Preliminary Analysis of Contamination of Emerging Concern using observations and machine learning approaches

Tuesday, May 20, 2025

10:45 AM – 11:00 AM Alaska Daylight Time (AKDT)

Room: Room 6

Author(s)

Venkataramana Sridhar

Associate Professor
Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, Virginia, United States

Co-Author (not attending)(s)

Raghavan Srinivasan

Blackland Research & Extension Center, Texas A&M University, Temple, TX, United States
Arun Bawa

Blackland Research & Extension Center, Texas A&M University, Temple, TX, United States

Co-Author(s)

Aditya Tyagi, PhD, PE (he/him/his)

Principal Technologist
Jacobs Engineering, Texas, United States

Co-Author (not attending)(s)

Manoj Kumar Jha, n/a

Manoj Kumar Jha Professor/Chair, Civil, Architectural, and Environmental Engineering
NCAT, North Carolina, United States
Biswanath Dari

North Carolina Agricultural and Technical State University, Greensboro, NC, United States

Abstract Submission: Intensive agriculture supports food security through the use of pesticides, fertilizers and chemicals, and helps to meet the goal set by the United Nations of providing enough food worldwide (UN Sustainable Development Goal 2); however, overuse of these substances can harm soil and water quality and contribute to land degradation, habitat loss and climate change. One of our long-term goals is to protect water quality by developing improved predictive and hindcasting decision-aids and tools that can better assess the impacts of changing climate and contaminants of emerging concerns (CECs) on water quality. However, lack of data particularly, Per- and polyfluoroalkyl substances (PFAS) can be an impediment to enhance our understanding of source and transport of PFAS in the watershed. We will focus on two watersheds, the South Yadkin River and the High Rock Lake watersheds in North Carolina, where water availability and water quality—including excess nutrients, pesticides and PFAS— are a matter of concern. For this presentation, we have also identified a few sites of PFAS contamination in the basin from the following interactive map published by the Environmental Working Group and perform a limited scale observational analysis and evaluate the trends and variability using both statistical and machine learning approaches. This analysis will subsequently help us inform the gaps in data as well as the development of a module with the Soil and Water Assessment Tool (SWAT) model in the next step. Point-scale preliminary results of the extent of PFAS pollution will be presented.

Learning Objectives/Expected Outcome (Optional) :