Suspended Sediment Load Estimation Research Articles

Estimation of suspended sediment load to the Volta Lake under changing climate using empirical discharge-sediment equations

ABSTRACT Reliable estimates of sediment fluxes into rivers provide valuable information for current and future water resource development and management. In this study, empirical equations relating discharge to suspended sediment loads were developed for the three major tributaries (Black Volta, White Volta, and Oti) of the Volta Lake. The empirical equations were fed with projected future river discharges for the three tributaries, to estimate the suspended sediment fluxes into the Volta Lake for the future (2051–2080) relative to a baseline period (2014–2016). The Soil and Water Assessment Tool (SWAT) was used to model the projected discharges. The SWAT model was driven by downscaled climate projections from two regional climate models (RCA4 and RACMO22 T) forced by two emission scenarios (RCP 4.5 and RCP 8.5) from the IPCC Fifth Assessment. Compared to the baseline, the estimated future sediment fluxes show average increments of about 19% and 28%, respectively under RCP 4.5 and RCP 8.5. This has negative consequences as less water is stored for hydropower generation and the lake water becomes more turbid, which inhibits zooplankton and fishery production. The result underscores the need to improve the management of the Lake's watershed by ensuring proper landcover/use planning, afforestation and enforcing regulations on land degradation and general environmental conservation.

Suspended sediment load modeling using Hydro-Climate variables and Machine learning

Machine learning (ML) models have the potential to improve the accuracy of Suspended Sediment Load (SSL) predictions. However, their ability to incorporate climate data to enhance SSL predictions remains underexplored. This study investigates the efficacy of ML models in modeling sediment transport using hydro-climate variables as input data. Boosted decision tree models, including Xgboost, LightGBM, and Catboost, were employed to estimate SSL using hydro-climate variables such as streamflow and precipitation, as well as maximum and minimum temperatures. The models were trained, optimized, and evaluated using either climate/streamflow data or a combination of both, encompassing 47 gauges across the contiguous United States (CONUS) between 1950 and 2022. The results demonstrate that streamflow data alone is insufficient for accurate SSL estimation in all rivers. Consequently, integrating climate data and streamflow data proved effective in predicting SSL at the daily scale across diverse climate zones with commendable accuracy. The models achieved an average Nash-Sutcliffe Efficiency (NSE) of > 0.66, Kling-Gupta Efficiency (KGE) of > 0.67, and Willmott Index (WI) of > 0.83 during the unseen (test) phase. These findings highlight the potential of ML models, particularly when combining climate data with streamflow, to enhance SSL predictions and improve our understanding of sediment transport dynamics.

Applying a transfer function model to improve the sediment rating curve

ABSTRACT Information relating to suspended sediment load is of particular importance in terms of water management and environmental protection schemes. The sediment rating curve (SRC) is used to express the common relation between suspended sediment load and flow discharge in a basin. SRC, however, does not consider the effect of time lag in flow discharge; instead, it makes a kind of bias in its estimation, toward overestimation or underestimation of a sediment load. In this direction, the current research aims to improve and extend the SRC model by using a transfer function method. The proposed method was applied to a specific basin to better capture the effect of time lag on suspended sediment load estimation. This function estimated the current sediment load in current time. Three case studies were conducted in different climate conditions of Iran to evaluate the application of transfer function with SRC in assessing the relation between flow and suspended sediment discharge. The results revealed that the transfer function method demonstrated high precision in forecasting suspended sediment load, as evidenced by its mean absolute difference (MADIF) and bias criterion calculations. The findings indicated that the proposed method could enhance the accuracy of sediment load estimations in basins with similar characteristics. This approach was based on an unbiased future forecast and provided a high level of confidence. The results suggested that this method could be successfully applied to other basins to improve the accuracy of sediment load estimations.

Mode decomposition based large margin distribution machines for sediment load prediction

