Modelling reservoir sedimentation and estimating historical deposition rates using a data-based mechanistic (DBM) approach

Abstract

The data-based mechanistic (DBM) modelling methodology is applied to the study of reservoir sedimentation. A lumped-parameter, discrete-time model has been developed which directly relates rainfall to suspended sediment load (SSL) at the reservoir outflow from the two years of measured data at Wyresdale Park Reservoir (Lancashire, UK). This nonlinear DBM model comprises two components: a rainfall to SSL model and a second model, relating the SSL at the reservoir inflow to the SSL at the reservoir spillway. Using a daily measured rainfall series as the input, this model is used to reconstruct daily deposition rates between 1911 and 1996. This synthetic sediment accretion sequence is compared with the variations in sand content within sediment cores collected from the reservoir floor. These profiles show good general agreement, reflecting the importance of low reoccurrence, high magnitude events. This preliminary study highlights the potential of this DBM approach, which could be readily applied to other sites.     

Complex behaviour of suspended sediment grain size downstream from a reservoir: an example from the Hanjiang River,China

Abstract

A study on the suspended sediment transportation downstream from the Danjiangkou Reservoir in China has shown that the dynamics of suspended sediment grain size are complicated. During the period when the reservoir was used for flood retention, the suspended sediment median size decreased gradually; after entering the period when the reservoir was used for water storage, the median size started to increase, reaching a maximum, and then decreased again. These variations correspond to different stages of channel adjustment. At the stage with dominant downcutting, most of the downstream reservoir sediment comes from bed downcutting, and thus the suspended sediment median size becomes coarser and coarser; at the succeeding stage with dominant channel widening, a majority of the suspended sediment comes from bank erosion, and so its median size becomes finer. This phenomenon can be regarded as a reflection of the complex response of channel adjustment in the characteristics of suspended sediment transportation downstream from a reservoir.     

Bias and variance of some estimators of suspended sediment load

Abstract

Walling & Webb (1981) made empirical comparisons amongst a number of methods for the estimation of suspended sediment loads. This paper presents expressions for the means and variances of three of the estimates compared by Walling & Webb, on the assumption that suspended sediment concentration, c, and mean daily discharge, q, are bivariate lognormally distributed. With this assumption, not inconsistent with the literature and with data, one of the three estimates is unbiased and the other two are biassed, the bias being an exponential function of the correlation and standard deviations in the lognormal distribution. This confirms the empirical finding of Walling & Webb and supports theoretical results of Ferguson (1987) which were established without distributional assumptions. Expressions for the variances of the three estimators of suspended sediment loads also confirm the empirical result of Walling & Webb, the unbiassed estimate having variance of order 1/n whilst the biassed estimates have variances of order 1/n².     

Temporal trend of the suspended sediment transport of the Amazon River (1984–2016)

ABSTRACT

We discuss the claim that the Amazon River has been subjected to a noticeable increase in suspended sediment transport (SST) in response to both climate and land-use changes. To study this, both satellite imagery and in situ data were compiled to produce a 32-year time series (1984–2016) of suspended sediment concentration. Both parametric and nonparametric statistics were applied to examine the SST time trend. The results indicate that there has been no statistically significant increase in SST in the last 32 years, independent of the statistical approach. The results indicate that, over the last 32 years at the Óbidos station, in Brazil, a recurring pattern of increase and decrease in SST has occurred, rather than a unidirectional systematic trend. This further explains the increasing trend reported in the literature and indicates that short time series are not recommended for time trend analyses due to the large inter-annual variability.     

Assessment of sediment yield estimations for large watershed areas: a case study for the Seyhan,Demirköprü and Hirfanlı reservoirs in Turkey

Abstract

Analyses of data from reservoir surveys and sediment rating curves are compared to predict sediment yield in three large reservoir watershed areas in Turkey. Sediment yield data were derived from reservoir sedimentation rates and suspended sediment measurements at gauging stations. The survey data were analysed to provide the volume estimates of sediment, the time-averaged sediment deposition rates, the long-term average annual loss rates in the reservoir storage capacity, and the long-term sediment yield of the corresponding watershed areas. Four regression methods, including linear and nonlinear cases, were applied to rating curves obtained from gauging stations. Application of the efficiency test to a power function form of a rating curve with nonlinear regression yielded the highest efficiency values. Based on the analysis of the sediment rating curves, sediment load fluxes were calculated by using average daily discharge data at each gauging station. Comparison of these two sediment yield values for each reservoir showed that the sediment yields from the suspended sediment measurements, SY_GS, are 0.99 to 3.54 times less than those obtained from the reservoir surveys, SY_RS. The results from the reservoir surveys indicate that all three reservoirs investigated have lost significant storage capacity due to high sedimentation rates.     

