Discovery of multi-temperature coronal mass ejection signatures from a young solar analogue

Abstract

Coronal mass ejections (CMEs) on the early Sun may have profoundly influenced the planetary atmospheres of early Solar System planets. Flaring young solar analogues serve as excellent proxies for probing the plasma environment of the young Sun, yet their CMEs remain poorly understood. Here we report the detection of multi-wavelength Doppler shifts in Far-Ultraviolet (FUV) and optical lines during a flare on the young solar analog EK Draconis. During and before a Carrington-class ($\sim$10$^{32}$ erg) flare, warm FUV lines ($\sim$10$^5$ K) exhibit blueshifted emission at 300-550 km s$^{-1}$, indicative of a warm eruption. 10 minutes later, the H$α$ line shows slow (70 km s$^{-1}$), long-lasting ($\gtrsim$2 hrs) blueshifted absorptions, suggesting a cool ($\sim$10$^4$ K) filament eruption. This provides evidence of multi-temperature and multi-component nature of a stellar CME. If Carrington-class flares/CMEs occurred frequently on the young Sun, they may have cumulatively impacted the early Earth's magnetosphere and atmosphere.

Figures (11)

• Figure 1: The FUV spectral lines showing blueshifted emission components. (a) Light curves of pre-flare subtracted luminosity of the top five strongest FUV spectral lines in the third HST's orbit during the March 29, 2024. We divided the light curve into five phases--pre-flare (PF), pre-flare brightening (PFB), flare, post-flare 1 (PS1), and post-flare 2 (PS2)--to extract phase-specific spectra. (b, c) Temporal evolution of the C III 1175.7 Å (log$T$=4.85) and Si IV 1393.8 Å (log$T$=4.85) spectra, which exhibited prominent blueshifted components. As a reference, error bars are shown for the flare-phase data. To enhance the signal-to-noise ratio (S/N), the spectra were binned over eight wavelength pixels (i.e., 0.080 Å = 20 km s$^{-1}$ for C III and 0.080 Å = 17 km s$^{-1}$ for Si IV). (d) Gaussian fitting of the C III spectral line during the PFB phase. The flare component at the line center is fitted with a single Gaussian (green dashed line), while the residual blueshifted component is fitted separately with another Gaussian (magenta dashed line). The blue line represents the total line profile, obtained by combining the two Gaussian components. (e) Gaussian fitting of the C III spectral line during the flare phase. The blueshifted component is a factor of 0.024 smaller than the rest line component. (f) Gaussian fitting of the Si IV spectral line. To reproduce the complex redshifted profile, a double-Gaussian fit was applied to the central component as indicated by the green dashed and purple dash-dotted lines. The blueshifted component is a factor of 0.016 smaller than the rest line component.
• Figure 1: Whole flaring profiles for C III and Si IV lines associated with flares. The spectra were obtained for five phases--pre-flare (PF), pre-flare brightening (PFB), flare, post-flare 1 (PS1), and post-flare 2 (PS2). The plotted data are the same spectra as Figure \ref{['fig:fuv']}, but are shown with a much larger flux and wavelength range to illustrate the overall spectral shape and line blending. A black dashed outline indicated the x-y range of Figure \ref{['fig:fuv']}. As a result, the blue-wing emission appears less visible in this figure, but Figure \ref{['fig:fuv']} shows that it is indeed significant above the flux errors.
• Figure 2: H$\alpha$ spectrum exhibiting a blueshifted absorption component following the FUV flare. (a) Light curve of the H$\alpha$ equivalent width (EW) in units of Å (black squares), overplotted with the normalized light curve of Si IV 1393 Å emission (blue line). The error bars of H$\alpha$ EW are calculated as the square root of the sum of squared standard deviations of the continuum spectrum relative to the template spectrum. The FUV flare shown in Figure \ref{['fig:fuv']} is indicated in red. Notably, the H$\alpha$ line shows an earlier, long-duration flare beginning approximately 3 hours before the FUV flare and nearly decaying by the time the FUV flare occurs. (b) Dynamic spectrum of the H$\alpha$ line after subtraction of the quiescent (pre-flare) spectrum, obtained during the period marked by black arrows and labeled "QS." Orange and blue colors indicate excess emission and absorption, respectively, relative to the reference