Statistical modelling of groundwater contamination monitoring data: A comparison of spatial and spatiotemporal methods

Interpretable machine learning for predicting the fate and transport of pentachlorophenol in groundwater

Pentachlorophenol (PCP) is a commonly found recalcitrant and toxic groundwater contaminant that resists degradation, bioaccumulates, and has a potential for long-range environmental transport. Taking proper actions to deal with the pollutant accounting for the life cycle consequences requires a better understanding of its behavior in the subsurface. We recognize the huge potential for enhancing decision-making at contaminated groundwater sites with the arrival of machine learning (ML) techniques in environmental applications. We used ML to enhance the understanding of the dynamics of PCP transport properties in the subsurface, and to determine key hydrochemical and hydrogeological drivers affecting its transport and fate. We demonstrate how this complementary knowledge, provided by data-driven methods, may enable a more targeted planning of monitoring and remediation at two highly contaminated Swedish groundwater sites, where the method was validated. We evaluated 6 interpretable ML methods, 3 linear regressors and 3 non-linear (i.e., tree-based) regressors, to predict PCP concentration in the groundwater. The modeling results indicate that simple linear ML models were found to be useful in the prediction of observations for datasets without any missing values, while tree-based regressors were more suitable for datasets containing missing values. Considering that missing values are common in datasets collected during contaminated site investigations, this could be of significant importance for contaminated site planners and managers, ultimately reducing site investigation and monitoring costs. Furthermore, we interpreted the proposed models using the SHAP (SHapley Additive exPlanations) approach to decipher the importance of different drivers in the prediction and simulation of critical hydrogeochemical variables. Among these, sum of chlorophenols is of highest significance in the analyses. Setting that aside from the model, tetra chlorophenols, dissolved organic carbon, and conductivity found to be of highest importance. Accordingly, ML methods could potentially be used to improve the understanding of groundwater contamination transport dynamics, filling gaps in knowledge that remain when using more sophisticated deterministic modeling approaches.

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

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 data-to-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 open-source PyLEnM package, we aim to improve the transparency of data analytics at contaminated sites, empowering concerned citizens as well as improving public relations.

Groundwater quality

Groundwater is a vital component of water supply for residential, industrial, and agricultural purposes. However, many groundwater basins are being used unsustainably and groundwater contamination is a growing water quality problem. Although anthropogenic activities and natural processes have been increasing the contamination in this valuable water resource, several remediation techniques have been developed in the last few decades to reduce these contamination levels. This review paper focuses on the recent studies developed on groundwater pollutions, remediation practices, and groundwater quality management. PRACTITIONER POINTS: Groundwater pollution is mainly due to anthropogenic activities and it is considered as a growing water quality problem. Groundwater bioremediation is one of the sustainable long-term solutions that uses the microorganisms to degrade the complex environmental pollutants. Groundwater quality management techniques play a significant role to restore or maintain water quality, which is critical for the sustainable development.

Comparing Bayesian spatial models: Goodness-of-smoothing criteria for assessing under- and over-smoothing

BACKGROUND

Many methods of spatial smoothing have been developed, for both point data as well as areal data. In Bayesian spatial models, this is achieved by purposefully designed prior(s) or smoothing functions which smooth estimates towards a local or global mean. Smoothing is important for several reasons, not least of all because it increases predictive robustness and reduces uncertainty of the estimates. Despite the benefits of smoothing, this attribute is all but ignored when it comes to model selection. Traditional goodness-of-fit measures focus on model fit and model parsimony, but neglect "goodness-of-smoothing", and are therefore not necessarily good indicators of model performance. Comparing spatial models while taking into account the degree of spatial smoothing is not straightforward because smoothing and model fit can be viewed as opposing goals. Over- and under-smoothing of spatial data are genuine concerns, but have received very little attention in the literature.

METHODS

This paper demonstrates the problem with spatial model selection based solely on goodness-of-fit by proposing several methods for quantifying the degree of smoothing. Several commonly used spatial models are fit to real data, and subsequently compared using the goodness-of-fit and goodness-of-smoothing statistics.

