The increase in harmful algal blooms (HABs) globally has been linked to climate change and anthropogenic activities such as agricultural runoff and urbanization. This study focused on analyzing the impact of these factors on HAB occurrences in the East China Sea and the Yellow Sea, identifying influential factors, and predicting future HAB events. For this study, random forest and numerical modeling were employed, with datasets encompassing physical and chemical properties of river water, seawater, and precipitation to assess the impact of discharge on HABs. Additionally, climate change scenarios derived from the fifth phase of the Coupled Model Intercomparison Project (CMIP5) were employed to predict future HAB occurrences, supported by a sensitivity analysis to identify influential factors affecting HAB occurrence. This study demonstrated that the growth rate and occurrence of HABs in the East China Sea (ECS) and Korean coastal waters (KCW) distinctively increased in July and November after the operation of the Three Gorges Dam (TGD). It is likely affected by the decreasing discharge from the Yangtze River (YR) owing to the operation of the TGD. Using the Random Forest model, future HAB events were predicted in good agreement with observations. The sensitivity results revealed that environmental properties, such as precipitation, water temperature, and salinity are major features affecting the HAB trends in both the KCW and YR basins. Moreover, based on the random forest model and climate change scenarios, HAB events were predicted to increase in frequency in July, September, and October. Therefore, the findings can contribute to preventing biological pollution of the ocean system in the ECS and KCW by supporting efficient environmental management.

The Mananga River in the Philippines has been classified as a Class A river in 1997. Since then, the river has significantly deteriorated due to pollution, especially in the downstream area around Talisay in the Cebu region. In this study, the water quality parameters – namely, dissolved oxygen (DO), biological oxygen demand (BOD), and total suspended solids (TSS) were assessed using principal component analysis (PCA). Water quality data were obtained from the Department of Environmental and Natural Resources–Environmental Management Bureau Region 7. The water samples were collected on a quarterly basis from 2016–2019. The effect of weather conditions on water quality and significant relationships between the water quality parameters were determined. In addition, the comparison of water quality in each sampling station was investigated. Results show that about 84.3% of the total variance in water quality can be attributed to two significant principal component scores. Evident correlations were observed such as DO and BOD are negatively correlated, whereas a positive correlation exists between DO to RH and wind speed, BOD to temperature, and TSS to wind speed. Furthermore, negative correlations are observed between DO to temperature and wind direction, BOD to rainfall, and TSS to wind direction and rainfall. In the overall analysis of the results, the heavier the influence of a variable, the more likely it is to contribute to affecting the water quality. Therefore, this can alter the overall distribution within the plots and the correlations among the variables. The findings are a predictive measure of future changes and trends in the Mananga River. Therefore, the results of the present work will be used in environmental monitoring, environmental management, and assistance for the rehabilitation of the Mananga River using PCA.

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.

The complexity associated with water quality models (WQMs) has increased owing to the introduction of numerous physical and biological mechanisms in the models. Sensitivity analysis (SA) is conducted to identify influential parameters in these mechanisms. However, enormous computational power and time are required to obtain numerical solutions from thousands of model simulations. Therefore, a cloud-based toolbox is developed for performing SA of WQMs by implementing a cloud computing system using grab sampling data and hyperspectral images (HSI) of waterbodies. Cloud computing can provide high-performance computation by adjusting the scale of the computational power according to user preference. The developed toolbox with the cloud system can reduce the computation time for SA by approximately 20 times compared to that of a desktop computer.

In recent years, as agricultural activities and types of crops have become diverse, the occurrence of micro-pollutants has been reported more frequently in rural areas. These pollutants have detrimental effects on human health and ecological systems; thus, it is important to manage and monitor their presence in the environment. The modeling approach could be an effective way to understand and manage these pollutants. This study predicts the concentrations of micro-pollutants (MPs) using deep learning (DL) models, and the results are then compared with simulation results obtained from the soil water assessment tool (SWAT) model. The SWAT model showed an unacceptable performance owing to the resulting negative Nash–Sutcliffe efficiency (NSE) values for the simulations. This may be caused by the limitations of SWAT, which pertains to adopting simplified equations to simulate micro-pollutants. In addition, the ambiguous plan of pesticide application increased the model uncertainty, thereby deteriorating the model result. Here, we developed two different DL models: long short-term memory (LSTM) and convolutional neural network (CNN). LSTM exhibited the highest model performance, with NSE values of 0.99 and 0.75 for the training and validation steps, respectively. In the multi-target MP model, the error decreased as the number of simulated pollutants increased. The simulation of the four pollutants had the highest error, while the six-target simulation had the lowest error. In conclusion, this study demonstrated that the LSTM model has the potential to improve the prediction of MPs in aquatic systems.

