Daily Suspended Sediment Concentration Research Articles

Estimation of daily suspended sediment concentration in the Ca River Basin using a sediment rating curve, multiple regression, and long short-term memory model

Abstract This study presents a sediment rating curve (SRC), multiple regression (MR), and long short-term memory (LSTM) model for estimating daily suspended sediment concentration (SSC). The data of daily SSC at Yen Thuong and daily flow at five locations in the Ca River Basin, Vietnam are used to demonstrate multiple approaches. Using the daily flow and SSC data in the period from 2009 to 2019, appropriate coefficients in each method are identified carefully using five popular criteria. The results showed that SRC and MR approaches reproduced acceptably the observed values, with the values of RMSE, MAE, and ME of daily SSC being less than 5% of daily SSC magnitude observed at the station, while NSE ranges from 0.47 to 0.63 and r coefficient varies between 0.69 and 0.80. The LSTM model represented the observed values of daily SSC very well. The values of two dimensionless criteria are greater than 0.94 and its values of three-dimensional criteria are smaller than 2.0% of the observed magnitude of daily SSC in both training and validation steps. The LSTM model is found to be the best among the three investigated approaches. Then, the model is applied to estimate daily SSC values for the period from 1969 to 2008 and the year 2020.

Daily suspended sediment concentration forecast in the upper reach of Yellow River using a comprehensive integrated deep learning model

The precise prediction of suspended sediment concentration (SSC) is of great importance for river reservoir construction planning, water resource management, and ecological environment restoration. This research aims to improve SSC prediction accuracy by constructing a comprehensive and integrated deep learning model Wavelet-MGGP-CNN-LSTM (ICNN-LSTM), combining wavelet transformation (WT), multi-gene genetic programming (MGGP), convolutional neural network (CNN), and long short-term memory (LSTM) simultaneously. In ICNN-LSTM, the WT decomposes the signal and extracts time and frequency domain information, while the MGGP filters out redundant information. Then, the CNN and LSTM are integrated in a parallel and loosely coupled manner to form an initial combined model CNN-LSTM (CNN combined with LSTM) to process filtered information by WT and MGGP. Furthermore, this study compares the performance of ICNN-LSTM with CNN, LSTM, CNN-LSTM, ICNN (CNN embedded with WT and MGGP), ILSTM (LSTM embedded with WT and MGGP), artificial neural network (ANN), and the traditional sediment rating curve (SRC). The evaluation of prediction accuracy for all models was conducted using root mean square error (RMSE), Nash-Sutcliffe coefficient (NSC), coefficient of determination (R2), and mean absolute error (MAE) as performance indicators. The daily discharge and suspended sediment concentration series data from Tangnaihai Hydrological Station in the upper reaches of the Yellow River spanning from 1977 to 1987 were selected to train and test the models. Results show that, first, deep learning networks such as CNN and LSTM outperform the shallow neural network ANN, with LSTM providing higher accuracy than CNN. Second, the CNN-LSTM hybrid outperforms both CNN and LSTM models, exhibiting a nearly 89% improvement in NSC value compared to SRC in the test phase. Third, deep learning models such as ICNN, ILSTM, and ICNN-LSTM show significantly higher NSC values than CNN, LSTM, and CNN-LSTM models in the test phase, with improvements of 13.8%, 5.7%, and 12.1%, respectively. Moreover, compared to SRC, the proposed ICNN-LSTM model improves NSC value by nearly 140% in the test phase. The proposed ICNN-LSTM model, integrating the advantages of WT, deep learning, and ensemble learning, provides accurate and reliable predictions and serves as a reference for time series prediction modeling.

