SolARED: Solar Active Region Emergence Dataset for Machine Learning Aided Predictions

TL;DR

The resulting ML-ready SolARED dataset is designed to support enhancements of predictive capabilities, enabling the development of operational forecasts for the emergence of active regions.

Abstract

The development of accurate forecasts of solar eruptive activity has become increasingly important for preventing potential impacts on space technologies and exploration. Therefore, it is crucial to detect Active Regions (ARs) before they start forming on the solar surface. This will enable the development of early-warning capabilities for upcoming space weather disturbances. For this reason, we prepared the Solar Active Region Emergence Dataset (SolARED). The dataset is derived from full-disk maps of the Doppler velocity, magnetic field, and continuum intensity, obtained by the Helioseismic and Magnetic Imager (HMI) onboard the Solar Dynamics Observatory (SDO). SolARED includes time series of remapped, tracked, and binned data that characterize the evolution of acoustic power of solar oscillations, unsigned magnetic flux, and continuum intensity for 50 large ARs before, during, and after their emergence on the solar surface, as well as surrounding areas observed on the solar disc between 2010 and 2023. The resulting ML-ready SolARED dataset is designed to support enhancements of predictive capabilities, enabling the development of operational forecasts for the emergence of active regions. The SolARED dataset is available at https://sun.njit.edu/sarportal/, through an interactive visualization web application.

Figures (6)

• Figure 1: Boxplots of the location, size, and observation time distributions for the 50 ARs included in SolARED. The left panel shows the distribution of starting ($\lambda_{start}$), ending ($\lambda_{end}$), and total traveled AR longitudes ($\lambda_{traveled}$) along with the constant latitude ($\phi$) in Stonyhurst coordinates. The middle panel shows the distribution of starting, ending, and maximum areas ($A_{start}, A_{end}$, and $A_{max}$) occupied by the ARs in millionths of a hemisphere. The right plot shows the distribution of the total time the ARs were visible on the disc($t_{traveled}$).
• Figure 2: The SDO/HMI observables processing pipeline used to create the SolARED dataset. The four different arrow colors (orange, red, green, and blue) represent the following processing steps: 1) tracking of SDO/HMI Observables, 2) production of acoustic power maps, 3) dimensionality reduction, and 4) data normalization, respectively. The blue rectangular regions on the spheres of Step 2.1 (Section \ref{['sec:tracking']}) represent the $30.66^{\circ} \times 30.66^{\circ}$ ($512 \times 512$ pixels) areas that were tracked, while the blue grid in Step 2.3 (Section \ref{['sec:tiling']}) is the 9 by 9 grid this area was split in.
• Figure 3: Snapshot of the data tracking from a movie for the magnetic evolution of AR 11158 using the SDO/HMI observations. The panels at right show (from top to bottom) the evolution of the continuum intensity, the line-of-sight magnetic field, and the acoustic power in three frequency ranges. The two columns correspond to the original data of the tracked region (left) and the tile averages calculated after splitting the tracked area into a 9-by-9 grid (right).
• Figure 4: The mean acoustic power time series for the 63 tiles of AR 13179, before (top) and after (bottom) the correction for the center-to-limb variations. The blue bold line represents a tile, where emergence is first observed.
• Figure 5: The interface of the SolARED portal. This web tool has two main components: the data selection panel (left) and the 1D data visualization panel (right).