Spectral Ground Motion Models for Himalayas Using Transfer Learning Technique

ABSTRACT

Predicting robust earthquake spectra is challenging, especially for data sparse regions such as India. Often, alternatives to the traditional data-driven regression analysis are used to develop empirical models for such regions. Advancing these efforts, the present study aims at exploring an alternative machine learning technique called Transfer learning, wherein a non-parametric deep neural network is trained for response (S_a) and Fourier spectra (FAS) of Himalayas, which uses network parameters that were derived for a large comprehensive database (NGA-West2). While the FAS is derived using magnitude, distance, focal depth, and site class, the S_a is scaled using FAS and significant duration.

KEYWORDS:

• Response spectra
• Fourier amplitude spectra
• effective amplitude spectra
• transfer learning
• deep neural network
• Western Himalayas

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Author Contributions

Professor STG has contributed to the conceptualization and supervision of this work and Ms Basu has contributed to data collection and simulation; whereas concept development, analysis and preparation of the manuscript were performed by the corresponding author Bhargavi Podili.

Data Availability Statement

The models developed in this study are given as electronic supplements. Further, the reader can obtain the required estimates for the Himalayan region, using the models derived in this study through the MATLAB codes published at https://github.com/Blu682/SpectralGMMcodes. A demo code showing the methodology is also provided here, which can be used for developing a DNN with transfer learning for other regions. On the other hand, the datasets generated during and/or analysed during the current study are openly available.

Supplementary Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/13632469.2024.2353261.

Additional information

Funding

The authors declare that no funds, grants, or other support were received for this project or during the preparation of this manuscript. The authors have no relevant financial or non-financial interests to disclose.