About

Kelsey Pieper is an Assistant Professor in the Department of Civil and Environmental Engineering at Northeastern University College of Engineering. Her research focuses on applied environmental chemistry, corrosion, drinking water quality, treatment, and infrastructure, as well as post-disaster drinking water recovery and public health engineering. She has been actively involved in projects such as aiding hurricane relief efforts in North Carolina by creating databases to assess the impact on buildings and wells affected by storms, and testing private well water to address contamination issues faced by residents. Dr. Pieper is also leading research initiatives to ensure the safety of well water, particularly after events like floods, and is part of collaborative efforts evaluating the impacts of social capital on access to safe water. She was named an inaugural Early-Career Research Fellow by the Gulf Research Program of the National Academies of Sciences, Engineering, and Medicine, highlighting her contributions to community resilience and human health related to water safety.

Research topics

• Computer Science
• Data Mining
• Machine Learning
• Ecology
• Artificial Intelligence
• Oceanography
• Geology
• Environmental engineering
• Engineering
• Biology
• Environmental science

Selected publications

• A Bayesian Belief Network Model Assessing the Risk to Wastewater Workers of Contracting Ebola Virus Disease During an Outbreak

  UNC Libraries · 2026-04-16

  articleOpen accessSenior author

  During an outbreak of Ebola virus disease (EVD), hospitals' connections to municipal wastewater systems may provide a path for patient waste bearing infectious viral particles to pass from the hospital into the wastewater treatment system, potentially posing risks to sewer and wastewater workers. To quantify these risks, we developed a Bayesian belief network model incorporating data on virus behavior and survival along with structural characteristics of hospitals and wastewater treatment systems. We applied the model to assess risks under several different scenarios of workers' exposure to wastewater for a wastewater system typical of a mid-sized U.S. city. The model calculates a median daily risk of developing EVD of approximately 6.1×10^-12 (90% confidence interval: 1.0×10^-12 to 5.4×10^-9 ; mean 1.8×10^-6 ) when no prior exposure conditions are specified. Under a worst-case scenario in which a worker stationed in the sewer adjacent to the hospital accidentally ingests several drops (0.35 mL) of wastewater, median risk is 5.8×10^-4 (90% CI: 8.8×10^-7 to 9.5×10^-2 ; mean 3.2×10^-2 ) . Disinfection of patient waste with peracetic acid for 15 minutes prior to flushing decreases the estimated median risk to 3.8×10^-7 (90% CI: 4.1×10^-9 to 8.6×10^-5 ; mean 2.9×10^-5 ). The results suggest that requiring hospitals to disinfect EVD patient waste prior to flushing may be advisable. The modeling framework can provide insight into managing patient waste during future outbreaks of highly virulent infectious pathogens.

  PublisherDOI

• <i>Legionella</i> and <i>Mycobacterium</i> populations exhibit geographic structuring across and within drinking water systems

  bioRxiv (Cold Spring Harbor Laboratory) · 2026-01-14 · 1 citations

  articleOpen access

  Abstract Opportunistic pathogens (OPs) within the Legionella and Mycobacterium can persist and sometimes proliferate in drinking water systems and pose a risk to public health. Most prior research has focused on isolated system components of the drinking water treatment and distribution system and has rarely examined spatiotemporal dynamics across the entire source water, treatment process, and distribution system continuum. This study addresses this critical knowledge gap by quantitative profiling of microbial communities with full length 16S rRNA gene sequencing and flow cytometry, and associated water chemistry parameters, including disinfection byproducts (DBPs), across five full-scale utilities. These utilities reflect varying source water types, geographic locations, treatment regimes, and climate zones. Microbial communities, including Legionella and Mycobacterium populations, in distribution system were shaped by source water type and exhibited significant community divergence across utilities. Within the same genus, strain-level analyses revealed highly distinct Legionella and Mycobacterium sequence variants unique to each utility. Interestingly, a substantial proportion of Legionella and Mycobacterium amplicon sequence variants were both utility specific and often specific to locations within the distribution system, indicating strong geographic structuring both across and within drinking water systems. Understanding the mechanistic underpinnings of this geographic structuring is critical to develop robust strategies for managing and monitoring Legionella and Mycobacterium populations in drinking water systems.