Variations in the Suspended Sediment Concentration in Mountain-Type Rivers Flowing Into the Sea in the Past 60 years—Taking Nanliu River in Beibu Gulf as an Example

Suspended sediment in the water body of rivers flowing into the sea is of great significance to the accumulation process in river basins and change pattern of landforms. In particular, small and medium rivers entering the sea in mountainous areas exhibit the characteristics of both mountains and streams, and the suspended sediment concentration (SSC) greatly contributes to the formation and evolution of deltas. However, scholars rarely give attention to the factors influencing changes in the SSC in small- and medium-sized mountainous rivers, and few studies have examined SSC changes in small- and medium-sized mountainous rivers. Here, based on daily SSC and flow data obtained at the Bobai Station and Changle Station, the percentile method and regression analysis method are employed to analyse the changes in SSC from the Nanliu River to the Beibu Gulf and possible influencing factors. The main research results indicate that 1) the SSC in river water bodies from 1965 to 2020 generally reveals a downwards trend, with significant annual variations. Specifically, the overall trend can be divided into three stages: 1) the SSC is the highest from 1965 to 1971, and the mean yearly SSC reaches 0.25 kg/m3; 2) the SSC is relatively high from 1972 to 2006, and the mean yearly SSC reaches 0.16 kg/m3; and 3) the SSC is the lowest from 2007 to 2020, and the mean yearly SSC reaches 0.11 kg/m3 2) High values of the SSC are mainly concentrated in the flood season, and low SSC values mostly occur in the dry season. The monthly average SSC in the flood season from April to September and the dry season from October to March exhibits the characteristics of a decrease in the flood season and an increase in the dry season. Moreover, the peak SSC value in the watershed obviously occurs out of sync with the peak flow value. Generally, the former precedes the latter. 3) The curve of the flow rate-sediment ratio is an irregular clockwise rhombus. Moreover, the impact of tropical cyclones, land cover changes, regional GDP per capita, and engineering construction are crucial reasons for the observed variations in the SSC in the Nanliu River Basin. In this paper, the obtained research results provide an important guiding significance for the planning and management of water and sediment resources in the Nanliu River and offer a reference for hydrological planning of other river basins discharging into the sea.

Modeling stage\u2013discharge\u2013sediment using support vector machine and artificial neural network coupled with wavelet transform

Many real water issues involve rivers’ sediment load or the load that rivers can bring without degrading the fluvial ecosystem. Therefore, the assessment of sediments carried by a river is also crucial in the planning and designing of various water resource projects. In the current study, five different data-driven techniques, namely artificial neural network (ANN), wavelet-based artificial neural network (WANN), support vector machine (SVM), wavelet-based support vector machine (WSVM), and multiple-linear regression (MLR) techniques, were employed for time-series modeling of daily suspended sediment concentration (SSC). Hydrological datasets containing the daily stage (h), discharge (Q), and SSC for 10 years (2004–2013) from June to October at Adityapur and Ghatshila station of Subernrekha river basin, Jharkhand, India, were considered for analysis. The Gamma test was used to determine the input variables in the first step. Various combinations were made by lagging the maximum three-day time step for predicting current-day SSC. The outcomes of ANN, SVM, WAAN, WSVM, and MLR models were evaluated with the actual values of SSC based on statistical metrics. Pearson correlation coefficient (PCC), root-mean-square error (RMSE), Nash–Sutcliffe efficiency (NSE), and Wilmot index (WI) as well as visual inspection of time variation, scatter plots, and Taylor diagrams. Our results stated that the WSVM model discovered the best trustworthy models among all existing models. PCC, RMSE, NSE, and WI values were 0.844 and 0.781, 0.096 g/l and 0.057 g/l, 0.711 and 0.591, 0.907 and 0.878, respectively, throughout the training and testing processes at the Adityapur site. Also, at the Ghatshila location, it was the most accurate model. During the training and testing stages, PCC, RMSE, NSE, and WI values were 0.928 and 0.751, 0.117 g/l and 0.095 g/l, 0.861 and 0.541, 0.962 and 0.859, respectively. Our findings showed that the WANN model was the second-best model during the testing phase for both sites. Hence, the WSVM technique can model SSC at this location and other similar (i.e., geomorphology and flow regime type) rivers.

