Water Quality Monitoring Stations Research Articles

Dissolved Oxygen Concentration Prediction in the Pearl River Estuary with Deep Learning for Driving Factors Identification: Temperature, pH, Conductivity, and Ammonia Nitrogen

Predicting the dissolved oxygen concentration and identifying its driving factors are essential for improved prevention and management of anoxia in estuaries. However, complex hydrodynamic conditions and the limitations in traditional methods result in challenges in the identification of the driving factors for the low dissolved oxygen (DO) phenomenon. The objective of our study is to develop a robust deep learning model using four-year in situ data collected from an automatic water quality monitoring station (AWQMS) in an estuary, for accurate identification and quantification of the driving factors influencing DO levels. Mitigations in hypoxia were observed during the initial two years, but a subsequent decline in DO concentrations was witnessed recently. The periodicity of DO concentrations in the Pearl River Estuary reduced with the increase in the hypoxic intensity. Maximal information coefficient (MIC) and extreme gradient boosting (XGBoost) were employed to determine the significance of input variables, which were subsequently validated by using the long- and short-term memory networks (LSTMs). The driving factors contributing to the hypoxia problem were shown as temperature, pH, conductivity, and NH4+-N concentrations. Notably, the evaluation index values of the hybrid model are MAPE = 0.0887 and R2 = 0.9208, which have been improved compared with the LSTM model by about 99.34% in MAPE reduction and 16.56% in R2 improvement, indicating that the MixUp-LSTM model was capable of effectively capturing nonlinear relationships between DO and other water quality indicators. Based on existing literature, three traditional statistical methods and four machine learning models were also performed to compare with the proposed MixUp-LSTM model, which outperformed other models in terms of prediction accuracy and robustness. Overall, the successful identification of the driving factors for the deoxygenation phenomenon would have important implications for the governance and regulation of low DO in estuaries.

Evaluating the suitability of large-scale datasets to estimate nitrogen loads and yields across different spatial scales

Decision makers are often confronted with inadequate information to predict nutrient loads and yields in freshwater ecosystems at large spatial scales. We evaluate the potential of using data mapped at large spatial scales (regional to global) and often coarse resolution to predict nitrogen yields at varying smaller scales (e.g., at the catchment and stream reach level). We applied the SPAtially Referenced Regression On Watershed attributes (SPARROW) model in three regions: the Upper Midwest part of the United States, New Zealand, and the Grande River Basin in southeastern Brazil. For each region, we compared predictions of nitrogen delivery between models developed using novel large-scale datasets and those developed using local-scale datasets. Large-scale models tended to underperform the local-scale models in poorly monitored areas. Despite this, large-scale models are well suited to generate hypotheses about relative effects of different nutrient source categories (point and urban, agricultural, native vegetation) and to identify knowledge gaps across spatial scales when data are scarce. Regardless of the spatial resolution of the predictors used in the models, a representative network of water quality monitoring stations is key to improve the performance of large-scale models used to estimate loads and yields. We discuss avenues of research to understand how this large-scale modelling approach can improve decision making for managing catchments at local scales, particularly in data poor regions.

The Temporal and Spatial Analysis and Assessment of Water Quality in the Tuo River (Suzhou Section) based on the CCME-WQI Method

To investigate the spatiotemporal variations and disparities in water quality within the Suzhou section of the Tuo River, as well as to identify the key pollution factors and influencing elements, this study utilized monitoring data from five water quality monitoring stations along the Tuo River in Suzhou, collected between 2020 and 2022. After analyzing the spatiotemporal variations in water quality indicators, the Canadian Council of Ministers of the Environment Water Quality Index (CCME-WQI) was employed to assess the river's water quality status. The study also examined the causes of water pollution and proposed appropriate pollution control measures. The research results indicate that:(1) Total Nitrogen (TN), Chemical Oxygen Demand (COD), Permanganate Index (CODMn), and Fluoride (F) were identified as the main pollution factors, with COD and TN being the primary pollutants exceeding standard limits, with exceedance rates of 55.66% and 37.14%, respectively;(2)From 2020 to 2022, the concentrations of CODMn, COD, and TN were significantly higher during the flood season compared to the non-flood season, suggesting that non-point source pollution is the major contributor to the pollution load;(3)Temporally, the study area exhibited significant seasonal variations in water quality, with water quality being better during the non-flood season than during the flood season. Spatially, the study area showed notable spatial differences in water quality, with the S1 monitoring station recording the poorest water quality, while the S3 station exhibited the best;(4)The water quality of the Tuo River was significantly influenced by precipitation, with increased rainfall during the flood season exacerbating urban runoff, agricultural non-point source pollution, and internal pollution within the river channels.