A.—A. I. H. S.—I. A. S. H.

Abstract

When a river flows into a reservoir, the solid materials it carries sink to the bottom thus reducing the potential reservoir capacity. The suspended load of eight major western rivers flowing into Kainji Lake was determined as a part of the investigation of the suspended sediment of the reservoir. The sediment content in each sample was determined in parts per million.

On the whole the results of this study is in agreement with the previous conclusion that the sediment load of the Niger was low. A range of 29–221 ppm was observed within the reservoir whereas the concentration in the inflowing rivers varied from 122–953 ppm. It was however not possible to determine the total sediment load because discharge data were lacking.

The data presented herein confirm the earlier conclusion that the rate of silting is slow but indicate that locally high rate of sediment supply should give cause for anxiety regarding the possible growth of deltas as the months of these rivers.

The full potential of this study is difficult to assess at this stage but it has revealed a number of problem areas worth looking into in future years. A number of future studies believed to be capable of giving the necessary data for operation and management of the reservoir are recommended.     

Development of a real-time forecasting model for turbidity current arrival time to improve reservoir desilting operation

ABSTRACT

Among various strategies for sediment reduction, venting turbidity currents through dam outlets can be an efficient way to reduce suspended sediment deposition. The accuracy of turbidity current arrival time forecasts is crucial for the operation of reservoir desiltation. A turbidity current arrival time (TCAT) model is proposed. A multi-objective genetic algorithm (MOGA), a support vector machine (SVM) and a two-stage forecasting technique are integrated to obtain more effective long lead-time forecasts of inflow discharge and inflow sediment concentration. The multi-objective genetic algorithm (MOGA) is applied for determining the optimal inputs of the forecasting model, support vector machine (SVM). The two-stage forecasting technique is implemented by adding the forecasted values to candidate inputs for improving the long lead-time forecasting. Then, the turbidity current arrival time from the inflow boundary to the reservoir outlet is calculated. To demonstrate the effectiveness of the TCAT model, it is applied to Shihmen Reservoir in northern Taiwan. The results confirm that the TCAT model forecasts are in good agreement with the observed data. The proposed TCAT model can provide useful information for reservoir sedimentation management during desilting operations.     

Announcements

Abstract

This paper investigates for a 25-year period the sediment distribution in a semi-arid Brazilian basin (2 × 10⁴ km²) with a network containing more than 4000 surface reservoirs. The methodology is based on rating curves and fitted parameters derived from field data. The results showed that suspended load corresponded to 70% of the total sediment yield (148 t km^-2 year^-1). The relatively low contribution of the suspended load (compared with other semi-arid regions) was attributed to the impact of the numerous upstream reservoirs, which retained 235 t km^-2 year^-1. The micro (<1 hm³), small (1–10 hm³), medium-sized (10–50 hm³), and large or strategic (>50 hm³) reservoirs responded to, respectively, 5, 17, 30 and 48% of the total sediment retention by the reservoir network. This indicates that retention in the non-strategic reservoirs has a positive impact on water availability, since siltation of the strategic reservoirs would be expected to more than double if only such reservoirs existed.

Citation Lima Neto, I. E., Wiegand, M. C. &; de Araújo, J. C. (2011) Sediment redistribution due to a dense reservoir network in a large semi-arid Brazilian basin. Hydrol. Sci. J. 56(2), 319–333.     

A METHODOLOGY FOR THE RAPID EVALUATION OF GROUNDWATER RESOURCES,SAO PAULO STATE,BRAZIL

ABSTRACT

Reservoirs are of necessity always built on the basis of incomplete hydrological information which introduces uncertainty into their design and operation. Since the advent of the electronic digital computer attempts have been made to reduce the uncertainty in hydrological design and reservoir management by the use of synthetic hydrology and simulation. It has been found by simulation that the expected benefits from a proposed reservoir system are often a function of the stochastic process selected for the synthetic hydrology, as well as depending upon the magnitude, and choice of driving parameters (commonly, the mean, variance, lag one serial correlation and Hurst's ‘h’). It is suggested that hydrological records are often two short and most statistical tests too weak for the hydrologist to be able to pick ‘the correct’ synthetic hydrological world with any reasonable degree of certainty. However, it would appear that for many problems and places that there is sufficient hydrological data for the hydrologist to assign probabilities to various prior distributions, and to optimize reservoir management and design by Bayesian decision theory.     

