Session 2- Monitoring Water and Health

Michigan’s Wastewater Monitoring Program for Public Health Use

2:15PM  Oral Presentation  
Authors: Nishita D'Souza and the Michigan Network for Environmental Health and Technology (MiNET) consortium
Presenting Author: Nishita D’Souza
Contact: dsouzan1@msu.edu 

Goals:

Wastewater monitoring is a valuable tool that can inform a variety of public health responses, including situations in which clinical data and resources are limited. The pandemic produced several large-scale research and surveillance system networks globally. Within the United States, Michigan was one of the first states to initiate the adoption of a larger scale state-wide WS program for SARS-CoV-2 in 2020. We discuss how Michigan’s state-wide laboratory network for monitoring built a growing knowledge base and expanded to include monitoring for other pathogens of concern. 

Methods:

The MiNET (Michigan Network for Environmental Health and Technology) conducts wastewater testing for the Michigan Department of Health and Human Services’ SEWER Network.  The network currently consists of 19 laboratories with community partners including 36 local health departments, 6 Native nations, 18 universities and 96 wastewater treatment plants (WWTPs) serving 58 counties. Michigan State University serves as the lead lab providing technical assistance, training and recommendations for testing methods. Since 2020, over 80,000 samples from 390 sites have been tested for SARS-CoV-2 using advanced molecular methods. Targeted testing for SARS-CoV-2 variants was implemented in 2021 to monitor the emergence of variants of concern.  Beginning in the Fall of 2023, the network began monitoring for norovirus and respiratory syncytial virus (RSV) and in Summer 2024 onboarded influenza A and B testing, with a focus on influenza A/H5. The data is provided to local stakeholders, state and federal agencies, and the public for public health use.

Key Findings:

Wastewater surveillance provided insights on community spread of viral infections since 2020. The data was able to provide early warning for SARS-CoV-2 and track the influx and spread of viral variants of concern from Alpha to Omicron. Norovirus, RSV and influenza trends were used to track prevalence and highlighted disease transmission patterns during peak seasons. The monitoring also supported influenza A/H5 tracking in Michigan during 2024.

Conclusions:

The SEWER Network monitoring program illustrates the power of collaboration between state agencies, academic laboratories, and local partners in wastewater monitoring and emphasizes the importance of building and capitalizing on advanced molecular capacity in Michigan for public health use. MiNET is currently poised to provide valuable support for state and federal agency decision-making and response by applying its competency in environmental monitoring to new and emerging pathogens of concern.

Presenter Bio

Nishita D’Souza is an Assistant Professor-Research in the Department of Fisheries and Wildlife at Michigan State University (MSU). Her research focuses on a One Health approach, applying culture and molecular based environmental and water quality monitoring tools to understand the burden of human pathogens; improve the safety of water, sustainability of resources and monitoring for public health impact. Dr. D’Souza works with the Michigan Network for Environmental Health and Technology (MiNET) and partners at the State of Michigan, MDHHS and EGLE to support funding acquisition, method optimization, quality assurance procedures, laboratory training and troubleshooting, data analysis and dissemination.

Implications of lateral groundwater flow on global land surface modeling across different spatial resolutions

2:30PM  Oral Presentation 
Authors: Tanjila Akhter, Yadu Pokhrel, Farshid Felfelani, Agnès Ducharne, Min-Hui Lo, and Robert Reinecke
Presenting Author: Tanjila Akhter
Contact: akhtert1@msu.edu

Groundwater, a vital freshwater source, is crucial for the hydrologic cycle. However, many Land Surface Models (LSMs) overlook lateral groundwater flow and aquifer pumping, especially at the global scale. This study uses an enhanced version of the Community Land Model (CLM5) to simulate lateral groundwater flow and conjunctive groundwater-surface water use for irrigation at three spatial resolutions: 0.5-degree, 0.25-degree, and 0.1-degree—validating the model against observed streamflow, terrestrial water storage from GRACE satellite, and annual groundwater dynamics. Results show that lateral groundwater flow increases with resolution, from 25 mm/year at 0.5-degree to 52 mm/year at 0.1-degree. Recharge is the dominant driver at 0.1-degree, while pumping plays a larger role at coarser resolutions. Lateral flow significantly influences land surface hydrology, particularly runoff in areas with high recharge and shallow water tables, and soil moisture and evapotranspiration (ET) in regions with deeper water tables and lower recharge. The increase in total runoff with higher resolution is more pronounced, while the effects on soil moisture and ET are relatively minor. This study emphasizes the implications of including lateral groundwater flow and pumping in different spatial resolution in the global land surface models.

