ABSTRACT
Monitoring suspended sediment concentration (SSC) in rivers is pivotal for water quality management and sustainable river ecosystem development. However, achieving continuous and precise SSC monitoring is fraught with challenges, including low automation, lengthy measurement processes, and high cost. This study proposes an innovative approach for SSC identification in rivers using multimodal data fusion. We developed a robust model by harnessing colour features from video images, motion characteristics from the Lucas-Kanade (LK) optical flow method, and temperature data. By integrating ResNet with a mixed density network (MDN), our method fused the image and optical flow fields, and temperature data to enhance accuracy and reliability. Validated at a hydropower station in the Xinjiang Uygur Autonomous Region, China, the results demonstrated that while the image field alone offers a baseline level of SSC identification, it experiences local errors under specific conditions. The incorporation of optical flow and water temperature information enhanced model robustness, particularly when coupling the image and optical flow fields, yielding a Nash-Sutcliffe efficiency (NSE) of 0.91. Further enhancement was observed with the combined use of all three data types, attaining an NSE of 0.93. This integrated approach offers a more accurate SSC identification solution, enabling non-contact, low-cost measurements, facilitating remote online monitoring, and supporting water resource management and river water-sediment element monitoring.
Subject(s)
Environmental Monitoring , Rivers , Temperature , Rivers/chemistry , Environmental Monitoring/methods , Geologic Sediments/analysis , China , Water QualityABSTRACT
In this article, a multivariate analysis of the parameters determining the transport and fate of selected heavy metals in the water - bottom sediment interface was carried out. The studies were carried out in the summer season of 2019 at Nielisz Reservoir (southeastern Poland, Lublin Voivodeship). Finally, a previously unknown factor related to the quality of organic matter was identified. Autochthonous organic matter was shown to promote the accumulation of the studied heavy metals. To date, the significance of the origin of organic matter in the context of the transport and fate of heavy metals in retention reservoirs has rarely been reported in the scientific literature. More than that, this factor was not considered an important component in the process of heavy metal deposition in bottom sediments. However, it turns out that not only the quantity of organic matter, but also its quality plays an important role in the circulation of heavy metals in retention reservoir ecosystems. It was found that autochthonous organic matter promotes the accumulation of the studied heavy metals. It can be assumed that, in a sense, it plays the role of a catenary ("hub") controlling the fate of heavy metals in the water-sediment system. It has also been conjectured that, in a sense, OMS may reflect the potential for heavy metal assimilation by aquatic vascular plants (mainly of the C3 group). Plants with a photosynthetic pathway similar to the C3 group generally have a much lower enrichment in the 13C isotope (δ13C from -38 to -22). In our case, the lowest δ13C-TOCS value was -24.05, and the average for the whole reservoir was -21.53. In addition, it was observed that quantitative changes in the isotopic composition of total organic carbon δ13C-TOCS, corresponded with changes in the content of the heavy metals studied in entrapped sediments.
ABSTRACT
- The sediment transport plays a major role in every aquatic ecosystem. However, the lack of instruments to monitor this process has been an obstacle to understanding its effects. We present the design of a single sensor built to measure water velocity, suspended sediment concentration and depth in situ, and how to associate the three variables to estimate and analyse sediment transport. During the laboratory calibrations, the developed instrument presented a resolution from 0.001 g/L to 0.1 g/L in the 0-12 g/L range for the measurement of suspended sediment concentration and 0.05 m/s resolution for 0-0.5 m/s range and 0.001 m/s resolution for 0.5-1 m/s range for the measurement of water velocity. The device was deployed for 6 days in an estuarine area with high sediment dynamics to evaluate its performance. During the field experiment, the sensor successfully measured the tidal cycles and consequent change of flow directions, and the suspended sediment concentration in the area. These measurements allowed to estimate water discharge and sediment transport rates during the different phases of tides, and the daily total volume of water and total amount of sediment passing through the estuary.