Precise suspended sediment load (SSL) prediction is essential for irrigation, hydropower and river management practices. But due to some external factors such as high altitude, heavy monsoon and tropical climate conditions, the data we collect may contain noisy samples. Hence, it becomes a challenging task to accurately predict the SSL in rivers. Therefore, to diminish the influence of noise, the empirical mode decomposition (EMD) -based techniques can be adopted. Moreover, the large margin distribution machine-based regression (LDMR) can deal efficiently with noisy datasets as it simultaneously minimizes the insensitive loss as well as quadratic loss. The significance of this work lies in its contribution to improving the accuracy of suspended sediment load (SSL) prediction in rivers, which has practical implications for various applications such as irrigation, hydropower, and river management. We recognize the challenges posed by external factors, which introduce noise into the collected data, making accurate prediction of SSL difficult. Overall, the significance of this work lies in its novel integration of EMD-based techniques and LDMR-based models, which address the challenges posed by noisy and non-stationary sediment load data. The findings contribute to the field of SSL prediction, offering practical solutions for managing and utilizing river resources more effectively in various domains. The main advantage of the least squares LDMR (LS-LDMR) approach is that it solves a system of linear equations rather than solving a large QPPs unlike SVR, TSVR, LDMR, which makes it computationally efficient. It is well known that the sediment load data is complex and non-stationary in nature. Therefore, for daily SSL prediction, the LDMR and least squares LDMR (LSLDMR) models are embedded with two different decomposition techniques, EMD and ensemble EMD (EEMD), to handle the nonlinear and non-stationary characteristics of sediment load data on the forecasted outputs and to increase prediction ability. The results of the proposed EMD-LDMR, EEMD-LDMR, EMD-LSLDMR and EEMD-LSLDMR are compared with the conventional support vector regression (SVR), twin SVR (TSVR), LDMR and LSLDMR. The performance of the evaluated using the best-fit model using the mean absolute error (MAE), root mean square error (RMSE), symmetric mean absolute percentage error (SMAPE), Willmott’s index (WI), correlation coefficient (CC) and R2. Better or comparable results specify the efficiency of the suggested models. For graphical visualization, scatterplots, prediction error plots and violin plots are shown. Numerical results show that these hybrid models show excellent prediction performance. The overall analysis of the results suggests using EEMD-LSLDMR for SSL estimation.

Satellite-based estimation of daily suspended sediment load using hybrid intelligent models

ABSTRACT This study uses a combination of support vector regression models, particle swarm optimization, and grey wolf optimization algorithms to predict suspended sediment load. For this purpose, The Satellite Precipitation of Climate Hazards Group InfraRed Precipitation with Station (CHIRPS) and Global Land Data Assimilation System (GLDAS) soil moisture products are utilized as the predictors. The prediction models are evaluated based on various visual and quantitative indicators. The Taylor and radar diagrams confirm that the support vector regression-particle swarm optimization best agrees with the observed values. Moreover, the obtained quantitative indices show that the support vector regression-particle swarm optimization model offers better performance than other models used in the present study. The values of the best indices are: Pearson correlation coefficient of 0.997, relative root mean square error of 13.17, percentage bias of 4.05, and Nash-Sutcliffe efficiency of 0.995.

Model identification and accuracy for estimation of suspended sediment load

In the present study, three widely used modeling approaches: (1) sediment rating curve (SRC) and optimized OSRC, (2) machine learning models (ML) (random forest (RF) and Dagging-RF (DA-RF)) and (3) the semi-physically based soil and water assessment tool (SWAT) are applied to predict suspended sediment load (Qs ) at the Talar watershed in Iran. Various graphical and quantitative methods were used to evaluate the goodness of fit. Results indicated that the RF model had the best prediction power in the training phase, while the dagging-RF hybrid algorithm outperformed all other models in the validation phase. The OSRC, RF and DA-RF had ‘very good’ performances based on the NSE in the validation phase, SRC showed ‘good’ performance, while the predicted values using SWAT were ‘satisfactory’. Our results suggest that the OSRC and ML models are more suitable for prediction of Qs in study catchments with poor data availability.

Suspended sediment load estimation in a severely eroded and data poor catchment

AbstractSoil erosion rates are high in many parts of Southern Africa, and are likely to rise because of climate change. Suspended sediment loads (SSL) and yields (SSY) are used to measure and benchmark soil erosion and/or sediment transport rates and determine trajectories of change. Some modelled SSY are available for Southern African catchments, but there is a dearth of contemporary observed data. Northern hemisphere approaches to suspended sediment measurement and the calculation of loads and yields are often unsuited to Southern Africa: locally appropriate methods are required. The manual, flood‐focused suspended sediment sampling programme that we implemented in the eroded and data‐scarce Tsitsa River catchment (Eastern Cape, South Africa) monitored four sub‐catchments from December 2015 to June 2019 at a sub‐daily timestep. We used a discharge‐weighted interpolation SSL estimator, investigating the effects of catchment area, hydrological regime, and sampling strategy on SSL, SSY, and variability, comparing our estimates with modelled results. Discharge increased with catchment area whilst flashiness (expressed by the Richards‐Baker Flashiness Index) mainly decreased, and was similar to that of North American catchments. The sampling frequency required to maintain precision was inversely related to catchment area. Mean annual SSL ranged from 18 121 t year−1 in the 204 km2 Gqukunqa River catchment to 984 267 t year−1 in the 1452 km2 Inxu River catchment. Mean annual SSY ranged from 61 t km2 year−1 in the 432 km2 Pot River catchment to 678 t km2 year−1 in the Inxu River catchment. Data stratification to limit sampling to wet‐season flows did not significantly impact SSL estimates, but year‐round sampling is required to maintain the citizen‐technician sampling network through regular income. Modelled SSY estimates were higher than our measured estimates. Our approach to suspended sediment sampling, load and yield estimation is robust, sustainable, precise, and can be adapted for similar, remote catchments.

