Fast Spectral Inversion of the H And Ca II 8542 Line Spectra Based on a Deep Learning Model
Abstract
Recently a multilayer spectral inversion (MLSI) model has been proposed to infer the physical parameters of plasmas in the solar chromosphere. The inversion solves a three-layer radiative transfer model using the H alpha and Ca II 8542 Å line profiles taken by the Fast Imaging Solar Spectrograph (FISS). The model successfully provides the physical plasma parameters, such as source functions, Doppler velocities, and Doppler widths in the layers of photosphere to chromosphere. However, it is quite expensive to apply the MLSI to a huge number of line profiles. For example, the calculating time is several hours for a scan raster. We apply deep-learning methods to the inversion code to reduce the cost of calculating the physical parameters. We train the models using pairs of absorption line profiles (H alpha and Ca II 8542 Å) from FISS and their 13 physical parameters (source functions, Doppler velocities, Doppler widths in the chromosphere, and the pre-determined parameters for photosphere) calculated from the spectral inversion code for 50 scan rasters (~2,000,000 dataset) including quiet and active regions. We use a fully connected dense layers for training the model. In addition, we utilize a skip connections to avoid a problem of vanishing gradients. We evaluate the model by comparing the pairs of absorption line profiles and their inverted physical parameters from other quiet and active regions. Our result shows that the deep learning model successfully reproduces physical parameter maps of a scan raster observation per second within 15% of mean absolute percentage error and the mean squared errors of 0.3 to 0.003 depending on the parameters. Taking this advantage of high-performance of the deep learning model, we plan to provide the physical parameter maps from the FISS observations to understand the chromospheric plasma conditions in various solar features.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFMSH44A..01L