Advanced Journal of Environmental Science and Technology

ISSN 2756-3251

Advanced Journal of Environmental Science and Technology | Vol. 17 (10), pp. 375-389, October, 2026
DOI: 10.46882/2026.AJEST.171026

Research Article

Acoustic Emission Tomography and Machine Learning Detect Subsurface Dam Seepage and Internal Piping Erosion in Earthen Embankments

Henrik Lindqvist¹*, Adesola Ogunleye², Mei-Ling Chen³

¹Department of Civil and Architectural Engineering, KTH Royal Institute of Technology, Stockholm, Sweden
²Department of Civil Engineering, Federal University of Technology Akure, Akure, Nigeria
³Department of Hydraulic and Ocean Engineering, National Cheng Kung University, Tainan, Taiwan

Corresponding Author: Henrik Lindqvist — [email protected]

Abstract

Internal backward erosion piping causes sudden catastrophic failures in aging earthen embankment dams before surface deformation becomes visible. We embedded fiber-optic distributed acoustic sensing cables and piezoelectric hydrophone arrays across a 12 m pilot embankment subjected to controlled hydraulic gradient escalation. A convolutional neural network trained on acoustic emission wavelets localized nascent seepage channels within ± 0.35 m spatial accuracy and classified particle detachment stages at 95.4% precision. Early detection occurred 48 hours prior to turbid tailwater discharge, enabling proactive grouting intervention.

Keywords: earthen dam safety; internal piping erosion; acoustic emission tomography; distributed fiber-optic sensing; machine learning; geotechnical monitoring

Manuscript Timeline: Received 18 September, 2026; Revised 20 October, 2026; Accepted 29 October, 2026; Published 31 October, 2026.

Citation: Lindqvist, H., Ogunleye, A., and Chen, M.-L. (2026). Acoustic Emission Tomography and Machine Learning Detect Subsurface Dam Seepage and Internal Piping Erosion in Earthen Embankments. Advanced Journal of Environmental Science and Technology, 17(10), pp. 375-389.