Subject(s)
Environmental Monitoring , Geologic Sediments , Geologic Sediments/analysis , Environmental Monitoring/methods , Water Movements , Estuaries , EcosystemABSTRACT
To reduce aquatic eutrophication, measurements of bioavailable phosphorus (BAP) rather than total phosphorus (TP) are deemed critical. However, current methods require much time to separate sediments from river water, which limits the routine measurement of BAP in rivers. Therefore, in this study, a simultaneous multisample ultrasonic extraction method is proposed to directly measure total BAP (TBAP) in river water without the separation of sediment and water. Spike-and-recovery assessments showed that at least three extractions are required to maintain efficiency. A process including 2-min extraction time and three extractions was suggested. The concentrations of TBAP extracted by this process showed no significant differences with the spike calculations. Furthermore, river water TBAP was quantified using the conventional and proposed method to examine the practicality of using the proposed method for simultaneous multisample ultrasonic extraction and to evaluate its adaptability to actual river water analysis. The extracted concentrations matched those obtained using the conventional method, in which total BAP is calculated as the sum of dissolved BAP and particulate BAP; no significant difference was observed between the concentrations. Ultrasonic extraction was considerably less time-consuming than the conventional method because more samples could be analyzed during a single run. Therefore, the simultaneous multisample ultrasonic extraction method proposed in this study can be used to directly quantify total BAP in river water.
Subject(s)
Environmental Monitoring , Phosphorus , Rivers , Water Pollutants, Chemical , Phosphorus/analysis , Rivers/chemistry , Water Pollutants, Chemical/analysis , Environmental Monitoring/methods , UltrasonicsABSTRACT
Wave-induced liquefaction is a geological hazard under the action of cyclic wave load on seabed. Liquefaction influences the suspended sediment concentration (SSC), which is essential for sediment dynamics and marine water quality. Till now, the identification of liquefaction state and the effect of liquefaction on SSC have not been sufficiently accounted for in the sediment model. In this study, we introduced a method for simulating the liquefaction-induced resuspension flux into an ocean model. We then simulated a storm north of the Yellow River Delta, China, and validated the results using observational data, including significant wave heights, water levels, excess pore water pressures, and SSCs. The liquefaction areas were mainly distributed in coastal zones with water depths less than 12 m, and the simulated maximum potential soil liquefaction depth was 1.39 m. The liquefaction-induced SSC was separated from the total SSC of both liquefaction- and shear-induced SSCs by the model, yielding a maximum liquefaction-induced SSC of 1.07 kg·m-3. The simulated maximum proportion of liquefaction-induced SSC was 26.2% in regions with water depths of 6-12 m, with a maximum significant wave height of 3.4 m along the 12 m depth contour. The erosion zone at water depths of 8-12 m was reproduced by the model. Within 52.5 h of the storm, the maximum erosion thickness along the 10 m depth contour was enhanced by 33.9%. The model is applicable in the prediction of liquefaction, and provides a new method to simulate the SSC and seabed erosion influenced by liquefaction. Model results show that liquefaction has significant effects on SSC and seabed erosion in the coastal area with depth of 6-12 m. The validity of this method is confined to certain conditions, including a fully saturated seabed exhibiting homogeneity and isotropic properties, small liquefaction depth, residual liquefaction dominating the development of pore pressures, no influence by structures, and the sediment composed of silt and mud that experiences frequent wave-induced liquefaction.
Subject(s)
Geologic Sediments , Models, Theoretical , Geologic Sediments/chemistry , China , Water MovementsABSTRACT
Prediction of suspended sediment load (SSL) in streams is significant in hydrological modeling and water resources engineering. Development of a consistent and accurate sediment prediction model is highly necessary due to its difficulty and complexity in practice because sediment transportation is vastly non-linear and is governed by several variables like rainfall, strength of flow, and sediment supply. Artificial intelligence (AI) approaches have become prevalent in water resource engineering to solve multifaceted problems like sediment load modelling. The present work proposes a robust model incorporating support vector machine with a novel sparrow search algorithm (SVM-SSA) to compute SSL in Tilga, Jenapur, Jaraikela and Gomlai stations in Brahmani river basin, Odisha State, India. Five different scenarios are considered for model development. Performance assessment of developed model is analyzed on basis of mean absolute error (MAE), root mean squared error (RMSE), determination coefficient (R2), and Nash-Sutcliffe efficiency (ENS). The outcomes of SVM-SSA model are compared with three hybrid models, namely SVM-BOA (Butterfly optimization algorithm), SVM-GOA (Grasshopper optimization algorithm), SVM-BA (Bat algorithm), and benchmark SVM model. The findings revealed that SVM-SSA model successfully estimates SSL with high accuracy for scenario V with sediment (3-month lag) and discharge (current time-step and 3-month lag) as input than other alternatives with RMSE = 15.5287, MAE = 15.3926, and ENS = 0.96481. The conventional SVM model performed the worst in SSL prediction. Findings of this investigation tend to claim suitability of employed approach to model SSL in rivers precisely and reliably. The prediction model guarantees the precision of the forecasted outcomes while significantly decreasing the computing time expenditure, and the precision satisfies the demands of realistic engineering applications.