Presenter Bio

Department of Civil and Environmental Engineering, Michigan State University, East Lansing, Michigan, USA

Process-Guided Ensemble Deep Learning Models for Predicting Lake Temperature Profiles

2:45PM  Oral Presentation  
Authors: Xin Lan, Lifeng Luo, Yue Deng and Pang-Ning Tan 
Presenting Author: Xin Lan
Contact: lanxin1@msu.edu 

Lakes act as sentinels of climate change, capturing its impacts both directly and indirectly. The recent rapid rise in global mean air temperatures has driven surface water warming in many freshwater lakes, as documented through in-situ and satellite observations. However, the scarcity of subsurface water temperature data leaves uncertainties about how subsurface temperatures and the vertical thermal structure of lakes have changed under climate change. The thermal structure is critical as it influences fish physiology, such as growth and reproduction, while also regulating the exchange of nutrients and oxygen between surface and deep lake waters. Deep learning offers a powerful method for simulating lake temperature profiles, enabling the analysis of long-term changes in subsurface water temperatures in response to climate change. In this study, we develop process-guided deep learning models combined with ensemble learning to simulate lake temperature profiles with improved accuracy and reliability.

The process-guided deep learning framework, integrated with ensemble learning, consists of two key components. First, the base models include four machine learning architectures: Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Transformer, and Recurrent Neural Network (RNN). These models are initially pre-trained with outputs from process-based models, such as the General Lake Model, to improve their predictive capabilities, followed by training with in-situ observational data. Second, the ensemble learning approach employs stacking, which combines these four base models. Within this stacking framework, a physics-guided loss function grounded in energy conservation principles is applied to ensure the deep learning models align with the physical dynamics of lake systems, enhancing both accuracy and interpretability

Presenter Bio

Xin Lan's research spans water resources, hydrology, machine learning, remote sensing, and coupled human-natural systems. Leveraging a strong interdisciplinary foundation, Xin Lan strives to tackle critical water challenges using advanced technologies and innovative approaches. Holding a Bachelor's degree in Marine Sciences from the China University of Geosciences (Beijing) and a Master's in Earth and Environmental Engineering from Columbia University, Xin Lan is currently pursuing a PhD dual major degree in Geography, Environment, and Spatial Sciences and Environmental Science and Policy. The focus is on investigating long-term trends in lake water temperatures across the United States using process-guided deep learning models. These models integrate physical principles with data-driven methods to enhance predictive accuracy and interpretability.

Representing reservoirs in Earth system models

3:00PM  Oral Presentation 
Authors: Ahmed Elkouk, Yadu Pokhrel, Lifeng Luo, Elizabeth Payton, and Ben Livneh
Presenting Author: Ahmed Elkouk
Contact: elkoukah@msu.edu 

Reservoir operation can profoundly impact river flow regimes and river basin hydrologic budget. However, reservoir operation is not represented in Earth system models (ESMs). Herein we evaluate the ability of a land model within an ESM with reservoir parametrization to capture reservoir inflows and storages in a highly managed region — the American Southwest. The model captures the dynamics of natural inflows to reservoirs across a wide range of hydroclimatic conditions in the Southwest. The model with a data-driven representation of reservoirs, based on reservoir record of operation, is capable of reproducing storage in some of the largest reservoirs in the region. However, the model performance is low in areas with substantial biases in modelled natural streamflow. Accurate modelling of reservoirs in these areas requires improving land model hydrologic performance. The ability of land models to accurately model reservoirs will allow for detailed assessments of the interplay between water availability, climate change, and human water management.

Presenter Bio

Ahmed is a PhD. student in the department of Civil and Environmental Engineering at Michigan State University. He has experience in assessing the hydrologic impacts of climate change, including studying soil moisture response to atmospheric warming across transitional environments. Currently, Ahmed's research, in collaboration with Western Water Assessment, focuses on a novel application of a state-of-the-science land model to investigate water and environmental sustainability in the American Southwest.

Assessing Agricultural Water Sustainability Using HYDRUS and Lysimeter for Deep Percolation in Sandy Soils

3:15PM  Oral Presentation 
Authors: Nawab Ali and Younsuk Dong
Presenting Author: Nawab Ali
Contact: alinawab@msu.edu 

