Daily Suspended Sediment Load Research Articles

Daily suspended sediment load prediction using artificial neural networks and support vector machines

In recent decades, development of artificial intelligence, as a predictor for hydrological phenomenon, has created a great change in predictions. This paper investigates the abilities of Support Vector Machine (SVM) and Artificial Neural Network (ANN) models to predict daily suspended sediment load (SSL) in Doiraj River, located in the west part of Iran. An 11-year data (1994–2004) was applied for predicting SSL. Streamflow and rainfall were used as the model inputs and SSL as the model output. The best input of SVM and ANN models was identified using combination of Gamma Test and Genetic Algorithm (GT–GA). Its results accuracy was compared with the results of conventional correlation coefficient analysis between input and output variables and the best combination was identified. Also, the present study explores Gamma Test to identify the length of the training dataset. Finally, in order to predict SSL, we used the nu-SVR (using the four kernels including linear, polynomial, sigmoid and Radial Basis Function (RBF)) and ANN models (based on BFGS algorithm and Conjugate algorithm). The reliability of SVM and ANN models were evaluated based on performance criteria such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Efficiency Index (EI) and correlation coefficient (R2). The obtained results show that ANN models and nu-SVR model using Gamma Test for input selection has better performance than regression combination. Also, the performance BFGS-ANN model were better than other models with RMSE value and R2 equal to 0.34 (ton) and 0.99, respectively. The nu-SVR model with RBF kernel has more capability in prediction of SSL than the other kernels (RMSE = 0.96 (ton) and R2 = 0.98). In addition, the results show M-test can be used as a new method to determine the number of required data for network training for creating a smooth model by nu-SVR and ANN models.

Suspended sediment modeling using genetic programming and soft computing techniques

Summary Modeling suspended sediment load is an important factor in water resources engineering as it crucially affects the design and management of water resources structures. In this study the genetic programming (GP) technique was applied for estimating the daily suspended sediment load in two stations in Cumberland River in U.S. Daily flow and sediment data from 1972 to 1989 were used to train and test the applied genetic programming models. The effect of various GP operators on sediment load estimation was investigated. The optimal fitness function, operator functions, linking function and learning algorithm were obtained for modeling daily suspended sediment. The GP estimates were compared with those of the Adaptive Neuro-Fuzzy Inference System (ANFIS), Artificial Neural Networks (ANNs) and Support Vector Machine (SVM) results, in term of coefficient of determination, mean absolute error, coefficient of residual mass and variance accounted for. The comparison results indicated that the GP is superior to the ANFIS, ANN and SVM models in estimating daily suspended sediment load.

Estimation of Daily Suspended Sediment Load by Using Wavelet Conjunction Models

Accurate estimation of sediment loads is important for the management and construction of water resources projects. In the first part of this study, the convenient gene expression programming (GEP), neuro-fuzzy (NF), and artificial neural network (ANN) techniques were applied to estimate suspended sediment loads by using recorded daily river discharge and sediment load data. These models were compared with one another in terms of the coefficient of determination, root mean square error, mean absolute error, variance accounted for, and Nash-Sutcliffe statistic criteria. It was found that the GEP model performed better than the NF and ANN models. In the second part of this study, the discrete wavelet conjunction models with convenient GEP, NF, and ANN techniques were constructed and compared with one another. Comparison results indicated that the wavelet conjunction models significantly increased the accuracy of single GEP, NF, and ANN models in suspended sediment estimation. The wavelet-GEP model performed better than the wavelet-NF and wavelet-ANN models.

Wavelet and ANN combination model for prediction of daily suspended sediment load in rivers

In this research, a new wavelet artificial neural network (WANN) model was proposed for daily suspended sediment load (SSL) prediction in rivers. In the developed model, wavelet analysis was linked to an artificial neural network (ANN). For this purpose, daily observed time series of river discharge ( Q) and SSL in Yadkin River at Yadkin College, NC station in the USA were decomposed to some sub-time series at different levels by wavelet analysis. Then, these sub-time series were imposed to the ANN technique for SSL time series modeling. To evaluate the model accuracy, the proposed model was compared with ANN, multi linear regression (MLR), and conventional sediment rating curve (SRC) models. The comparison of prediction accuracy of the models illustrated that the WANN was the most accurate model in SSL prediction. Results presented that the WANN model could satisfactorily simulate hysteresis phenomenon, acceptably estimate cumulative SSL, and reasonably predict high SSL values.

