PyLEnM: A Machine Learning Framework for Long-Term Groundwater Contamination Monitoring Strategies

Abstract

In this study, we have developed a comprehensive machine learning (ML) framework for long-term groundwater contamination monitoring as the Python package PyLEnM (Python for Long-term Environmental Monitoring). PyLEnM aims to establish the seamless datato-ML pipeline with various utility functions, such as quality assurance and quality control (QA/QC), coincident/colocated data identification, the automated ingestion and processing of publicly available spatial data layers, and novel data summarization/visualization. The key ML innovations include (1) time series/multianalyte clustering to find the well groups that have similar groundwater dynamics and to inform spatial interpolation and well optimization, (2) the automated model selection and parameter tuning, comparing multiple regression models for spatial interpolation, (3) the proxy-based spatial interpolation method by including spatial data layers or in situ measurable variables as predictors for contaminant concentrations and groundwater levels, and (4) the new well optimization algorithm to identify the most effective subset of wells for maintaining the spatial interpolation ability for long-term monitoring. We demonstrate our methodology using the monitoring data at the Savannah River Site F-Area. Through this opensource PyLEnM package, we aim to improve the transparency of data analytics at contaminated sites, empowering concerned citizens as well as improving public relations.

Cr(VI) Reduction and Sequestration by FeS Nanoparticles Formed in situ as Aquifer Material Coating to Create a Regenerable Reactive Zone

“…Hexavalent chromium [Cr­(VI)] is one of the most commonly detected metal contaminants in groundwater. , Approximately 11% of the polluted sites on the U.S. National Priorities List are contaminated by Cr­(VI), and persistence of Cr­(VI) contamination at concentrations above the maximum contaminant level (MCL) is common, even after decades of remediation. ,, Cost-effective management of long-term contamination of groundwater at low concentrations is particularly challenging. − Most of the remedial technologies designed for fast removal of Cr­(VI) (e.g., in situ precipitation or chemical reduction) are less efficient for lasting large plumes of low contaminant concentrations. − Permeable reactive barriers have been used, and yet the loss of permeability due to biofouling or mineral precipitation limits their applicability. , Currently, pump and treat is still the most commonly used approach in dealing with low-level residual Cr­(VI) at many contaminated sites, but the generation of large quantities of wastewater and the long-term operating costs hinder its feasibility. , …”

An Artificial Intelligence Based Self-Adaptive Dynamic Process Control System for Enhancing In-Situ Bioremediation of Benzene-Contaminated Groundwater

“…Because the ISB prediction model is computationally costly, proxy modeling may be a good means of solving this challenge, i.e., to produce a set of proxy models to replace initial ones through statistical or artificial intelligence (AI) methods. Usually, proxy models have the advantages of computation-rapid, result-stable, and error-tolerable (Gopalakrishnan et al, 2011;Gorelick and Zheng 2015;He et al, 2008a;He et al, 2008b;Meray et al, 2022;Siade et al, 2020;Stramer et al, 2010). Second, generation of optimal operating conditions within a given time period by conventional DPC relies on the difference (or error) between the predicted remediation performance and pre-determined level.…”