Daily suspended sediment concentrations and load variability in karst watersheds

A better understanding of sediment dynamics and variability may provide helpful insights for improving the efficiency of community-based monitoring strategies, enhancing sediment yield estimation, and optimizing soil conservation practices. However, this issue is rarely studied in karst areas, which are experiencing severe soil erosion, and sediment concentrations and load are highly variable. The objectives of this study were to characterize the daily suspended sediment concentrations (SSC) and sediment load (SL) variability, and to test whether the watershed scale and lithology exerted substantial influence on this variability in 40 karst watersheds (420–3.4 × 105 km2) in southwest China. Several indicators of variability were used for daily SSC and SL including standard deviation, coefficient of deviation, inhomogeneity coefficient, the percentage of annual SL exported in 2% of the time (Ms2%), and the percentage of time necessary to transport 50% of the annual SL (Ts50%). Results indicated that the daily SSC and SL had high temporal variability, and the SL demonstrated stronger variability than the SSC. On average, half of the annual sediment load was discharged in<2.3% of the time in the period 2009–2012, implying that a few erosive events dominated the annual sediment load. The temporal variability of daily SSC and SL were scale dependent, and a larger watershed generally had a lower temporal variability. The Ms2% and Ts50% were relatively stable when the area of the watershed was greater than 0.15 million km2. However, no significant relationship was detected between karst coverage and Ms2% or Ts50% since karst coverage might not completely represent the characteristics of the 3-D hydrogeological structure in karst regions. The results from this study can help improve river flow management and control soil loss in karst watersheds.

Integration of ANFIS with PCA and DWT for daily suspended sediment concentration prediction

Quantifying sediment load is vital for aquatic and riverine biota and has been the subject of various environmental studies since sediment plays a key role in maintaining ecological integrity, river morphology and agricultural productivity. However, predicting sediment concentration in rivers is difficult because of the non-linear relationships of flow rates, geophysical characteristics and sediment loads. It is thus very important to propose suitable statistical methods which can provide fast, accurate and robust prediction of suspended sediment concentration (SSC) for management guidance. In this study, we developed coupled models of discrete wavelet transform (DWT) with adaptive neuro-fuzzy inference system (ANFIS), named DWT-ANFIS, and principal component analysis (PCA) with ANFIS, named PCA-ANFIS, for SSC time-series modeling. The coupled models and single ANFIS model were trained and tested using long-term daily SSC and river discharge which were measured on the Schuylkill and Iowa Rivers in the United States. The findings showed that the PCA-ANFIS performed better than the single ANFIS and the coupled DWT-ANFIS. Further applications of the PCA-ANFIS should be considered for simulation and prediction of other indicators relating to weather, water resources, and the environment.

Suspended sediment dynamics and associated hydro-meteorological interrelations in east rathong glacier, eastern himalaya, india

The records of suspended sediments and associated hydro-meteorology of the glacierized catchments are very limited in Sikkim Himalaya. Therefore, we carried out the first-ever estimate of suspended sediment transport from the meltwater stream of East Rathong Glacier for June-September during three consecutive melt seasons (2013–2015). These records were used to estimate the suspended sediment concentration (SSC), load (SSL), yield (SSY) and erosion rate (ER). The daily mean SSC was 105.75, 84.20, and 91.51 mg/l during melt seasons of 2013–2015 respectively. The monthly mean SSC of the combined melt season of 2013–2015 was 92.72, 107.14, 105.93 and 60.46 mg/l for June-September, respectively. The observation shows that July-August contributed to maximum SSC, accounting 70.45% of total SSC. Similarly, the monthly mean SSL for June-September was 55.64, 69.61, 65.75 and 25.72 tons, respectively. We observed high variability in SSL (CV = 0.54) than SSC (CV = 0.38) since the estimation of SSL depends on both discharge (Q) and SSC. The relationship between monthly mean SSC and Q was highly significant (R2 = 0.98) than daily SSC and Q (R2 = 0.25) due to reduced variability on a monthly scale. Further, mean SSY in the catchment was high during peak melting season (July-August), which substantially reduced during the end of the melting season. We estimated the mean SSY of 281.93 ton/yr km2 during the combined melt season (2013–2015). The average ER for the East Rathong Glacier basin was 1.04 mm/yr, which is lower than the ER of most of the Himalayan glaciers. We observed significant meteorological control of Tmean, Tmax and Rainfall in the transportation of suspended sediments from meltwater stream at a monthly scale. These in-situ based records on dynamics of suspended sediment and associated hydro-meteorological parameters are relevant for the planning and management of freshwater resources and planning for hydropower projects under a changing climate.

