Replacing the internal standard to estimate micropollutants using deep and machine learning

Assessing Event-Driven Dynamics of Pesticides and Transformation Products in an Agricultural Stream Using Comprehensive Target, Suspect, and Nontarget Analysis

A comprehensive assessment of pesticide transport in surface waters is challenging due to discharge characteristics and the occurrence of transformation products (TPs). Detailed long-term sampling of pesticide concentrations, including rainfall and pesticide application events, is still lacking to better predict pesticide transport pathways and toxicity within agricultural catchments. In the present study, pesticide and TP transport dynamics were evaluated over a three-year monitoring period, which included 12 stormwater events and 7 dry events. An extensive target screening for 328 pesticides was conducted, while simultaneously performing suspect and nontarget analysis (SNTA) using liquid chromatography high-resolution mass spectrometry. Twenty-one pesticides and two TPs associated with the main crop, rice, were identified as the major pollutants. The risk assessment results, based on the stepwise toxicity data collection, suggested that insecticides, primarily neonicotinoids, exhibited severe ecological risk. Additionally, SNTA revealed the presence of 8 parent compounds and 46 TPs. TPs occurred following parent peak periods, indicating that integrated pesticide monitoring is a practical approach to risk assessment. A precautionary approach using SNTA of parent pesticides and TP identification suggests that the potential aquatic effects of pesticide TPs may be underestimated by a conventional pesticide monitoring strategy.

Developing a real-time water quality simulation toolbox using machine learning and application programming interface

Rivers are vital for sustaining human life as they foster social development, provide drinking water, maintain aquatic ecosystems, and offer recreational spaces. However, most rivers are being increasingly contaminated by pollutants from non-point sources, urbanization, and other sources. Consequently, real-time river water quality modeling is essential for managing and protecting rivers from contamination, and its significance is growing across various sectors, including public health, agriculture, and water treatment systems. Therefore, a real-time river water quality simulation toolbox was developed using machine learning (ML) and an application program interface (API). To create the toolbox, models that simulated water quality parameters such as chlorophyll a (Chl-a), dissolved oxygen (DO), total nitrogen (TN), total organic carbon (TOC), and total phosphorus (TP) at each point in the Nakdong River were constructed. The models were constructed using Artificial neural network (ANN), Random Forest (RF), support vector machines (SVM), and data from API. Subsequently, hyperparameter optimization was conducted to enhance the model's performance. During training, the models' performances were evaluated and compared based on the data sampling method and ML algorithms. Models trained with random sampling data outperformed those trained with time-series data. Among the algorithm models that used random sampling data, the RF exhibited the best performance. The average coefficient of determination (R2) values for each water quality simulation with randomly sampled data using RF for DO, TN, TP, Chl-a, and TOC were 0.79, 0.65, 0.74, 0.45, and 0.48, respectively. For ANN, they were 0.7, 0.51, 0.64, 0.35, and 0.35, respectively, and for SVM, they were 0.73, 0.51, 0.59, 0.21, and 0.3, respectively. The Chl-a and TOC models exhibited relatively poor performance, whereas the DO, TN, and TP models demonstrated superior performance. Diversifying the input data variables is necessary to improve the performance of the Chl-a and TOC models. Sensitivity and uncertainty analyses were conducted to evaluate and enhance the models' understanding. Furthermore, using a graphic user interface (GUI) toolbox, user convenience was maximized.

