Crust Macrofracturing as Evidence of Last Deglaciation
This repository contains data and code used for the research.
2D-Code
2D-Code contains scipts and input data for both Moho map computation (moho.py) and map of low S-velocity layer presence (lvsl.py).
Input data
- moho_stations_deg.csv - list of stations in the following format:
StationCode;Longitude(degrees);Latitude(degrees);MohoDepth(km) - moho_class_ext2.csv - list of stations in the following format:
StationCode;Longitude(degrees);Latitude(degrees);0/1
Output
moho.py
-
moho_cv_score.csv - Mean absolute error (MAE) and root mean squared error (RMSE) vs. neighboor count (NN):
NN;MAE;RMSE -
moho_map_data.csv - Moho depth values with rows as X and columns as Y. First column are Xs.
lvsl.py
-
res2_df.csv - ROC AUC value vs. neighboor count (NN):
NN;ROC_AUC -
Vs_low_layer_map_ext_class_nn4_v10.csv - Low S-velocity layer presence map
3D-Code
Input data
-
sveka_xyz_vs_0p10.asc - Vs values under the stations:
x y Depth Vs -
rftomo-hyp.py - generates root mean squared error (RMSE) for various NN and Scale parameter values. Outputs sveka_Vs_rmse.csv
n,Scale,NNb,RMSE -
rftomo-horz.py - generates horizontal profiles for various NN and Scale parameter values. Outputs sveka_Vs_horz_0_40_40.csv
n,x,y,z,Vs,lon,lat -
rftomo-vert.py - generates vertical profiles for various NN and Scale parameter values between given points A(lon1,lat1) and B(lon2,lat2). Units are degrees. Outputs sveka_Vs_vert_J10-J01.csv
n,x,y,z,Vs,lwhere l is distance from A to current point in km.