Catchment size and contribution of the largest daily events to suspended sediment load on a continental scale

Abstract The classic approach defines an extreme event as a rare event identified by magnitude–frequency analysis and quantified by its deviation from a central value. They are key to understand geomorphological dynamics, since they are responsible for a considerable amount of work and “time compression”. Time compression means that most of the geomorphic work (particularly sediment transport) is produced in very short temporal intervals (i.e. in few events). Moreover, it is well known from magnitude–frequency analyses that events not necessarily extreme by magnitude could be responsible for a large amount of geomorphic work. To analyse the time compression of geomorphological processes, a new approach has recently been suggested, based on the effects of “largest events”, defined by rank, whatever the magnitude. In this paper, we analyse the relation between the relative contribution of the largest daily event to the total suspended sediment load in relation to catchment size at the continental scale. For this purpose, we used two massive datasets of daily suspended sediment load facilitated by the United States Geological Survey (i.e. USGS Ancillary database), and the Hydrological Service of Canada (i.e. HYDAT dataset). The whole study area covers most of the temperate and sub-polar latitudinal climate conditions. Original data include information from more than 1800 catchments comprising appr. 3 × 10 6 daily events. The data show that the contribution of few events represents a high proportion of the total basin suspended sediment load. For instance, the mean contribution of the 25-largest daily events varies between 63% and 46% of the total load depending on the basin area, while the mean contribution of just the 5-largest events varies between 39% and 23%. The analysis suggests that the relative contribution of different n-largest aggregated daily events on total suspended sediment load decreases as catchment size increases. This is a general pattern and means that, for instance, time compression of suspended load is exacerbated in small catchments; hence, relatively few events are the key to explain the catchment sediment export. In addition, the relative contribution of the n-largest daily event to the suspended sediment load seems to be independent of climate conditions in small catchments, particularly in those 2 .

River Suspended Sediment Load Prediction: Application of ANN and Wavelet Conjunction Model

Accurate suspended sediment prediction is an integral component of sustainable water resources and environmental systems. This study considered artificial neural network (ANN), wavelet analysis and ANN combination (WANN), multilinear regression (MLR), and sediment rating curve (SRC) models for daily suspended sediment load (S) modeling in the Iowa River gauging station in the United States. In the proposed WANN model, discrete wavelet transform was linked to the ANN method. For this purpose, observed time series of river discharge (Q) and S were decomposed into several subtime series at different scales by discrete wavelet transform. Then these subtime series were imposed as inputs to the ANN method to predict one-day-ahead S. The results showed that the WANN model was in good agreement with the observed S values and that it performed better than the other models. The coefficient of efficiency was 0.81 for the WANN model and 0.67, 0.6, and 0.39 for the ANN, MLR, and SRC models, respectively. In addition, the WANN model presented relatively reasonable predictions for extreme S values, acceptably simulated the hysteresis phenomenon, and satisfactorily estimated the cumulative suspended sediment load. Wavelet transforms provide useful decompositions of primary time series, so that wavelet-transformed data improve the ability of a predicting model by capturing useful information on various resolution levels. The proposed WANN model can be considered as a relatively new application of a combined wavelet and ANN model for suspended sediment prediction.

Prediction of daily suspended sediment load using wavelet and neurofuzzy combined model

This study investigated the prediction of suspended sediment load in a gauging station in the USA by neuro-fuzzy, conjunction of wavelet analysis and neuro-fuzzy as well as conventional sediment rating curve models. In the proposed wavelet analysis and neuro-fuzzy model, observed time series of river discharge and suspended sediment load were decomposed at different scales by wavelet analysis. Then, total effective time series of discharge and suspended sediment load were imposed as inputs to the neuro-fuzzy model for prediction of suspended sediment load in one day ahead. Results showed that the wavelet analysis and neuro-fuzzy model performance was better in prediction rather than the neuro-fuzzy and sediment rating curve models. The wavelet analysis and neuro-fuzzy model produced reasonable predictions for the extreme values. Furthermore, the cumulative suspended sediment load estimated by this technique was closer to the actual data than the others one. Also, the model could be employed to simulate hysteresis phenomenon, while sediment rating curve method is incapable in this event.