Artificial intelligence for suspended sediment load prediction: a review

The estimation of sediment yield concentration is crucial for the development of stream ventures, watershed management, toxins estimation, soil disintegration, floods, and so on. In this study, we summarize various existing artificial intelligence (AI)-based suspended sediment load (SSL) estimation models to calculate the suspended sediment load, to our knowledge to date. The artificial neural network (ANN), generalized regression neural network (GRNN), neuro-fuzzy (NF), genetic algorithm (GA), gene expression programming (GEP), classification and regression tree (CART), linear regression (LR), multilinear regression (MLR), Chi-squared automatic interaction detection (CHAID), extreme learning machine (ELM), and support vector machine (SVM) are among the many AI-based models that have been successfully implemented for sediment load prediction. In this paper, we describe a few popular AI-based models that have been used for SSL prediction. ANN, SVM, and NF had overcome each other in different circumstances of prediction; and all three can be said as good predictors. Models using ANN with ELM or wavelet analysis in some ways are good predictors as their predicted values generally lie closer to the measured value. Performances of the algorithms are usually evaluated by applying various types of performance assessment methods most commonly RMSE, R2, MAE, etc. This review is required to bear some significance to the researchers and hydrologists while seeking models that have been effectively actualized inSSLestimation or in hydrology related aspects, however, mainly focused on the researches between January 2015 and November 2020.

Tidal Asymmetry in Ocean-Boundary Flux and In-Estuary Trapping of Suspended Sediment Following Watershed Storms: San Francisco Estuary, California, USA

Suspended-sediment flux at the ocean boundary of the San Francisco Estuary—the Golden Gate—was measured over a tidal cycle following peak watershed runoff from storms to the estuary in two successive years to investigate sediment transport through the estuary. Observations were repeated during low-runoff conditions, for a total of three field campaigns. Boat-based measurements of velocity and acoustic backscatter were used to calculate water and suspended-sediment flux at a location 1 km landward of the Golden Gate. Suspended-sediment concentration (SSC) and salinity data from up-estuary sensors were used to track watershed-sourced sediment plumes through the estuary. Estimates of suspended-sediment load from the watershed and net suspended-sediment flux for one up-estuary subembayment were used to infer in-estuary trapping of sediment. For both post-storm field campaigns, observations at the ocean boundary were conducted on the receding limb of the watershed hydrograph. At the ocean boundary, peak instantaneous suspended-sediment flux was tidally asymmetric and was greater on flood tides than on ebb tides for all three field campaigns, due to higher average SSC in the cross-section on flood tides. Shear-induced sediment resuspension was greater on flood tides and suggests the presence of an erodible pool outside the estuary. The storms in 2016 led to less export of discharge and sediment from the watershed and greater sediment trapping within one up-estuary subembayment compared to that observed in 2017. Results suggest that substantial trapping of watershed sediments occurred during both storm events, likely due to the formation of estuarine turbidity maxima (ETM) at different locations in the estuary. ETM locations were forced nearer the ocean boundary in 2017. Additional measurements and modeling are required to quantify the long-term sediment flux at the Golden Gate.