ABSTRACT
After the Fukushima Daiichi nuclear power plant accident, the terrestrial environment became severely contaminated with radiocesium. Consequently, the river and lake water in the Fukushima area exhibited high radiocesium levels, which declined subsequently. The partition coefficient of 137Cs between the suspended sediment (SS) and dissolved phases, Kd, was introduced to better understand the dynamic behavior of 137Cs in different systems. However, the Kd values in river water, ranging from 2 × 104 to 7 × 106 L kg-1, showed large spatiotemporal variability. Therefore, the factors controlling the 137Cs partition coefficient in natural water systems should be identified. Herein, we introduce a chemical model to explain the variability in 137Cs Kd in natural water systems. The chemical model includes the complexation of Cs+ with mineral and organic binding sites in SS, metal exchange reactions, and the presence of colloidal species. The application of the chemical model to natural water systems revealed that Cs+ is strongly associated with binding sites in SS, and a major chemical interaction between 137Cs and the binding sites in SS is the isotope exchange reaction between stable Cs and 137Cs, rather than metal exchange reactions with other metal ions such as potassium ions. To explain the effect of the SS concentration on Kd, the presence of colloidal 137Cs passing through a filter is significant as the dominant dissolved species of 137Cs in river water. These results suggest that a better understanding of stable Cs dissolved in natural water is important for discerning the geochemical and ecological behaviors of 137Cs in natural water.
Subject(s)
Cesium Radioisotopes , Fukushima Nuclear Accident , Radiation Monitoring , Water Pollutants, Radioactive , Cesium Radioisotopes/analysis , Water Pollutants, Radioactive/analysis , Water Pollutants, Radioactive/chemistry , Models, Chemical , Japan , Rivers/chemistry , Geologic Sediments/chemistryABSTRACT
Understanding particle size distribution (PSD) of total suspended sediments in urban runoff is essential for pollutant fate and designing effective stormwater treatment measures. However, the PSDs from different land uses under different weather conditions have yet to be sufficiently studied. This research conducted a six-year water sampling program in 15 study sites to analyze the PSD of total suspended sediments in runoff. The results revealed that the median particle size decreased in the order: paved residential, commercial, gravel lane residential, mixed land use, industrial, and roads. Fine particles less than 125 µm are the dominant particles (over 75%) of total suspended sediments in runoff in Calgary, Alberta, Canada. Roads have the largest percentage of particles finer than 32 µm (49%). Gravel lane residential areas have finer particle sizes than paved residential areas. The results of PSD were compared with previous literature to provide more comprehensive information about PSD from different land uses. The impact of rainfall event types can vary depending on land use types. A long antecedent dry period tends to result in the accumulation of fine particles on urban surfaces. High rainfall intensity and long duration can wash off more coarse particles. The PSD in spring exhibits the finest particles, while fall has the largest percentage of coarse particles. Snowmelt particles are finer for the same land use than that during rainfall events because the rainfall-runoff flows are usually larger than the snowmelt flows.
Subject(s)
Particle Size , Rain , Seasons , Geologic Sediments/analysis , Geologic Sediments/chemistry , Water Movements , Environmental Monitoring , AlbertaABSTRACT
Suspended sediment concentration prediction is critical for the design of reservoirs, dams, rivers ecosystems, various operations of aquatic resource structure, environmental safety, and water management. In this study, two different machine models, namely the cascade correlation neural network (CCNN) and feedforward neural network (FFNN) were applied to predict daily-suspended sediment concentration (SSC) at Simga and Jondhara stations in Sheonath basin, India. Daily-suspended sediment concentration and discharge data from 2010 to 2015 were collected and used to develop the model to predict suspended sediment concentration. The developed models were evaluated using statistical indices like Nash and Sutcliffe efficiency coefficient (NES), root mean square error (RMSE), Willmott's index of agreement (WI), and Legates-McCabe's index (LM), supplemented by a scatter plot, density plots, histograms and Taylor diagram for graphical representation. The developed model was evaluated and compared with CCNN and FFNN. Nine input combinations were explored using different lag-times for discharge (Qt-n) and suspended sediment concentration (St-n) as input variables, with the current suspended sediment concentration as the desired output, to develop CCNN and FFNN models. The CCNN4 model with 4 lagged inputs (St-1, St-2, St-3, St-4) outperformed the other developed models with the lowest RMSE = 95.02 mg/l and the highest NES = 0.0.662, WI = 0.890 and LM = 0.668 for the Jondhara Station while the same CCNN4 model secure as the best with the lowest RMSE = 53.71 mg/l and the highest NES = 0.785, WI = 0.936 and LM = 0.788 for the Simga Station. The result shows the CCNN model was better than the FFNN model for predicting daily-suspended sediment at both stations in the Sheonath basin, India. Overall, CCNN showed better forecasting potential for suspended sediment concentration compared to FFNN at both stations, demonstrating their applicability for hydrological forecasting with complex relationships.