RESULTS

The proposed goodness-of-smoothing statistics show substantial agreement in the task of model selection, and tend to avoid models that over- or under-smooth. Conversely, the traditional goodness-of-fit criteria often don't agree, and can lead to poor model choice. In particular, the well-known deviance information criterion tended to select under-smoothed models.

CONCLUSIONS

Some of the goodness-of-smoothing methods may be improved with modifications and better guidelines for their interpretation. However, these proposed goodness-of-smoothing methods offer researchers a solution to spatial model selection which is easy to implement. Moreover, they highlight the danger in relying on goodness-of-fit measures when comparing spatial models.

Investigating Atrazine Concentrations in the Zwischenscholle Aquifer Using MODFLOW with the HYDRUS-1D Package and MT3DMS

Simulation models that describe the flow and transport processes of pesticides in soil and groundwater are important tools to analyze how surface pesticide applications influence groundwater quality. The aim of this study is to investigate whether the slow decline and the stable spatial pattern of atrazine concentrations after its ban, which were observed in a long-term monitoring study of pesticide concentrations in the Zwischenscholle aquifer (Germany), could be explained by such model simulations. Model simulations were carried out using MODFLOW model coupled with the HYDRUS-1D package and MT3DMS. The results indicate that the spatial variability in the atrazine application rate and the volume of water entering and leaving the aquifer through lateral boundaries produced variations in the spatial distribution of atrazine in the aquifer. The simulated and observed water table levels and the average annual atrazine concentrations were found to be comparable. The long-term analysis of the simulated impact of atrazine applications in the study area shows that atrazine persisted in groundwater even 20 years after its ban at an average atrazine concentration of 0.035 µg/L. These results corroborate the findings of the previous monitoring studies.

A probabilistic model of decision making regarding the use of chemical dispersants to combat oil spills in the German Bight

Oil spills are one of the major threats to the marine environment in the German Bight (North Sea). In case of an accident, application of chemical dispersants would be one response option among others. Dispersion breaks oil slicks into small droplets which get then mixed into the water column. Removal of the oil from the water surface may reduce contamination of the coast. However, the window of opportunity for effective dispersant application is short and there are concerns about potential effects to the marine life. We propose a Bayesian network (BN) as an interactive and intuitive tool for responders to justify decisions on using chemical dispersants and possibly the provision of appropriate assets. The BN combines detailed sub-BNs for different criteria that govern the decision process. Expected drift trajectories are estimated based on comprehensive numerical ensemble simulations of hypothetical oil spills. Ecological impacts are represented prototypically, focusing on vulnerability of seabird concentrations to pollution in coastal areas. Dispersant effectiveness is estimated considering oil properties and weather conditions. Decision making is supposed to be based on expected satisfaction. The definition of what is considered satisfactory is of central importance for the whole analysis.

Health risk assessment of nitrate in groundwater resources of Iranshahr using Monte Carlo simulation and geographic information system (GIS)

Because of exposure to a wide range of chemical contaminants such as nitrate via potable water resources, the use of the approaches to set standards for drinking water quality and also to do a risk assessment is necessary for maintaining the public health. High levels of nitrate in drinking water can have adverse health effects; primarily for infants and pregnant women. So, the present study aimed to the assessment of nitrate health risk in drinking water resources of the Iranshahr city, Sistan and Baluchistan province and also, evaluation of the uncertainty of nitrate and the probability of contamination occurrence by Monte-Carlo Simulation (MCS) technique. Besides, the geographic information system (Arc GIS, Ver 10.3) was applied to mapping the nitrate concentration in groundwater resources of the studied area. For these aims, the numbers of 66 samples were collected from rural groundwater resources, and nitrate concentration was measured using a Spectrophotometer in wavelength of 220 nm. According to the results, the nitrate concentration was in the range of 6.49 mg/L, and its average level was 6.15 mg/L. Also, the simulation results with 90% confidence showed that the hazard equitant (HQ) in the infant groups, children-teenagers and adults was less than 0.331, 0.311, 0.312, and 0.3, respectively.