Farmland's silent threat: Comprehensive multimedia assessment of micropollutants through non-targeted screening and targeted analysis in agricultural systems

Intricate agricultural ecosystems markedly influence the dynamics of organic micropollutants, posing substantial threats to aquatic organisms and human health. This study examined the occurrence and distribution of organic micropollutants across soils, ditch sediment, and water within highly intensified farming setups. Using a non-targeted screening method, we identified 405 micropollutants across 10 sampling sites, which mainly included pesticides, pharmaceuticals, industrial chemicals, and personal care products. This inventory comprised emerging contaminants, banned pesticides, and controlled pharmaceuticals that had eluded detection via conventional monitoring. Targeted analysis showed concentrations of 3.99-1021 ng/g in soils, 4.67-2488 ng/g in sediment, and 12.5-9373 ng/L in water, respectively, for Σ40pesticides, Σ8pharmaceuticals, and Σ3industrial chemicals, indicating notable spatial variability. Soil organic carbon content and wastewater discharge were likely responsible for their spatial distribution. Principal component analysis and correlation analysis revealed a potential transfer of micropollutants across the three media. Particularly, a heightened correlation was decerned between soil and sediment micropollutant levels, highlighting the role of sorption processes. Risk quotients surpassed the threshold of 1 for 13-23 micropollutants across the three media, indicating high environmental risks. This study highlights the importance of employing non-targeted and targeted screening in assessing and managing environmental risks associated with micropollutants.

Profiling emerging micropollutants in urban stormwater runoff using suspect and non-target screening via high-resolution mass spectrometry

Urban surface runoff contains chemicals that can negatively affect water quality. Urban runoff studies have determined the transport dynamics of many legacy pollutants. However, less attention has been paid to determining the first-flush effects (FFE) of emerging micropollutants using suspect and non-target screening (SNTS). Therefore, this study employed suspect and non-target analyses using liquid chromatography-high resolution mass spectrometry to detect emerging pollutants in urban receiving waters during stormwater events. Time-interval sampling was used to determine occurrence trends during stormwater events. Suspect screening tentatively identified 65 substances, then, their occurrence trend was grouped using correlation analysis. Non-target peaks were prioritized through hierarchical cluster analysis, focusing on the first flush-concentrated peaks. This approach revealed 38 substances using in silico identification. Simultaneously, substances identified through homologous series observation were evaluated for their observed trends in individual events using network analysis. The results of SNTS were normalized through internal standards to assess the FFE, and the most of tentatively identified substances showed observed FFE. Our findings suggested that diverse pollutants that could not be covered by target screening alone entered urban water through stormwater runoff during the first flush. This study showcases the applicability of the SNTS in evaluating the FFE of urban pollutants, offering insights for first-flush stormwater monitoring and management.

Estimating the temporal and spatial distribution and threats of bisphenol A in temperate lakes using machine learning models

Bisphenol A (BPA) is easily enriched in many human-disturbed watersheds, particularly lakes with poor water mobility, which is posing a threat to aquatic biota. While previous studies have focused on the concentration of BPA in water and its toxicity to aquatic organisms, a small amount of measured data is not enough to reveal the temporal and spatial distribution and threats of BPA, and estimate the ecological risk in watersheds. Therefore, we collected 164 measured BPA data points from Taihu Lake to develop machine learning models using random forest (RF), support vector machine (SVM) and least square regression (LSR) and created month-by-month watershed prediction maps in temperate lakes to estimate the spatiotemporal distribution and threats of BPA. Due to RF's superior robustness to noisy data, the RF model exhibits the best performance among the three algorithms. The RF model showed acceptable predictive performance on the modeling dataset (coefficients of determination and root-mean-square error for the training set were 0.927 and 17.499, respectively, and 0.607, 39.645 for the validation set, respectively). The maps indicated that areas susceptible to anthropogenic activities were more severely polluted by BPA, and rainy climate may favor the migration of BPA to aquatic ecosystems. The model was also applied to predict 42 data points of BPA collected from Dianchi Lake, and the results showed that most predicted data were within a factor of 10 of the measured data, but the prediction accuracy of the model has declined. The ecological risks in the two lakes were evaluated and attention should be paid to the regions with higher risks. Our study provided a novel idea for comprehensive monitoring of an unconventional trace pollutant with endocrine disrupting effects in aquatic ecosystems and analyzing their spatiotemporal distribution, which will contribute to the scientific assessment of the ecological risk of BPA.