Quantifying seasonal variations in pollution sources with machine learning-enhanced positive matrix factorization

As the pace of industrialization and urbanization accelerates, water quality management faces increasing challenges, with traditional methods for pollutant source apportionment often proving inadequate in handling complex environmental data. This study enhances the precision and reliability of pollutant source identification by integrating Positive Matrix Factorization (PMF) models with diverse machine learning techniques. Utilizing data from 17 water quality monitoring stations along the Wuding River from 2017 to 2021, we employed Random Forest (RF), Support Vector Machine (SVM), Elastic Net (EN), and Extreme Gradient Boosting (XGBoost) models to predict the Water Quality Index (WQI) during dry and wet seasons. Results indicate that the RF model exhibited optimal performance in the dry season (R2 = 0.93), while the SVM was superior in the wet season (R2 = 0.94). SHAP (SHapley Additive exPlanations) value analysis identified CODMn and NH3-N as significant influencers on WQI in the dry season, whereas COD, BOD, and TP gained prominence during the wet season. SHAP values reveal the contribution of each feature to the model output, thereby enhancing the model’s transparency and interpretability. Additionally, feature importance identified by machine learning was utilized as weights to optimize the contribution rates predicted by the PMF model. The optimised model was able to identify the contribution of domestic and farm effluent discharges more accurately in the dry season, with a significant increase in the percentage of identification from 19.4 % to 45.4 %, and an increase in the percentage of contribution from agricultural non-point sources and domestic effluent in the rainy season. This research offers a novel perspective on the characteristics of river water pollution and holds significant implications for formulating data-driven environmental management strategies.

Environmental implication identification on water quality variation and human health risk assessment in the middle and lower reaches of the Hanjiang River, China

As the longest tributary of the Yangtze River, the Hanjiang River has been a “lifeline” for the “Yangtze River Economic Belt of China”. The water quality of the middle and lower reaches of the Hanjiang River (MLHR) has received considerable attention due to environmental degradation, increasing population, and water diversion project operations during the past decade. In this study, 16 water quality parameters from 14 national water quality monitoring stations along the MLHR were collected monthly from January 2021 to December 2022. Water quality index (WQI) methods coupling multivariate statistical techniques, e.g., seasonal Mann-Kendall test (SKT) and Mantel’s test, were applied to analyze and evaluate the spatiotemporal variations of water quality under different environmental and hydrological conditions. The Hazard Quotients (HQs) were used to assess the non-carcinogenic and carcinogenic risk for adults and children via ingestion and dermal pathways. The results showed that the overall water quality in the MLHR was classified as “excellent”, with the average WQI higher than “91”, and each single water quality indicator met Class I or Class II surface water quality standards of China. Ammonia (NH3-N), Total phosphorus (TP), Permanganate (CODMn), pH, and Arsenic (As) were significantly correlated with water level variations. CODMn and TP were decisive in explaining the WQI variations of the MLHR. NH3-N, water temperature, pH, and dissolved oxygen were key parameters for explaining WQI variations in most regions of the MLHR. As, Cadmium (Cd), and Chromium (Cr) presented potential health risks in the MLHR and should warrant close monitoring due to their significant health implications. Due to the continuous construction and operation of cascade hydro-projects, the synchronous monitoring of water quality and hydrological data of the MLHR should be conducted in the future. This study provides a scientific reference for managing water quality in the MLHR and similar natural riverine water bodies.

