Access Data
Special Data Sets
Training and Validation Data Sets for Deep Learning
- P Wave Picking and First Motion Polarity
- Generalized Phase Detection
- Seismic Signal/Noise Discrimination
P Wave Arrival Picking and First‐Motion Polarity Determination With Deep Learning
These files are supplemental material for "P Wave Arrival Picking and First‐Motion Polarity Determination With Deep Learning” doi.org/10.1029/2017JB015251. hdf5 files correspond to the training and validation data sets used in the paper. The trained model and model architecture are also included. For additional information, please contact Zachary Ross (zross@caltech.edu).
Files
- scsn_p_2000_2017_6sec_0.5r_fm_test.hdf5
- scsn_p_2000_2017_6sec_0.5r_fm_train.hdf5
- scsn_p_2000_2017_6sec_0.5r_pick_test.hdf5
- scsn_p_2000_2017_6sec_0.5r_pick_train.hdf5
- model_fm.json
- model_fm_best.hdf5
- model_pick.json
- model_pick_best.hdf5
Generalized Seismic Phase Detection with Deep Learning
These files are supplementary material for “Generalized Seismic Phase Detection with Deep Learning” by Ross et al. (2018), BSSA (doi.org/10.1785/0120180080). The models were trained using keras and TensorFlow, and can be used with these libraries. The training dataset contains 4.5 million seismograms evenly split between P-waves, S-waves, and pre-event noise classes. We encourage the use of this hdf5 dataset for training deep learning models, and hope that it and the model architecture in the paper can serve as a benchmark for future studies. For additional information please contact Zachary Ross (zross@caltech.edu).
Code
Files
Reliable Real-time Seismic Signal/Noise Discrimination with Machine Learning
These files are supplementary material for "Rapid and Reliable Real-time Seismic Signal/Noise Discrimination with Machine Learning." by Meier et al. (2019), Journal of Geophysical Research (doi.org/10.1029/2018JB016661). The data set included here only contains the noise, quake and teleseismic data from the Caltech/USGS Southern California Seismic Network (SCSN). If you use these data from the SCSN please acknowledge:
Analyzed SCSN data; doi: 10.7914/SN/CI; stored at the Southern California Earthquake Data Center. doi:10.7909/C3WD3xH1.
We do not have permission to redistribute the NIED data from Japan, which were used in this study. If you are interested in the Japanese data they are available from: https://www.kyoshin.bosai.go.jp (Aoi, S., Kunugi, T. and Fujiwara, H., 2004. "Strong-motion seismograph network operated by NIED: K-NET and KiK-net." Journal of Japan Association for Earthquake Engineering, 4(3), pp.65-74).
The authors can provide guidance for how to download and process the data into the same format.
For additional information, please contact Men-Andrin Meier (mmeier@caltech.edu).
Files