Cokriging estimation of daily suspended sediment loads

Daily suspended sediment loads (S) were estimated using cokriging (CK) of S with daily river discharge based on weekly, biweekly, or monthly sampled sediment data. They were also estimated with ordinary kriging (OK) and a rating curve method. The estimated daily loads were compared with the daily measured values over a nine-year-period. The results show that the estimated daily sediment loads with the CK using the weekly measured data best matched the measured daily values. The rating curve method based on the same data provides a fairly good match but it tends to underestimate the peak and overestimate the low values. The CK estimation was better than the rating curve because CK considers the temporal correlation among the data values and honors the measured points whereas the rating curve method does not. For the site studied, weekly sampling may be frequent enough for estimating daily sediment loads with CK when daily discharge data is available. The estimated daily loads with CK were less reliable when the sediment samples were taken less frequently, i.e., biweekly or monthly. The OK estimates using the weekly measured data significantly underestimates the daily S because unlike CK and the rating curve, OK makes no use of the correlation of sediment loads with frequently measured river discharge.

Suspended sediment load estimation and the problem of inadequate data sampling: a fractal view

AbstractSuspended sediment load estimation at high resolutions is an extremely difficult task, because: (1) it depends on the availability of high‐resolution water discharge and suspended sediment concentration measurements, which are often not available; (2) any errors in the measurements of these two components could significantly influence the accuracy of suspended sediment load estimation; and (3) direct measurements are very expensive. The purpose of this study is to approach this sampling problem from a new perspective of fractals (or scaling), which could provide important information on the transformation of suspended sediment load data from one scale to another. This is done by investigating the possible presence of fractal behaviour in the daily suspended sediment load data for the Mississippi River basin (at St. Louis, Missouri). The presence of fractal behaviour is investigated using five different methods, ranging from general to specific and from mono‐fractal to multi‐fractal: (1) autocorrelation function; (2) power spectrum; (3) probability distribution function; (4) box dimension; and (5) statistical moment scaling function. The results indicate the presence of multi‐fractal behaviour in the suspended sediment load data, suggesting the possibility of transformation of data from one scale to another using a multi‐dimensional model. Copyright © 2005 John Wiley & Sons, Ltd.

Geomorphic implications of subglacial drainage configuration: rates of basal sediment evacuation controlled by seasonal drainage system evolution

Although basal sediment evacuation by subglacial meltwater is widely assumed to dominate the sediment budget of most temperate glaciers and ice caps, its potential influence for glacial geomorphic processes and landforms has largely been overlooked in the glacial literature. Data collected at Haut Glacier d'Arolla, Switzerland during the 1998 ablation season are used to test theoretical predictions that subglacial drainage configuration should exert a primary control on rates of basal sediment evacuation by subglacial meltwater. Seasonal evolution of the subglacial drainage system enables observation of sediment evacuation from both distributed and channelised drainage systems. Daily suspended sediment load exhibits an extreme, nonlinear increase with respect to discharge between winter and summer periods, which correspond to periods of predominantly distributed and channelised subglacial drainage, respectively. These results suggest that subglacial drainage configuration should significantly impact upon: (a) rates of subglacial erosion and debris production; (b) the nature of debris transport and the proportion of sediment in glacial versus fluvioglacial transport pathways; and (c) rates and styles of ice-marginal sedimentation. In contrast to their distributed counterparts, channelised drainage configurations are predicted to encourage subglacial erosion through the removal of basal sediment from the ice–bedrock interface and, with the exception of supraglacial transport, will discourage sediment transport by alternative glacial mechanisms. Direct sedimentation at the glacier margin may therefore be reduced, resulting in smaller moraines despite an overall increase in sediment production. An appreciation of the coupling between glacier hydrology and geomorphic processes suggests new frameworks for the understanding of rates and styles of glacial erosion, sediment transfer and deposition, and the complex geomorphology of ice-marginal environments.