Machine Learning of Sparse Site Investigation Data for Landslide Risk Assessment

Chao Shi; Yu Wang

doi:10.1061/9780784484975.005

Machine Learning of Sparse Site Investigation Data for Landslide Risk Assessment

Department of Architecture and Civil Engineering

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

Abstract

The landslide is a major natural hazard that can cause catastrophic economic and life losses. Previous studies mainly focus on the evaluation of spatial variability of soil parameters on slope stability, and the effects of stratigraphic variations on slope failure mechanisms have often been overlooked. In this study, a Bayesian supervised learning algorithm, called multiple point statistics (MPS), is adopted to determine the spatial distribution of stratigraphic boundaries from sparse site-specific data and a single training image. Multiple generated realizations of slope stratigraphy serve as input for slope stability analysis. Subsequently, a unified indicator is defined for the landslide risk assessment considering both the exceedance probability and sliding mass. A synthetic slope example is simulated to illustrate the proposed framework. Results indicate that stratigraphic uncertainty has a significant effect on the slope risk, and the calculated risk may be significantly underestimated when only sparse site investigation data are available. © 2023 American Society of Civil Engineers (ASCE).

Original language	English
Title of host publication	GEO-RISK 2023: INNOVATION IN DATA AND ANALYSIS METHODS
Subtitle of host publication	SELECTED PAPERS FROM SESSIONS OF GEO-RISK 2023
Editors	Jianye Ching, Shadi Najjar, Lei Wang
Place of Publication	Reston, Virginia
Publisher	American Society of Civil Engineers
Pages	38-48
ISBN (Electronic)	978-0-7844-8497-5
DOIs	https://doi.org/10.1061/9780784484975.005
Publication status	Published - 2023
Event	Geo-Risk 2023: Advances in Theory and Innovation in Practice - DoubleTree by Hilton, Arlington, United States Duration: 23 Jul 2023 → 26 Jul 2023 https://www.geo-risk.org/

Publication series

Name	Geotechnical Special Publication
Number	345
ISSN (Print)	0895-0563

Conference

Conference	Geo-Risk 2023: Advances in Theory and Innovation in Practice
Country/Territory	United States
City	Arlington
Period	23/07/23 → 26/07/23
Internet address	https://www.geo-risk.org/

Funding

The work described in this paper was supported by a grant from the Research Grant Council of Hong Kong Special Administrative Region (Project no. CityU 11202121) and a grant from Shenzhen Science and Technology Innovation Commission (Shenzhen-Hong Kong-Macau Science and Technology Project (Category C) No: SGDX20210823104002020), China. The financial support is gratefully acknowledged.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access Output

10.1061/9780784484975.005

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Shi, C., & Wang, Y. (2023). Machine Learning of Sparse Site Investigation Data for Landslide Risk Assessment. In J. Ching, S. Najjar, & L. Wang (Eds.), GEO-RISK 2023: INNOVATION IN DATA AND ANALYSIS METHODS: SELECTED PAPERS FROM SESSIONS OF GEO-RISK 2023 (pp. 38-48). (Geotechnical Special Publication; No. 345). American Society of Civil Engineers. https://doi.org/10.1061/9780784484975.005

Shi, Chao ; Wang, Yu. / Machine Learning of Sparse Site Investigation Data for Landslide Risk Assessment. GEO-RISK 2023: INNOVATION IN DATA AND ANALYSIS METHODS: SELECTED PAPERS FROM SESSIONS OF GEO-RISK 2023. editor / Jianye Ching ; Shadi Najjar ; Lei Wang. Reston, Virginia : American Society of Civil Engineers, 2023. pp. 38-48 (Geotechnical Special Publication; 345).

@inproceedings{b489b097db474f2abdc78517cefe0732,

title = "Machine Learning of Sparse Site Investigation Data for Landslide Risk Assessment",

abstract = "The landslide is a major natural hazard that can cause catastrophic economic and life losses. Previous studies mainly focus on the evaluation of spatial variability of soil parameters on slope stability, and the effects of stratigraphic variations on slope failure mechanisms have often been overlooked. In this study, a Bayesian supervised learning algorithm, called multiple point statistics (MPS), is adopted to determine the spatial distribution of stratigraphic boundaries from sparse site-specific data and a single training image. Multiple generated realizations of slope stratigraphy serve as input for slope stability analysis. Subsequently, a unified indicator is defined for the landslide risk assessment considering both the exceedance probability and sliding mass. A synthetic slope example is simulated to illustrate the proposed framework. Results indicate that stratigraphic uncertainty has a significant effect on the slope risk, and the calculated risk may be significantly underestimated when only sparse site investigation data are available. {\textcopyright} 2023 American Society of Civil Engineers (ASCE).",

author = "Chao Shi and Yu Wang",

year = "2023",

doi = "10.1061/9780784484975.005",

language = "English",

series = "Geotechnical Special Publication",

publisher = "American Society of Civil Engineers",

number = "345",

pages = "38--48",

editor = "Jianye Ching and Shadi Najjar and Lei Wang",

booktitle = "GEO-RISK 2023: INNOVATION IN DATA AND ANALYSIS METHODS",

address = "United States",

note = "Geo-Risk 2023: Advances in Theory and Innovation in Practice ; Conference date: 23-07-2023 Through 26-07-2023",

url = "https://www.geo-risk.org/",

}

Shi, C & Wang, Y 2023, Machine Learning of Sparse Site Investigation Data for Landslide Risk Assessment. in J Ching, S Najjar & L Wang (eds), GEO-RISK 2023: INNOVATION IN DATA AND ANALYSIS METHODS: SELECTED PAPERS FROM SESSIONS OF GEO-RISK 2023. Geotechnical Special Publication, no. 345, American Society of Civil Engineers, Reston, Virginia, pp. 38-48, Geo-Risk 2023: Advances in Theory and Innovation in Practice, Arlington, Virginia, United States, 23/07/23. https://doi.org/10.1061/9780784484975.005