Long short-term memory models of water quality in inland water environments

Water quality is substantially influenced by a multitude of dynamic and interrelated variables, including climate conditions, landuse and seasonal changes. Deep learning models have demonstrated predictive power of water quality due to the superior ability to automatically learn complex patterns and relationships from variables. Long short-term memory (LSTM), one of deep learning models for water quality prediction, is a type of recurrent neural network that can account for longer-term traits of time-dependent data. It is the most widely applied network used to predict the time series of water quality variables. First, we reviewed applications of a standalone LSTM and discussed its calculation time, prediction accuracy, and good robustness with process-driven numerical models and the other machine learning. This review was expanded into the LSTM model with data pre-processing techniques, including the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise method and Synchrosqueezed Wavelet Transform. The review then focused on the coupling of LSTM with a convolutional neural network, attention network, and transfer learning. The coupled networks demonstrated their performance over the standalone LSTM model. We also emphasized the influence of the static variables in the model and used the transformation method on the dataset. Outlook and further challenges were addressed. The outlook for research and application of LSTM in hydrology concludes the review.

Predicting micropollutant fate during wastewater treatment using refined classical kinetic model based on quantitative monitoring in multi-metropolitan regions of South Korea

This study aimed to implement an extensive prediction model for the fate of micropollutants (MPs) in wastewater treatment plants (WWTPs). Five WWTPs equipped with seven different biological treatment processes were monitored from 2020 to 2022 with three to four sampling events in each year, and 27 datasets for 20 MPs were collected. Among these datasets, 12 were used to investigate the behavior and fate of MPs in WWTPs in South Korea. Metformin, acetaminophen, caffeine, naproxen, and ibuprofen were the MPs with the highest influent concentrations (ranging from 3,933.3-187,637.0 ng L-1) at all WWTPs. More than 90% of MPs were removed by biological treatment processes in all WWTPs. The Kruskal-Wallis test verified that their efficacy did not differ statistically (p-value > 0.05). Meanwhile, to refine the performance of the prediction model, this study optimized the biodegradation rate constants (kbio) of each MP according to the variation of seasonal water temperature. As a result, compared to the original prediction model, the mean difference between the actual data and predicted results (MEAN) decreased by 6.77%, while the Nash-Sutcliffe efficiency (NSE) increased by 0.226. The final MEAN and NSE for the refined prediction model were calculated to be 5.09% and 0.964, respectively. The prediction model made accurate predictions, even for MPs exhibiting behaviors different from other cases, such as estriol and atrazine. Consequently, the optimization strategy proposed in this study was determined to be effective because the overall removal efficiencies of MPs were successfully predicted even with limited reference datasets.

A data-driven approach for understanding the structure dependence of redox activity in humic substances

Humic substances (HS) can facilitate electron transfer during biogeochemical processes due to their redox properties, but the structure-redox activity relationships are still difficult to describe and poorly understood. Herein, the linear (Partial Least Squares regressions; PLS) and nonlinear (artificial neural network; ANN) models were applied to monitor the structure dependence of HS redox activities in terms of electron accepting (EAC), electron donating (EDC) and overall electron transfer capacities (ETC) using its physicochemical features as input variables. The PLS model exhibited a moderate ability with R² values of 0.60, 0.53 and 0.65 to evaluate EAC, EDC and ETC, respectively. The variable influence in the projection (VIP) scores of the PLS identified that the phenols, quinones and aromatic systems were particularly important for describing the redox activities of HS. Compared with the PLS model, the back-propagation ANN model achieved higher performance with R² values of 0.81, 0.65 and 0.78 for monitoring the EAC, EDC and ETC, respectively. Sensitivity analysis of the ANN separately identified that the EAC highly depended on quinones, aromatics and protein-like fluorophores, while the EDC depended on phenols, aromatics and humic-like fluorophores (or stable free radicals). Additionally, carboxylic groups were the best indicator for evaluating both the EAC and EDC. Good model performances were obtained from the selected features via the PLS and sensitivity analysis, further confirming the accuracy of describing the structure-redox activity relationships with these analyses. This study provides a potential approach for identifying the structure-activity relationships of HS and an efficient machine-learning model for predicting HS redox activities.

