Using machine learning to characterize solar wind driving of convection in the terrestrial magnetotail lobes
Journal article, Peer reviewed
Published version
![Thumbnail](/bora-xmlui/bitstream/handle/11250/3106662/fspas-10-1180410.pdf.jpg?sequence=6&isAllowed=y)
View/ Open
Date
2023Metadata
Show full item recordCollections
- Department of Physics and Technology [2153]
- Registrations from Cristin [10149]
Original version
Frontiers in Astronomy and Space Sciences. 2023, 10, 1180410. 10.3389/fspas.2023.1180410Abstract
In order to quantitatively investigate the mechanism of how magnetospheric convection is driven in the region of magnetotail lobes on a global scale, we analyzed data from the ARTEMIS spacecraft in the deep tail and data from the Cluster spacecraft in the near and mid-tail regions. Our previous work revealed that, in the lobes near the Moon’s orbit, the convection can be estimated by using ARTEMIS measurements of lunar ions’ velocity. Based on that, in this paper, we applied machine learning models to these measurements to determine which upstream solar wind parameters significantly drive the lobe convection in magnetotail regions, to help us understand the mechanism that controls the dynamics of the tail lobes. The results demonstrate that the correlations between the predicted and measured convection velocities for the machine learning models (>0.75) are superior to those of the multiple linear regression model (∼0.23–0.43) in the testing dataset. The systematic analysis shows that the IMF and magnetospheric activity play an important role in influencing plasma convection in the global magnetotail lobes.