ABSTRACT
The prediction of suspended sediment load (SSL) within riverine systems is critical to understanding the watershed's hydrology. Therefore, the novelty of our research is developing an interpretable (explainable) model based on deep learning (DL) and Shapley Additive ExPlanations (SHAP) interpretation technique for prediction of SSL in the riverine systems. This paper investigates the abilities of four DL models, including dense deep neural networks (DDNN), long short-term memory (LSTM), gated recurrent unit (GRU), and simple recurrent neural network (RNN) models for the prediction of daily SSL using river discharge and rainfall data at a daily time scale in the Taleghan River watershed, northwestern Tehran, Iran. The performance of models was evaluated by using several quantitative and graphical criteria. The effect of parameter settings on the performance of deep models on SSL prediction was also investigated. The optimal optimization algorithms, maximum iteration (MI), and batch size (BC) were obtained for modeling daily SSL, and structure of the model impact on prediction remarkably. The comparison of prediction accuracy of the models illustrated that DDNN (with R2 = 0.96, RMSE = 333.46) outperformed LSTM (R2 = 0.75, RMSE = 786.20), GRU (R2 = 0.73, RMSE = 825.67), and simple RNN (R2 = 0.78, RMSE = 741.45). Furthermore, the Taylor diagram confirmed that DDNN has the highest performance among other models. Interpretation techniques can address the black-box nature of models, and here, SHAP was applied to develop an interpretable DL model to interpret of DL model's output. The results of SHAP showed that river discharge has the strongest impact on the model's output in estimating SSL. Overall, we conclude that DL models have great potential in watersheds to predict SSL. Therefore, different interpretation techniques as tools to interpret DL model's output (DL model is as black-box model) are recommended in future research.
Subject(s)
Deep Learning , Geologic Sediments , Rivers , Rivers/chemistry , Iran , Neural Networks, Computer , Environmental Monitoring/methods , Models, TheoreticalABSTRACT
Anthropogenic stressors such as agriculture and urbanization can increase river turbidity, which can negatively impact fish gill morphology and growth due to reduced oxygen in the benthic environment. We assessed the gill morphology, field metabolic rate (FMR), and two hypoxia tolerance metrics (oxygen partial pressure at loss of equilibrium, PO2 at LOE, and critical oxygen tension, Pcrit) of eastern sand darter (Ammocrypta pellucida), a small benthic fish listed as threatened under the Species at Risk Act in Canada, from rivers in southern Ontario. Field trials were conducted streamside in the Grand River (August 2019; mean NTU 8) and in the comparatively more turbid Thames River (August 2020; mean NTU 94) to test the effect of turbidity on each physiological endpoint. Gills were collected from incidental mortalities and museum specimens, and were assessed using hematoxylin and eosin and immunofluorescent staining. The between-river comparison indicated that turbidity significantly increased interlamellar space and filament width but had no significant influence on other gill morphometrics or FMR. Turbidity significantly increased PO2 at LOE (i.e., fish had a lower hypoxia tolerance) but did not significantly impact Pcrit. Therefore, although turbidity influences hypoxia tolerance through LOE, turbidity levels were not sufficiently high in the study rivers to contribute to measurable changes in gill morphology or metabolism in the wild. Determining whether changes in gill morphology or metabolism occur under higherturbidity levels would help resolve the ecological importance of turbidity on species physiology in urban and agricultural ecosystems.
