TY - DATA T1 - SeismicDataNeNet PY - 2022/06/24 AU - Stefan Nielsen UR - https://data.4tu.nl/articles/dataset/SeismicDataNeNet/20101901/1 DO - 10.4121/20101901.v1 KW - Earthquakes KW - Neural Network Prediction KW - forecasting method KW - artificail intelligence (AI) N2 -
In this DATASET we replicate seismological broadband data from IRIS Web Services (https://service.iris.edu/) including the IU network. This is the data used for training and testing of the neural network for the paper "Temporal earthquake forecasting".
We provide both originally downloaded data and the version that has been processed (normlaised, standardised) and classified before being used as input for the network.
Original input files are mseed (miniseed)are named TESTxx.mseed (xx is number):
-All files contain information on time interval, the recording station and data parameters in the mseed hearder.
-Events used for training are TESTxx.mseed where xx=(7, 8, 3, 4, 5, 6, 9, 10, 11, 16, 17, 20, 21, 22, 24, 25, 27, 28, 31, 33, 34, 35, 37, 38)
-Events used for testing are TESTyy.msedd where yy=(12, 13, 14, 15, 29, 30, 39)
The standardized, normalized and classified data are in files normpre.p and normnoise.p, containing widows classified as precursor and noise, respectively, for the network training. These files are in format pickle, and are used as input for the network
-File normnoise.p contains the formatted data from all windows classified as noise in all events (first 1000 s of 10 hours file).
-File normpre.p contains the formatted data from all windows classified as precursor in all events (last 1000 s of 10 hours file).
ER -