Hindcasting and Forecasting Total Suspended Sediment Concentrations Using a NARX Neural Network

Estimating and forecasting suspended sediments concentrations in streams constitutes a valuable asset for sustainable land management. This research presents the development of a non-linear autoregressive exogenous neural network (NARX) for forecasting sediment concentrations at the exit of Francia Creek watershed (Valparaiso, Chile). Details are presented on input data selection, data splitting, selection of model architecture, determination of model structure, NARX training (optimization of model parameters), and model validation (hindcasting and forecasting). The study explored if the developed artificial neural network model is valid for forecasting daily suspended sediment concentrations for a complete year, capturing seasonal trends, and maximum and baseflow concentrations. Francia Creek watershed covers approximately 3.24 km2. Land cover within the catchment consists mainly of native and exotic vegetation, eroded soil, and urban areas. Input data consisting of precipitation and stream flow time-series were fed to a NARX network for forecasting daily suspended sediments (SST) concentrations for years 2013–2014, and hindcasting for years 2008–2010. Training of the network was performed with daily SST, precipitation, and flow data from years 2012 and 2013. The resulting NARX net consisted of an open-loop, 12-node hidden layer, 100 iterations, using Bayesian regularization backpropagation. Hindcasting of daily and monthly SST concentrations for years 2008 through 2010 was successful. Daily SST concentrations for years 2013 and 2014 were forecasted successfully for baseflow conditions (R2 = 0.73, NS = 0.71, and Kling-Gupta efficiency index (K-G) = 0.84). Forecasting daily SST concentrations for year 2014 was within acceptable statistical fit and error margins (R2 = 0.53, NS = 0.47, K-G = 0.60, d = 0.82). Forecasting of monthly maximum SST concentrations for the two-year period (2013 and 2014) was also successful (R2 = 0.69, NS = 0.60, K-G = 0.54, d = 0.84).

Effectiveness of heuristic approach for daily sediment flow prediction of Koyna river basin

The aim of this study was to test the applicability of heuristic approach for daily suspended sediment concentration (SSC) prediction of Koyna river basin situated in western Maharashtra region of India. Daily rainfall, runoff and SSC data were used for development of different models. Heuristic based models like artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS), statistical model like multiple linear regressions (MLR) and traditional sediment rating curve (SRC) models were developed for SSC prediction. In order to avoid time consuming and tedious trial and error method by varying inputs, Gamma test was applied to select best inputs for development of different SSC prediction models. 39 ANN models were developed using tansig, logsig and purelin transfer function. 48 ANFIS models were developed using subtractive clustering method of FIS generation. The best model was selected on the basis performance criteria such as higher correlation coefficient (r) and coefficient of determination (R2). It was found that ANN (6-7-1) model with purelin transfer function outperformed all other models with r and R2 values of 0.79 and 0.62, respectively for daily SSC prediction. The traditional SRC model performed worst and could not able to catch the overall shape of observed SSC time series. The coefficient of determination (R2) was increased from 0.42 to 0.62 using heuristic approach than traditional SRC method. It was found that the heuristic approach could be successfully applied for cumulative SSC load estimation which is essential in reservoir operations and managements. It proves the applicability of heuristic approach in hydrologic modelling for better watershed planning and management.

