Solar wind in situ data suitable for machine learning (python numpy structured arrays): STEREO-A/B, Wind, Parker Solar Probe, Ulysses, Venus Express, MESSENGER
These are solar wind in situ data arrays in python pickle format suitable for machine learning, i.e. the arrays consist only of numbers, no strings and no datetime objects. See AAREADME_insitu_ML.txt for more explanation. If you use these data for peer reviewed scientific publications, please get in touch concerning usage and possible co-authorship by the authors (C. Möstl, A. J. Weiss, R. L. Bailey, R. Winslow, A. Isavnin, D. Stansby): christian.moestl@oeaw.ac.at or twitter @chrisoutofspace Made with https://github.com/cmoestl/heliocats Load in python with e.g. for Parker Solar Probe data: > import pickle> filepsp='psp_2018_2019_sceq_ndarray
Resource information
- Data purpose
- Research dataset
- Task
- AI research
- Subject
- 20109 Space and Solar Physics, FOS: Physical sciences, FOS: Physical sciences
- Dataset size
- 2814521120 Bytes
- DOI
- 10.6084/m9.figshare.12058065.v7
- Credit basis
- Creators in DataCite DOI metadata
- Publisher
- figshare
- Upstream last modified
- Jun 1, 2025
- Access
- Files remain with the original publisher
- Last checked
- Nov 6, 2026
- License
- MIT License
- Last record update
- Nov 6, 2026
About this resource
These are solar wind in situ data arrays in python pickle format suitable for machine learning, i.e. the arrays consist only of numbers, no strings and no datetime objects. See AAREADME_insitu_ML.txt for more explanation. If you use these data for peer reviewed scientific publications, please get in touch concerning usage and possible co-authorship by the authors (C. Möstl, A. J. Weiss, R. L. Bailey, R. Winslow, A. Isavnin, D. Stansby): christian.moestl@oeaw.ac.at or twitter @chrisoutofspace Made with https://github.com/cmoestl/heliocats Load in python with e.g. for Parker Solar Probe data: > import pickle> filepsp='psp_2018_2019_sceq_ndarray.p'> [psp,hpsp]=pickle.load(open(filepsp, "rb" ) ) plot time vs total field> import matplotlib.pyplot as plt> plt.plot(psp['time'],psp['bt']) Times psp[:,0 ] or psp['time'] are in matplotlib format. Variable 'hpsp' contains a header with the variable names and units for each column. Coordinate systems for magnetic field components are RTN (Ulysses), SCEQ (Parker Solar Probe, STEREO-A/B, VEX, MESSENGER), HEEQ (Wind) available parameters: bt = total magnetic fieldbxyz = magnetic field componentsvt = total proton speedvxyz = velocity components (only for PSP)np = proton densitytp = proton temperaturexyz = spacecraft position in HEEQr, lat, lon = spherical coordinates of position in HEEQ