An Optimized Stacking Machine Learning Estimator for Sediment Transport Estimation

Emami, Hojjat; Emami, Somayeh; Achite, Mohammed; Smerat, Aseel

doi:10.22115/scce.2025.2151

An Optimized Stacking Machine Learning Estimator for Sediment Transport Estimation

Document Type : Regular Article

Authors

Hojjat Emami ¹

Somayeh Emami ²

Mohammed Achite ³^{, 4}

Aseel Smerat ⁵^{, 6}

¹ Associate Professor, Computer Engineering Department, Faculty of Engineering, University of Bonab, Bonab, Iran

² Ph.D., Department of Water Science and Engineering, University of Tabriz, Tabriz, Iran

³ Professor, Laboratory of Water and Environment, Faculty of Nature and Life Sciences, Hassiba Benbouali University of Chlef, 02180 Chlef, Algeria

⁴ University of Oran 2 Mohamed Ben Ahmed, Georessources, Environment, and Natural Risks Laboratory, Oran, Algeria

⁵ Faculty of Educational Sciences, Al-Ahliyya Amman University, Amman 19328, Jordan

⁶ Centre for Research Impact and Outcome, Chitkara University, Punjab, India

10.22115/scce.2025.2151

Abstract

Accurate estimation of sediment transport is essential for effective water resources management, as sedimentation can decrease dam reservoir capacity, impact water quality, and cause significant changes to river ecosystems. This study introduces a two-level optimized ensemble machine learning model to estimate sediment transport at Algeria’s Oued El Abtal station. The proposed model combines four machine learning algorithms—random forest regression, categorical boosting, gradient boosting regression, and k-nearest neighbor—as base learners, with ridge regression serving as a meta-learner to improve accuracy. Multivariate inputs, including height, instantaneous concentration, and water discharge, are used to estimate sediment discharge. Testing on a dataset of 8,792 hydrological records shows that the stacking model significantly outperforms individual learners and alternative methods. The model achieved superior results on the test data, with an R² of 0.973, an RMSE of 671.147 kg/s, and an MAE of 68.887 kg/s. These results demonstrate a clear advantage over the next best method (DNN), which had an R² of 1.00 and an MAE of 102.175 kg/s. This research highlights the potential of stacking ensembles to accurately estimate sediment transport, providing a dependable method for sustainable water management.

Graphical Abstract

An Optimized Stacking Machine Learning Estimator for Sediment Transport Estimation

Highlights

Accurate estimation of sediment transport is essential for sustainable water resources management.
A two-level optimized ensemble machine learning model is proposed to predict sediment transport.
This model integrates RF, CatBoost, GBR, and KNN as base learners with ridge regression as a meta-learner.
Using 8792 records from Oued El Abtal station (Algeria), the model outperformed individual learners.
The combined model achieved high accuracy (R² = 0.975, RMSE = 642.801, MAE = 38.736).

Keywords

Sediment transport estimation

Ensemble learning

Stacking model

Data analysis

Subjects

Machine Learning

[1] Emami H, Emami S. Application of Whale Optimization Algorithm Combined with Adaptive Neuro-Fuzzy Inference System for Estimating Suspended Sediment Load. J Soft Comput Civ Eng 2021;5:1–14.

[2] E D, İ Y, Ö K. Estimation of total sedimentload concentration obtained by experimentalstudyusingartificial neural networks. Env FluidMech 2007;7:271–288.

[3] Fakharian P, Rezazadeh Eidgahee D, Akbari M, Jahangir H, Ali Taeb A. Compressive strength prediction of hollow concrete masonry blocks using artificial intelligence algorithms. Structures 2023;47:1790–802. https://doi.org/10.1016/j.istruc.2022.12.007.

[4] Ghanizadeh AR, Ziaie A, Khatami SMH, Fakharian P. Predicting resilient modulus of clayey subgrade soils by means of cone penetration test results and back-propagation artificial neural network. J Rehabil Civ Eng 2022;10:146–62. https://doi.org/10.22075/JRCE.2022.25013.1568.

[5] Jain SK. Development of integrated sediment rating curves using ANNs. J Hydraul Eng 2001;127:30–7.

[6] Ciǧizoǧlu HK. Suspended sediment estimation for rivers using artificial neural networks and sediment rating curves. Turkish J Eng Environ Sci 2002;26:27–36.

[7] Tayfur G. Artificial neural networks for sheet sediment transport. Hydrol Sci J 2002;47:879–92. https://doi.org/10.1080/02626660209492997.

