Machine Learning-Based Early Warning Level Prediction for Cyanobacterial Blooms Using Environmental Variable Selection and Data Resampling.

Many countries have attempted to mitigate and manage issues related to harmful algal blooms (HABs) by monitoring and predicting their occurrence. The infrequency and duration of HABs occurrence pose the challenge of data imbalance when constructing machine learning models for their prediction. Furthermore, the appropriate selection of input variables is a significant issue because of the complexities between the input and output variables. Therefore, the objective of this study was to improve the predictive performance of HABs using feature selection and data resampling. Data resampling was used to address the imbalance in the minority class data. Two machine learning models were constructed to predict algal alert levels using 10 years of meteorological, hydrodynamic, and water quality data. The improvement in model accuracy due to changes in resampling methods was more noticeable than the improvement in model accuracy due to changes in feature selection methods. Models constructed using combinations of original and synthetic data across all resampling methods demonstrated higher prediction performance for the caution level (L-1) and warning level (L-2) than models constructed using the original data. In particular, the optimal artificial neural network and random forest models constructed using combinations of original and synthetic data showed significantly improved prediction accuracy for L-1 and L-2, representing the transition from normal to bloom formation states in the training and testing steps. The test results of the optimal RF model using the original data indicated prediction accuracies of 98.8% for L0, 50.0% for L1, and 50.0% for L2. In contrast, the optimal random forest model using the Synthetic Minority Oversampling Technique-Edited Nearest Neighbor (ENN) sampling method achieved accuracies of 85.0% for L0, 85.7% for L1, and 100% for L2. Therefore, applying synthetic data can address the imbalance in the observed data and improve the detection performance of machine learning models. Reliable predictions using improved models can support the design of management practices to mitigate HABs in reservoirs and ultimately ensure safe and clean water resources.

Synthesis and applications of bismuth-impregnated biochars originated from spent coffee grounds for efficient adsorption of radioactive iodine: A mechanism study.

The adsorption of radioactive iodine, which is capable of presenting high mobility in aquatic ecosystems and generating undesirable health effects in humans (e.g., thyroid gland dysfunction), was comprehensively examined using pristine spent coffee ground biochar (SCGB) and bismuth-impregnated spent coffee ground biochar (Bi@SCGB) to provide valuable insights into the variations in the adsorption capacity and mechanisms after pretreatment with Bi(NO3)3. The greater adsorption of radioactive iodine toward Bi@SCGB (adsorption capacity (Qe) = 253.71 µg/g) compared to that for SCGB (Qe = 23.32 µg/g) and its reduced adsorption capability at higher pH values provide evidence that the adsorption of radioactive iodine with SCGB and Bi@SCGB is strongly influenced by the presence of bismuth materials and the electrostatic repulsion between their negatively charged surfaces and negatively charged radioactive iodine (IO3-). The calculated R2 values for the adsorption kinetics and isotherms support that chemisorption plays a crucial role in the adsorption of radioactive iodine by SCGB and Bi@SCGB in aqueous phases. The adsorption of radioactive iodine onto SCGB was linearly correlated with the contact time (h1/2), and the diffusion of intra-particle predominantly determined the adsorption rate of radioactive iodine onto Bi@SCGB (Cstage II (129.20) > Cstage I (42.33)). Thermodynamic studies revealed that the adsorption of radioactive iodine toward SCGB (ΔG° = -8.47 to -7.83 kJ/mol; ΔH° = -13.93 kJ/mol) occurred exothermically and that for Bi@SCGB (ΔG° = -15.90 to -13.89 kJ/mol; ΔH° = 5.88 kJ/mol) proceeded endothermically and spontaneously. The X-ray photoelectron spectroscopy (XPS) analysis of SCGB and Bi@SCGB before and after the adsorption of radioactive iodine suggest the conclusion that the change in the primary adsorption mechanism from electrostatic attraction to surface precipitation upon the impregnation of bismuth materials on the surfaces of spent coffee ground biochars is beneficial for the adsorption of radioactive iodine in aqueous phases.

Downstream Transport of Geosmin Based on Harmful Cyanobacterial Outbreak Upstream in a Reservoir Cascade.