Subject(s)
Gills , Oxygen , Rivers , Animals , Gills/anatomy & histology , Gills/physiology , Ontario , Oxygen/metabolism , Hypoxia , Perciformes/physiology , Perciformes/anatomy & histologyABSTRACT
Channel-forming discharge (Dcf) is an important parameter in river management and reservoir flood regulation. Applying the methods for calculating Dcf to reaches downstream reservoirs characterized by drastic changes in water and sediment conditions and long-term scouring status is difficult. Based on the riverbed-shaping principle of sediment-laden water flow, while simultaneously considering the active action of water flow and response of the riverbed, this study proposes a new method for calculating Dcf by identifying the extreme value of the suspended sediment-carrying capacity index. The application of this method to the middle and lower reaches of the Yangtze River showed that after the impoundment of the Three Gorges Reservoir, Dcf in this section was reduced by an amplitude between 2500 and 4700 m3/s. The results can be used to guide the operation of the Three Gorges Reservoir and the management of the middle and lower reaches of the Yangtze River, thus providing reference for other river channels downstream of the reservoir.
ABSTRACT
Previous research methodologies for quantifying Suspended Sediment Concentration (SSC) have encompassed in-situ observations, numerical simulations, and analyses of remote sensing datasets, each with inherent constraints. In this study, we have harnessed Convolutional Neural Networks (CNNs) to create a deep learning model, which has been applied to the remote sensing data procured from the Geostationary Ocean Color Imager (GOCI) spanning April 2011 to March 2021. Our research indicates that on a small time scale, wind and hydrodynamic forces both have a significant impact on the prediction results of CNNs model. Considering both wind and hydrodynamic forces can effectively improve the model's prediction efficiency for SSC. Moreover, we have employed CNNs to interpolate absent values within the remote sensing datasets, yielding enhancements superior to those attained via linear or multivariate regression techniques. Finally, the correlation coefficient between CNN-derived SSC estimates for Jiaozhou Bay (JZB) and its corresponding remote sensing data is 0.72. Correlation coefficient and root mean square error differ in different regions. In the shallow water of JZB, due to water level changes, there is limited data, and the correlation coefficient in this area is about 0.5-0.6. In the central region of JZB with sufficient data, the correlation coefficient is generally higher than 0.75. Therefore, we believe that this CNNs model can be used to predict the hourly variation of SSC. When juxtaposed with alternative methodologies, the CNN approach is found to economize computational resources and enhance processing efficiency.
Subject(s)
Deep Learning , Environmental Monitoring , Environmental Monitoring/methods , Bays , Water , China , Geologic SedimentsABSTRACT
Suspended sediment (SS) is a natural component of aquatic environments. It is characterized by the adsorption of pollutants, and its physical properties can affect water volume quality. In this study, SS dynamics were simulated using a 2D hydrodynamic model in the Nanji Mountain Nature Reserve (NNR), and the fluxes of pollutants caused by SS were calculated to assess the biological risks during the wet (May-August) and dry (November-March) seasons. High spatial and temporal variability in SS load within the NNR was found in this study. The average SS load in the reserve increased and then decreased during the year, and the SS input from Ganjiang significantly affected the SS load in the NNR (p < 0.01). The SS load uptrend in the NNR occurred later than that of Ganjiang during the wet season because of the SS sedimentation in the NNR. And the suspension of SS in the NNR during the dry season resulted in a later SS load downtrend compared to Ganjiang. High SS load from Ganjiang during the wet season was responsible for the high nutrient and microplastic fluxes in the NNR, which were 8.38 and 10.61 times higher than those in the dry season, respectively. And the pollutant fluxes during the wet season were almost all from Ganjiang. In contrast, higher waterbird diversity and population during the dry season is the main reason for the increased biological risk of contaminants. Therefore, monitoring and managing SS and its contamination concentrations in rivers entering the lake is helpful for the protection of ecologically sensitive areas and key species in the lake.