Artificial intelligence and machine learning-based monitoring and design of biological wastewater treatment systems

Artificial intelligence (AI) and machine learning (ML) are currently used in several areas. The applications of AI and ML based models are also reported for monitoring and design of biological wastewater treatment systems (WWTS). The available information is reviewed and presented in terms of bibliometric analysis, model's description, specific applications, and major findings for investigated WWTS. Among the applied models, artificial neural network (ANN), fuzzy logic (FL) algorithms, random forest (RF), and long short-term memory (LSTM) were predominantly used in the biological wastewater treatment. These models are tested by predictive control of effluent parameters such as biological oxygen demand (BOD), chemical oxygen demand (COD), nutrient parameters, solids, and metallic substances. Following model performance indicators were mainly used for the accuracy analysis in most of the studies: root mean squared error (RMSE), mean square error (MSE), and determination coefficient (DC). Besides, outcomes of various models are also summarized in this study.

Hierarchical deep learning model to simulate phytoplankton at phylum/class and genus levels and zooplankton at the genus level

Harmful algal blooms (HABs) have become a global issue, affecting public health and water industries in numerous countries. Because funds for monitoring HABs are limited, model development may be an alternative approach for understanding and managing HABs. Continuous monitoring based on grab sampling is time-consuming, costly, and labor-intensive. However, improving simulation performance remains a major challenge in modeling, and current methods are limited to simulating phytoplankton (e.g., Microcystis sp., Anabaena sp., Aulacoseira sp., Cyclotella sp., Pediastrum sp., and Eudorina sp.) and zooplankton (e.g., Cyclotella sp., Pediastrum sp., and Eudorina sp.) at the genus level. The traditional modeling approach is limited for evaluating the interactions between phytoplankton and zooplankton. Recently, deep learning (DL) models have been proposed for solving modeling problems because of their large data handling capabilities and model structure flexibilities. In this study, we evaluated the applicability of DL for simulating phytoplankton at the phylum/class and genus levels and zooplankton at the genus level. Our work was an explicit representation of the taxonomic and ecological hierarchy of the DL model structure. The prerequisite for this model design was the data collection at two taxonomic and hierarchical levels. Our model consisted of hierarchical DL with classification transformer (TF) and regression TF models. These DL models were hierarchically connected; the output of the phylum/class level model was transferred to the genus level simulation model, and the output of the genus level model was fed into the zooplankton simulation model. The classification TF model determined the phytoplankton occurrence initiation date, whereas the regression TF model quantified the cell concentration of plankton. The hierarchical DL showed potential to simulate phytoplankton at the phylum/class and genus levels by producing average R^2, and root mean standard error values of 0.42 and 0.83 [log(cells mL^-1)], respectively. All simulated plankton results closely matched the measured concentrations. Particularly, the simulated cyanobacteria showed good agreement with the measured cell concentration, with an R² value of 0.72. In addition, our simulated result showed good agreement in peak concentration compared to observations. However, a limitation remained in following the temporal variation of Tintinnopsis sp. and Bosmia sp. Using an importance map from the TF model, water temperature, total phosphorus, and total nitrogen were identified as significant variables influencing phytoplankton and zooplankton blooms. Overall, our study demonstrated that DL can be used for modeling HABs at the phylum/class and genus levels.

