in strong correlations between precipitation and wastewater trends, higher spatial and temporal variation was not observed, and strong correlations were observed between viral signal and viral infection trends. Finally, in chapter 5, we assessed the relationship between monthly Norovirus GII, Rotavirus, and SARS-CoV-2 wastewater trends with seasonal infection trends for each of the viruses to ascertain whether WBE could be used in a rural sewershed of this size with substantial IandI impacts to track and potentially predict population level infection trends. Though up-sewer, or near-source sampling, at sites with permanent IandI impacts did not exhibit a clear relationship with seasonal infection trends for Rotavirus, SARS-CoV-2, and Norovirus GII, WWTP influent signal and consensus signals aggregated from multiple up-sewer sites largely mirrored expected seasonal trends. Findings also suggested that for more ubiquitous viral targets, such as SARS-CoV-2, viral trends measured at WWTP influent in a small IandI impacted system may still provide a sufficiently useful measure of infection trends to inform the use of WBE (assuming appropriate normalization to sewershed population). These findings elucidate the potential utility and relative robustness of wastewater testing to ascertain community-level circulation of pathogens in small, rural sewersheds even those compromised by extensive IandI inputs. Overall, this dissertation examined drinking water and wastewater as critical metrics for assessing contaminant exposures and infectious disease trends in rural communities, particularly in the context of small, rural communities which tend to have more limited health infrastructure and lower-resource wastewater systems. Overall, findings underscore the need for baseline data on drinking water quality by identifying gaps in current knowledge and calling for further research to better understand drinking water contaminant exposure levels in rural areas. Wastewater as a non-invasive, population-level health metric was evaluated in the context of a small, rural sewer system overall, and by varying observed levels of IandI, as well as associated tradeoffs for normalization adoption. By evaluating these environmental surveillance metrics using both desk-based and field-based research study designs, findings from this dissertation offer valuable insights and practical recommendations for improving baseline drinking water quality monitoring and wastewater pathogen testing, all with the overarching goal of supporting more targeted public health interventions in rural settings. / Doctor of Philosophy / In the United States, there is a significant health and healthcare gap between rural and urban areas. Rural communities often face worse health outcomes, poorer health behaviors, and have less access to affordable and nearby healthcare services compared to their urban and peri-urban counterparts. Additionally, rural areas are exposed to higher risks for infectious diseases due to closer proximity to wildlife and livestock and proportionately lower access to regulated drinking water sources. Chronic conditions like diabetes and hypertension, which are more common in rural populations, can exacerbate the severity and duration of symptoms for infectious diseases, potentially leading to more serious illness and hospitalizations. Despite these heightened risks, data on health behaviors, outcomes, and healthcare services in rural areas is often lacking and less comprehensive compared to urban regions. This dissertation investigates two promising avenues of improving monitoring to provide information needed to better understand and address contaminant exposures and health trends in rural communities: drinking water and wastewater. Firstly, this dissertation underscores the importance of establishing baseline data on drinking water quality. This is essential for accurately estimating exposure levels and understanding the health impacts associated with elevated levels of drinking water contaminants, particularly in rural areas where a higher share of primary drinking water sources is unregulated by the federal government compared to urban areas. This study reveals significant gaps in current knowledge and highlights the need for more research to provide a clearer picture of drinking water quality in these communities. Secondly, this dissertation explores the use of wastewater as a non-invasive tool for assessing community health. This method, known as wastewater-based surveillance (WBS) or wastewater-based epidemiology (WBE), offers a way to measure population-level health trends without relying on clinical testing, which can be limited by factors such as access to healthcare and delays in testing. The dissertation evaluates how effective wastewater monitoring can be in small, rural sewer systems, even when these systems face challenges like aging infrastructure and significant inflow and infiltration (IandI) from groundwater and surface water. It examines how different normalization strategies for wastewater data can influence the reliability of this method and how wastewater testing can be adapted to account for varying levels of IandI. Overall, the dissertation provides valuable insights into the effectiveness of using drinking water and wastewater as environmental metrics for informing public health intervention strategies in rural settings. It offers justifications for improving drinking water quality monitoring and wastewater testing practices, aiming to support more targeted and effective public health interventions in rural communities. By addressing the challenges and limitations associated with these environmental monitoring strategies this research contributes to a better understanding of how to reduce health data disparities in rural areas.