spectrum. The dotted line marks the central wavelength of the H$\alpha$ line. Blue and orange points denote the velocities of blueshifted components in the C III and Si IV lines, respectively. Note that the C III line is a blended multiplet; thus, velocities for both the strongest (1175.7 Å) and bluest (1174.9 Å) components are plotted using different symbols. The velocity error bars indicate the velocity dispersion derived from Gaussian fits to the blueshifted components while the time error bars are the integration time of spectra. (c) Spectral fitting of the H$\alpha$ line. To improve the S/N, multiple frames were binned. The red and blue lines represent Gaussian fits to the emission and absorption components, respectively, while the green line shows their combined profile, fitted simultaneously. The central velocity and line width for each component are annotated.
• Figure 2: FUV spectral lines associated with flares other than the one shown in Figure \ref{['fig:fuv']}. (a) Light curves of the narrow and broad wing components of the hydrogen Ly$\alpha$ line, as the wavelength ranges are defined as the black error ranges in panel (c). (b) Time evolution of the blue-to-red wing flux ratio for both the narrow and broad wings for Ly$\alpha$ line. (c–f) Temporal evolution of the spectral line profiles for hydrogen Ly$\alpha$, N V 1238 Å, the C II doublet, and Si IV 1402 Å, respectively. The gray shaded regions indicate wavelength ranges flagged with bad Data Quality (DQ) in the HST dataset. N V does not show a significant blueshifted enhancement. In panel (d), the black vertical line indicates the velocity of the Si IV blueshift from Figure \ref{['fig:fuv']}. The black horizontal line represents the continuum level, measured by integrating over the range 1238.8 + (1–3) Å. The red dashed line shows the estimated level of the continuum plus a possible blueshift component, calculated by scaling the Si IV blueshift signal by the N V / Si IV flaring flux ratio. Since the red dashed line is below or comparable to the noise level, even if a blueshifted component is present, it would be difficult to detect in these weaker lines. C II appears to exhibit an extended blue wing component; however, this region is likely contaminated by a nearby blended emission line and affected by poor data quality on the blue side, requiring cautious interpretation. Si IV 1402 Å also shows an extended blue wing, but due to blending with O IV lines, this feature was excluded from the velocity analysis. The red arrows indicate the possible locations of the blueshifted emission components.
• Figure 3: Time evolution of blueshifted velocities in FUV and optical wavelengths. (a) Each data point represents the central velocity derived from a Gaussian fit, where the vertical error bars correspond to the Gaussian standard deviation. The horizontal error bars indicate the time intervals over which the spectra were integrated for the fitting. The C III line is a blended multiplet; therefore, velocities for both the strongest (1175.7 Å) and bluest (1174.9 Å) components are plotted using distinct symbols. The time axis is referenced to the onset of the FUV flare, and the normalized Si IV light curve is shown as a shaded gray curve. The black dotted line indicates a simple exponential fit to the data. (b) Schematic illustration of possible explanations for the time evolution of blueshifts in the FUV and H$\alpha$. The left panel depicts a warm plasma eruption occurring during the "PFB" and "Flare" phases. The middle and right panels illustrate scenarios relevant to the period $\sim$10 minutes after the flare, when only H$\alpha$ filament eruption starts to be observed. The middle panel represents the possibility of a multi-temperature, multi-layer structure within a single eruptive event. The right panel illustrates an alternative scenario involving multi-temperature eruptions at different locations, which are likely physically connected, analogous to "sympathetic" events observed on the Sun. Here, sympathetic eruptions means phenomena where a eruption caused by another eruption in a nearby or connected region. In ten minutes later, the warm plasma became faint and undetectable in the FUV line wings, likely due to its expansion. However, it is illustrated in the panels as still present, either continuing to expand or influencing the surrounding environment.