A review of the application of machine learning in water quality evaluation

With the rapid increase in the volume of data on the aquatic environment, machine learning has become an important tool for data analysis, classification, and prediction. Unlike traditional models used in water-related research, data-driven models based on machine learning can efficiently solve more complex nonlinear problems. In water environment research, models and conclusions derived from machine learning have been applied to the construction, monitoring, simulation, evaluation, and optimization of various water treatment and management systems. Additionally, machine learning can provide solutions for water pollution control, water quality improvement, and watershed ecosystem security management. In this review, we describe the cases in which machine learning algorithms have been applied to evaluate the water quality in different water environments, such as surface water, groundwater, drinking water, sewage, and seawater. Furthermore, we propose possible future applications of machine learning approaches to water environments.

Insights into the Crucial Role of Electron and Spin Structures in Heteroatom-Doped Covalent Triazine Frameworks for Removing Organic Micropollutants

The water shortage crisis, characterized by organic micropollutants (OMPs), urgently requires new materials and methods to deal with it. Although heteroatom doping has been developed into an effective method to modify carbon nanomaterials for various heterogeneous adsorption and catalytic oxidation systems, the active source regulated by intrinsic electron and spin structures is still obscure. Here, a series of nonmetallic element-doped (such as P, S, and Se) covalent triazine frameworks (CTFs) were constructed and applied to remove organic pollutants using the adsorption-photocatalysis process. The external mass transfer model (EMTM) and the homogeneous surface diffusion model (HSDM) were employed to describe the adsorption process. It was found that sulfur-doped CTF (S-CTF-1) showed a 25.6-fold increase in saturated adsorption capacity (554.7 μmol/g) and a 169.0-fold surge in photocatalytic kinetics (5.07 h^-1), respectively, compared with the pristine CTF-1. A positive correlation between electron accumulation at the active site (N1 atom) and adsorption energy was further demonstrated with experimental results and theoretical calculations. Meanwhile, the photocatalytic degradation rates were greatly enhanced by forming a built-in electric field driven by spin polarization. In addition, S-CTF-1 still maintained a 98.3% removal of 2,2',4,4'-tetrahydroxybenzophenone (BP-2) micropollutants and 97.6% regeneration after six-cycle sequencing batch treatment in real water matrices. This work established a relation between electron and spin structures for adsorption and photocatalysis, paving a new way to design modified carbon nanomaterials to control OMPs.

A novel method for micropollutant quantification using deep learning and multi-objective optimization

Micropollutants (MPs) released into aquatic ecosystems have adverse effects on public health. Hence, monitoring and managing MPs in aquatic systems are imperative. MPs can be quantified by high-resolution mass spectrometry (HRMS) with stable isotope-labeled (SIL) standards. However, high cost of SIL solutions is a significant issue. This study aims to develop a rapid and cost-effective analytical approach to estimate MP concentrations in aquatic systems based on deep learning (DL) and multi-objective optimization. We hypothesized that internal standards could quantify the MP concentrations other than the target substance. Our approach considered the precision of intra-/inter-day repeatability and natural organic matter information to reduce instrumental error and matrix effect. We selected standard solutions to estimate the concentrations of 18 MPs. Among the optimal DL models, DarkNet-53 using nine standard solutions yielded the highest performance, while ResNet-50 yielded the lowest. Overall, this study demonstrated the capability of DL models for estimating MP concentrations.

Predicting cytotoxicity of binary pollutants towards a human cell panel in environmental water by experimentation and deep learning methods

