The widely used inversion code HAZEL was modified to analyze He I Stokes parameters observed during flares. We implemented a new parameter (beta) into the code, which addresses the emission in the profiles. The upgraded code was uploaded to github and is available to the scientific community.
The modified inversion code was applied to data of an M-class flare using a two-fold strategy. Initially, all the data was inverted with one atmospheric component, followed by inversions with two atmospheric components. The Bayesian Information Criterion (BIC) was then employed for each pixel to choose between one or two atmospheric components.
For the first time, the magnetic field was inferred using the He I triplet during a high-energetic flare. The results of the inversions are smooth, assuring that the inversion process was successful. Enhancements of up to 1000 G in the magnetic field strength in the active part of the flare are detected. The line-of-sight inclination shows significant changes at the borders of the flare.
For the analysis of the second data set #2 (X-class flare), a secondment at the University of Bern was carried out. The data was inverted using the same strategy as for data set #1. A partial success was achieved in at least half of the field-of-view. The reason might be some additional contamination of cross-talk and, even more likely, the complexity of the spectral profiles.
The M-class flare inversions were disseminated at the following international conferences/ workshops as an oral presentation: Spanish National Solar Physics Meeting (2023), Methods and techniques used in NLTE inversions Workshop (2023), Annual Meeting of the German Astronomical Society (2023), CLASP 2.1 Science Meeting (2023), and as a poster at the European Space Weather Week (2023).
In replacement of the secondment at University College London (UCL), additional ground-based data in the lower chromosphere of the Ca II 8542 Å line, simultaneously observed with the He I triplet during the M-class flare (data set #1), was analyzed. The data were inverted using the novel NLTE inversion code (DeSIRe). We successfully extracted information regarding the line-of-sight velocities of the plasma before, during, and after the flare.
These results were presented at several conferences: XV Reunion Cientifica de la SEA (2022), Spanish National Solar Physics Meeting (2023), and at the Methods and techniques used in NLTE inversions Workshop (2023).
An instrumentation upgrade for acquiring flare data with high-speed recording cameras was completed and integrated into the GREGOR telescope, Europe's largest solar telescope, now accessible to the scientific community.
Ground-based telescope images require image restoration to correct for wavefront distortion caused by Earth's atmosphere, only achieved with computationally expensive methods like MOMFBD. To address this, we devised a novel approach leveraging neural networks to accelerate image restoration, enabling significantly faster processing times compared to the conventional MOMFBD code.
New observations were conducted during the project using the GREGOR telescope. Through competitive calls, we were awarded a total of 43 observing days over the project's duration. Four days were successful to record new flare data, albeit with relatively low intensity (C-class flare).