Deep percolation (DP) is a challenging task in hydrology particularly in humid regions experiencing winter rainfall and snowmelt. The complex interaction of water balance components including precipitation, snow melt, bottom flux, surface runoff and soil hydraulic characteristics often impedes accurate estimation. Weighing lysimeter with soil moisture sensors were used to estimate DP in common sandy soils. The CS655 soil moisture sensors were installed at 15 cm, 30 cm, and 45 cm depths to estimate DP under different irrigation depths (mm) through change in soil moisture contents with time. Multilinear regression using irrigation depth (mm), change in soil water contents and time until stability (min) were used to predict DP across soil types. HYDRUS-1D was calibrated with time series sensor data to assess winter season (Nov-April) DP dynamics from rainfall and snowmelt across common soil types using the 10 years (2014-15 to 2023-24) climate data of two different locations. Maximum accuracy for DP estimation through sensor was observed in coarse textured soils. HYDRUS predicted well the DP which showed an interannual variability for Oakville Fine Sand (37-76%) and Riddles-Hillsdale Sandy Loam (75-89%) that of the water budget inputs (rainfall and snowmelt). Higher DP in wet years than dry years was observed with variation for soil types and locations. Besides these, certain location specific factors like vegetation, crop type, roots characteristics, soil cover and land use impact the soil water dynamics. HYDRUS modeling provides an understanding about fluxes under diverse locations, soils and climates which supports informed decision making for sustainable water management in agricultural fields. Certain management practices like precision irrigation, soil amendment, soil organic matter, soil cover, terracing and cover cropping enhance the water infiltration, sustainable agricultural water use and recharge.

Presenter Bio

I am Dr. Nawab Ali, PhD in Agronomy with research area of precision irrigation and nutrients management. Currently, I am working as Research Associate in Irrigation Lab, Biosystem and Agricultural Engineering, Michigan State University with research projects on Agriculture return flow and groundwater recharge estimation through sensor technology and empirical models and Precision irrigation with nutrients management through innovative techniques, IoT sensors and simulation modeling for enhancing water and nutrients use efficiencies, reducing losses and improving crop productivity of the specialty crops.

Session 2- Monitoring Water and Health

Michigan’s Wastewater Monitoring Program for Public Health Use

2:15PM  Oral Presentation 
Authors: Nishita D'Souza and the Michigan Network for Environmental Health and Technology (MiNET) consortium
Presenting Author: Nishita D’Souza
Contact: dsouzan1@msu.edu 

Goals:

Wastewater monitoring is a valuable tool that can inform a variety of public health responses, including situations in which clinical data and resources are limited. The pandemic produced several large-scale research and surveillance system networks globally. Within the United States, Michigan was one of the first states to initiate the adoption of a larger scale state-wide WS program for SARS-CoV-2 in 2020. We discuss how Michigan’s state-wide laboratory network for monitoring built a growing knowledge base and expanded to include monitoring for other pathogens of concern. 

Methods:

The MiNET (Michigan Network for Environmental Health and Technology) conducts wastewater testing for the Michigan Department of Health and Human Services’ SEWER Network.  The network currently consists of 19 laboratories with community partners including 36 local health departments, 6 Native nations, 18 universities and 96 wastewater treatment plants (WWTPs) serving 58 counties. Michigan State University serves as the lead lab providing technical assistance, training and recommendations for testing methods. Since 2020, over 80,000 samples from 390 sites have been tested for SARS-CoV-2 using advanced molecular methods. Targeted testing for SARS-CoV-2 variants was implemented in 2021 to monitor the emergence of variants of concern.  Beginning in the Fall of 2023, the network began monitoring for norovirus and respiratory syncytial virus (RSV) and in Summer 2024 onboarded influenza A and B testing, with a focus on influenza A/H5. The data is provided to local stakeholders, state and federal agencies, and the public for public health use.

Key Findings:

Wastewater surveillance provided insights on community spread of viral infections since 2020. The data was able to provide early warning for SARS-CoV-2 and track the influx and spread of viral variants of concern from Alpha to Omicron. Norovirus, RSV and influenza trends were used to track prevalence and highlighted disease transmission patterns during peak seasons. The monitoring also supported influenza A/H5 tracking in Michigan during 2024.

Conclusions:

The SEWER Network monitoring program illustrates the power of collaboration between state agencies, academic laboratories, and local partners in wastewater monitoring and emphasizes the importance of building and capitalizing on advanced molecular capacity in Michigan for public health use. MiNET is currently poised to provide valuable support for state and federal agency decision-making and response by applying its competency in environmental monitoring to new and emerging pathogens of concern.

Presenter Bio

Nishita D’Souza is an Assistant Professor-Research in the Department of Fisheries and Wildlife at Michigan State University (MSU). Her research focuses on a One Health approach, applying culture and molecular based environmental and water quality monitoring tools to understand the burden of human pathogens; improve the safety of water, sustainability of resources and monitoring for public health impact. Dr. D’Souza works with the Michigan Network for Environmental Health and Technology (MiNET) and partners at the State of Michigan, MDHHS and EGLE to support funding acquisition, method optimization, quality assurance procedures, laboratory training and troubleshooting, data analysis and dissemination.