Utbyggnation av städer och allt större andel hårdgjorda ytor leder till problem med dagvattenhanteringen. Flödena blir snabbare samtidigt som större mängder än tidigare bildar avrinning och då är det viktigt att de befintliga systemen klarar av att hantera dem. Dagvatten från tak och hårdgjorda ytor och dräneringsvatten från källarfastigheter kan vara kopplat till spillvattennätet och leda till att vattenflödena i ledningarna vid stora regn kan bli betydligt större än de är dimensionerade för, med kapacitetsproblem som följd. En lösning som många VA-huvudmän har implementerat är att använda öppna dagvattenlösningar istället för att välja det mer kostnadskrävande alternativet att bygga ut ledningssystemet. Att koppla om stuprör, brunnar och dräneringsledningar och gräva diken för att leda om vattnet är både kostnads- och resurskrävande. Syftet var därför att undersöka om multikriterieanalys kan fungera som ett lämpligt verktyg för att allokera resurserna till de områden som ger mest nytta per satsad krona. Detta genom att ta fram en metodik för detta syfte och testa dess robusthet för att avgöra om den är lämplig att applicera eller om osäkerheterna i parametervärdena blir för stora för att några slutsatser ska kunna dras. Metodiken som togs fram testades på Bjursås, ett litet samhälle två mil utanför Falun, eftersom omfattande utredningar gjorts i området tidigare, vilket innebar att mycket data fanns att tillgå. Undersökningar gjordes av var mängderna kunde förväntas bli stora, var det fanns stor risk för källaröversvämningar och vilka områden som bidrog mest till bräddningar. Detta utvärderades tillsammans med förväntade åtgärdskostnader och en samlad bedömning gjordes av var nyttan per satsad krona bör bli störst. Stora osäkerheter återfanns i bedömningen. Slutsatsen blev att metoden kan fungera som stöd vid beslutsunderlag, men att den inte är tillräckligt robust för att kunna användas uteslutande utan att efterföljande utvärderingar och kritisk granskning av resultaten måste göras. Detta kan lämpligtvis ske genom en känslighetsanalys. För att resultaten ska utgöra ett användbart verktyg för VA-huvudmannen måste också kostnaderna utvärderas noggrannare än de har gjorts i den här studien. / Urbanization along with a greater amount of hardened surfaces affects the storm water management. When the flows get faster and larger amounts create runoff, it is crucial that the available systems are able to handle the water. Storm water from roofs and asphalt surfaces and drainage water can be connected to the sewage water pipelines and make the flows in the conduits at rainfall much greater than they are designed for, leading to capacity issues. Many municipalities have implemented open storm water solutions instead of choosing the more cost-ineffective way of expanding the conduit system. Reconnecting downspouts, wells and drainage pipelines and digging trenches to lead the water is both cost and resource demanding. Therefore the aim of this study was to investigate whether multi criteria analysis can be used as an appropriate tool in order to allocate the resources to the most beneficial areas. This was done by developing a method for this aim and testing its robustness in order to determine if it is suitable to use in this context or if the uncertainties make the method too unreliable. The robustness in the method developed can be questioned since the uncertainties can be substantial. To be able to use this method, a lot of data is needed and the method needs to be updated relatively often in order to contain relevant information. It is recommended that a sensitivity analysis is performed along with the method, since the use of only one set of parameters can make the result relatively arbitrary. Using a Monte Carlo procedure with the uncertainties defined can reduce the time needed to measure and identify the values. The sensitivity analysis showed that the parameters that have the largest impact on the results are the number of residents living in every real estate with a basement, the catchment areas, the roughness parameters of the pipelines and the use of energy and chemicals in the system. In the future, the urbanization is expected to increase as well as the amount of rainfall and problems related to inflow and infiltration are expected to become more common. The hope is that the method used and its results will be useful for the municipalities’ future planning and to inspire to more studies on this topic.