Estimation of Suspended Sediment Load Using Artificial Intelligence-Based Ensemble Model

Suspended sediment modeling is an important subject for decision-makers at the catchment level. Accurate and reliable modeling of suspended sediment load (SSL) is important for planning, managing, and designing of water resource structures and river systems. The objective of this study was to develop artificial intelligence- (AI-) based ensemble methods for modeling SSL in Katar catchment, Ethiopia. In this paper, three single AI-based models, that is, support vector machine (SVM), adaptive neurofuzzy inference system (ANFIS), feed-forward neural network (FFNN), and one conventional multilinear regression (MLR) modes, were used for SSL modeling. Besides, four different ensemble methods, neural network ensemble (NNE), ANFIS ensemble (AE), weighted average ensemble (WAE), and simple average ensemble (SAE), were developed by combining the outputs of the four single models to improve their predictive performance. The study used two-year (2016-2017) discharge and SSL data for training and verification of the applied models. Determination coefficient (DC) and root mean square error (RMSE) were used to evaluate the performances of the developed models. Based on the performance measure results, the ANFIS model provides higher efficiency than the other developed single models. Out of all developed ensemble models, the nonlinear ANFIS model combination method was found to be the most accurate method and could increase the efficiency of SVM, MLR, ANFIS, and FFNN models by 19.02%, 37%, 9.73%, and 16.3%, respectively, at the verification stage. Overall, the proposed ensemble models in general and the AI-based ensemble in particular provide excellent performance in SSL estimation.

Implementation of hybrid particle swarm optimization-differential evolution algorithms coupled with multi-layer perceptron for suspended sediment load estimation

River suspended sediment load (SSL) estimation is of importance in water resources engineering and hydrological modeling. In this study, a novel hybrid approach is recommended for SSL estimation in which multi-layer perceptron (MLP) is hybridized with particle swarm optimization (PSO) and then, integrated with differential evolution algorithm (DE) called as MLP-PSODE. The hybrid MLP-PSODE model is implemented to model the SSL of Mahabad river located at northwest of Iran. For the sake of examination of the MLP-PSODE model performance, several techniques including multi-layer perceptron (MLP), multi-layer perceptron integrated with particle swarm optimization (MLP-PSO), radial basis function (RBF) and support vector machine (SVM) are selected as benchmarks. For this purpose, five different scenarios are considered for the modeling. The results indicated that the new hybrid model of MLP-PSODE is successful in estimating SSL by considering single input of discharge (Q) with high accuracy as compared to its alternatives with RMSE = 1794.4 ton·day−1, MAPE = 41.50% and RRMSE = 107.09%, which were much lower than those of MLP based model with RMSE = 3133.7 ton·day−1, MAPE = 121.40% and RRMSE = 187.03%. The developed MLP-PSODE model, not only outperforms its counterparts in terms of accuracy in extreme values estimation, but also it is found as a parsimonious model that incorporates lower number of input parameters in its structure for SSL estimation.

Comparative study of multilayer perceptron-stochastic gradient descent and gradient boosted trees for predicting daily suspended sediment load: The case study of the Mississippi River, U.S.

Monitoring sediment transport is essential for managing and maintaining rivers. Estimation of the sediment load in rivers is fundamental for the study of sediment movement, erosion, and flood control. In the current study, three machine learning models—multi-layer perceptron (MLP), multi-layer perceptron-stochastic gradient descent (MLP-SGD), and gradient boosted tree (GBT)—were utilized to estimate the suspended sediment load (SSL) at the St. Louis (SL) and Chester (CH) stations on the Mississippi River, U.S. Four evaluation criteria including the Correlation Coefficient (CC), Nash Sutcliffe Efficiency (NSE), Scatter Index (SI), and Willmott's Index (WI) were utilized to evaluate the performance of the used models. A sensitivity analysis of the models to the input variables revealed that the current day discharge variable had the most effect on the SSL at both stations, but in the absence of current-day discharge data (Qt), a combination of input parameters including SSLt−3,SSLt−2,SSLt−1,Qt−3,Qt−2,Qt−1 can be used to estimate the SSL. The comparative outcomes indicated the high accuracy of MLP-SGD-5 model with a CC of 0.983, SI of 0.254, WI of 0.991, and NSE of 0.967 at station CH and the MLP-SGD-6 model with a CC of 0.933, SI of 0.576, WI of 0.961, and NSE of 0.867, respectively, at station SL. The results of MLP models were improved by SGD optimization. Therefore, the MLP-SGD method is recommended as the most accurate model for SSL estimation.