Daily Suspended Sediment Prediction Using Seasonal Time Series and Artificial Intelligence Techniques

Estimating the amount of suspended sediment in rivers correctly is important due to the adverse impacts encountered during the design and maintenance of hydraulic structures such as dams, regulators, water channels and bridges. The sediment concentration and discharge currents have usually complex relationship, especially on long term scales, which can lead to high uncertainties in load estimates for certain components. In this paper, with several data-driven methods, including two types of perceptron support vector machines with radial basis function kernel (SVM-RBF), and poly kernel learning algorithms (SVM-PK), Library SVM (LibSVM), adaptive neuro-fuzzy (NF) and statistical approaches such as sediment rating curves (SRC), multi linear regression (MLR) are used for forecasting daily suspended sediment concentration from daily temperature of water and streamflow in the river. Daily data are measured at Augusta station by the US Geological Survey. 15 different input combinations (1 to 15) were used for SVM-PK, SVM-RBF, LibSVM, NF and MLR model studies. All approaches are compared to each other according to three statistical criteria; mean absolute errors (MAE), root mean square errors (RMSE) and correlation coefficient (R). Of the applied linear and nonlinear methods, LibSVM and NF have good results, but LibSVM generates a slightly better fit under whole daily sediment values.

Stochastic Simulation of Daily Suspended Sediment Concentration Using Multivariate Copulas

An estimation of daily suspended sediment concentration (SSC) is required for water resource and environmental management. The traditional methods for simulating daily SSC focus on modeling the SSCs themselves, whereas the cross-correlation structure between SSC and streamflow has received only minor attention. To address this issue, we propose a stochastic method to generate long-term daily SSC using multivariate copula functions that account for temporal and cross dependences in daily SSCs. We use the conditional copula method to construct daily multivariate distributions to alleviate the complications and workload of parameter estimations using high-dimensional copulas. The observed daily streamflow and SSC data are normalized using the normal quantile transform method to relax the computationally intensive model of building daily marginal distributions. Daily SSCs can thus be simulated through the multivariate conditional distribution using previous daily SSC and concurrent daily streamflow values. The proposed method is rigorously examined by application to a case study at the Pingshan station in the Jinsha River Basin, China, and compared with the bivariate copula method. The results show that the proposed method has a high degree of accuracy, in preserving the statistics and temporal correlation of daily SSC observations, and better preserves the lag-0 cross correlation compared with the bivariate copula method. The multivariate copula framework proposed here can accurately and efficiently generate long-term daily SSC data for water resource and environmental management, which play a critical role in accurately estimating the frequency and magnitude of extreme SSC events.

Sediment modeling of a large-scale basin supported by remote sensing and in-situ observations

Sediment models are important for estimating soil erosion, sediment transport, and deposition at different scales and for disparate scenarios; however, the availability of in-situ measurements may not be sufficient for the appropriate evaluation of model performance. For this reason, this study investigated how surrogate data (water quality [WQ] and remote sensing) could be used as a proxy for suspended sediment concentration (SSC) in order to evaluate the performance of a large-scale erosion and sediment transport model. To achieve this, the daily SSC was simulated, using the Large Scale Sediment Model (MGB-SED), a semi-distributed and conceptual model in which the Modified Universal Soil Loss Equation (MUSLE) is used to estimate sediment yield and an advection-diffusion equation is used to route suspended loads along rivers. Four experiments were performed that compared simulated SSC against: (1) in-situ SSC; (2) turbidity; (3) total suspended solids (TSS); and (4) surface spectral reflectance (SSR). Correlation analysis showed that r-values above 0.5 were found in 62%, 91%, 89% and 83% of the SSC, turbidity, TSS and SSR stations, respectively. Spatially, the model was evaluated in smaller drainage areas, using WQ data. Temporally, the number of SSR observations was greater than the number of in-situ measurements, allowing a better analysis of the temporal dynamics of the simulated SSC. Also, the SSR data improved the spatial coverage and could be employed for even large-scale, ungauged basins. These results bring a new perspective to the application of sediment models in poorly or ungauged basins, since surrogate data can improve the spatial and temporal suspended sediment coverage when compared to traditional SSC measurements, which is a common issue in large-scale sediment modeling.