Spatio-Temporal Dynamics Coupling between Land Use/Cover Change and Water Quality in Dongjiang Lake Watershed Using Satellite Remote Sensing

With rapid social and economic development, land use/land cover change (LUCC) has intensified with serious impacts on water quality in the watershed. In this study, we took Dongjiang Lake watershed as the study area and obtained measured data on water quality parameters from the watershed’s water quality monitoring stations. Based on Landsat-5, Landsat-8, or Sentinel-2 remote sensing data for multiple periods per year between 1992 and 2022, the sensitive satellite bands or band combinations of each water quality parameter were determined. The Random Forest method was used to classify the land use types in the watershed into six categories, and the area proportion of each type was calculated. We established machine learning regression models and polynomial regression models with WQI as the dependent variable and the area proportion of each land use type as the independent variable. Accuracy test results showed that, among them, the quadratic cubic polynomial regression model with grassland, forest land, construction land, and unused land as its independent variables was the best model for coupling watershed water quality with LUCC. This study’s results provide a scientific basis for monitoring spatial and temporal changes in water quality caused by LUCC in the Dongjiang Lake watershed.

Understanding watershed sources of pollution in Vinh Long Province, Vietnamese Mekong Delta

We explore point source (PS) and non-point source (NPS) pollution impacts on surface water quality in Vinh Long Province, Vietnam. We used data from 60 surface water quality monitoring stations across dry, transitional, and wet seasons from 2017 to 2021, in addition to sampling data collected from 12 wastewater outlets from June to October 2021 to determine the sources of pollution loads. Surface water quality was assessed using both the water quality index (WQI) and the Vietnamese standards QCVN 08:2015, whilst cluster analysis (CA) and principal components analysis (PCA) were used to evaluate spatial variation and key influencing factors. We observed seasonal variation in surface water quality, with a decline in quality during the rainy season. Moreover, the water quality parameters such as chemical oxygen demand (COD), biochemical oxygen demand (BOD5), total nitrogen (TN), and total phosphorus (TP) frequently exceeded the QCVN 08:2015 standard. Across, the different monitoring fixed points and seasons, COD levels were found to range from 8.94 to 15.14 mg/L, while TN levels varied between 0.24 and 0.53 mg/L. The cluster analysis categorized the monitoring fixed points into three groups, based on their water quality parameters, while PCA identified four principal components that explain 69% of the variance, distinguishing between pollution sources and seasonal factors. Our findings emphasize that poor water quality in many areas is affected by non-point source pollution, underscoring the need for watershed and land management. The results and applied methodologies provide insights for watershed management, policy development, and adaptation, applicable to regions facing similar environmental challenges.

Developing an ensembled machine learning model for predicting water quality index in Johor River Basin

Currently, the Water Quality Index (WQI) model becomes a widely used tool to evaluate surface water quality for agriculture, domestic and industrial. WQI is one of the simplest mathematical tools that can assist water operator in decision making in assessing the quality of water and it is widely used in the last years. The water quality analysis and prediction is conducted for Johor River Basin incorporating the upstream to downstream water quality monitoring station data of the river. In this research, the numerical method is first used to calculate the WQI and identify the classes for validating the prediction results. Then, two ensemble and optimized machine learning models including gradient boosting regression (GB) and random forest regression (RF) are employed to predict the WQI. The study area selected is the Johor River basin located in Johor, Peninsular Malaysia. The initial phase of this study involves analyzing all available data on parameters concerning the river, aiming to gain a comprehensive understanding of the overall water quality within the river basin. Through temporal analysis, it was determined that Mg, E. coli, SS, and DS emerge as critical factors affecting water quality in this river basin. Then, in terms of WQI calculation, feature importance method is used to identify the most important parameters that can be used to predict the WQI. Finally, an ensemble-based machine learning model is designed to predict the WQI using three parameters. Two ensemble ML approaches are chosen to predict the WQI in the study area and achieved a R2 of 0.86 for RF-based regression and 0.85 for GB-based ML technique. Finally, this research proves that using only the biochemical oxygen demand (BOD), the chemical oxygen demand (COD) and percentage of dissolved oxygen (DO%), the WQI can be predicted accurately and almost 96 times out of 100 sample, the water class can be predicted using GB ensembled ML algorithm. Moving forward, stakeholders may opt to integrate this research into their analyses, potentially yielding economic reliability and time savings.