[8] Cigizoglu HK, Alp M. Generalized regression neural network in modelling river sediment yield. Adv Eng Softw 2006;37:63–8. https://doi.org/10.1016/j.advengsoft.2005.05.002.

[9] Tayfur G, Guldal V. Artificial neural networks for estimating daily total suspended sediment in natural streams. Nord Hydrol 2006;37:69–79. https://doi.org/10.2166/nh.2005.031.

[10] Raghuwanshi NS, Singh R, Reddy LS. Runoff and Sediment Yield Modeling Using Artificial Neural Networks: Upper Siwane River, India. J Hydrol Eng 2006;11:71–9. https://doi.org/10.1061/(asce)1084-0699(2006)11:1(71).

[11] Alp M, Cigizoglu HK. Suspended sediment load simulation by two artificial neural network methods using hydrometeorological data. Environ Model Softw 2007;22:2–13. https://doi.org/10.1016/j.envsoft.2005.09.009.

[12] Melesse AM, Ahmad S, McClain ME, Wang X, Lim YH. Suspended sediment load prediction of river systems: An artificial neural network approach. Agric Water Manag 2011;98:855–66. https://doi.org/10.1016/j.agwat.2010.12.012.

[13] Mustafa MR, Rezaur RB, Saiedi S, Isa MH. River suspended sediment prediction using various multilayer perceptron neural network training algorithms-A case study in Malaysia. Water Resour Manag 2012;26:1879–97. https://doi.org/10.1007/s11269-012-9992-5.

[14] Mustafa MR, Isa MH, Rezaur RB. Artificial Neural Networks Modeling in Water. Nternational J Civ Environ Eng 2012;6:128–36.

[15] Panahi F, Ehteram M, Emami M. Suspended sediment load prediction based on soft computing models and Black Widow Optimization Algorithm using an enhanced gamma test. Env Sci Pollut Res 2021;21:48253–48273.

[16] Samantaray S, Sahoo A, Satapathy DP, Oudah AY, Yaseen ZM. Suspended sediment load prediction using sparrow search algorithm-based support vector machine model. vol. 14. Nature Publishing Group UK; 2024. https://doi.org/10.1038/s41598-024-63490-1.

[17] Bezak N, Lebar K, Bai Y, Rusjan S. Using Machine Learning to Predict Suspended Sediment Transport under Climate Change. Water Resour Manag 2025;39:3311–26. https://doi.org/10.1007/s11269-025-04108-7.

[18] Kundu S, Swarnkar S, Agarwal A. Bayesian-optimized recursive machine learning for predicting human-induced changes in suspended sediment transport. Environ Monit Assess 2025;197:603. https://doi.org/10.1007/s10661-025-14039-w.

[19] Zeyneb T, Nadir M, Boualem R. Modeling of suspended sediment concentrations by artificial neural network and adaptive neuro fuzzy interference system method–study of five largest basins in Eastern Algeria. Water Pract Technol 2022;17:1058–1081.

[20] Achite M, Katipoğlu OM, Elshaboury N, Tuğrul T, Pandey K. Intercomparison of sediment transport curve and novel deep learning techniques in simulating sediment transport in the Wadi Mina Basin, Algeria. Environ Earth Sci 2025;84. https://doi.org/10.1007/s12665-024-12051-w.

[21] Anaraki MV, Kadkhodazadeh M, Morshed-Bozorgdel A, Farzin S. Predicting rainfall response to climate change and uncertainty analysis: Introducing a novel downscaling CMIP6 models technique based on the stacking ensemble machine learning. J Water Clim Chang 2023;14:3671–91. https://doi.org/10.2166/wcc.2023.477.

[22] Kadkhodazadeh M, Anaraki MV, Kachoueiyan F, Farzin S. Introducing a Novel Double Hybrid Algorithm (DHA) and Developing Its Application for Predicting Air Temperature Under Climate Change Conditions (A Case Study of Iran Country). Eng J 2024;28:31–67. https://doi.org/10.4186/ej.2024.28.11.31.

[23] Farzin S, Valikhan Anaraki M, Kadkhodazadeh M, Morshed-Bozorgdel A. Novel methodology for prediction of missing values in river flow based on convolution neural networks: Principles and application in Iran country. Phys Chem Earth 2025;138. https://doi.org/10.1016/j.pce.2025.103875.

