Full seismic waveform analysis combined with transformer neural networks improves coseismic landslide prediction

Type
Article

Authors
Dahal, Ashok
Tanyaş, Hakan
Lombardo, Luigi

KAUST Grant Number
CRG
URF/1/4338-01-01

Date
2024-02-09

Abstract
Seismic waves can shake mountainous landscapes, triggering thousands of landslides. Regional-scale landslide models primarily rely on shaking intensity parameters obtained by simplifying ground motion time-series into peak scalar values. Such an approach neglects the contribution of ground motion phase and amplitude and their variations over space and time. Here, we address this problem by developing an explainable deep-learning model able to treat the entire wavefield and benchmark it against a model equipped with scalar intensity parameters. The experiments run on the area affected by the 2015 M$_{w}$7.8 Gorkha, Nepal earthquake reveal a 16% improvement in predictive capacity when incorporating full waveforms. This improvement is achieved mainly on gentle (~25°) hillslopes exposed to low ground shaking (~0.2 m/s). Moreover, we can largely attribute this improvement to the ground motion before and much after the peak velocity arrival. This underscores the limits of single-intensity measures and the untapped potential of full waveform information.

Acknowledgements
This work used the Dutch national e-infrastructure with the support of the SURF Cooperative using grant No. EINF-7984. This work was funded by the NATO Science for Peace and Security Program (SPS project G6190). We would also like to thank the KAUST competitive research grant (CRG) office for funding support for this research under grant URF/1/4338-01-01. We would like to thank Laura Gnesko from the University of Canterbury for proofreading the paper.

Publisher
Springer Science and Business Media LLC

Journal
Communications Earth & Environment

DOI
10.1038/s43247-024-01243-8

Additional Links
https://www.nature.com/articles/s43247-024-01243-8