Long short-term memory for predicting daily suspended sediment concentration

Frequent and accurate estimation of suspended sediment concentration (SSC) in surface waters and hydraulic schemes is of prime importance for proper design, operation and management of many hydraulic projects. in the present study, a long short-term memory (LSTM) was considered for predicting daily suspended sediment concentration in a river. The LSTM extends recurrent neural network with memory cells, instead of recurrent units, to store and output information, easing the learning of temporal relationships on long time scales. To build the model, daily observed time series of river discharge (Q) and SSC in the Schuylkill River in the United States were used. The results of the proposed model were evaluated and compared with the feedforward neural network and the adaptive neuro fuzzy inference system models which were trained using three different learning algorithms and widely used in the literature for prediction of daily SSC. The comparison of prediction accuracy of the models demonstrated that the LSTM model could satisfactory predict SSC time series, and adequately estimate cumulative suspended sediment load (SSL).

Evaluating the performance of four different heuristic approaches with Gamma test for daily suspended sediment concentration modeling.

Accurate prediction of suspended sediment concentration (SSC) carried by a river or watershed basin is essential for understanding the hydrology of basin in terms of water quality, river bed sustainability and aquatic habitats. In this study, four heuristic methods, namely, radial basis neural network (RBNN), self-organizing map neural network (SOMNN), least square support vector regression (LSSVR), and multivariate adaptive regression spline (MARS) were employed for daily SSC modeling at Ashti, Bamini, and Tekra stations located in Godavari River basin, Andhra Pradesh, India. The Gamma test (GT) was utilized for identifying the most significant input variables for the applied heuristic approaches. The results obtained by RBNN, SOMNN, LSSVR, and MARS models were compared with those of the traditional sediment rating curve (SRC). The performance of the models was evaluated based on the root mean square error (RMSE), coefficient of efficiency (COE), Pearson correlation coefficient (PCC), Willmott index (WI), and pooled average relative error (PARE) indices, as well as the visual inspection using line diagram, scatter diagram, and Taylor diagram (TD). The results of comparison revealed that the four heuristic methods gave higher accuracy than the SRC model. Among the heuristic models, the RBNN-3 (RMSE = 0.045, 0.062, 0.131g/l; COE = 0.884, 0.883, 0.914; PCC = 0.955, 0.961, 0.958; and WI = 0.970, 0.963, 0.976) outperformed the other models in simulating daily SSC records in the studied stations.

Time-frequency analysis and simulation of the watershed suspended sediment concentration based on the Hilbert-Huang transform (HHT) and artificial neural network (ANN) methods: A case study in the Loess Plateau of China

Suspended sediment concentration (SSC) time series are highly nonlinear and nonstationary due to numerous influencing factors that can be characterized by specific time scales, thereby increasing the difficulty of performing SSC simulations. Analyzing the spectral and temporal information of the SSC and its contributing factors can improve the resulting simulation efforts. In this study, the Hilbert-Huang transform (HHT) was used to identify the time scales and change trends of the daily SSC, rainfall, runoff and normalized difference vegetation index (NDVI) of the Kuye River in the Loess Plateau of China. Artificial neural network (ANN), multiple linear regression (MLR), hybrid ensemble empirical mode decomposition (EEMD)-ANN and EEMD-MLR models were chosen to simulate the SSC time series. The periods of the rainfall intrinsic mode functions (IMFs, except IMF9 and 10) were generally consistent with those of the IMFs corresponding to the SSC and runoff. In comparison, the periods of the IMFs corresponding to the NDVI were longer. The minimal period was 3 days for the SSC, rainfall and runoff and 10 days for the NDVI. The minimal period displays the greatest variance contribution and plays the most important role in the SSC, rainfall and runoff changes. The annual cycle has the largest contribution to the NDVI variability. Residual components indicated that the SSC of the Kuye River decreased from 2006 to 2010, while the regional rainfall, runoff and NDVI increased. The R2 values of the ANN and MLR models were 0.34 and 0.39, respectively. Compared with the MLR and ANN models, the R2 values of the EEMD-ANN and EEMD-MLR models increased by 52.9% and 41.0%, respectively, indicating that the ANN model was more robust than the MLR model when simulating nonlinear SSC time series. Compared with EMD, EEMD can alleviate the mode mixing problem and significantly increase the simulation accuracy with the ANN and MLR models. Significance tests were employed to determine the strength of the physical meaning in each IMF component. Taking IMFs with significant (p < 0.05) period oscillations as input data, the EEMD-ANN model had a high R2 of 0.64. Using IMFs having significant (p < 0.05) correlations with the SSC as input variables, the EEMD-ANN model achieved the largest R2 of 0.66. The results of this study supply effective means for simulating and predicting soil erosion. However, it still remains a remarkable challenge to accurately simulate the SSC. Additional factors, such as topographic characteristics and land use, should be considered in future research.