  PublisherOA PDFDOI

• Interpretable Machine Learning Reveals Integrated Water Chemistry and Parameter-Specific Nonlinear Responses Shaping <i>Legionella</i> spp. and <i>Mycobacterium</i> spp. in Drinking Water

  medRxiv · 2026-04-27

  article

  Abstract Traditionally, studies have explored the impacts of individual water chemistry parameters on the persistence of Mycobacterium spp. and Legionella spp. in isolation with the underlying assumption that these associations are likely monotonic in nature. Yet chemical and microbiological changes are complex, and associations are likely highly combinatorial. In this study, we use interpretable machine learning models to disentangle the integrative and nonlinear associations between water chemistry and occurrence/abundance of Mycobacterium spp. and Legionella spp. Seasonal data from source water, point-of-entry and distribution systems of eight full-scale drinking water systems demonstrated that shifts in overall water chemistry were associated with the changes in microbial abundance during treatment and distribution. Machine learning models indicated moderate predictive ability of integrated water chemistry towards Legionella spp. abundance and towards the occurrence of both Legionella spp. and Mycobacterium spp., whereas predictive performance for Mycobacterium spp. abundance was limited. The association between nitrate and Legionella spp. abundance was disinfectant regimes dependent, while dissolved organic carbon exhibited a concentration dependent response type (i.e., positive and negative association). In chloraminated systems, Legionella spp. abundance was positively associated with ammonia and nitrate, highlighting the critical role of nitrification. Here, it appears that pH likely influences the initial colonization of Legionella spp. while ammonia governs its abundance in drinking water. Overall, this study demonstrates that integrated water chemistry and parameter-specific nonlinear effects collectively explain persistence of Mycobacterium spp. and Legionella spp. in drinking water systems. Synopsis This study elucidates the integrative impact of water chemistry and the nonlinear responses of individual water chemistry parameters on the occurrence and abundance of Mycobacterium spp. and Legionella spp. in drinking water using interpretable machine learning. TOC

  PublisherDOI

• Quantifying Improvements in Derived Storm Events from Version 07 of GPM IMERG Early, Late, and Final Data Products over North Carolina

  Remote Sensing · 2025-07-24

  articleOpen access

  In North Carolina (NC), roughly 1 in 4 residents rely on private wells for drinking water. Given the potential for flooding to impact well water quality, which poses serious health hazards to well users, accurate near real-time precipitation estimates are vital for guiding outreach and mitigation efforts. GPM IMERG precipitation data provides a solution for this need. Previous studies have shown that IMERG version 06 performs well throughout NC for capturing event totals. This study investigates changes in precipitation performance from IMERG version 06 to version 07 in NC and surrounding regions. There was significant improvement pertaining to errors quantifying the magnitude of precipitation events; the mean error in event precipitation decreased 75–85%, bias decreased 65–80%, and the root mean square error decreased 15–30% for Early, Late, and Final products as compared to event totals from in situ precipitation gauges. V07 shows improved performance during events in colder conditions, in mountainous regions, and with higher, prolonged intensities. During Hurricane Florence (September 2018), v07 improved precipitation estimates in regions with higher rainfall totals. These findings demonstrate the potential of the IMERG v07 Early and Late data products for the creation of accurate and timely flood models in emergency response applications.