Application of soft-computing techniques in forecasting sediment load and concentration

ABSTRACT This paper investigates conventional and soft-computing methods for the estimation of suspended sediment concentration (SSC) and load (SSL) in rivers. Frequently used methods of sediment rate curve (SRC) and multi-nonlinear regression, and soft-computing methods of multi-layer perceptron, multi-linear regression and adaptive neuro-fuzzy inference system are implemented using various hydrological and hydraulic parameters for the Little Kickapoo Creek Watershed, Illinois, USA. All methods performed equally well in the estimation of SSL, without any noticeable outperformance from any from the methods. However, the application of soft-computing methods decreased SSC estimation errors considerably as compared to the results of SRC. The results are significant in the way they reconcile traditionally used hydrological parameters into the soft-computing methods. Overall, soft-computing methods are recommended for the estimation of SSC in rivers because of their reasonably better performance and ease of implementation.

A comparative study of artificial intelligence models for predicting monthly river suspended sediment load

When high precision modelling is required, for example, with the estimation of suspended sediment load (SSL), data-driven models are preferred over physically-based numerical models for their real-time, short-horizon prediction ability. The investigation of SSL, as an important index in engineering practices assessment, like design and operation of the hydraulic structures not only shows the hydrological behaviour of the river, but also illustrates the valuable information about the water quality deterioration, surface-groundwater interaction and land-use changes of the watershed. The following data-driven methods were compared in order to predict SSL at the Seyra gauging station on the Karaj River in Iran: Fuzzy logic (FL), two adaptive neuro-fuzzy inference systems (i.e., ANFIS-GP and ANFIS-FCM models), an artificial neural network (ANN), and least squares support vector machine (LSSVM). Monthly average river flow and SSL data for 50 years were obtained from the Tehran Regional Water Authority (TRWA). The data was first divided into training, validation and test sets and the SSL was then predicted using the ANN, FL, ANFIS, and LSSVM models. The reliability of the applied models was evaluated by the correlation coefficient (R), root mean square error (RMSE), and mean absolute error (MAE). The results showed that the ANFIS models outperformed the ANN, FL, and LSSVM models for predicting SSL using the given input and output data. Overall, the performances of the artificial intelligence models used in the present study were satisfac-tory in predicting the non-linear behaviour of the SSL.

Monthly Suspended Sediment Load Estimation Using Artificial Neural Network and Traditional Models in Krishna River Basin, India

The measurement of sediment yield is essential for getting the information of the mass balance between sea and land. It is difficult to directly measure the suspended sediment because it takes more time and money. One of the most common pollutants in the aquatic environment is suspended sediments. The sediment loads in rivers are controlled by variables like canal slope, basin volume, precipitation seasonality and tectonic activity. Water discharge and water level are the major controlling factor for estimate the sediment load in the Krishna River. Artificial neural network (ANN) is used for sediment yield modeling in the Krishna River basin, India. The comparative results show that the ANN is the easiest model for the suspended sediment yield estimates and provides a satisfactory prediction for very high, medium and low values. It is also noted that the Multiple Linear Regressions (MLR) model predicted an many number of negative sediment outputs at lower values. This is entirely unreality because the suspended sediment result can not be negative in nature. The ANN is provided better results than traditional models. The proposed ANN model will be helpful where the sediment measures are not available.

Artificial Neural Network Model for Estimation of Suspended Sediment Load in Krishna River Basin, India

The correct assessment of amount of sediment during design, management and operation of water resources projects is very important. Efficiency of dam has been reduced due to sedimentation which is built for flood control, irrigation, power generation etc. There are traditional methods for the estimation of sediment are available but these cannot provide the accurate results because of involvement of very complex variables and processes. One of the best suitable artificial intelligence technique for modeling this phenomenon is artificial neural network (ANN). In the current study ANN techniques used for simulation monthly suspended sediment load at Vijayawada gauging station in Krishna river basin, Andhra Pradesh, India. Trial & error method were used during the optimization of parameters that are involved in this model. Estimation of suspended sediment load (SSL) is done using water discharge and water level data as inputs. The water discharge, water level and sediment load is collected from January 1966 to December 2005. This approach is used for modelled the SSL. By considering the results, ANN has the satisfactory performance and more accurate results in the simulation of monthly SSL for the study location.

