Home > Conference Contributions & Seminars > Conference Talks > Contributed Talks > Performance analysis of AI generated solar farside magnetograms in EUHFORIA |
Senthamizh Pavai, Valliappan ; Magdalenic, Jasmina ; Jeong, Hyun-Jin
Talk presented at Machine Learning and Computer Vision in Heliophysics, Sofia, Bulgaria on 2023-04-21
Abstract: EUHFORIA 2.0 (EUropean Heliospheric FORecasting Information Asset, Pomoell & Poedts, 2018) is a 3D MHD model of solar wind and CMEs (coronal mass ejections) which reconstructs the evolution of inner heliospheric plasma over the time. The lack of accurate background solar wind modelling affects the prediction of arrival time of CME in all similar models. The validation of EUHFORIA at multiple radial distances and at multiple vantage points could improve its background solar wind modelling potential. With recent flyby missions like Parker Solar Probe, the performance of EUHFORIA at short radial distances from the Sun can be analysed and improved. Our comparative study of solar wind simulations by EUHFORIA with the in situ data of PSP, shows that using GONG synoptic magnetograms as input to EUHFORIA does not always provide accurate modelling of the solar wind parameters. The inaccuracy of modeled plasma parameters is more often found for the times when the PSP is at the farside of the Sun. Artificial Intelligence (AI)-generated Solar Farside Magnetograms (AISFMs) created using Solar Terrestrial Relations Observatory (STEREO) and Solar Dynamics Observatory (SDO) were made publicly available recently (Jeong et al., 2022). We run EUHFORIA employing the synoptic magnetograms, constructed from the HMI observations and synthetic AISFMs, with an aim to study and improve modelling results at multiple vantage points.
Keyword(s): EUHFORIA ; AI-farside magnetogram
Funding: 2BRAIN_SWIM/2BRAIN_SWIM/2BRAIN_SWIM
The record appears in these collections:
Conference Contributions & Seminars > Conference Talks > Contributed Talks
Royal Observatory of Belgium > Solar Physics & Space Weather (SIDC)