Measurements of the solar wind propagation delay for L1 to Earth based on ACE and ground-based magnetometer data

This database is the basis for the analysis described in the manuscript

'Timing of the solar wind propagation delay between L1 and Earth based on machine learning'

published in Journal of Space Weather and Space Climate.

https://doi.org/10.1051/swsc/2021026

The database contains the times of 380 interplanetary shocks detected at ACE (T_ACE) which also caused a sudden impulse (T_SI) in the magnetosphere based on ground-based magnetometer data. This information can be found in 'measurement_SW_propagation.txt'.

final_learningset_SWdelay.pickle contains the corresponding ACE data (solar wind speed, ACE position) at time T_ACE for each of the 380 cases.

The datafile can be loaded with Python as follows:

import pickle
with open('final_learningset_SWdelay.pickle', 'rb') as f:
    [learnvector_o,learnvector_m,learnvector_s,learnvector,timevector]=pickle.load(f)

The content is described as follows:

learnvector_o - contains a list of ACE data in its original form, ordering ['rx','ry','rz','vx','vy','vz']

learnvector_m - median of each feature

learnvector_s - standard deviation of each feature

learnvector - contains an array of the standardized data, which have been used to train the ML models.

timevector - contains an array with the vector delay in seconds(first column), flat delay in seconds (second column), and measured solar wind propagation delay in seconds (third column)