Enhancing Influenza A Virus Sequencing in Wastewater

2:30PM  Oral Presentation 
Authors: Anwar Kalalah and Joan B. Rose 
Presenting Author: Anwar Kalalah
Contact: kalalaha@msu.edu 

Background: Influenza A virus (IAV) poses a significant public health concern due to its high mutation rate and rapid transmission rate. Wastewater-based surveillance offers an innovative approach for early detection and monitoring of viral pathogens to understand disease prevalence in communities. While reverse transcription-droplet digital PCR (RT-ddPCR) enables sensitive quantification of IAV RNA in wastewater, it lacks the detailed genomic characterization provided by next-generation sequencing (NGS). NGS, though less quantitative, can identify both known and emerging IAV variants. In this study, we evaluated NGS approaches to improve the detection of IAV variants in wastewater. Method: Wastewater samples were collected between March 2024 and January 2025. The samples were concentrated, and RNA was extracted using the commercial kits, and IAV as well as H5 subtype were subsequently quantified in the RNA extracts using an RT-ddPCR-based method. Positive samples were selected for NGS, initially using Illumina Microbial Amplicon Prep (IMAP) Influenza A/B kit and sequenced on Illumina MiSeq platform. To improve genome recovery, genomic and environmental DNA was digested prior to RT-PCR, and targeted amplification of cDNA was performed using universal influenza primers (Tuni 12, Tuni 12.4, and Tuni 13), followed by long-read sequencing on the MinION platform. Sequencing reads were imported to Galaxy platform and aligned to reference IAV genome using Minimap2 tool, and sequence variants were identified using the LoFreq tool. Additionally, taxonomic classification was performed using Kraken to assess microbial community composition and potential background contamination. Results and conclusion: A total of 13 wastewater samples positive for IAV (H5) were collected and analyzed. Initial sequencing using the Illumina MiSeq platform resulted in poor viral genome recovery, with viral reads comprising less than 0.01% of total reads. The low recovery suggests a strong bias toward bacterial and archaeal sequences and emphasizes the challenge of detecting IAV in wastewater using short-read sequencing alone. In response, genomic and environmental DNA was digested prior to RT-PCR, and targeted amplification using Tuni 12, Tuni 12.4, and Tuni 13 primers was performed before sequencing on the MinION platform. Further analyses are ongoing to assess genomic coverage in IAV segments in samples with high and low viral abundance. Our findings will demonstrate the limitations of amplicon-based short-read sequencing for IAV surveillance in wastewater and highlight the potential of targeted long-read sequencing for improved genome characterization. Future work will focus on evaluating additional enrichment methods and expanding sequencing efforts to assess IAV genomic diversity over time.

Presenter Bio

Anwar Kalalah is a microbial genomics researcher specializing in microbial genomics. He holds a Ph.D. in Biomedical Sciences from Auburn University and currently is a postdoc with Dr. Joan Rose at MSU, focusing on wastewater-based epidemiology using ddPCR and NGS. His current research interests include Influenza A sequencing to monitor viral evolution and diversity. Previously, he studied Shiga toxin-producing Escherichia coli at UTSA, where he investigated genomic diversity and antibiotic resistance.

Metagenomic insights into diversity of viruses in untreated wastewater: from virus concentration to bioinformatic analysis

2:45PM  Oral Presentation 
Authors: Yabing Li (liyabing@msu.edu), Pankaj Bhatt (bhattpan@msu.edu), Brijen Miyani (miyanibr@msu.edu),  and Irene Xagoraraki (xagorara@egr.msu.edu)
Presenting Author: Yabing Li
Contact: liyabing@msu.edu 

Monitoring of potentially pathogenic human viruses in wastewater is of crucial importance to understand disease trends in communities, predict potential outbreaks, and boost preparedness and response by public health departments. High throughput metagenomic sequencing opens an opportunity to expands the capabilities of wastewater surveillance. However, there are major bottlenecks in the metagenomic enabled wastewater surveillance, including the complexities in selecting appropriate sampling and concentration/virus enrichment methods as well as in bioinformatic analysis of complex samples with low human virus concentrations. In this presentation, we compared the effects of Virus Adsorption-Elution (VIRADEL) and the PolyEthylene Glycol (PEG) precipitation methods on identifying viruses and human viruses using sequencing in untreated wastewater samples in Detroit, Michigan metropolitan area.  To connect the wastewater sequencing data and clinical data in the given community, we conducted a broad wastewater screening for virus-related diseases using untreated wastewater samples collected from the metropolitan Detroit area in Michigan for 1-year period with metagenomics and bioinformatics. Metagenomic analysis of virus in wastewater is limited by the low abundance of viral genetic material in the sample. To address this limitation, we compared the performance of untargeted and targeted sequencing for detecting virus diversity in wastewater. Comparison of human viruses concentrated with two methods revealed that more viral reads, contigs, and human viral-related contigs were obtained in almost all wastewater samples collected and concentrated with the VIRADEL method as compared to samples collected and concentrated with the PEG precipitation method. Nearly complete draft genomes of Astrovirus, Betapolyomavirus, Norovirus, and Enterovirus were recovered in a few of the collected 48 samples, showing that this method can pinpoint circulating pathogens at the species or genotype level. The presence of some human viruses in wastewater was associated with reported clinical disease cases in the community. Some of the detected viral-related sequences belonged to human viruses that are not reported by the local health department. With the untargeted and targeted sequencing, 45 viruses that are potentially associated with human health were found in wastewater, 8 of them were unique to the targeted methods and 7 of them were unique to the untargeted method. More foodborne RNA viruses were identified with the targeted method. Integrating untargeted and targeted sequencing methods is essential to identify probable human pathogens in wastewater accurately and sensitively. Optimization of the current genomic and bioinformatic methods are essential to further broaden the capabilities of wastewater surveillance using metagenomics.