Probabilistic Prediction and Forecast of Daily Suspended Sediment Concentration on the Upper Yangtze River

AbstractSediment transport in suspension can represent more than 90% of a river's total annual flux of sediment. In the case of the Yangtze River, more than 99% of the sediment supplied to the sea is suspended load. Suspended sediment is thus an important component of the total sediment load, with implications for channel dynamics, landscape evolution, ecology, and human‐related activities. For hydrological management of large basins such as the Yangtze River, knowledge of the processes governing suspended sediment concentration (SSC) is essential. An analysis of the temporal variation of SSC for the Upper Yangtze basin (defined at Pingshan station) is presented here. For this purpose, a database of 50 years of concurrent discharge and SSC measurements, made by the Yangtze River Commission, is used. The analysis is made using a novel probabilistic data‐driven technique, the Generalized Pareto Uncertainty (GPU). This technique allows for the testing of several strategies of prediction and forecast applied to a time series of SSC and streamflow. Changing between local or seasonal variables to feed these strategies, we inferred that although the main driver of the SSC transport is flow (as reported by previous authors), sediment storage is also a major control. Furthermore, the maximum necessary time lag for forecasts made with the data is on the order of one week, which provides one indication of the time scale of the local processes of SSC transport in the Upper Yangtze. In this paper, limitations and data requirements of the GPU methodology are also discussed.

Hydroclimatic control on suspended sediment dynamics of a regulated Alpine catchment: a conceptual approach

Abstract. We analyse the control of hydroclimatic factors on suspended sediment concentration (SSC) in Alpine catchments by differentiating among the potential contributions of erosion and suspended sediment transport driven by erosive rainfall, defined as liquid precipitation over snow-free surfaces, ice melt from glacierized areas, and snowmelt on hillslopes. We account for the potential impact of hydropower by intercepting sediment fluxes originated in areas diverted to hydropower reservoirs, and by considering the contribution of hydropower releases to SSC. We obtain the hydroclimatic variables from daily gridded datasets of precipitation and temperature, implementing a degree-day model to simulate spatially distributed snow accumulation and snow–ice melt. We estimate hydropower releases by a conceptual approach with a unique virtual reservoir regulated on the basis of a target-volume function, representing normal reservoir operating conditions throughout a hydrological year. An Iterative Input Selection algorithm is used to identify the variables with the highest predictive power for SSC, their explained variance, and characteristic time lags. On this basis, we develop a hydroclimatic multivariate rating curve (HMRC) which accounts for the contributions of the most relevant hydroclimatic input variables mentioned above. We calibrate the HMRC with a gradient-based nonlinear optimization method and we compare its performance with a traditional discharge-based rating curve. We apply the approach in the upper Rhône Basin, a large Swiss Alpine catchment heavily regulated by hydropower. Our results show that the three hydroclimatic processes – erosive rainfall, ice melt, and snowmelt – are significant predictors of mean daily SSC, while hydropower release does not have a significant explanatory power for SSC. The characteristic time lags of the hydroclimatic variables correspond to the typical flow concentration times of the basin. Despite not including discharge, the HMRC performs better than the traditional rating curve in reproducing SSC seasonality, especially during validation at the daily scale. While erosive rainfall determines the daily variability of SSC and extremes, ice melt generates the highest SSC per unit of runoff and represents the largest contribution to total suspended sediment yield. Finally, we show that the HMRC is capable of simulating climate-driven changes in fine sediment dynamics in Alpine catchments. In fact, HMRC can reproduce the changes in SSC in the past 40 years in the Rhône Basin connected to air temperature rise, even though the simulated changes are more gradual than those observed. The approach presented in this paper, based on the analysis of the hydroclimatic control of suspended sediment concentration, allows the exploration of climate-driven changes in fine sediment dynamics in Alpine catchments. The approach can be applied to any Alpine catchment with a pluvio-glacio-nival hydrological regime and adequate hydroclimatic datasets.