Suspended sediment load estimation in an ungauged river in south‐eastern Australia

AbstractSoil erosion poses a significant threat to river health as well as the sustainability of soil resources. Management of this issue requires catchment‐specific data to be developed; however, many large catchments are poor in terms of hydrological and sediment transport data. Regional scale (thousands of square kilometres), computer‐based modelling methods are a way to generate such data. This study aims to apply the SedNet model to estimate a sediment budget for a 575 km2, ungauged, agricultural catchment of south‐eastern Australia. The model results are then compared with field measured erosion rates for the catchment, in order to assess the sustainability of soil loss and redistribution across the catchment. SedNet estimated average suspended sediment concentrations between 70 and 120 mg/L, under conditions deemed most representative of the river. These model estimates were comparable with monitoring data, showing suspended sediment concentrations between 30 and 350 mg/L. It was found that soil loss and redistribution across the catchment is low; however, the estimated base sediment loads indicate that river health may be negatively impacted by this. SedNet accurately represented current catchment and river conditions and provided a reliable estimation of sediment yield, demonstrating the ability to estimate sediment loss and redistribution across data‐poor catchments using a multifaceted modelling approach. Methodologies, such as the one presented here, offer the ability to better assess the impacts of erosion on sediment loads to develop strategies to effectively manage excess suspended sediment.

Sediment assessment for a watershed in arid region via neural networks

In the present study, the estimation of suspended sediment load is computed by four Artificial Neural Networks (ANNs) algorithms, Cascade Forward Back Propagation (CFBP), Feed Forward Back Propagation (FFBP), Radial Basis Function (RBF), and Recurrent Neural Network (RNN). Five cases of model input are calibrated to establish the relationship among precipitation, discharge and suspended sediment load. While discharge and rainfall up to four previous days as employed for input, model gives pre-eminent performance. Sensitivity of all models is appraised concerning Nash-Sutcliffe coefficient (ENS) and coefficient of determination (R2) for predicting sediment load. Among all ANNs, MMF (Morgan-Morgan-Finney) model when trained with stream flow as the input in RNN, gives best result with coefficient of determination, R2 as 0.9474, while the values for FFBP, CFBP and RBF are 0.9115, 0.8766 and 0.8511, respectively. Performance of all results show that an MMF model is superior to conventional SRC (Sediment Rating Curve) and MLR (Multiple Linear Regression) models in determining the complex relationship between discharge and suspended sediment load.

Development of the FCM-SVR Hybrid Model for Estimating the Suspended Sediment Load

The accurate estimation of suspended sediment load (SSL) carried by a river is one of the primary issues in river engineering, water resources, and environment projects. Boundary condition and simplification of some important parameters lead to some limitation on the empirical sediment equations which are based on flow and sediment properties. In this study, the potential of a developed Fuzzy C-mean clustering-support vector regression model, as a FCM-SVR hybrid model, was investigated in comparison with sediment rating curve (SRC), artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), and support vector regression (SVR) models for estimating the daily SSL in Sistan River in Iran. Root mean square error (RMSE), mean absolute error (MAE), determination coefficient (R2), and Nash-Sutcliffe efficiency coefficient (NSE) statistics were utilized to evaluate the accuracy of the models. The result showed that FCM-SVR model estimates SSL (with RMSE, MAE, R2, and NSE equal to 34,415.52 ton/day, 12,256.28 ton/day, 0.922, and 0.918 respectively in testing period) more accurate than other models. The RMSE value, which was 50% lower compared to other models, reveals that this model possesses the lowest error than the other models. Also, the obtained results indicated that the estimated SSLs, using the best FCM-SVR, were in good agreement and linear dependence with observed values. Unlike other models, FCM-SVR model appropriately estimates extreme values of SSL.

Suspended sediment load prediction using non-dominated sorting genetic algorithm II

Awareness of suspended sediment load (SSL) and its continuous monitoring plays an important role in soil erosion studies and watershed management. Despite the common use of the conventional model of the sediment rating curve (SRC) and the methods proposed to correct it, the results of this model are still not sufficiently accurate. In this study, in order to increase the efficiency of SRC model, a multi-objective optimization approach is proposed using the Non-dominated Sorting Genetic Algorithm II (NSGA-II) algorithm. The instantaneous flow discharge and SSL data from the Ramian hydrometric station on the Ghorichay River, Iran are used as a case study. In the first part of the study, using self-organizing map (SOM), an unsupervised artificial neural network, the data were clustered and classified as two homogeneous groups as 70% and 30% for use in calibration and evaluation of SRC models, respectively. In the second part of the study, two different groups of SRC model comprised of conventional SRC models and optimized models (single and multi-objective optimization algorithms) were extracted from calibration data set and their performance was evaluated. The comparative analysis of the results revealed that the optimal SRC model achieved through NSGA-II algorithm was superior to the SRC models in the daily SSL estimation for the data used in this study. Given that the use of the SRC model is common, the proposed model in this study can increase the efficiency of this regression model.