Presenter Bio

Yabing Li, is a postdoc in Environmental Engineering at Michigan State University (MSU). Her research interest is advancing the application of microbial technologies, bioinformatics and statistical models to address the real-world issues associated with public health and environmental sustainability. Currently, she works on environmental surveillance for identification and prediction of infectious disease outbreaks. She received her Ph.D. in Environmental Engineering from the University of Tennessee, Knoxville in 2020. Before this, she graduated from Nanjing University in China with a M.S. in 2016 and Wuhan University of Technology with a B.S. in 2013.

Influenza A in wastewater correlates to health outcomes in two different sub-populations (Michigan, USA): Analysis of Influenza A & B, RSV, and SARS-CoV-2

3:00PM  Oral Presentation 
Authors: Alexis M. Porter (porteal@gvsu.edu) and Ryan R. Otter (otterr@gvsu.edu)
Presenting Author: Alexis Porter
Contact: porteal@gvsu.edu 

Introduction

Early warning systems, such as wastewater monitoring, provide valuable information for public health guidance. Previous research has shown that data generated via wastewater monitoring is most useful when combined with other data (e.g., geographical context, clinical health outcomes). In this study we combined independently collected local health outcomes with wastewater monitoring data (influenza, RSV, SARS-CoV-2) collected from two wastewater treatment plants (WWTP) throughout the influenza season (October 2023 – April 2024) in Michigan, USA.  The goal of this study was to investigate seasonal and correlative relationships between health outcomes and wastewater monitoring data. 

Methods

Weekly samples (24-hour composites) from two WWTP were extracted for RNA and analyzed via ddPCR for influenza A & B, RSV, and SARS-CoV-2. Both WWTPs served two similar sized populations (~25,000 residents) that are separated by less than 50 km but had significantly different age structures. Site U served a skewed population due to a university (~15,000 students) and Site B served a more balanced and older population, due to its residential location. Human health outcomes were acquired from the local Public Health Departments. Weekly health outcomes included influenza positivity (A & B), Influenza-like Illnesses (ILI), and emergency department visits (ED). Analysis of data consisted of seasonal and correlation comparisons among and between all endpoints. 

Key Findings

Across the entire study period Site U had higher viral loads compared to Site B (influenza A:1.6x; influenza B:1.9x; SARS-CoV-2: 1.8x; RSV: 1.3x). Seasonal trends were observed for all wastewater monitoring endpoints, though when seasonality occurred varied by virus. Temporal analysis revealed the influenza A spike aligned ~1-2 week prior to the clinical peak of influenza A positivity. Correlation analysis revealed that significant positive relationships exist between influenza A in wastewater and clinical data at both WWTPs, after adjusting for the single spike observed at Site U (Site U: R2=0.58; p=0.001; Site B: R2=0.56; p=0.013). 

Discussion

In this study population age structure appears to have a significant influence on the total viral load captured via wastewater monitoring. Human health outcomes for influenza A correlated well with influenza A wastewater monitoring endpoints. Sensitivity to large viral spikes was noted, as these pose unique perspectives regarding human health outcomes versus mathematical analyses. Our analysis highlights importance of having interdisciplinary viewpoints when combining and interpreting molecular and public health data. 

Conclusion

Wastewater monitoring data correlated well with human health data for influenza A, but not influenza B, SARS-CoV-2, and RSV.

Presenter Bio

Alexis Porter is a research scientist at Annis Water Resources Institute at Grand Valley State University specializing in environmental health, waterborne epidemiology, and infectious disease pathogen surveillance. Alexis received MPH and BS degree in Biomedical Sciences from Grand Valley State University. Alexis is the project manager for the AWRI beach monitoring project that focuses on bacteria contamination and the S.E.W.E.R. Project that focuses on surveillance of pathogens in wastewater, identification of emerging variants and science communication to public health partners.

