Earthquake Damage Assessment Empowered by AI Remote Sensing: Case Studies in 2023 M7.8 Kahramanmaraş Earthquake, 2023 M6.0 Jishishan Earthquake and 2025 M7.1 Southern Tibetan Plateau Earthquake
Description:
Big earthquakes can result in multiple types of secondary damage, including surface ruptures, structural cracks or failures, landslides, and liquefaction. These damages present immediate risks to human safety. Nevertheless, assessing earthquake damage is often limited by site accessibility and subjective assessment. Remote sensing offers a crucial means to conduct rapid and large-scale earthquake damage assessment. Utilizing various electromagnetic frequencies, remote sensing exhibits different sensitivities to surface features, with microwave capable of penetrating rains and clouds. Here we extract indices of surface disturbances from a range of remote sensing images that cover the periods of the earthquakes. We synergize these damage proxies with PGA, Vs30, geology, topography, and proximity to fault strands. Utilizing specialized machine learning techniques, we quantitatively analyze 24k+ building pixels (approximately 30 m posting) affected by M7.8 Kahramanmaraş earthquake around Turkey-Syria borders on February 6, 2023, 16k+ landslides from M6.0 earthquake in Jishishan County, located between the Qinghai-Tibet Plateau and the Loess Plateau on December 18, 2023, and hundreds of landslides along with tens of thousands of liquefaction events from M7.1 earthquake in Tingri County, Southern Tibetan Plateau on January 7, 2025. In particular, the damage proxy obtained from InSAR coherence differences effectively illustrates surface disturbances caused by ground shaking. At a compound hazard site triggered by the Jishishan earthquake, variations in soil moisture identified through optical imagery indicate that winter flood irrigation on terraced fields may exacerbate the liquefaction process and mudflows upon shaking. The compound hazard resulting from the Jishishan earthquake offers insights into the devastation experienced by the Lajia Settlement, located less than 5 km away, which is associated with the Qijia Culture from the Neolithic to Bronze Age. Both locations bury houses and individuals, with evidence of adults shielding children with their bodies, suggesting the abruptness and similarity of the events.
Session: Earthquake-triggered Ground Failure: Data, Hazards, Impacts and Models [Poster]
Type: Poster
Date: 4/17/2025
Presentation Time: 08:00 AM (local time)
Presenting Author: Xie
Student Presenter: No
Invited Presentation:
Poster Number: 4
Authors
Xie Hu Presenting Author Corresponding Author hu.xie@pku.edu.cn Peking University |
Xiao Yu xyu521@illinois.edu Illinois State Water Survey |
Yueren Xu xuyr@ief.ac.cn China Earthquake Administration |
Yuqi Song yqsong@stu.pku.edu.cn Peking University |
Feng Lin linfeng@stu.pku.edu.cn Peking University |
|
|
|
|
Earthquake Damage Assessment Empowered by AI Remote Sensing: Case Studies in 2023 M7.8 Kahramanmaraş Earthquake, 2023 M6.0 Jishishan Earthquake and 2025 M7.1 Southern Tibetan Plateau Earthquake
Session
Earthquake-triggered Ground Failure: Data, Hazards, Impacts and Models