The seawater battery (SWB) is a promising desalination technology that utilizes abundant sodium ions as an energy storage medium. Recently, the alternative desalination system, seawater battery desalination (SWB-D), was developed by placing an SWB next to the desalination compartment. This SWB-D system can desalt water while charging the SWB next to it. However, only a fixed catholyte solution has been investigated, although the catholytes impact the overall SWB-D performance. Therefore, we evaluated the effect of different catholytes on the desalination performance. High-saline reverse osmosis (RO) concentrate or brackish water exhibited excellent salt removal capability (>85.3% of sodium and >76.6% of chloride ions) with relatively short operation times (36.4 h for RO concentrate and 39.5 h for brackish water) upon charging, whereas the relatively low-saline river water showed the longest operation time (81.0 h), implying that river water should be excluded as a potential catholyte. The amount of desalinated water was marginally reduced due to osmosis through the anion exchange membrane; however, the amount of treated salt was >82.9% even after the reduction in water volume. These findings suggest that the catholyte with a resistance of >0.041 kΩ·cm can be ideal for the SWB-D.

A marine outfall can be a wastewater management system that discharges sewage and stormwater into the sea; hence, it is a source of microbial pollution on recreational beaches, including antibiotic resistant genes (ARGs), which lead to an increase in untreatable diseases. In this regard, a marine outfall must be efficiently located to mitigate these risks. This study aimed to 1) investigate the spatiotemporal variability of Escherichia coli (E. coli) and ARGs on a recreational beach and 2) design marine outfalls to reduce microbial risks. For this purpose, E. coli and ARGs with influential environmental variables were intensively monitored on Gwangalli beach, South Korea in this study. Environmental fluid dynamic code (EFDC) was used and calibrated using the monitoring data, and 12 outfall extension scenarios were explored (6 locations at 2 depths). The results revealed that repositioning the marine outfall can significantly reduce the concentrations of E. coli and ARGs on the beach by 46–99%. Offshore extended outfalls at the bottom of the sea reduced concentrations of E. coli and ARGs on the beach more effectively than onshore outfalls at the sea surface. These findings could be helpful in establishing microbial pollution management plans at recreational beaches in the future.

Data assimilation (DA) techniques are powerful means of dynamic natural system modeling that allow for the use of data as soon as it appears to improve model predictions and reduce prediction uncertainty by correcting state variables, model parameters, and boundary and initial conditions. The objectives of this review are to explore existing approaches and advances in DA applications for surface water quality modeling and to identify future research prospects. We first reviewed the DA methods used in water quality modeling as reported in literature. We then addressed observations and suggestions regarding various factors of DA performance, such as the mismatch between both lateral and vertical spatial detail of measurements and modeling, subgrid heterogeneity, presence of temporally stable spatial patterns in water quality parameters and related biases, evaluation of uncertainty in data and modeling results, mismatch between scales and schedules of data from multiple sources, selection of parameters to be updated along with state variables, update frequency and forecast skill. The review concludes with the outlook section that outlines current challenges and opportunities related to growing role of novel data sources, scale mismatch between model discretization and observation, structural uncertainty of models and conversion of measured to simulated vales, experimentation with DA prior to applications, using DA performance or model selection, the role of sensitivity analysis, and the expanding use of DA in water quality management.

Recent intensification in climate change have resulted in the rise of hydrological extreme events. This demands modeling of hydrological processes at high temporal resolution to better understand flow patterns in catchments. To model surface and sub-surface flows in a catchment we utilized a physically based model called Hydrological Simulated Program-FORTRAN and two deep learning-based models. One deep learning model consisted of only one long short-term memory (simple LSTM), whereas the other model simulated processes in each hydrological response unit (HRU) by defining one separate LSTM for each HRU (HRU-based LSTM). The models use environmental time-series data and two-dimensional spatial data to predict surface and sub-surface flows at 6-minute time step simultaneously. We tested our models in a tropical humid headwater catchment in northern Lao PDR and compared their performances. Our results showed that the simple LSTM model outperformed the other models on surface runoff prediction with the lowest MSE (7.4e−5 m3 s−1), whereas HRU-based LSTM model better predicted patterns and slopes in sub-surface flow in comparison with the other models by having the smallest MSE value (3.2e−4 m3 s−1). This study demonstrated the performance of a deep learning model when simulating hydrological cycle with high temporal resolution.