Subject(s)
Environmental Pollutants , Water Pollutants, Chemical , Environmental Monitoring/methods , Water , Plastics , Geologic Sediments , Rivers , Water Pollutants, Chemical/analysis , Seasons , ChinaABSTRACT
At sites that have been sampled for decades, changes in field and laboratory methods happen over time as instrumentation and protocols improve. Here, we compare the influence of depth- and point-integrated sampling on total, fine (< 0.0625 mm), and coarse (≥ 0.0625 mm) suspended sediment (SS) concentrations in the Lower Mississippi and Atchafalaya Rivers. Using historical field method information, we identified seven sites to test such differences. We found SS samples collected using point-integration tended to have higher concentrations than those collected using depth-integration. However, the presence and magnitude of the bias were inconsistent across sites. Bias was present at the site with less-than-ideal conditions (i.e., non-trapezoidal channel, non-uniform flow) and non-existent at the ideal site location, indicating the bias between sampling methods depends on site sampling conditions. When present, the bias is greater at higher concentrations and at moderate to high flows. At the less-than-ideal site, point-integrated samples can have 16% (total) and 34% (coarse) higher concentrations than depth-integrated samples. When flow effects are removed, this translates to a bias of 19, 9, and 8 mg per liter for total, fine, and coarse SS. When a change in field methods occurs, comparison samples and a rigorous evaluation of those samples are warranted to determine the proper course of action for a particular site. Often, the effect and solution will not be known until several years of comparison samples have been collected under a variety of hydrologic conditions.
Subject(s)
Environmental Monitoring , Rivers , Environmental Monitoring/methods , Mississippi , Geologic SedimentsABSTRACT
Excess sediment is a common reason water bodies in the USA become listed as impaired resulting in total maximum daily loads (TMDL) that require municipalities to invest millions of dollars annually on management practices aimed at reducing suspended-sediment loads (SSLs), yet monitoring data are rarely used to quantify SSLs and track TMDL progress. A monitoring network was created to quantify the SSL from the City of Roanoke, Virginia, USA (CoR), to the Roanoke River and Tinker Creek and help guide TMDL assessment and implementation. Suspended-sediment concentrations were estimated between 2020 and 2022 from high-frequency turbidity data using surrogate linear-regression models. Sixty-one percent of the total three-year SSL resulted from five large storm events. The average suspended-sediment yield from the CoR (58.1 metric tons/km2/year) was similar to other urban watersheds in the Eastern United States; however, the yield was nearly five times larger than the TMDL allocation (12.2 metric tons/km2/year). The TMDL allocated load was modeled based on a predominantly forested reference watershed and may not be a practical target for highly impervious watersheds within the CoR. The TMDL model used daily input data which likely does not capture the full range of SSLs during storm events, particularly from flashy urban streams. The average SSL following the five large storm events doubled that of the CoR's annual allocated load from the TMDL. The results of this study highlight the importance of using high-frequency monitoring data to accurately estimate SSLs and evaluate TMDLs in urban areas.
Subject(s)
Environmental Monitoring , Goals , United States , Environmental Monitoring/methods , Cities , Virginia , RiversABSTRACT
Due to the regional differences between the North and South Yellow Sea, and under the influence of winter winds, the relative changes in the coastal current and the Yellow Sea warm current will lead to the instability of the front, which will lead to the cross-front transport of sediment. Therefore, the study of sediment exchange between the North and South Yellow Sea has become an indispensable part of the study of the Yellow Sea environment. In this study, the current field and sediment concentration in the southern part of Chengshantou, a representative area of the Yellow Sea, were observed in winter in order to analyze the sediment exchange process between the North Yellow Sea and the South Yellow Sea in winter. The observation results show that in the southern sea area of Chengshantou, in winter, the current velocity does not change with the water depth when it exceeds 15 m, and the tides are regular semi-diurnal tides. When the water depth is less than 15 m, the current direction changes clockwise with the increase in the water depth. The turbidity increases rapidly when the wind direction is offshore and the bottom residual current is onshore, which may cause the sediment transported offshore under the action of wind and ocean current to settle under the obstruction of the Yellow Sea warm current, resulting in the rise of bottom turbidity. This also indicates that the change in residual current direction at different water depths may also lead to an increase in suspended sediment concentration. Based on this, in the estuarine area, the relative change in the current direction between the wind current and the coastal current may also be the cause of the change in the maximum turbidity zone.