  PublisherOA PDFDOI

• Exploring Demographic Disparities in Private Well Water Testing in North Carolina

  Carolina Digital Repository (University of North Carolina at Chapel Hill) · 2025-01-09

  articleOpen accessSenior author

  The natural, built, and social environments shape drinking water quality supplied by private wells. However, the combined effects of these factors are not well understood. Using North Carolina as a case study, we (i) estimate the demographic characteristics of the private well population; (ii) evaluate representation in well testing records; and (iii) demonstrate how spatial scale influences knowledge of well-using household demographics and representation in testing. We leverage a statewide database of 117,960 well testing records collected over 20 years and a national model predicting well locations. An estimated 25% well-using households identify as Black, Indigenous, and Persons of Color (BIPOC) and 15% have incomes below the poverty threshold. While there is robust well sampling (an average of 4,269 wells tested annually), we observed that most testing records were from predominately White block groups (BGs). Well-using households that did not participate in state testing were 2.4 times more likely to be from predominately BIPOC BGs compared predominately White BGs. Due to the spatial heterogeneity of the well population, demographic differences in well populations were more evident using higher resolution data. Multifaceted testing approaches that couple government-driven efforts with localized studies that engage underrepresented communities are needed to facilitate evidence-based management.

  PublisherDOI

• Responding after Hurricane Helene: Rapidly Estimating Impacts to Environmental Health Services in North Carolina

  Environmental Science & Technology Letters · 2025-06-23 · 4 citations

  articleOpen access1st authorCorresponding

  Hurricane Helene caused catastrophic flooding and infrastructure damage across the mountainous regions of western North Carolina. Responding agencies had to make real-time decisions about emergency response, infrastructure repair, and aid allocation. Here, we describe how our decade-long transdisciplinary research program supported data-driven recovery decisions in the days following a storm through the development of a novel emergency response decision support system (DSS). Integrating publicly available and geospatial data sets, we estimated that 4% of the total land area across the initial 25 disaster declared counties was flooded during Helene. While some areas did not experience a 100-year flood event, others had more severe flooding. We estimated that approximately 19 600 private wells, 34 300 businesses, and 500 fire stations were flooded. This type of real-time information was critical for supporting local health departments (LHDs) and state governments in their requests for emergency relief funding and their planning for emergency needs and assistance. Lessons learned through this effort highlight the importance of codeveloping knowledge and resources and providing actionable data and insights to enhance future disaster response efforts. Overall, our rapidly conceptualized and executed DSS demonstrated how providing actionable intelligence to responding LHDs and state governments can enable more effective distribution of real-time emergency resources.

  PublisherDOI

• Household Point-of-Use Faucet Filters for Lead Removal: Field Performance and User Experiences

  ACS ES&T Water · 2025-05-09 · 1 citations

  articleOpen access

  Lead certified point of use (POU) faucet filters were field tested in two Louisiana cities in a total of 21 occupied homes during normal use and in two unoccupied homes with very high risk of water lead (Pb) contamination. In an unoccupied home with a 28 m lead service line (LSL) with average lead of 17 μg/L in flushed water, treatment by POU filters generally produced water with <5 μg/L Pb even when tested to 200% of rated capacity. In the unoccupied home with a disturbed LSL, the water had erratic influent particulate lead of 9-3053 μg/L, and the POUs did not consistently produce water with <10 μg/L Pb despite very high percentage removals. When testing POUs in occupied homes known to occasionally have lead over 5 μg/L, POUs always reduced lead to <1 μg/L. In addition, POUs were also effectively removing high levels of iron and manganese present in the water of one city, but this also caused clogging before reaching half of their rated capacity. Laboratory experiments confirmed POU susceptibility to iron clogging.

