yinjiuxun / das-strain-scaling

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

das_strain_scaling

Code to reproduce results of manuscript "Earthquake magnitude with DAS: a transferable data-based scaling relation"

Data can be downloaded from Caltech DATA: https://doi.org/10.22002/rxtp0-38405

The scripts for regression are in the folder ./regression Scripts for results validation and visualization are in the folder ./validation_prediction

1. Download the data from Caltech DATA: https://data.caltech.edu/records/sk6em-th949, upzip it and rename to data_files in the current directory

2. Download the utility folder, which constains the modules and functions, from https://github.com/yinjiuxun/DASscaling-modules. Move the utility folder to the current folder.

3. Run the Python script in regression to get the results

  • iter_regression.py gives the regression results from the given data sets: (1) combined dataset of 3 California arrays; (2) individual array of Ridgecrest, Long Valley North and Long Valley South, Sanriku

  • The results will be put to the new directories named: iter_results, iter_results_Ridgecrest, iter_results_LongValley_N, iter_results_LongValley_S and iter_results_Sanriku

  • transfer_regression.py gives the transfered regression results: using the coefficients from the combined dataset, and the measurements from 5 randomly chosen events from Sanriku dataset to calibrate the site terms.

4. The scripts and notebooks in the directory validation_prediction can reproduce the figures of the paper

  • check_peak_amplitude_info.ipynb: notebook to reproduce Figure 1 and 2

  • magnitude_estimation.py: script to reproduce Figure 3

  • real_time_estimation.py: script to reproduce Figure 4

  • Others scripts can reproduce Figures in the Supporting Information

About


Languages

Language:Jupyter Notebook 99.5%Language:Python 0.5%