Browning from headwaters to coastal areas in the boreal region: Trends and drivers

Browning of freshwaters, mainly caused by increased terrestrial organic carbon loading, has been widely studied during the last decades. However, there are still uncertainties regarding both the extent of browning in different aquatic ecosystems and the actual importance of different driving forces and mechanisms. To refine understanding of the extent and causes of browning and its temporal variation, we gathered a comprehensive dataset including 746 Finnish water quality monitoring stations representing various waterbody types: streams, rivers, lakes, and coastal waters. Monotonic trend analyses revealed that TOC concentrations increased in all waterbody types during the study period from 1990 to 2020, whereas non-linear trends indicated that upward trends in TOC concentrations have substantially decreased since the mid-2000s. However, despite the upward trends levelling off, non-linear analyses also indicated decreases in TOC concentrations at only a few stations. As a result, the TOC contents of the majority of Finnish waterbody types in 2020 were at a higher level than in 1990. To examine the driving forces of increasing TOC concentrations, we selected 100 riverine catchments and linked the detected trends to 24 different drivers, including both hydrometeorological and catchment characteristics. The increased TOC concentrations in surface waters could be connected to diverse human impacts: hydrometeorological variables impacted by climate change, decreased acidic deposition, and land use in terms of peatland drainage. The importance of increased temperatures was emphasized, and its role as a driver of increased leaching of organic carbon in the forthcoming years is expected to grow with climate change.

Effects of Land Use Types on Water Quality at Different Buffer Scales: Tianjin Section of the Haihe River Basin as an Example

It is important to explore the relationship between land use types and water quality to improve the surface water environment. Based on monthly water quality monitoring data from 16 nationally controlled surface water quality monitoring stations in Tianjin and land use data in 2021, GIS spatial analysis and mathematical and statistical methods were used to study the influence of land use types on surface water quality in buffer zones at different scales. The results showed that:① the land use types in the study area were mainly construction land, farmland, and water areas, which had significant effects on river water quality. Except for water temperature (WT) and pH, the farmland, construction land, and water areas were negatively correlated with each water quality indicator; forest land and grassland were positively correlated with dissolved oxygen (DO) and total nitrogen (TN) and negatively correlated with other water quality indicators. ② The water quality indicators showed obvious spatial differences in different seasons. The pH, DO and TN concentrations were higher in the dry season, whereas the permanganate index, ammonia nitrogen (NH4+-N), and total phosphorus (TP) concentrations were higher in the rainy season. ③ The results of the RDA analysis showed that the 800 m buffer zone land use had the greatest explanatory power for water quality changes in the dry season (50.4%), whereas the 3 000 m buffer zone land use could explain the water quality changes in the rainy season to the greatest extent (49.6%); from the average explanation rate of the dry and rainy seasons, the 3 000 m buffer zone was the best impact scale (50.0%) on water quality indicators in Tianjin. ④ The partial least squares regression (PLSR) analysis showed that the most important variables affecting surface water quality changes were construction land, farmland, and water areas. The predictive ability of the PLSR model of most water quality indicators was stronger in the dry season than that in the rainy season. In the dry season, all water quality indicators, except WT and pH, were most influenced by farmland. In the rainy season, construction land had the greatest influence on WT and NH4+-N concentrations, and the most important influencing factor for the remaining water quality indicators was still farmland. This study showed that the rational planning of land use types within 3 000 m of rivers or lakes was beneficial to improving the water quality of surface water.