ABSTRACT
Wet ponds have been extensively used for controlling stormwater pollutants, such as sediment and nutrients, in urban watersheds. The removal of pollutants relies on a combination of physical, chemical, and biological processes. It is crucial to assess the performance of wet ponds in terms of removal efficiency and develop an effective modeling scheme for removal efficiency prediction to optimize water quality management. To achieve this, a two-year field program was conducted at two wet ponds in Calgary, Alberta, Canada to evaluate the wet ponds' performance. Additionally, machine learning (ML) algorithms have been shown to provide promising predictions in datasets with intricate interactions between variables. In this study, the generalized linear model (GLM), partial least squares (PLS) regression, support vector machine (SVM), random forest (RF), and K-nearest neighbors (KNN) were applied to predict the outflow concentrations of three key pollutants: total suspended solids (TSS), total nitrogen (TN), and total phosphorus (TP). Generally, the concentrations of inflow pollutants in the two study ponds are highly variable, and a wide range of removal efficiencies are observed. The results indicate that the concentrations of TSS, TN, and TP decrease significantly from the inlet to outlet of the ponds. Meanwhile, inflow concentration, rainfall characteristics, and wind are important indicators of pond removal efficiency. In addition, ML algorithms can be an effective approach for predicting outflow water quality: PLS, GLM, and SVM have shown strong potential to capture the dynamic interactions in wet ponds and predict the outflow concentration. This study highlights the complexity of pollutant removal dynamics in wet ponds and demonstrates the potential of data-driven outflow water quality prediction.
ABSTRACT
Cesium-137, discharged by nuclear installations under normal operations and deposited in watersheds following atmospheric testing and accidents (i.e. Chernobyl, Fukushima ), has been studied for decades. Thus, modelling of 137Cs concentration in rivers have been developed based on geochemical approaches and equilibrium assumptions (solid/liquid ratio) as this radionuclide has moved into rivers and oceans due to soil erosion. Recently a new approach is possible to model these concentrations with the popularization of data-driven models based on data acquired in the environment by monitoring networks. In this study, the concentrations of particulate cesium-137 measured near the mouth of the Rhône River (France), a highly nuclearized river, are simulated using two data-driven models, a Hierarchical Attention-Based Recurrent Highway Networks (HRHN) and a Random Forest Regressor (RF). The data-driven predictions were done using only hydrological data (water discharge and suspended solid fluxes) and industrial input of 137Cs. Although the data-driven models provided a better prediction than a recent empirical model, the best prediction (R2 = 0.71) was obtained with HRHN, a model that considers the temporal aspect of the monitoring data. The most important predictors were the hydrological data at the monitoring station and of the tributary that generate the most sediment flux (Durance River). In fact, the concentration of 137Cs in the perimeter of this study was more related to hydrology than to nuclear release, as there were few events with high 137Cs concentrations (concomitant nuclear release and low water discharge). However, the HRHN approach, which is more complex to implement than RF, can predict the concentrations of such events correctly despite their low representation of these events. The results of this study demonstrate the usefulness of data-driven models to assist monitoring programs by filling in gaps or helping to understand observed concentrations.
Subject(s)
Deep Learning , Fukushima Nuclear Accident , Radiation Monitoring , Water Pollutants, Radioactive , Water Pollutants, Radioactive/analysis , Rivers , Cesium Radioisotopes/analysis , Dust , Machine Learning , Water , JapanABSTRACT
Suspended sediment concentration (SSC) in water increases temperature and turbidity, limits the photosynthesis of aquatic plants, and reduces biologically available oxygen. It is important to study SSC in the coastal waters of the Arabian Gulf. Thus, this study mapped the SSC of coastal water between Al Arish and Al Ghariyah in northern Qatar using the spectral bands of the MultiSpectral Imager (MSI) of Sentinel-2 by calculating the Normalized Difference Suspended Sediment Index and Normalized Suspended Material Index. The results are studied using the Normalized Difference Turbidity Index and Modified Normalized Difference Water Index. The mapping of SSC in the water using NDSSI showed the presence of a high concentration of suspended sediments between Al Arish and Al Mafjar and a low concentration between Al Mafjar and Al Ghariyah. The mapping of NSMI showed values between 0.012 (clear water) and 0.430 (more suspended material) for the occurrence of suspended materials and supported the results of NDSSI. The study of turbidity using an NDTI image showed turbidity index values ranging from -0.44 (clear water) to 0.12 (high turbidity) and confirmed the occurrence and distribution of suspended sediments and materials in the water. The MNDWI image was able to discriminate clear water with bright pixels from silty sand and mud flats. The relationships between NDSSI, NSMI, and NDTI were correlated with in-situ measurements and studied to find suitable indices to map SSC. Regression analyses showed the strongest relationship between NSMI and NDTI (R2 = 0.95) next to NDSSI and NDTI, where NDTI had the strongest effect on NDSSI (R2 = 0.86). The satellite data results were evaluated by studying the physical parameters and spatial distribution of suspended sediments in the surface and bottom waters. In addition, the grain size distributions, mineral identification, and chemical element concentrations in the bottom sediment samples were studied.