WDC for Geophysics, Beijing(中国地球物理学科中心)
 
   

Author-submitted data information


ID 615
Title Partial codes and data for paper 'Global Distribution of Low Frequency Family Marsquakes From Deep-Learning-Based Polarization Estimation'
Creator Quanhong Li
Subject Deep-learning-based SWIFT method, Single-station polarization estimation, Marsquake
Publisher Xiukuan Zhao
Description Codes and data for demonstrating the process of using SWIFT method to obtain the final prediction value for marsquakes data in the paper 'Global Distribution of Low Frequency Family Marsquakes From Deep-Learning-Based Polarization Estimation'.
'SWIFT_prediction.ipynb' is the script for predicting the polarization of marsquakes and plotting Figures S6-S8.
'ML_Models' is the trained model for predicting the polarization of events. It includes 8 trained models, with filenames in the format: '[Quality]quality_[filter_range]_model.h5'. '[filter_range]' is the frequency band range used for data filtering, such as '0.3_0.5Hz' and '0.3_1Hz'.
'Marsquake_data' includes three component marsquake data, with filenames in the format: '[event_label]_[filter_range]_ [Quality]_zne.mseed'. [event_label] is the event label, such as 'S1000a', 'S1094b'. The raw data was downloaded from IRIS and preprocessed according to the description in the article. Here, 56 events are provided as inputs for the model.
Contributor Zhuowei Xiao, JinLai Hao, Juan Li
Date 2023-2025
Type Python scripts, seismic data and trained deep-learning models
Format Source codes are in python ipynb file format. Data are in miniseed formats. Trained models are in hdf5 format.
URL http://www.geophys.ac.cn/ArticleData/20250403CodesAndMarsquakesData.zip
DOI 10.12197/2024GA016
Source
Language eng
Relation
Coverage
Rights Institute of Geology and Geophysics, Chinese Academy of Sciences