Evaluating the Effectiveness of the “River Chief System”: An Empirical Study Based on the Water Quality Data of Coastal Rivers in Guangdong Province

The river chief system (RCS) is an innovative reform in China for strengthening the management of rivers and lakes. It is an important means of curbing the current severe water-environment situation. However, the policy impact of the RCS is still inconclusive in the existing literature. Using monthly data spanning from January 2015 to March 2022 from 25 water quality monitoring stations in rivers flowing into the sea across 13 prefecture-level cities in Guangdong Province, this study adopted regression discontinuity to evaluate the policy effects of the RCS on water quality. The results show that after the RCS’s full implementation in Guangdong Province, the concentrations of dissolved oxygen (DO) increased and water quality indicators, such as permanganate index (CODMn), biochemical oxygen demand (BOD), ammonia nitrogen (NH3-N), chemical oxygen demand (COD), and total phosphorus (TP), decreased; NH3-N showed the largest decrease. These findings indicate that the RCS may contribute to a measurable improvement in reducing water pollution. However, no statistically significant changes in pH and total nitrogen (TN) were found, which indicates that the RCS fell short of achieving the policy effect of comprehensive water-pollution control. Therefore, in order to improve the RCS, it is necessary to refine the existing water-quality assessment indicators and to establish an evaluation system centered on the ecological health of rivers and lakes. Additionally, a paradigm shift from an administrative-boundary-based river management model to an overarching, holistic river-basin-based management approach is crucial for actualizing the holistic governance goals of the RCS.

Water quality emergency monitoring networks: A method for identifying non-critical variables based on Shannon's entropy

Abstract In the occurrence of environmental disasters involving water resources, deploying an emergency monitoring network for assessing water quality is within the first measures to be taken. Emergency networks usually cover a large set of water quality variables and monitoring stations along the watershed. Focusing on variables that represent greater risk to the environment and have less predictable spatial and temporal distribution is a strategy to optimize efforts on monitoring. The goal of this study is to assess the use of Shannon's entropy to identify non-critical water quality variables in an emergency monitoring network implemented in a watershed impacted by the collapse of a mining iron tailing dam, the Doce River watershed (Brazil). Monitoring stations were grouped into water quality subregions through cluster analysis and Shannon's entropy was used to estimate information redundancy of monitored variables. From information redundancy and after checking for compliance with environment normative, non-critical water quality variables were identified. Results indicated that non-critical variables represent 32–50% of the variables monitored. Emergency network managers find in this method a robust tool to improve the network performance. However, special attention should be paid to outliers' presence that can bias analyses based on Shannon's entropy.

Bottom water hypoxia suppresses fish chorusing in estuariesa).

Hypoxia in coastal ecosystems is increasing as a result of water quality declines from nutrient pollution. Hypoxia negatively affects fish populations and marine life, limiting their spawning habitats, population size, and growth. In this study, two approaches were used to understand the effect of hypoxia on the chorusing and reproductive behavior of fishes in estuaries. One approach used a water quality meter integrated with a prototype passive acoustic recorder, developed to monitor dissolved oxygen and fish chorusing simultaneously and continuously at sites with normoxic and hypoxic conditions. In a second approach, passive acoustic recorders were deployed near ambient water quality monitoring stations, monitored by the North Carolina agencies in estuaries where hypoxia occurs periodically. In both approaches, when hypoxia (dissolved oxygen < 4.0 mg/L) occurred, fish chorusing was diminished or ceased. A strong correlation was observed between bottom water dissolved oxygen and the power spectral density in a 100-200 Hz frequency band associated with red drum (Sciaenops ocellatus, Sciaenidae) calling. Passive acoustic monitoring stations and integrated passive acoustic and water quality meters should be used in estuarine hypoxia monitoring efforts to examine the expanding areas of hypoxia and its impact on fish critical spawning habitats.

Distance makes a difference: The role of water pollution regulation on firms’ innovation in China

Throughout the imminent water crisis era and stringent water environment management, enterprises play a vital role in environmental governance. However, there is seldom analysis on how the water environment regulation influences an enterprise's innovation due to data deficiency. Using our database from 2000 to 2022, we combine geographic information from water quality monitoring stations with the data on listed companies in China to pinpoint what distance firms are exposed to the monitoring stations. Results from a spatial regression discontinuity design along the geographic distances demonstrate that water environmental regulation has a significant positive impact on the quantity and quality of enterprise innovation. This paper found that it is primarily driven by the decreased usage of intermediate goods, rising industry profits from the exit of low-productivity firms, and enhanced enterprise research and development (R&D) investment. Furthermore, the positive impact is more pronounced for privately-owned firms, firms closer to suppliers, polluting firms, and politically unaffiliated firms. Our study offers scientifically valuable insights for emerging market countries in formulating effective water environmental regulation policies to incentivize firms to innovate in environmental protection. Additionally, it provides novel ideas on synergizing water environmental regulation and firm innovation to foster a mutually beneficial relationship between environmental protection and sustainable development.

