Three-dimensional Propagation of the Global EUV Wave associated with a solar eruption on 2021 October 28

Over the past two decades, coronal wave-like phenomena (i.e., extreme ultraviolet waves, ‘EUV waves’ for short) have been extensively studied (Liu & Ofman, 2014; Chen, 2016). These EUV waves usually appear as propagating bright or dark fronts in coronal EUV images (e.g., Wills-Davey & Thompson, 1999; Thompson et al., 1999; Thompson & Myers, 2009; Liu et al., 2010; Chen & Wu, 2011; Tian et al., 2012; Shen & Liu, 2012a; Shen et al., 2014b, 2018; Ying et al., 2018). They are sometimes simultaneously observed with Moreton-Ramsey waves (Warmuth et al., 2001; Eto et al., 2002; Okamoto et al., 2004; Asai et al., 2012; Shen & Liu, 2012a; Shen et al., 2019; Long et al., 2019), which are chromospheric waves observed in H$\alpha$ images (Moreton, 1960; Moreton & Ramsey, 1960; Smith & Harvey, 1971; Shibata et al., 2011). The EUV waves are often associated with large-scale energetic eruptions, e.g., coronal mass ejections (CMEs) and flares (Biesecker et al., 2002; Nitta et al., 2013; Liu et al., 2018). However, small-scale events such as surges, jets, mini-filaments, and mini-CMEs may also drive EUV waves (e.g., Zheng et al., 2011, 2012a, 2012b, 2012c, 2012d, 2013a, 2013b; Shen et al., 2017).

The large-scale EUV waves have often been interpreted as fast-mode MHD waves (Thompson et al., 1999; Wang, 2000; Warmuth et al., 2001; Ofman & Thompson, 2002; Li et al., 2012; Liu et al., 2018) driven by CME expansion (Asai et al., 2012; Liu et al., 2012; Shen & Liu, 2012b; Shen et al., 2013, 2018) and often followed by non-wave components due to the CME-caused magnetic field reconfiguration (Chen et al., 2002, 2005; Shen & Liu, 2012a; Downs et al., 2012; Liu et al., 2012). The EUV waves generally propagate as a single wavefront, while sometimes appear in the form of multiple wavefronts (Liu et al., 2012; Shen et al., 2019). Their propagating velocities are measured to be 50–700 km s^-1 with typical values of 200-400 km s^-1 (Thompson & Myers, 2009) from the observations of the Extreme-ultraviolet Imaging Telescope (EIT, Delaboudinière et al., 1995) onboard the Solar and Heliospheric Observatory (SOHO, Moses et al., 1997; Thompson et al., 1998), and 200–1500 km s^-1 with an averaged value of 644 km s^-1 (Nitta et al., 2013) from the high-resolution observations with the Atmospheric Imaging Assembly (AIA, Lemen et al., 2012) onboard the Solar Dynamics Observatory (SDO, Pesnell et al., 2012). Previous studies have shown that the EUV waves often experience a deceleration process with a deceleration of an order of several hundred m s^-2 (e.g., Warmuth et al., 2001; Vršnak et al., 2002, 2005; Veronig et al., 2008, 2011; Nitta et al., 2013). As the EUV waves propagate, they could interact with coronal structures. For instance, the wavefronts could not only exhibit reflections(Li et al., 2012; Olmedo et al., 2012; Shen et al., 2013; Yang et al., 2013; Liu et al., 2018; Zhou et al., 2021b), transmissions(Olmedo et al., 2012; Liu et al., 2012; Shen et al., 2013; Liu et al., 2018), and refractions(Thompson et al., 2000; Shen & Liu, 2012b; Shen et al., 2013), but also result in oscillations of coronal loops (Liu et al., 2012; Shen & Liu, 2012b; Zheng et al., 2013b) and filaments (Okamoto et al., 2004; Asai et al., 2012; Shen et al., 2014a, b, 2017). EUV waves are often accompanied by quasi-periodic fast-mode propagating wave trains (Liu & Ofman, 2014; Shen et al., 2021; Duan et al., 2022). Signatures of coronal heating have also been found near wave fronts (Wills-Davey & Thompson, 1999; Downs et al., 2012; Liu et al., 2018).

Limb observations have shown that an EUV wave does have a dome-shaped wavefront (Veronig et al., 2010; Li et al., 2012; Liu et al., 2012; Selwa et al., 2012; Shen et al., 2014a; Zhou et al., 2021a). In the studies of Veronig et al. (2010), Li et al. (2012), and Shen et al. (2014a), the upward expansions of wavefronts were found to be faster than the lateral expansions. With AIA observations, Liu et al. (2012) studied an off-limb EUV wave with an initial height of 110 Mm driven by a dome-shaped CME front and found that the wavefront propagates forwardly inclined towards the surface of the Sun (Liu et al., 2018).

With stereoscopic observations in dual perspectives and multi-passbands, here we report one global EUV wave observed at the beginning of Solar Cycle 25, and investigate the three-dimensional (3D) propagation of the wavefront as well as its impact on the coronal plasma temperature. We describe our observations in Section 2, present the analysis results in Section 3, discuss the results in Section 4, and summarize our findings in Section 5.