Wavelet and Neuro‐fuzzy Conjunction Approach for Suspended Sediment Prediction

Simulation approaches employed in suspended sediment processes are important in the areas of water resources and environmental engineering. In the current study, neuro‐fuzzy (NF), a combination of wavelet transform and neuro‐fuzzy (WNF), multi linear regression (MLR), and the conventional sediment rating curve (SRC) models were considered for suspended sediment load (S) modeling in a gauging station in the USA. In the proposed WNF model, the discrete wavelet analysis was linked to a NF approach. To achieve this aim, the observed time series of river flow discharge (Q) and S were decomposed to sub time series at different scales by discrete wavelet transform. Afterwards, the effective sub time series were added together to obtain a useful Q and S time series for prediction. Eventually, the obtained total time series were imposed as inputs to the NF method for daily S prediction. The results illustrated that the predicted values by the proposed WNF model were in good agreement with the observed S values and gave better results than other models. Furthermore, the WNF model satisfactorily estimated the cumulative suspended sediment load and produced relatively reasonable predictions for extreme values of S, while NF, MLR, and SRC models provided unacceptable predictions.     

Simulation of sediment transport during flood events: laboratory work and field experiments

Abstract

This paper aims at initiating a fundamental understanding of the suspended load transport of river sediment in unsteady flow. Laboratory erosion tests as well as artificial flood experiments are used to evaluate the influence of the transient regime on the transport efficiency of the flow. The erosion experiments reveal that the transport capacity is augmented when the unsteadiness of the flow increases. However, the influence of the transient regime is counteracted by the cohesive properties of the river bed. Field experiments with artificial floods released from a reservoir into a small canal confirm these findings and show a relationship between the friction velocity and the suspended load transport. An appropriate parameter β is proposed to evaluate the impact of the transient regime on the transport of suspended sediment.     

Process-based suspended sediment carrying capacity of silt-sand sediment in wave conditions

The sediment carrying capacity is one of the fundamental issues in sediment simulation.It is of great importance both in theory and practice to develop process-based approaches for the sediment carrying capacity for a wider range of silt-sand sediment.The current study focuses on the approach for depth-averaged concentration of silt-sand sediment under non-breaking wave conditions.By integrating process-based suspended sediment concentration(SSC) profiles,new synthetic expressions for depth-aver...     

Sensitivity analysis of sediment flux derived by laser diffraction and acoustic backscatter within a reservoir

A combination of two indirect methods to measure sediment flux is presented in this study to evaluate suspended sediment transport in a hydropower reservoir. The acoustic backscatter signal (ABS) from an Acoustic Doppler Current Profiler (ADCP) is therefore applied in pre-defined transects within the reservoir in combination with a Laser In-Situ Scattering Transmissometry – stream lined device (LISST-SL). The stationary LISST-SL derived suspended sediment concentration (SSC) measurements are used to calibrate the ABS. From the LISST-SL measurements a time series of SSC is obtained. This enables, in addition, a comprehensive data analysis to evaluate the influence of natural fluctuations of the SSC on the calculated sediment flux, which should be taken into account when assessing sediment transport. Furthermore SSC measurements are done with the LISST-SL close to the reservoir bed. In areas close to the bed no information regarding the ABS is available from the ADCP measurements due to the side-lobe interference. In various studies the information from the last three valid cells is used for extrapolation. However, as result of a comparison of the LISST-SL measurements with extrapolated SSC values from the ADCP measurements it can be seen that, especially in deep reservoirs, this method has to be adapted to the in-situ conditions.     

An investigation of lag times for rainfall–runoff–sediment yield events in small river basins / Une analyse des temps de réponse d'événements pluie–débit–transport solide au sein de petits bassins versants

Abstract

River basin lag time (LAG), defined as the elapsed time between the occurrence of the centroids of the effective rainfall intensity pattern and the storm runoff hydrograph, is an important factor in determining the time to peak and the peak value of the instantaneous unit hydrograph, IUH. In the procedure of predicting a sedimentgraph (suspended sediment load as a function of time), the equivalent parameter is the lag time for the sedimentgraph (LAG_s ), which is defined as the elapsed time between the occurrence of the centroids of sediment production during a storm event and the observed sedimentgraph at the gauging station. Results of analyses of rainfall, runoff and suspended sediment concentration event data collected from five small Carpathian basins in Poland and from a 2.31-ha agricultural basin, in central Illinois, USA have shown that LAG_s was, in the majority of cases, smaller than LAG, and that a significant linear relationship exists between LAG_s and LAG.     