Non-traditional abiotic drivers explain variability of chlorophyll-a in a shallow estuarine embayment

Understanding the factors influencing eutrophication, as represented by concentrations of chlorophyll-a (Chl-a), is needed to inform effective management and conservation strategies promoting ecological resilience. The objective of this study was to evaluate a unique combination of abiotic explanatory factors to describe Chl-a concentrations within the study estuary (North Biscayne Bay, Florida, USA). Multiple linear regression determined the strength and direction of influence of factors using data from 10 water quality monitoring stations. The analysis also considered time scales for evaluating cumulative effects of freshwater inflow and wind. Results show that dominant drivers of Chl-a were temperature, freshwater volume (whose cumulative effects were evaluated up to a 60-day time scale), and turbidity, which were statistically significant at 60, 60, and 70 % of the investigated stations, respectively. All drivers collectively accounted for 22 to 63 % of the variability of Chl-a measurements. Of the nine variables evaluated, nutrient concentrations (orthophosphate and ammonia) were not among the top three overall drivers. Despite nutrients historically being cited in the literature as the most significant factor, this study asserts that non-nutrient factors often govern Chl-a levels, necessitating a paradigm shift in management strategies to bolster estuarine resilience against climate change.

Simulation and Application of Water Environment in Highly Urbanized Areas: A Case Study in Taihu Lake Basin

In the wake of frequent and intensive human activities, highly urbanized areas consistently grapple with severe water environmental challenges. It becomes imperative to establish corresponding water environment models for simulating and forecasting regional water quality, addressing the associated environmental risks. The distributed framework water environment modeling system (DF-WEMS) incorporates fundamental principles, including the distributed concept and node concentration mass conservation. It adeptly merges point source and non-point source pollution load models with zero-dimensional, one-dimensional, and two-dimensional water quality models. This integration is specifically tailored for various Hydrological Feature Units (HFUs), encompassing lakes, reservoirs, floodplains, paddy fields, plain rivers, and hydraulic engineering structures. This holistic model enables the simulation and prediction of the water environment conditions within the watershed. In the Taihu Lake basin of China, a highly urbanized region featuring numerous rivers, lakes and gates, the DF-WEMS is meticulously constructed, calibrated, and validated based on 26 key water quality monitoring stations. The results indicate a strong alignment between the simulation of water quality indicators (WQIs) and real-world conditions, demonstrating the model’s reliability. This model proves applicable to the simulation, prediction, planning, and management of the water environment within the highly urbanized watershed.

Contributions of the vegetation index (NDVI) in water quality prediction models in a semi-arid tropical watershed

In this work, a new approach to using NDVI as a predictor of water quality parameters in arid environments is proposed. Our focus was the Araçuaí river basin, which has a predominance of native cerrado vegetation and is subject to seasonal variations in rainfall and vegetation cover. MODIS images (MOD13Q1) from 2000 to 2018 were used to calculate the NDVI of the contributing areas of the water quality monitoring stations and to analyze its relationship with fourteen water quality parameters. The NDVI showed significant seasonality, with high values in the rainy season, and temporal trends of increase in stretches related to the main river. A strong and new relationship was observed between NDVI and six water quality parameters: color, total dissolved solids, total suspended solids, total solids, nitrate and turbidity, this being stronger at the local scale, with better performance at from models that use incremental NDVI, capturing local variations in vegetation cover, instead of regional NDVI, which reflects the general state of vegetation. The results demonstrate the potential of using these indices to develop and improve prediction models of water quality parameters in river basins and to expand the spatial and temporal coverage of water quality monitoring.

