Speaker
Description
Contemporary solar physics deals with increasing volumes of high-dimensional data from observations and realistic theoretical models alike. Machine Learning (ML) is rapidly integrated into solar and heliophysics research, facilitating dimensionality reduction, visualization and analysis. We developed a pipeline, which employed carefully selected unsupervised ML algorithms for classification and cluster analysis, to extract information regarding the physical properties of the solar atmosphere contained in the large variety of spectral profiles. The pipeline is tested on the synthetic spectra of the Fe I $\lambda\,7090.38$ Å photospheric absorption line, computed with the CO$^5$BOLD radiation hydrodynamics code. This line is also part of the observing setup for the Fast Multi-line Universal Spectrograph (FaMuLUS) camera system at the Vacuum Tower Telescope (VTT). CO$^5$BOLD snapshot time-series serve as a simulation of high-resolution, fast-cadence solar observations, thus confronting the pipeline with the scenario of dynamic solar feature evolution. This project aims to deliver a robust classification scheme with minimal user interaction, which also prepares the spectral dataset for further analysis, such as spectral inversions. Successful classification allows quick identification of structure and dynamics in the region of interest, as well as diagnostics of the ambient plasma based on spectra line parameters.
| Submit to 'solar physics' topical issue? | Yes |
|---|