Monitoring baseline suspended sediment in forested basins: the effects of sampling on suspended sediment rating curves

Abstract

Rating curves are widely used for directly assessing changes in the suspended sediment delivery process and indirectly for estimating total yields. Four sampling methods were simulated over a 31-day record of suspended sediment from the North Fork of the Mad River near Korbel, California. The position and size of the four groups of plotted slope/intercept pairs indicated differences in bias and variance among the methods. Estimates of total yield for the 31-day period and for storms of three sizes were also biased according to sampling method. A standard bias-correcting technique improved yield estimates, but did not remove sampling bias uniformly. Methods of data collection have a large and systematic effect on the estimation of rating-curve parameters and on estimates of suspended sediment yield. Differences attributed to land management may, in fact, result from changes in sampling methods.     

Estimation of daily suspended sediments using support vector machines

Abstract

The use of support vector machines—a new regression procedure in water resources—was investigated for predicting suspended sediment concentration/load in rivers. The method was applied to the observed streamflow and suspended sediment data of two rivers in the USA, which have already been used in earlier studies using soft computing techniques. The estimated suspended sediment values were found to be in good agreement with the observed ones. Negative sediment estimates, which were encountered in the soft computing calculations, are not produced by this method. The results indicate that this approach may give better performance than those described in the literature using different methodologies.     

Estimation of suspended sediment concentration in the Saint John River using rating curves and a machine learning approach

ABSTRACT

Sedimentation in navigable waterways and harbours is of concern for many water and port managers. One potential source of variability in sedimentation is the annual sediment load of the river that empties in the harbour. The main objective of this study was to use some of the regularly monitored hydro-meteorological variables to compare estimates of hourly suspended sediment concentration in the Saint John River using a sediment rating curve and a model tree (M5?) with different combinations of predictors. Estimated suspended sediment concentrations were multiplied by measured flows to estimate suspended sediment loads. Best results were obtained using M5? with four predictors, returning an R² of 0.72 on calibration data and an R² of 0.46 on validation data. Total load was underestimated by 1.41% for the calibration period and overestimated by 2.38% for the validation period. Overall, the model tree approach is suggested for its relative ease of implementation and constant performance.

EDITOR M.C. Acreman; ASSOCIATE EDITOR B. Touaibia     

Modelling daily suspended sediment of rivers in Turkey using several data-driven techniques / Modélisation de la charge journalière en matières en suspension dans des rivières turques à l'aide de plusieurs techniques empiriques

Abstract

The transport of sediment load in rivers is important with respect to pollution, channel navigability, reservoir filling, longevity of hydroelectric equipment, fish habitat, river aesthetics and scientific interest. However, conventional sediment rating curves cannot estimate sediment load accurately. An adaptive neuro-fuzzy technique is investigated for its ability to improve the accuracy of the streamflow—suspended sediment rating curve for daily suspended sediment estimation. The daily streamflow and suspended sediment data for four stations in the Black Sea region of Turkey are used as case studies. A comparison is made between the estimates provided by the neuro-fuzzy model and those of the following models: radial basis neural network (RBNN), feed-forward neural network (FFNN), generalized regression neural network (GRNN), multi-linear regression (MLR) and sediment rating curve (SRC). Comparison of results reveals that the neuro-fuzzy model, in general, gives better estimates than the other techniques. Among the neural network techniques, the RBNN is found to perform better than the FFNN and GRNN.     

Time‐varying,nonlinear suspended sediment rating curves to characterize trends in water quality: An application to the Upper Hudson and Mohawk Rivers,New York

This work proposes two modelling frameworks for diagnosing temporal variations in nonlinear rating curves that describe suspended sediment–discharge relationships. A variant of the weighted regression on time, discharge, and season model is proposed and is compared against dynamic nonlinear modelling, a newly developed nonlinear time series filter based on sequential Monte Carlo sampling. Both approaches estimate a time series of rating curve parameters, with uncertainty, that can be used to diagnose variability in the sediment–discharge relationship over time. We evaluate the models with a variety of synthetic scenarios to highlight their ability to estimate signals of known rating curve change. Results reveal important bias‐variance trade‐offs unique to each approach, and in general, suggest that dynamic nonlinear modelling is better suited for rapid rating curve changes, whereas the weighted regression on time, discharge, and season variant more precisely estimates slow change. The techniques are then applied in two case studies in the Upper Hudson and Mohawk Rivers in New York. We conclude with a discussion of the implications of dynamic rating curves for the management of water quality in riverine and estuary systems.