Machine learning approach for 2D abrasion mapping in Sediment Bypass Tunnels - a case study of Koshibu SBT, Japan

Aug 5, 2026

Sediment Bypass Tunnels (SBTs) mitigate reservoir sedimentation by diverting flood-laden flows, yet hydro-abrasive erosion threatens their sustainability. Predicting abrasion is challenging because of complex coupling between flow hydraulics and sediment transport and the scarcity of high-quality data. This study applies the XGBoost machine learning algorithm to predict spatial abrasion in 2D, using the Koshibu SBT in Japan (about 4 km) as a case study with three experimental scenarios and laser-scanned topography data.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages19

Reading Time48 min

AccessFree to download

Abstract

Paper highlights

Article history

Keywords

Introduction

Engineering Applications of Computational Mechanics Fluid ISSN: (Print) (Online) Journal [homepage: ](homepage: www.tandfonline.com/journals/tcfm20)[www.tandfonline.com/journals/tcfm20](homepage: www.tandfonline.com/journals/tcfm20) Machine learning approach for 2D abrasion mapping in Sediment Bypass Tunnels: a case ﬆudy of Koshibu SBT, Japan Ahmed Emara, Sameh A. Kantoush, Mohamed Saber, Tetsuya Sumi, Vahid Nourani & Emad Mabrouk To cite this article: Ahmed Emara, Sameh A. Kantoush, Mohamed Saber, Tetsuya Sumi, Vahid Nourani & Emad Mabrouk (2025) Machine learning approach for 2 D abrasion mapping in Sediment Bypass Tunnels: a case study of Koshibu SBT, Japan, Engineering Applications of Computational Fluid Mechanics, 19:1, 2444419, DOI: 10.1080/19942060.2024.2444419 To link to this article: [https://doi.org/10.1080/19942060.2024.2444419](https://doi.org/10.1080/19942060.2024.2444419) © 2024 The Author(s) . Published by Informa UK Limited, trading as Taylor & Francis Group. | Published online: 26 Dec 2024. | | | --- | --- | | | Submit your article to this journal | | | Article views: 656 | | | View related articles | | View Crossmark data | | Full Terms & Conditions of access and use can be found at [https://www.tandfonline.com/action/journalInformation?journalCode=tcfm20](https://www.tandfonline.com/action/journalInformation?journalCode=tcfm20) Machine learning approach for 2D abrasion mapping in Sediment Bypass Tunnels: a case study of Koshibu SBT, Japan Ahmed Emara a, b, c, Sameh A. Kantoushb, Mohamed Saberb, Tetsuya Sumib, Vahid Nouranid, e and Emad Mabroukf, g, h a Department of Urban Management, Graduate School of Engineering, Kyoto University, Kyoto, Japan;b Disaster Prevention Research Institute (DPRI), Kyoto University, Kyoto, Japan; c Irrigation Engineering and Hydraulics Department, Alexandria University, Alexandria, Egypt;dCenter of Excellence in Hydro Informatics, Faculty of Civil Engineering, University of Tabriz, Tabriz, Iran; e Faculty of Civil and Environmental Engineering, Near East University, Nicosia, Turkey;fCollege of Engineering and Technology, American University of the Middle East, Egaila, Kuwait; g Department of Mathematics, Faculty of Science, Assiut University, Assiut, Egypt;h Department of Computer Science, Faculty of Computers & Information, Assiut University, Assiut, Egypt ABSTRACT Sediment Bypass Tunnels (SBTs) effectively mitigate reservoir sedimentation by diverting floodladen flows, but they face significant challenges due to hydroabrasive erosion, which compromises their sustainability. Predicting this abrasion is complex duetothe intricate interactions between flow hydraulics and sediment transport, along with limited high-quality data. In this study, we explore, for the first time, the potential of using the XGBoost machine learning algorithm to predict the spatial abrasion of SBTs. The Koshibu SBT in Japan, extending approximately 4 km, was selected as the case study. Three experimental scenarios were evaluated: the entire tunnel, the straight section, and the curved section. A spatial abrasion topography was measured using laser scanning tools with a spatial resolution of 2 cm. The controlling factors for abrasion were developed based on geometric and hydraulic features. The abrasion inventory map, consisting of over 1 million data points indicating damaged and non-damaged sites, was divided equally for training and testing the XGBoost algorithm. Results indicate that the XGBoost model effectively predicts 2D spatial abrasions in SBTs, achieving an overall accuracy of 0.864, exceeding 0.9 in some sections. The developed abrasion map accurately captures various complex patterns throughout the tunnel but has some limitations in areas with small wave-like patterns. Overall, this study demonstrates the potential of machine learning algorithms for predicting tunnel abrasion in SBTs. Paper highlights • This study introduces a validated 2D model for tunn

FAQs

Why is predicting hydro-abrasion in sediment bypass tunnels difficult?

It involves intricate interactions between flow hydraulics and sediment transport, and it is hindered by limited high-quality data.

Which machine learning method is used to predict 2D abrasion mapping in the study?

The study uses the XGBoost machine learning algorithm to predict spatial 2D abrasion.

How was the Koshibu SBT case study set up and evaluated?

The Koshibu SBT in Japan (about 4 km) was analyzed under three scenarios (entire tunnel, straight section, curved section), using laser scanning data at 2 cm resolution and an abrasion inventory map split into training and testing sets.

Aug 5, 2026

Upload by River Wang