A Comparison of Variant Shifting of the SARS-CoV-2 Virus at Six West Michigan Locations

3:15PM  Oral Presentation  
Authors: Leah Wilson (wilsonlo@gvsu.edu),  Alexis Porter (porteal@gvsu.edu), Ryan Otter (otterr@gvsu.edu), Brendan May (maybre@mail.gvsu.edu), Brian Scull (scullbr@gvsu.edu)
Presenting Author: Leah Wilson
Contact: wilsonlo@gvsu.edu 

Introduction

High-throughput sequencing is a tool that has been utilized to understand viral changes in communities. Analyzing the whole genome of viruses allows for an accurate representation of viral dynamics through populations. Since 2020, the Otter Lab has been monitoring the SARS-CoV-2 virus in both Muskegon and Ottawa counties in an attempt to further understand how disease spreads and to help health officials make informed decisions. As SARS-CoV-2 evolved, this monitoring was adapted to include variant identification and analysis. Comparing sequencing data from different locations over one year showed that the COVID-19 disease moves differently through populations. The purpose of this presentation is to serve as a comparison study of COVID-19 variant shifting at six site locations over one year of sample collection in West Michigan.

Methods

Wastewater samples were collected from both wastewater treatment plants (WWTP) and manholes. RNA was extracted and analyzed for SARS-CoV-2 viral concentrations through droplet digital PCR (ddPCR). Samples with concentrations that exceeded 10,000 gene copies/100 mL were further analyzed utilizing the Qiagen QIAseq DIRECT protocol for library preparation and the Illumina MiSeq for whole genome analysis. Generated files were aligned with the SARS-CoV-2 parent sequence using the Qiagen QLC Workbench and variant calling was conducted using the Freyja pipeline to further address variant lineages and abundances.

Key Findings

Sequencing analysis of six sites revealed a total of 223 variants of the SARS-CoV-2 virus. Muskegon County’s WWTP (site M1) had a total of 29 variants, the City of Allendale’s WWTP (site A4) had 45 variants, and the Grand Haven and Spring Lake Area’s WWTP (site O1) had 37 variants. The Muskegon Heights location (site M5) revealed 48 variants, the Spring Lake Station (site O5) had 45 variants, and Manhole 4340 in Allendale (site A3) had a total of 55 variants. Each site location had differences in variants present from December of 2023- September of 2024. WWTP sites (M1, O1, and A4) were expected to have more variants present due to the large populations they serve. The manhole locations were expected to have fewer variants due to serving smaller populations, but often revealed more variant shifting than treatment plants. This could be due to manhole samples being less dilute than those from city or county wide treatment plants.

Conclusion

High-throughput sequencing analysis revealed the presence of variant shifting and offered information on viral dynamics of  COVID-19 disease that were present at 6 different locations in West Michigan.

Presenter Bio

Leah Wilson is an emerging career scientist who specializes in environmental and molecular work. She recently earned a bachelor's degree in Biology from Grand Valley State University. Currently, Leah is a research assistant in the Otter Lab at GVSU's Annis Water Resources institute. She is involved with multiple long term monitoring projects, including beach monitoring at Lake Michigan beaches, pathogen fate in wastewater, and whole genome sequencing of pathogens in wastewater.

Session 2- Water Quality and Impacts 

Flocculent Organic Sediments in Shallow Freshwater Ecosystems: Characteristics, Ecological Functions, and Research Opportunities

2:15PM  Oral Presentation 
Authors: Nudrat Fatima(fatimanu@msu.edu) and Jay Zarnetske (Jpz@msu.edu)
Presenting Author: Nudrat Fatima
Contact: fatimanu@msu.edu 

Flocculent Organic sediments (floc) may be an integral part of the ecosystem but are mostly overlooked or ignored in freshwater ecosystems. These loosely aggregated sediments are composed of organic matter, microbial communities, and inorganic materials. Floc has all the ingredients to act as a significant biogeochemical hotspot in the landscape as they contain high organic content and exist at the interface between surface and groundwaters where water, matter, energy, and organisms preferential mix and interact. Hence, floc has many potential functions from acting as a natural sink for pollutants and source of greenhouse gases. Yet, after more than 70 years since floc research started, the topic remains a mystery with research scattered across many disciplines and communities. Further, the logistics of sampling and quantifying floc characteristics are complex because they exist in many ecosystems and methods remain underdeveloped. Thus, a more wholistic approach to floc research is needed to overcome the current logistical constraints and reveal its role in ecosystems and society.

This review synthesizes decades of fragmented floc studies. It explores the biogeochemical characteristics of floc, alongside its spatial and temporal variability. This review consolidates many disciplines of research on floc and that without a clear definition and unified methodology for floc research, future floc studies will remain siloed. Here, we build upon more than seven decades of work, to help unify future floc research by introducing a common research framework and definition for floc. The standardized framework for studying floc is centered on its function in the environment and the definition integrates common and quantifiable physical, chemical, and biological attributes. With this definition and research framework, floc will be recognized more wholistically in freshwater research and ecological management. 