Understanding water quality events in a multiple-impoundment series is important but seldom presented comprehensively. Therefore, this study was conducted to systematically understand the explosion event of geosmin (GSM) in the North Han River (Chuncheon, Soyang, Euiam, and Cheongpyeong Reservoirs) and Han River (Paldang Reservoir), which consists of a cascade reservoir series, the largest drinking water source system in South Korea. We investigated the spatiotemporal relationship of harmful cyanobacterial blooms in the upstream reservoir (Euiam) with the water quality incident event caused by the GSM in the downstream reservoir (Paldang) from January to December 2011. The harmful cyanobacterial bloom occurred during August−September under a high water temperature (>20 °C) after a heavy-rainfall-based flood runoff event. The high chlorophyll-a (Chl-a) concentration in the upper Euiam Reservoir was prolonged for two months with a maximum concentration of 1150.5 mg m−3, in which the filamentous Dolichospermum circinale Kütz dominated the algal community at a rate of >99%. These parameters remarkably decreased (17.3 mg Chl-a m−3) in October 2011 when the water temperature decreased (5 °C) and soluble reactive phosphorus was depleted. However, high and unprecedented GSM concentrations, with a maximum value of 1640 ng L−1, were detected in the downstream reservoirs (Cheongpyeong and Paldang); the level was 11 times higher than the value (10 ng L−1) recommended by the World Health Organization. The concentrations of GSM gradually decreased and had an adverse effect on the drinking water quality until the end of December 2011. Our study indicated that the time lag between the summer−fall cyanobacterial outbreak in the upstream reservoir and winter GSM explosion events in the downstream reservoirs could be attributed to the transport and release of GSM through the effluent from hydroelectric power generation in this multiple-reservoir system. Therefore, we suggest that a structural understanding of the reservoir cascade be considered during water quality management of drinking water sources to avoid such incidents in the future.

Improving the performance of machine learning models for early warning of harmful algal blooms using an adaptive synthetic sampling method.

Many countries have attempted to monitor and predict harmful algal blooms to mitigate related problems and establish management practices. The current alert system-based sampling of cell density is used to intimate the bloom status and to inform rapid and adequate response from water-associated organizations. The objective of this study was to develop an early warning system for cyanobacterial blooms to allow for efficient decision making prior to the occurrence of algal blooms and to guide preemptive actions regarding management practices. In this study, two machine learning models: artificial neural network (ANN) and support vector machine (SVM), were constructed for the timely prediction of alert levels of algal bloom using eight years' worth of meteorological, hydrodynamic, and water quality data in a reservoir where harmful cyanobacterial blooms frequently occur during summer. However, the proportion imbalance on all alert level data as the output variable leads to biased training of the data-driven model and degradation of model prediction performance. Therefore, the synthetic data generated by an adaptive synthetic (ADASYN) sampling method were used to resolve the imbalance of minority class data in the original data and to improve the prediction performance of the models. The results showed that the overall prediction performance yielded by the caution level (L1) and warning level (L2) in the models constructed using a combination of original and synthetic data was higher than the models constructed using original data only. In particular, the optimal ANN and SVM constructed using a combination of original and synthetic data during both training (including validation) and test generated distinctively improved recall and precision values of L1, which is a very critical alert level as it indicates a transition status from normalcy to bloom formation. In addition, both optimal models constructed using synthetic-added data exhibited improvement in recall and precision by more than 33.7% while predicting L-1 and L-2 during the test. Therefore, the application of synthetic data can improve detection performance of machine learning models by solving the imbalance of observed data. Reliable prediction by the improved models can be used to aid the design of management practices to mitigate algal blooms within a reservoir.

A machine learning approach for early warning of cyanobacterial bloom outbreaks in a freshwater reservoir.

Understanding the dynamics of harmful algal blooms is important to protect the aquatic ecosystem in regulated rivers and secure human health. In this study, artificial neural network (ANN) and support vector machine (SVM) models were used to predict algae alert levels for the early warning of blooms in a freshwater reservoir. Intensive water-quality, hydrodynamic, and meteorological data were used to train and validate both ANN and SVM models. The Latin-hypercube one-factor-at-a-time (LH-OAT) method and a pattern search algorithm were applied to perform sensitivity analyses for the input variables and to optimize the parameters of the models, respectively. The results indicated that the two models well reproduced the algae alert level based on the time-lag input and output data. In particular, the ANN model showed a better performance than the SVM model, displaying a higher performance value in both training and validation steps. Furthermore, a sampling frequency of 6- and 7-day were determined as efficient early-warning intervals for the freshwater reservoir. Therefore, this study presents an effective early-warning prediction method for algae alert level, which can improve the eutrophication management schemes for freshwater reservoirs.