Biological assays are useful in water quality evaluation by providing the overall toxicity of chemical mixtures in environmental waters. However, it is impossible to elucidate the source of toxicity and some lethal combination of pollutants simply using biological assays. As facile and cost-effective methods, computation model-based toxicity assessments are complementary technologies. Herein, we predicted the human health risk of binary pollutant mixtures (i.e., binary combinations of As(III), Cd(II), Cr(VI), Pb(II) and F(I)) in water using in vitro biological assays and deep learning methods. By employing a human cell panel containing human stomach, colon, liver, and kidney cell lines, we assessed the human health risk mimicking cellular responses after oral exposures of environmental water containing pollutants. Based on the experimental cytotoxicity data in pure water, multi-task deep learning was applied to predict cellular response of binary pollutant mixtures in environmental water. Using additive descriptors and single pollutant toxicity data in pure water, the established deep learning model could predict the toxicity of most binary mixtures in environmental water, with coefficient of determination (R²) > 0.65 and root mean squared error (RMSE) < 0.22. Further combining the experimental data on synergistic and antagonistic effects of pollutant mixtures, deep learning helped improve the predictive ability of the model (R² > 0.74 and RMSE <0.17). Moreover, predictive models allowed us identify a number of toxicity source-related physiochemical properties. This study illustrates the combination of experimental findings and deep learning methods in the water quality evaluation.

Machine learning in natural and engineered water systems

Water resources of desired quality and quantity are the foundation for human survival and sustainable development. To better protect the water environment and conserve water resources, efficient water management, purification, and transportation are of critical importance. In recent years, machine learning (ML) has exhibited its practicability, reliability, and high efficiency in numerous applications; furthermore, it has solved conventional and emerging problems in both natural and engineered water systems. For example, ML can predict various water quality indicators in situ and real-time by considering the complex interactions among water-related variables. ML approaches can also solve emerging pollution problems with proven rules or universal mechanisms summarized from the related research. Moreover, by applying image recognition technology to analyze the relationships between image information and physicochemical properties of the research object, ML can effectively identify and characterize specific contaminants. In view of the bright prospects of ML, this review comprehensively summarizes the development of ML applications in natural and engineered water systems. First, the concept and modeling steps of ML are briefly introduced, including data preparation, algorithm selection and model evaluation. In addition, comprehensive applications of ML in recent studies, including predicting water quality, mapping groundwater contaminants, classifying water resources, tracing contaminant sources, and evaluating pollutant toxicity in natural water systems, as well as modeling treatment techniques, assisting characterization analysis, purifying and distributing drinking water, and collecting and treating sewage water in engineered water systems, are summarized. Finally, the advantages and disadvantages of commonly used algorithms are analyzed according to their structures and mechanisms, and recommendations on the selection of ML algorithms for different studies, as well as prospects on the application and development of ML in water science are proposed. This review provides references for solving a wider range of water-related problems and brings further insights into the intelligent development of water science.

Prediction of Oxidant Exposures and Micropollutant Abatement during Ozonation Using a Machine Learning Method

Oxidation of micropollutants (MPs) by ozonation proceeds via the reactions with molecular ozone (O₃) and hydroxyl radicals (^•OH). To predict MP abatement during ozonation, a model that can accurately predict oxidant exposures (i.e., ∫0t[O3]dt⁢  and⁢  ∫0t[O•H]dt) needs to be developed. This study demonstrates machine learning models based on the random forest (RF) algorithm to output oxidant exposures from water quality parameters (input variables) that include pH, alkalinity, dissolved organic carbon concentration, and fluorescence excitation-emission matrix (FEEM) data (to characterize organic matter). To develop the models, 60 different samples of natural waters and wastewater effluents were collected and characterized, and the oxidant exposures in each sample were determined at a specific O₃ dose (2.5 mg/L). Four RF models were developed depending on how FEEM data were utilized (i.e., one model free of FEEM data, and three other models that used FEEM data of different resolutions). The regression performance and Akaike information criterion (AIC) were evaluated for each model. The models using high-resolution FEEM data generally exhibited high prediction accuracy with reasonable AIC values, implying that organic matter characteristics quantified by FEEM can be important factors to improve the accuracy of the prediction model. The developed models can be applied to predict the abatement of MPs in drinking water and wastewater ozonation processes and to optimize the O₃ dose for the intended removal of target MPs. The machine learning models using higher-resolution FEEM data offer more accurate prediction by better calculating the complex nonlinear relationship between organic characteristics and oxidant exposures.