Presenter Bio

Nudrat Fatima is a dual Ph.D. candidate in Environmental Science and the Environmental Science and Policy Program at Michigan State University. Her research focuses on freshwater ecosystems, particularly the role of flocculent sediments (Floc) as ecosystem control points at the interface of groundwater and surface water.

Pharmaceutical and personal care products (PPCPs) in the sludge and biosolids matrix: distribution, persistence and innovation mitigation strategies

2:30PM  Oral Presentation  
Authors: Qianqian Dong
Presenting Author: Qianqian Dong
Contact: dongqia4@msu.edu 

PPCPs are the ‘pseudo-persistent’ emerging contaminants because of their continuous release to the environment from the municipal wastewater treatment plants (WWTP). Conventional wastewater treatment processes are not adequate to effectively remove PPCPs from the effluents, resulting in their accumulation in biosolids. When these biosolids are used as fertilizer in agriculture, the enriched PPCPs may pose a potential threat to the environment. In this study, we collected sludge and biosolids samples from twelve WWTPs across the United States and analyzed 36 PPCPs using liquid chromatography coupled to tandem mass spectrometry. The result revealed that over 72% of the PPCPs were detected in biosolids, with concentrations ranging from 3 ng/g to 10 µg/g. Ofloxacin and ciprofloxacin were among the most abundant PPCPs, with concentrations similar to those found in previous studies. A significant reduction in triclosan and triclocarban was observed, likely due to regulatory phase-outs of these chemicals in over-the-counter products. Experiments with 13C-labeled PPCPs showed enhanced adsorption of diphenhydramine, caffeine, and estrogen on biosolids after anaerobic digestion. A mass balance based on volatile suspended solids (VSS) loss indicated that the removal of PPCPs during anaerobic digestion was slower than the reduction of biomass. Cluster analysis revealed that PPCPs with varying structures may be adsorbed at different locations within the sludge, contributing to differences in removal efficiency. Two-stage anaerobic digestion was found to be more effective at removing PPCPs compared to one-stage systems, and a higher ratio of primary sludge facilitated PPCP transformation as well. Additionally, direct sanitation and dewatering processes reduced the PPCP load by up to 80%. Finally, Incineration of biosolids proved to be a highly effective method for the complete removal of most PPCPs. The implications of this work highlight that the elimination of trace organic compounds, such as PPCPs, requires a collaborative approach between governance and technology.

Kinetics for mass transfer of TCDD from contaminated soils to activated carbon amendments

2:45PM  Oral Presentation  
Authors: Lili Tian (tianlili@msu.edu), Kirankumar P. S.(pushpak1@msu.edu), Hui Li (lihui@msu.edu) and Brian Teppen (teppen@msu.edu)
Presenting Author: Lili Tian
Contact: tianlili@msu.edu 

Dioxin compounds are notorious anthropogenic environmental toxicants and standard remediation practice is to landfill contaminated soils to minimize potential risks. A potentially much less expensive soil remediation practice would be to minimize dioxin bioavailability by adding activated carbon (AC). Previous studies showed that 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) added to AC had bioavailability of zero to mice. Thus, AC amendments show promise for the remediation of dioxin-contaminated soils, but the kinetics of dioxin uptake by AC are still unclear. The present study aims to quantify the mass transfer kinetics of TCDD from soil to AC amendments. We used 14C-TCDD because TCDD is difficult to extract from AC, and we used a magnetite-AC material to isolate AC for analysis. Complex environmental transfer processes were examined as functions of soil mixing, soil water content, and soil organic matter content. The results showed that 46% TCDD transferred from constructed soil to AC in 26 d under static condition, high water to soil ratio decreased the transfer of TCDD under static state while increased the transfer under shaking state. OM increased the rate of mass transfer of TCDD, especially with low water content. Shaking increase the mass transfer of TCDD only with high water content. The present study provide fundamental information for the application of activated carbon in dioxin-contaminated sites.

Using stream chemistry as a tracer of thaw depth across Arctic Alaskan permafrost catchments

3:00PM  Oral Presentation 
Authors: Amelia Grose (groseame@msu.edu), Jay Zarnetske (jpz@msu.edu), Abigail Rec (Abigail.Rec@uvm.edu), Arial Shogren (ashogren@ua.edu), Jonathan O’Donnell (Jonathan_O'Donnell@nps.gov), Benjamin Abbott (benabbott@byu.edu), Breck Bowden (Breck.Bowden@uvm.edu), Arsh Grewal (grewala7@msu.edu)
Presenting Author: Amelia Grose
Contact: groseame@msu.edu 