Machine Learning of Sparse Site Investigation Data for Landslide Risk Assessment. / Shi, Chao; Wang, Yu.
GEO-RISK 2023: INNOVATION IN DATA AND ANALYSIS METHODS: SELECTED PAPERS FROM SESSIONS OF GEO-RISK 2023. ed. / Jianye Ching; Shadi Najjar; Lei Wang. Reston, Virginia: American Society of Civil Engineers, 2023. p. 38-48 (Geotechnical Special Publication; No. 345).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Machine Learning of Sparse Site Investigation Data for Landslide Risk Assessment

AU - Shi, Chao

AU - Wang, Yu

PY - 2023

Y1 - 2023

N2 - The landslide is a major natural hazard that can cause catastrophic economic and life losses. Previous studies mainly focus on the evaluation of spatial variability of soil parameters on slope stability, and the effects of stratigraphic variations on slope failure mechanisms have often been overlooked. In this study, a Bayesian supervised learning algorithm, called multiple point statistics (MPS), is adopted to determine the spatial distribution of stratigraphic boundaries from sparse site-specific data and a single training image. Multiple generated realizations of slope stratigraphy serve as input for slope stability analysis. Subsequently, a unified indicator is defined for the landslide risk assessment considering both the exceedance probability and sliding mass. A synthetic slope example is simulated to illustrate the proposed framework. Results indicate that stratigraphic uncertainty has a significant effect on the slope risk, and the calculated risk may be significantly underestimated when only sparse site investigation data are available. © 2023 American Society of Civil Engineers (ASCE).

AB - The landslide is a major natural hazard that can cause catastrophic economic and life losses. Previous studies mainly focus on the evaluation of spatial variability of soil parameters on slope stability, and the effects of stratigraphic variations on slope failure mechanisms have often been overlooked. In this study, a Bayesian supervised learning algorithm, called multiple point statistics (MPS), is adopted to determine the spatial distribution of stratigraphic boundaries from sparse site-specific data and a single training image. Multiple generated realizations of slope stratigraphy serve as input for slope stability analysis. Subsequently, a unified indicator is defined for the landslide risk assessment considering both the exceedance probability and sliding mass. A synthetic slope example is simulated to illustrate the proposed framework. Results indicate that stratigraphic uncertainty has a significant effect on the slope risk, and the calculated risk may be significantly underestimated when only sparse site investigation data are available. © 2023 American Society of Civil Engineers (ASCE).

UR - https://www.webofscience.com/wos/woscc/full-record/WOS:001058586800005

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85184386481&origin=recordpage

UR - http://www.scopus.com/inward/record.url?scp=85184386481&partnerID=8YFLogxK

U2 - 10.1061/9780784484975.005

DO - 10.1061/9780784484975.005

M3 - RGC 32 - Refereed conference paper (with host publication)

T3 - Geotechnical Special Publication

SP - 38

EP - 48

BT - GEO-RISK 2023: INNOVATION IN DATA AND ANALYSIS METHODS

A2 - Ching, Jianye

A2 - Najjar, Shadi

A2 - Wang, Lei

PB - American Society of Civil Engineers

CY - Reston, Virginia

T2 - Geo-Risk 2023: Advances in Theory and Innovation in Practice

Y2 - 23 July 2023 through 26 July 2023

ER -

Shi C, Wang Y. Machine Learning of Sparse Site Investigation Data for Landslide Risk Assessment. In Ching J, Najjar S, Wang L, editors, GEO-RISK 2023: INNOVATION IN DATA AND ANALYSIS METHODS: SELECTED PAPERS FROM SESSIONS OF GEO-RISK 2023. Reston, Virginia: American Society of Civil Engineers. 2023. p. 38-48. (Geotechnical Special Publication; 345). Epub 2023 Jul 20. doi: 10.1061/9780784484975.005

Machine Learning of Sparse Site Investigation Data for Landslide Risk Assessment

Abstract

Publication series

Conference

Funding

UN SDGs

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

GRF: Multiscale Machine Learning of Subsurface Stratigraphy from Limited Site-specific Measurements and Prior Geological Knowledge using Iterative Convolutional Neural Networks (CNN)

Cite this

Machine Learning of Sparse Site Investigation Data for Landslide Risk Assessment

Abstract

Publication series

Conference

Funding

UN SDGs

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Projects

GRF: Multiscale Machine Learning of Subsurface Stratigraphy from Limited Site-specific Measurements and Prior Geological Knowledge using Iterative Convolutional Neural Networks (CNN)

Cite this