Automatic detection of Ellerman bombs using Deep Learning

Abstract

Ellerman bombs (EBs) are observable signatures of photospheric small-scale magnetic reconnection events. The reliable automatic detection of EBs would enable the study of the impact of magnetic reconnection on the Sun's dynamics. We aim to develop a method to automatically detect EBs in H$\alpha$ observations from the Swedish 1-m Solar Telescope (SST) and in SDO/AIA observations using the 1600Å, 1700Å, 171Å and 304Å passbands. We trained models based on neural networks (NNs) to perform automatic detection of EBs. Additionally, we used different types of NNs to study how different properties contribute to the detection of EBs. We find that for SST observations, the NN-based models are proficient at detecting EBs. With sufficiently high spectral resolution, the spatial context is not required to detect EBs. However, as we degrade the spectral and spatial resolution, the spatial information becomes more important. Models that include both dimensions perform best. For SDO/AIA, the models struggle to reliably distinguish between EBs and bright patches of different origin. Permutation feature importance revealed that the H$\alpha$ line wings (around $\pm$1 Å from line center) are the most informative features for EB detection. For the SDO/AIA case, the 1600Å channel is the most relevant one when used in combination with 171Å and 304Å. The combination of the four different SDO/AIA passbands is not informative enough to accurately classify EBs. From our analysis of a few sample SDO/AIA 1600Å and 1700Å light curves, we conclude that inclusion of the temporal variation may be a significant step towards establishing an effective EB detection method that can be applied to the extensive SDO/AIA database of observations. Abstract modified for ArXiv purposes.

References in corpus (33)

• The Solar Orbiter mission -- Science overview
• CRISP Spectropolarimetric Imaging of Penumbral Fine Structure
• Solar Flare Prediction Using SDO/HMI Vector Magnetic Field Data with a Machine-Learning Algorithm
• CRISPRED: A data pipeline for the CRISP imaging spectropolarimeter
• Ellerman Bombs at high resolution: I. Morphological evidence for photospheric reconnection
• Ellerman bombs at high resolution III. Simultaneous observations with IRIS and SST
• Solar ultraviolet bursts
• Ellerman Bombs at high resolution: II. Visibility, triggering and effect on upper atmosphere
• Ellerman bombs: fallacies, fads, usage
• The European Solar Telescope
• Ellerman bombs and UV bursts: transient events in chromospheric current sheets
• SSTRED: Data- and metadata-processing pipeline for CHROMIS and CRISP
• Magnetic Flux Cancellation in Ellerman Bombs
• The spectral content of SDO/AIA 1600 and 1700 Å filters from flare and plage observations
• Segmentation of coronal holes in solar disk images with a convolutional neural network
• Statistical Analysis of Small Ellerman Bomb Events
• Fast Solar Image Classification Using Deep Learning and its Importance for Automation in Solar Physics
• Dissecting bombs and bursts: non-LTE inversions of low-atmosphere reconnection in SST and IRIS observations
• An observationally-constrained model of strong magnetic reconnection in the solar chromosphere. Atmospheric stratification and estimates of heating rates
• Solar image denoising with convolutional neural networks
• Multi-channel coronal hole detection with convolutional neural networks
• Signatures of ubiquitous magnetic reconnection in the lower solar atmosphere
• Solar Filament Recognition Based on Deep Learning
• Ultra-high resolution observations of plasmoid-mediated magnetic reconnection in the deep solar atmosphere
• Automating Ellerman bomb detection in ultraviolet continua
• Occurrence and Statistics of IRIS Bursts
• Exploring spectropolarimetric inversions using neural fields. Solar chromospheric magnetic field under the weak-field approximation
• Signatures of ubiquitous magnetic reconnection in the deep atmosphere of sunspot penumbrae
• Formation of H$\epsilon$ in the solar atmosphere
• Transition region response to Quiet Sun Ellerman Bombs
• Reconstructing solar magnetic fields from historical observations VIII. AIA 1600 Å contrast as a proxy of solar magnetic fields
• FarNet-II: An improved solar far-side active region detection method
• Designing wavelength sampling for Fabry-Pérot observations. Information-based spectral sampling

Cited by in corpus (1)

• Automatic detection of Ellerman bombs in the H$\alpha$ line