  PublisherDOI

• Building plumbing influences the microdiversity and community assembly of the drinking water microbiome

  Water Research · 2025-02-06 · 5 citations

  article
  PublisherDOI

• Exploring Demographic Disparities in Private Well Water Testing in North Carolina

  Environmental Science & Technology · 2025-01-09 · 6 citations

  articleOpen accessSenior authorCorresponding

  The natural, built, and social environments shape drinking water quality supplied by private wells. However, the combined effects of these factors are not well understood. Using North Carolina as a case study, we (i) estimate the demographic characteristics of the private well population; (ii) evaluate representation in well testing records; and (iii) demonstrate how spatial scale influences knowledge of well-using household demographics and representation in testing. We leverage a statewide database of 117,960 well testing records collected over 20 years and a national model predicting well locations. An estimated 25% well-using households identify as Black, Indigenous, and Persons of Color (BIPOC) and 15% have incomes below the poverty threshold. While there is robust well sampling (an average of 4,269 wells tested annually), we observed that most testing records were from predominately White block groups (BGs). Well-using households that did not participate in state testing were 2.4 times more likely to be from predominately BIPOC BGs compared predominately White BGs. Due to the spatial heterogeneity of the well population, demographic differences in well populations were more evident using higher resolution data. Multifaceted testing approaches that couple government-driven efforts with localized studies that engage underrepresented communities are needed to facilitate evidence-based management.

  PublisherDOI

• Lead occurrence in North Carolina well water: importance of sampling representation and collection techniques

  Environmental Research Letters · 2024-02-20 · 4 citations

  articleOpen accessSenior authorCorresponding

  Abstract Private wells often lack centralized oversight, drinking water quality standards, and consistent testing methodologies. For lead in well water, the lack of standardized data collection methods can impact reported measurements, which can misinform health risks. Here, we conducted a targeted community science testing of 1143 wells across 17 counties in North Carolina (USA) and compared results to state testing data primarily associated with new well construction compiled in the NCWELL database. The goal of our study was to explore the impacts of sampling methodology and household representation on estimated lead exposures and subsequent health risks. At the household scale, we illustrated how sampling and analytical techniques impact lead measurements. The community science testing first draw samples (characterizing drinking water) had a 90th percentile lead value of 12.8 μ g l −1 while the NCWELL database flushed samples (characterizing groundwater) had a value below the reporting level of 5 μ g l −1 . As lead was associated with the corrosion of premise plumbing, flushing prior to collection substantially reduced lead concentrations. At the community scale, we examined how the lack of representation based on household demographics and well construction characteristics impacted the knowledge of lead and blood lead level (BLL) occurrence. When simulating representative demographics of the well populations, we observed that the 90th percentile lead level could differ by up to 6 μ g l −1 , resulting in communities being above the USEPA action level. This translated to a 1.0–1.3 μ g dl −1 difference in predicted geometric mean BLL among infants consuming reconstituted formula. Further, inclusion of less common well construction types also increased lead in water occurrence. Overall, under- and overestimations of lead concentrations associated with differences in sampling techniques and sample representation can misinform conclusions about risks of elevated BLLs associated with drinking water from private wells which may hinder investigations of waterborne lead exposure.

  PublisherOA PDFDOI

Recent grants

• RAPID: Extreme water use patterns and their impact on the microbial and chemical ecology of drinking water.

  NSF · $199k · 2020–2022

Frequent coauthors

• Marc Edwards

  40 shared

• Adrienne Katner

  Louisiana State University Health Sciences Center New Orleans

  25 shared

• William J. Rhoads

  19 shared

• Jeffrey Parks

  12 shared

• Min Tang

  Anhui University of Science and Technology

  9 shared

• Ameet Pinto

  Georgia Institute of Technology

  9 shared

• Dongjuan Dai

  Dalian Medical University

  9 shared

• Amy Pruden

  Virginia Tech

  8 shared

Labs

• Northeastern University College of Engineering - Kelsey Pieper LabPI

Education

• Doctor of Philosophy , Biological Systems Engineering

  Virginia Tech

  2015

• Masters of Science , Civil and Environmental Engineering

  Virginia Tech

  2011

• Bachelor of Science , Mechanical Engineering

  Binghamton University

  2009

Awards & honors

• Kelsey Pieper Named Gulf Research Program Inaugural Early-Ca…
• Spring 2022 PEAK Experiences Awardees for Undergrad Research