[24] Hancock JT, Khoshgoftaar TM. CatBoost for big data: an interdisciplinary review. J Big Data 2020;7:94. https://doi.org/10.1186/s40537-020-00369-8.

[25] Tahmouresi MS, Niksokhan MH, Ehsani AH. Enhancing spatial resolution of satellite soil moisture data through stacking ensemble learning techniques. Sci Rep 2024;14:1–25. https://doi.org/10.1038/s41598-024-77050-0.

[26] Ileri K. Comparative analysis of CatBoost, LightGBM, XGBoost, RF, and DT methods optimised with PSO to estimate the number of k-barriers for intrusion detection in wireless sensor networks. Int J Mach Learn Cybern 2025:1–20. https://doi.org/10.1007/s13042-025-02654-5.

[27] Srisuradetchai P, Suksrikran K. Random kernel k-nearest neighbors regression. Front Big Data 2024;7:1402384. https://doi.org/10.3389/fdata.2024.1402384.

[28] Fernández-Delgado M, Sirsat MS, Cernadas E, Alawadi S, Barro S, Febrero-Bande M. An extensive experimental survey of regression methods. Neural Networks 2019;111:11–34. https://doi.org/10.1016/j.neunet.2018.12.010.

[29] Aldin Shojaeezadeh S, Al-Wardy M, Reza Nikoo M. Suspended sediment load modeling using Hydro-Climate variables and Machine learning. J Hydrol 2024;633:130948. https://doi.org/10.1016/j.jhydrol.2024.130948.

[30] Stull T, Ahmari H. Estimation of Suspended Sediment Concentration along the Lower Brazos River Using Satellite Imagery and Machine Learning. Remote Sens 2024;16:1727. https://doi.org/10.3390/rs16101727.

[31] Chen L, Nouri Y, Allahyarsharahi N, Naderpour H, Rezazadeh Eidgahee D, Fakharian P. Optimizing compressive strength prediction in eco-friendly recycled concrete via artificial intelligence models. Multiscale Multidiscip Model Exp Des 2025;8:24. https://doi.org/10.1007/s41939-024-00641-x.

[32] Chen L, Fakharian P, Rezazadeh Eidgahee D, Haji M, Mohammad Alizadeh Arab A, Nouri Y. Axial compressive strength predictive models for recycled aggregate concrete filled circular steel tube columns using ANN, GEP, and MLR. J Build Eng 2023;77:107439. https://doi.org/10.1016/j.jobe.2023.107439.

[33] Dadrasajirlou Y, Ghazvinian H, Heddam S, Ganji M. Reference Evapotranspiration Estimation Using ANN, LSSVM, and M5 Tree Models (Case Study: of Babolsar and Ramsar Regions, Iran). J Soft Comput Civ Eng 2022;6:101–18. https://doi.org/10.22115/scce.2022.342290.1434.

[34] Ziari M, Karami H, Ostadi A, Ghazvinian H. Simulation and prediction of hydraulic jump characteristics over expanding rough beds using FLOW-3D and soft computing techniques. J Hydroinformatics 2025;27:88–106. https://doi.org/10.2166/hydro.2025.270.

[35] Mir AA, Patel M. A Comprehensive Review on Sediment Transport, Flow Dynamics, and Hazards in Steep Channels. J Water Manag Model 2024;32. https://doi.org/10.14796/JWMM.C517.

[36] Komasi M, Alizadefard A, Ahmadi M. Examination of players ’ strategies in determining the optimal groundwater exploitation by game theory. Hydrogeol J 2024;32:691–704. https://doi.org/10.1007/s10040-024-02770-6.

[37] Sharghi A, Komasi M, Ahmadi M. Variable sensitivity analysis in groundwater level projections under climate change adopting a hybrid machine learning algorithm. Environ Model Softw 2025;183:106264. https://doi.org/10.1016/j.envsoft.2024.106264.

[38] Shariati M, Saeed M, Behzad M, Fazel G, Masoud A. A novel hybrid extreme learning machine – grey wolf optimizer ( ELM ‑ GWO ) model to predict compressive strength of concrete with partial replacements for cement. Eng Comput 2020. https://doi.org/10.1007/s00366-020-01081-0.

[39] Delcey M, Cheny Y, Keck JB, Gans A, de Richter SK. Identification of settling velocity with physics informed neural networks for sediment Laden flows. Comput Methods Appl Mech Eng 2024;432:117389. https://doi.org/10.1016/j.cma.2024.117389.