Estimating the biodegradability of treated sewage samples using synchronous fluorescence spectra.

Synchronous fluorescence spectra (SFS) and the first derivative spectra of the influent versus the effluent wastewater samples were compared and the use of fluorescence indices is suggested as a means to estimate the biodegradability of the effluent wastewater. Three distinct peaks were identified from the SFS of the effluent wastewater samples. Protein-like fluorescence (PLF) was reduced, whereas fulvic and/or humic-like fluorescence (HLF) were enhanced, suggesting that the two fluorescence characteristics may represent biodegradable and refractory components, respectively. Five fluorescence indices were selected for the biodegradability estimation based on the spectral features changing from the influent to the effluent. Among the selected indices, the relative distribution of PLF to the total fluorescence area of SFS (Index II) exhibited the highest correlation coefficient with total organic carbon (TOC)-based biodegradability, which was even higher than those obtained with the traditional oxygen demand-based parameters. A multiple regression analysis using Index II and the area ratio of PLF to HLF (Index III) demonstrated the enhancement of the correlations from 0.558 to 0.711 for TOC-based biodegradability. The multiple regression equation finally obtained was 0.148 × Index II - 4.964 × Index III - 0.001 and 0.046 × Index II - 1.128 × Index III + 0.026. The fluorescence indices proposed here are expected to be utilized for successful development of real-time monitoring using a simple fluorescence sensing device for the biodegradability of treated sewage.

Multivariate analysis for spatial distribution of dissolved organic matters in a large river-type dam reservoir.

In contrast to extensive studies of dissolved organic matters (DOM) in natural lakes, the distributions and the characteristics of DOM in artificial dam reservoirs have not been well documented despite a growing demand for the construction worldwide. For this study, spatial variations in the concentrations and the characteristics of DOM in Lake Paldang, a large river-type dam reservoir, were investigated using the concentrations, the specific UV absorbance (SUVA), the synchronous fluorescence spectra and the molecular weight (MW(w)) values. In addition, environmental factors determining the DOM spatial distribution were examined based on a principal component analysis (PCA). Variations in the DOM characteristics were greater than those for the concentrations (1.1-2.4 mg C/L). In contrast to typical lakes, vertical variations with a depth were much smaller than those observed among horizontal sampling sites within the reservoir. Irrespective of the depth, four individual sampling locations were easily distinguished by comparison of some selected DOM characteristics. The protein-like fluorescence (PLF), MW(w) and SUVA values observed at the location near the dam exceeded the corresponding values for the sampling locations near major influent rivers, suggesting that, even for the river-type dam reservoir, the downstream DOM characteristics may be governed by in-lake DOM production processes such as the release from sediments and algal activities. The results of principal component analysis (PCA) revealed that approximately 61% of the variance in DOM distribution might be explained by allochthonous/autochthonous carbon sources and predominant presence of either total nitrogen or total phosphorous over the other.

Spectroscopic distribution of dissolved organic matter in a dam reservoir impacted by turbid storm runoff.

Spectroscopic characteristics of dissolved organic matter (DOM) in a large dam reservoir were determined using ultraviolet absorbance and fluorescence spectroscopy to investigate spatial distribution of DOM composition after turbid storm runoff. Water samples were collected along a longitudinal axis of the reservoir at three to four depths after a severe storm runoff. Vertical profiles of turbidity data showed that a turbid water layer was located at a middle depth of the entire reservoir. The spectroscopic characteristics of DOM samples in the turbid water layer were similar to those of terrestrial DOM, as demonstrated by the higher specific UV absorbance (SUVA) and the lower fluorescence emission intensity ratio (F(450)/F(500)) compared to other surrounding DOM samples in the reservoir. Synchronous fluorescence spectroscopy revealed that higher content of humic-like DOM composition was contained in the turbid water. Fluorescence excitation-emission matrix (EEM) showed that lower content of protein-like aromatic amino acids was present in the turbid water DOM. The highest protein-like fluorescence was typically observed at a bottom layer of each sampling location. The bottom water DOM exhibited extremely high protein-like florescence near the dam site. The particular observation was attributed to the low water temperature and the isolation of the local bottom water due to the upper location of the withdrawal outlet near the dam. Our results suggest that the distribution of DOM composition in a dam reservoir is strongly influenced by the outflow operation, such as selective withdrawal, as well as terrestrial-origin DOM inputs from storm runoff.