The Arctic is rapidly changing due to increasing temperatures and hydrologic intensification. The Arctic is also data-limited, necessitating the development of new tools to document and quantify ecosystem responses to these changes. Some of the hardest changes to observe are in the subsurface, including thaw depth conditions in continuous permafrost regions. Thaw depth is dynamic across the thaw season as well as on a longer, interannual scale as the Arctic warms and permafrost degrades. Measuring thaw depth often requires intensive sampling or remote sensing capabilities that have spatiotemporal limitations; therefore, little is known about complex subsurface dynamics and how they will affect Arctic ecosystems in the future. Often, surface waters are our best proxy for subsurface dynamics, as streams integrate signals from the landscape as water travels through the subsurface. In permafrost systems, select solutes increase in concentration with depth in the subsurface. As thaw depth increases, deeper water flowpaths are activated and increased concentrations of these solutes in the stream may signal increasing thaw depth. Prior work suggests there are multiple soil-derived solutes that may be tracers of thaw in Arctic catchments. We want to know if this tracer approach works at different spatiotemporal scales, including across catchments with varying characteristics (e.g., slope, vegetation).

To study this, we sampled the stream outlets of three catchments underlain by continuous permafrost on Alaska’s North Slope across three thaw seasons (2021-2023). We measured continuous discharge and analyzed nine different ions to identify seasonal patterns in stream chemistry, as these elemental concentrations change with soil depth in this region. As discharge could impact instream solute concentrations, we also analyzed concentration-discharge relationships to determine whether discharge was influencing concentration significantly and reducing the utility of the tracer to signal thaw depth. 

We found that the efficacy of the stream tracer approach to detect thaw on a seasonal scale is seemingly dependent on catchment characteristics, as we suspected. No single solute worked in all catchments to track thaw depth. Solute concentration was often impacted by discharge, especially in our low-gradient catchments. Overall, understanding thaw depth dynamics will become increasingly important with climate change, necessitating the development of tools to document and predict thaw depth at a range of scales. Here, we find that stream tracers of thaw at the catchment scale show promise, but it is much more nuanced and complex than preliminary studies indicated.

Presenter Bio

Amelia Grose is a PhD Candidate in the Department of Earth and Environmental Sciences at MSU, where she studies Arctic hydrology and stream biogeochemistry. Throughout her PhD, she has completed fieldwork for a total of 12 months in Arctic Alaska, based at Toolik Field Station.

Are there differences in microbial community composition within the Saginaw Aquifer, the predominant source of drinking water for mid-Michigan?

3:15PM  Oral Presentation 
Authors: Carol Hogan, Katie Quinlan (quinla37@msu.edu) and Matt Schrenk (schrenkm@msu.edu)
Presenting Author: Carol Hogan
Contact: hoganca2@msu.edu 

The goal of this research project is to study how the geochemistry of groundwater and groundwater age is shaping the microbial communities in the Saginaw aquifer. The aquifer is primarily made from sandstone, which allows groundwater to move through the pores, carrying dissolved chemicals and bacterial cells. The bedrock of Michigan Basin, contains multiple types of bedrock each with different geochemistry. Additionally, the groundwater in the lower peninsula flows from mid- Michigan to Lake Huron towards Saginaw Bay, displaying different ages of water, young to old. I hypothesize that each well will have different microbial communities but the geographically close wells will have similar biogeochemical data, and the variation is due to difference in location or the groundwater age.

Over the summer of 2023, the groundwater samples were collected from private wells throughout the communities of Grand Ledge, Lansing, Okemos, Saginaw City and Bay City, MI from the Saginaw aquifer. The water was filtered for PCR and DNA extraction, then for Amplicon 16s rRNA sequencing. Additionally, the samples were observed under epifluorescent microscope for biomass quantification and microbial characterization. The geochemistry of groundwater was measured in situ and through ion analysis. The groundwater age is determined by tritium and carbon dating with the help of collaborators. All the biological and geochemical data are statistically and spatially analyzed using R and GIS software. 

Most of the groundwater appeared orange in color, likely an indication of high iron content. This inhibited PCR amplification and only half the samples were successful in Amplicon sequencing. Each well has a different microbial community from the other wells, and the difference is majorly driven by the amount of total nitrogen and total carbon, and by groundwater age. Surprisingly, measurements such as pH or electric conductivity have very little influence on the community. In addition, the number of species and the species richness have a correlation, and there tends to be more diversity in wells from the Lansing area than from the Grand Ledge area. 

In conclusion, the groundwater age and some nutrients have the most influence on the difference of microbial communities in each well, and there are regional differences in diversity. The health of groundwater is a very important topic, as it’s the major drinking water source in Michigan and is vulnerable to anthropogenic influence. In the future, I’d like to study the spatial variables and how human activities may have affected the microbial communities.

Presenter Bio

Carol Hogan is an undergraduate student studying Environmental Biology/Microbiology. She has been working with Matt Schrenk on Michigan groundwater microbiology project. She is from Japan, and she is passionate about microscopy and sampling field trips.