Interpretable CEEMDAN-FE-LSTM-transformer hybrid model for predicting total phosphorus concentrations in surface water

The complexity of the biogeochemical cycle of phosphorus in lakes makes it challenging to produce efficient and accurate predictions of total phosphorus (TP) concentrations. In this study, a hybrid model is developed for TP predictions. This model combines the Complete ensemble empirical mode decomposition with adaptive noise, Fuzzy entropy, Long short-term memory, and Transformer (CF-LT). The introduction of data split-frequency reconstruction effectively solves the problems of over- and underfitting suffered by previous machine learning models in the face of high-dimensional data, while an attention mechanism overcomes the inability of these models to establish long-term dependencies between data when making long-term predictions. The CF-LT model is applied to predict TP concentrations from January 1, 2015, to December 31, 2020, at Yaoxiangqiao, Zhihugang, and Guanduqiao, three national water quality monitoring stations at the inlet of Taihu Lake, China. Moreover, the Shapley additive explanations are used to interpret the CF-LT model and identify the essential input features. The prediction results demonstrate that the CF-LT model achieves a coefficient of determination (R2) of 0.37–0.87 on the test dataset, representing an improvement of 0.05–0.17 (6%-85%) over the control models. In addition, the CF-LT model provides the best peak value predictions. The model interpretation results indicate that the turbidity and total nitrogen are the essential factors influencing TP predictions. This demonstrates that the TP concentrations at the inlet of Taihu Lake are closely related to the non-point pollution discharge and the status of aquatic plants. It's worth noting that these two indicators exert a more significant influence on the prediction of TP during wet season. This work provides a viable modeling strategy for predicting TP concentrations and guidance for early warning and treatment of surface water eutrophication in the Taihu Lake basin.

Mathematical forecasting simulation of salinity intrusion in Chao Phraya River

At present, at two water stations along Chao Phraya River, the main river of the central region of Thailand, the salinity in the river exceeds the raw water salinity standard for tap water production that must be lower than 0.25 g/L. During December to May, since 2013, the salinity value exceeded the raw water standard, strongly affecting the raw water management of Metropolitan Waterworks Authority. Surveillance and prediction of salinity level at the river mouth at Klong Lat Pho Water Quality Monitoring Station, downriver from Sam Lae station, in advance, will be beneficial to the Metropolitan Waterworks Authority’s raw water management. The purpose of this study was to forecast the hourly salinity levels at Khlong Lat Pho Water Quality Monitoring Station. The methods used were Exponential Smoothing and Winter’s Linear and Seasonal and Exponential Smoothing (EXPOW). Both forecasting methods showed a very low mean of square error (MSE): 0.1551 when the exponential smoothing method was used with alpha= 0.9995 and 0.4869 when the EXPOW was used with alpha = 0.0618, gamma = 0.9995, and beta = 0. 0001. Nevertheless, the exponential smooth method fell out of favor since, for long-term prediction, its predicted value converged to a constant value, 0.2405, useless for predicting seasonal variation. On the other hand, the EXPOW method, even though less effective in terms of MSE, produced more reliable short and long-term forecasts for salinity at Khlong Lad Pho Water Quality Monitoring Station.

New Graph-Based and Transformer Deep Learning Models for River Dissolved Oxygen Forecasting

Dissolved oxygen (DO) is a key indicator of water quality and the health of an aquatic ecosystem. Aspiring to reach a more accurate forecasting approach for DO levels of natural streams, the present work proposes new graph-based and transformer-based deep learning models. The models were trained and validated using a network of real-time hydrometric and water quality monitoring stations for the Credit River Watershed, Ontario, Canada, and the results were compared with both benchmarking and state-of-the-art approaches. The proposed new Graph Neural Network Sample and Aggregate (GNN-SAGE) model was the best-performing approach, reaching coefficient of determination (R2) and root mean squared error (RMSE) values of 97% and 0.34 mg/L, respectively, when compared with benchmarking models. The findings from the Shapley additive explanations (SHAP) indicated that the GNN-SAGE benefited from spatiotemporal information from the surrounding stations, improving the model’s results. Furthermore, temperature has been found to be a major input attribute for determining future DO levels. The results established that the proposed GNN-SAGE model outperforms the accuracy of existing models for DO forecasting, with great potential for real-time water quality management in urban watersheds.