Investigation of the Middle Corona with SWAP and a Data-Driven Non-Potential Coronal Magnetic Field Model

TL;DR

This study uses a global non-potential magnetic field model driven by HMI magnetograms to simulate the solar corona and compares the results with SWAP EUV observations, demonstrating good reproduction of large-scale structures and identifying areas for improvement.

Contribution

The paper presents a data-driven non-potential magnetic field model that effectively reproduces large-scale coronal structures observed by SWAP, advancing the understanding of coronal magnetic field evolution.

Findings

Model accurately reproduces large-scale off-limb structures.

Discrepancies mainly occur off the east limb due to unseen active regions.

Incorporating late active region emergences improves model accuracy.

Abstract

The large field-of-view of the Sun Watcher using Active Pixel System detector and Image Processing (SWAP) instrument on board the PRoject for Onboard Autonomy 2 (PROBA2) spacecraft provides a unique opportunity to study extended coronal structures observed in EUV in conjunction with global coronal magnetic field simulations. A global non-potential magnetic field model is used to simulate the evolution of the global corona from 1 September 2014 to 31 March 2015, driven by newly emerging bipolar active regions determined from Helioseismic and Magnetic Imager (HMI) magnetograms. We compare the large-scale structure of the simulated magnetic field with structures seen off-limb in SWAP EUV observations. In particular, we investigate how successful the model is in reproducing regions of closed and open structures; the scale of structures; and compare the evolution of a coronal fan observed over several rotations. The model is found to accurately reproduce observed large-scale off-limb structures. When discrepancies do arise they mainly occur off the east solar limb due to active regions emerging on the far side of the Sun, which cannot be incorporated into the model until they are observed on the Earth-facing side. When such ``late'' active region emergences are incorporated into the model, we find that the simulated corona self-corrects within a few days, so that simulated structures off the west limb more closely match what is observed. Where the model is less successful, we consider how this may be addressed, through model developments or additional observational products.