Inflow and infiltration (I/I) water in the sewer network originates from groundwater, precipitation, seawater as well as over-leakage from drinking water pipes. It is the proportion of water in the sewer system that is not wastewater and causes problems both at the treatment plant as well as in the sewer network. Through overflows, floods and bypassing at the treatment plant I/I causes discharges of wastewater to recipients and nature and thus have negative environmental effects. I/I also stand for an overload in the system which, in addition to environmental consequences, also has social, economic and technical consequences. Despite efforts to reduce and mitigate the effects of I/I water for a long time, the problem remains and a large proportion of all the wastewater in the network and the treatment plan still is I/I.  Historically, and for many cases at present, there is little to no methodology for how the work with I/I water takes place and is prioritized in the municipalities and responsible water and sewerage organizations. Measures have instead been implemented with a reaction based approach, in other words when something has broken down or there is a failure in the system. A new discussion about I/I water has arisen in recent years due to new requirements from supervisory authorities. As there is not much follow-up or methodology around the work with I/I water, the process of developing a strategy to meet these new requirements for municipalities has shown to be difficult. In cases where analysis of measures against I/I water has been carried out, cost and effects in the treatment plant (i.e. the proportion of I/I water) have most often been used and dimensions other than economic have been left out of the analysis. As I/I water affects social and environmental factors to a great extent, this should be taken into greater account.  In this work, a multi-criterion analysis has been performed in a case study where similar measures to reduce I/I water and its effects in two different areas have been analysed. The included measures were separation, lining, pipe bursting and installation of a stormwater cassette. The results of the case study show that measure A2 has the greatest positive effect. However, there are major uncertainties in the case study conducted and the results are not considered to be robust enough to be used. However, what has been important to include from the work is that by defining and using criteria from all sustainability dimensions, a comprehensive action choice analysis is created and that it is a structured approach to apply. Here, the importance of opening up for discussion within the municipality/organization is emphasized to achieve a coherent view of the prioritization of I/I water and to achieve efficient management. Problems and amounts of I/I water are complex issues and vary from place to place, and are strongly linked to local geohydrological and climate conditions and design of local sewage systems. Therefore, it is important to coordinate the planning of I/I water to use a workflow that is general and can be used, regardless of conditions, in a simple way where adjustments in accordance with local conditions are possible. This work has shown that multi-criteria analysis is a clear and adaptable tool for meeting the difficulties that exist. / Tillskottsvatten härstammar från grundvatten, nederbörd samt havsvatten genom felaktiga anslutningar eller läckage i ledningsnätet samt från överläckage från dricksvattenledningar. Det är andelen av vatten i avloppsnätet som inte är spillvatten och orsakar, genom bräddning, översvämningar och förbiledning i reningsverket, utsläpp av avloppsvatten till recipienter och natur samt står för en överbelastning i systemet. Trots att man arbetat med att minska och lindra effekterna av tillskottsvatten under lång tid så kvarstår problemet och en stor andel av allt samlat vatten i ledningsnätet och som kommer till reningsverket är just tillskottsvatten.  Historiskt, samt för många fall i nuläget, så existerar det lite till ingen metodik kring hur arbetet med tillskottsvatten sker i kommunerna och ansvariga VA-organisationer utan åtgärder har implementerats på måfå eller när något har gått sönder. En ny diskussion kring tillskottsvatten har uppstått de senaste åren på grund av uppkomna nya kravställningar från tillsynsmyndigheter. I och med att det inte finns mycket uppföljning eller metodik kring arbetet med tillskottsvatten så har processen med att ta fram en strategi för att möta dessa nya krav för kommuner ofta varit svår. I de fall som analys av åtgärder mot tillskottsvatten har utförts så har kostnad och effekter i reningsverket (det vill säga andel mängd tillskottsvatten) oftast används och andra dimensioner än ekonomiska har lämnats utanför analysen. Då tillskottsvatten påverkar sociala samt miljömässiga faktorer i hög grad bör detta tas större hänsyn till.  I det här arbetet så har en multikriterieanalys utförts i en fallstudie där liknande åtgärder för att minska tillskottsvatten och dess effekter i olika områden har analyserats. De ingående åtgärderna var separering, infodring, rörspräckning samt anläggning av en dagvattenkassett. Resultatet av fallstudien visar att åtgärdsalternativ 2 har störst positiv effekt. Det finns dock stora osäkerheter i den fallstudie som utförts och resultatet anses inte vara tillräckligt robust för att användas. Det som dock har varit viktigt att ta med från arbetet är att genom att definiera och använda kriterier från alla hållbarhetsdimensioner skapas en heltäckande åtgärdsvalsanalys och att det är ett strukturerat arbetssätt att applicera. Här understryks vikten av att öppna upp för diskussion inom kommunen/organisationen för att nå en sammanhållen syn på prioriteringen av tillskottsvatten samt för att nå en effektiv hantering. Problem och mängder tillskottsvatten i olika ställen är komplext och starkt kopplat till lokala klimat och geohydrologiska förhållanden samt lokala avloppssystem. Därför är det viktigt för att samordna planeringen av tillskottsvatten att använda en arbetsgång som är generell och kan användas, oavsett förutsättningar, på ett enkelt sätt där justeringar i enlighet med lokala förutsättningar är möjligt. Det här arbetet har visat att multikriterieanalys är ett tydligt och anpassningsbart verktyg för att möta de svårigheter som finns.