Toward multi-day-ahead forecasting of suspended sediment concentration using ensemble models.

This study explores two ideas to made an improvement on the artificial neural network (ANN)-based models for suspended sediment forecasting in several time steps ahead. In this regard, both observed and forecasted time series are incorporated as input variables of the models when applied for more than one lead time. Secondly, least-square ensemble models employing multiple wavelet-ANN models are developed to increase the performance of the single model. For this purpose, different wavelet families are linked with the ANN model and performance of each model is evaluated using error measures. The Skagit River near Mount Vernon in Washington county is selected as the case study. The daily flow discharge and suspended sediment concentration (SSC) in the current day are considered as input variables to predict suspended sediment concentration in the next day. For more lead times, the input structure is updated by adding the forecast of SSC in the previous time step. Results of this study demonstrate that incorporating both observed and predicted variables in the input structure improves performance of conventional models in which those only employ observed time series as input variables. Moreover, ensemble model developed for each lead time outperforms the best single wavelet-ANN model which indicates superiority of the ensemble model over the other one. Findings of this study reveal that acceptable forecasts of daily suspended sediment concentration up to 3days in advance can be achieved using the proposed methodology.

Daily suspended sediment concentration simulation using hydrological data of Pranhita River Basin, India

Simulation of suspended sediment concentration (SSC) in a river is very important for planning and management of water resources. In this study, co-active neuro-fuzzy inference system (CANFIS), multi-layer perceptron neural network (MLPNN), multiple linear and non-linear regressions (MLR and MNLR), and sediment rating curve (SRC) techniques were applied for simulating the daily SSC at Tekra gauging site on Pranhita River, a major tributary of Godavari River basin, Andhra Pradesh, India. The daily data of discharge (m3/s) and SSC (g/l) from June 2000 to November 2003 were used for SSC simulation. The appropriate combination of input variables for CANFIS, MLPNN, MLR and MNLR models were decided using the gamma test (GT). The outcomes from CANFIS, MLPNN, MLR, MNLR and SRC models were compared to observed values of SSC on the basis of root mean squared error (RMSE), Pearson correlation coefficient (r) and coefficient of efficiency (CE). The results indicate the superiority of CANFIS model than the MLPNN, MLR, MNLR and SRC models in simulating daily SSC for the study location.

A comparative study of three different learning algorithms applied to ANFIS for predicting daily suspended sediment concentration

The modeling and prediction of suspended sediment in a river are key elements in global water recourses and environment policy and management. In the present study, an Adaptive Neuro-Fuzzy Inference System model trained with the Levenberg-Marquardt learning algorithm is considered for time series modeling of suspended sediment concentration in a river. The model is trained and validated using daily river discharge and suspended sediment concentration data from the Schuylkill River in the United States. The results of the proposed method are evaluated and compared with similar networks trained with the common Hybrid and Back-Propagation algorithms, which are widely used in the literature for prediction of suspended sediment concentration. Obtained results demonstrate that models trained with the Hybrid and Levenberg-Marquardt algorithms are comparable in terms of prediction accuracy. However, the networks trained with the Levenberg-Marquardt algorithm perform better than those trained with the Hybrid approach.