Integrated early warning method for landslide acceleration and expansion based on GB-InSAR monitoring

TY - JOUR

T1 - Integrated early warning method for landslide acceleration and expansion based on GB-InSAR monitoring

AU - Xiao, Ting

AU - Tian, Weiming

AU - Segoni, Samuele

AU - Intrieri, Emanuele

AU - Deng, Yunkai

AU - Liao, Yunping

N1 - Publisher Copyright: © 2026 John Wiley & Sons Ltd.

PY - 2026/1

Y1 - 2026/1

N2 - Real-time monitoring and early warning systems for landslides are crucial for minimizing casualties and property losses. The tangential angle method, which assesses the deformation rate of the displacement–time curve at specific instances, has been successfully applied in some cases. However, this method often results in omissions, false alarms and frequent alerts due to its reliance on fixed time windows and single-point displacement measurements. Ground-Based Interferometric Synthetic Aperture Radar (GB-InSAR) is an advanced deformation monitoring technology offering high frequency and accuracy. Nonetheless, it currently lacks a quantitative early warning method that fully leverages surface scene information. To address these challenges, this paper proposes a hybrid intelligent early warning approach based on surface deformation monitoring, comprising a point-based early warning (PEW) method and an area-based early warning (AEW) method. The PEW method enhances the traditional tangential angle approach by adopting a self-adaptive time window, thereby reducing warning errors associated with fixed time intervals. The AEW method facilitates early warnings by detecting landslide expansion behaviours, effectively utilizing the extensive data from surface scene monitoring. The proposed early warning system was validated through a detailed case study of the Jianshan landslide monitored by GB-InSAR. The results demonstrate that both PEW and AEW methods perform effectively within their respective scopes, although each possesses certain information blind spots. The integrated method capitalizes on the strengths of both approaches while mitigating their individual limitations, thereby achieving more accurate and reliable early warnings.

AB - Real-time monitoring and early warning systems for landslides are crucial for minimizing casualties and property losses. The tangential angle method, which assesses the deformation rate of the displacement–time curve at specific instances, has been successfully applied in some cases. However, this method often results in omissions, false alarms and frequent alerts due to its reliance on fixed time windows and single-point displacement measurements. Ground-Based Interferometric Synthetic Aperture Radar (GB-InSAR) is an advanced deformation monitoring technology offering high frequency and accuracy. Nonetheless, it currently lacks a quantitative early warning method that fully leverages surface scene information. To address these challenges, this paper proposes a hybrid intelligent early warning approach based on surface deformation monitoring, comprising a point-based early warning (PEW) method and an area-based early warning (AEW) method. The PEW method enhances the traditional tangential angle approach by adopting a self-adaptive time window, thereby reducing warning errors associated with fixed time intervals. The AEW method facilitates early warnings by detecting landslide expansion behaviours, effectively utilizing the extensive data from surface scene monitoring. The proposed early warning system was validated through a detailed case study of the Jianshan landslide monitored by GB-InSAR. The results demonstrate that both PEW and AEW methods perform effectively within their respective scopes, although each possesses certain information blind spots. The integrated method capitalizes on the strengths of both approaches while mitigating their individual limitations, thereby achieving more accurate and reliable early warnings.

KW - GB-InSAR

KW - hybrid intelligent approach

KW - landslide early warning

KW - surface deformation monitoring

KW - tangential angle method

UR - https://www.scopus.com/pages/publications/105026347859

U2 - 10.1002/esp.70226

DO - 10.1002/esp.70226

M3 - Article

AN - SCOPUS:105026347859

SN - 0197-9337

VL - 51

JO - Earth Surface Processes and Landforms

JF - Earth Surface Processes and Landforms

IS - 1

M1 - e70226

ER -