Simulated ground motion dataset in the Azores plateau, Portugal, on median soil.

Karimzadeh, Shaghayegh; Hussaini, Sayed Mohammad Sajad; Caicedo Diaz, Daniel; Carvalho, Alexandra; Rezaeian, Sanaz; Lourenço, Paulo B.

doi:10.5281/zenodo.18055608

Simulated ground motion dataset in the Azores plateau, Portugal, on median soil.

Karimzadeh, Shaghayegh; Hussaini, Sayed Mohammad Sajad; Caicedo Diaz, Daniel; Carvalho, Alexandra; Rezaeian, Sanaz; Lourenço, Paulo B.

Publisher: Zenodo
DOI: 10.5281/zenodo.18055608
Repository: zenodo.org
Language: en

Abstract

Simulated Ground Motion Dataset for the Azores PlateauThis repository contains a simulated ground motion dataset for the Azores Plateau (Portugal), for median soil conditions. The dataset is generated using a stochastic source-based approach with randomized input parameters to capture aleatory uncertainty. Validation confirms that the dataset accurately represents real ground motion variability and inter-period correlations. This dataset supports seismic hazard analysis and risk assessment in the region.

Full text

SimulatedGroundMotionDataset Files 1. dataset_name.h5 (WaveformData) 2. metadata.csv (FlatfileIndex) 1.FileStructure HDF5Structure The .h5 fileusesaflatstructure(nohierarchyfolders).EachrecordisstoredasaGroupnamedafteritsunique simulationID. Root [Unique_Record_Name] (Group) ac (Dataset):1DArray( float32 ).Accelerationvalues. dt (Dataset):Scalar( float32 ).Timestepincrement. Note:Timearraysarenotstoredexplicitlytosavespace.Youmustconstructthetimevectorusing dt andthelength of ac . CSVStructure The .csv filecontainsmetadataforfilteringandquerying.Usethe Name columntolookupthecorrespondingdatain theHDF5file. KeyColumn:Name (MatchestheHDF5GroupName) MetadataColumns:Mw , Rjb , Vs30 , Scenario_ID ,etc. 2.Units&Formats Quantity Unit Precision Acceleration cm/s² float32 Time(dt) Seconds(s) float32 3.PythonUsageExample Requirements:h5py , pandas , numpy importh5py importpandasaspd importnumpyasnp #FILES h5_file='dataset_name.h5' csv_file='metadata.csv' #1.LOADMETADATA&FILTER #WeusetheCSVtofindthesimulationwewant meta=pd.read_csv(csv_file) target_record=meta[meta['Mw']>6.5].iloc[0]['Name'] print(f"Readingrecord:{target_record}") #2.READWAVEFORMFROMHDF5 withh5py.File(h5_file,'r')asf:  #AccesstheGroup(usingtheNamefromCSV) grp=f[target_record]  #ReadData #[:]readsthefullarray,[()]readsascalar acc=grp['ac'][:] dt=grp['dt'][()]  #ReconstructTimeVector time=np.arange(len(acc))*dt  print(f"DataLoaded:{len(acc)}points,dt={dt}s,Duration={time[-1]:.2f}s")