The global EUV wave of interest occurred on 2021 October 28, was simultaneously observed by AIA onboard SDO, the Extreme Ultraviolet Imager (EUVI, Howard et al., 2008) onboard the Solar Terrestrial Relations Observatory (STEREO, Kaiser et al., 2008), the Solar X-ray Extreme Ultraviolet Imager (X-EUVI) onboard the Fengyun-3E saterllite (FY-3E, Zhang et al., 2021), and the Global Oscillation Network Group (GONG, Harvey et al., 2011). From the Earth’s view, the full-disk EUV images taken by AIA and X-EUVI revealed the propagations of the EUV wave from the top-down view. Images of all the AIA EUV passbands, with a 12 s cadence and a $\sim$0.6${}^{{}^{\prime\prime}}$pixel size, were used in this study. Different from the AIA 193 Å passband centered at 193 Å, the EUV image taken by X-EUVI onboard FY-3E in an Early Morning Orbit is centered at 195.5 Å with a bandwidth of 7.5 Å. It is the only EUV passband of the X-EUVI instrument. The preprocessing procedure of the X-EUVI 195 Å image is described in Song et al. (2022a). The X-EUVI 195 Å image has a cadence of $\sim$14 s and a pixel size of $\sim$2.5${}^{{}^{\prime\prime}}$. The X-EUVI data will be released to the community after the satellite on-orbit commission test is completed. We also analyzed the H$\alpha$ images provided by GONG to investigate the chromospheric counterpart of the EUV wave. The H$\alpha$ images with a cadence of 60 s and a pixel size of 1.07${}^{{}^{\prime\prime}}$ were taken with a filter centered at 6562.8 Å.

In the meantime, on the western side of the Earth, STEREO-Ahead was separated about 37.5^∘ from the Earth-Sun line. The 195 Å and 304 Å images taken by STEREO-Ahead/EUVI revealed the propagation of the EUV wave from the edge-on view. The pixel size of these images is 1.58${}^{{}^{\prime\prime}}$, and the cadence is $\sim$2.5 min.

Since the X-EUVI 195 Å and the AIA 193 Å images have similar temperature responses, we aligned them through a linear Pearson correlation. For aligning the H$\alpha$ images and the AIA images, the sunspot and its surrounding plage region within AR 12887 in the H$\alpha$ and AIA 1600 Å images were used as the reference features.

Figure 1 presents an overview of the solar corona at 15:10 UT in the AIA 193 Å and X-EUVI 195 Å passbands, which includes four active regions (ARs), a polar coronal hole, and the rest quiet-Sun regions. Among them, NOAA AR 12887 was located in the southern hemisphere and close to AR 12889. AR 12887 was very active around Oct 28, generating a global EUV wave we study here as well as its associated X1.0 flare and halo CME. The X1.0 flare started and peaked at 15:17 UT and 15:35 UT, respectively. It is the first X-class flare that occurred around the disk center in Solar Cycle 25, and its eruption mechanism is investigated by Song et al. (2022b). The global EUV wave appeared at $\sim$15:28 UT and quickly propagated to the surrounding areas, affecting almost the entire front side of the Sun. From coronagraph images taken by SOHO/LASCO, we can see an associated halo CME that first appeared around 15:53 UT.

The stereoscopic observations of multiple spacecraft with high spatial and temporal resolutions revealed the evolution of this global EUV wave very well, allowing us to investigate its 3D propagation and its impact on the coronal plasma.

In the last twenty years, the EUV waves have been extensively studied. Despite intensive debates on the nature of the EUV waves, in recent years more and more researchers tend to interpret these waves by a ‘hybrid model’, in which an outer wave component is likely a fast-mode MHD wave driven by the CME expansion and an inner non-wave component is caused by the magnetic field reconfiguration during the CME eruption (Downs et al., 2012; Liu et al., 2012; Liu & Ofman, 2014; Chen, 2016). The primary EUV wave we studied is accompanied by an X1.0 flare and a halo CME. It originates from a distance of $\sim$150 Mm away from the eruption center, where the accompanying dimming region stops expanding and the wavefront continues to propagate with an initial velocity of 600–720 km s^-1 and a significant deceleration of 110–320 m s^-2. The initial velocity is larger than the local sound speed ($\sim$ 150 km s^-1), and possibly also larger than the coronal Alfvén speed (200–400 km s^-1, Gopalswamy et al., 2012). These indicate that the EUV wave is likely a nonlinear fast-mode MHD wave or shock driven by the CME expansion (Asai et al., 2012; Liu et al., 2012; Shen & Liu, 2012b; Shen et al., 2013, 2018). Assuming that this EUV wave decays to an ordinary fast-mode wave, its final velocity can be taken as a proxy for the fast-mode speed. Then, the fast-mode Mach number is estimated to be 1.52–1.82 for this event by dividing the initial velocity by the final one (estimated to be 380–440 km s^-1).

The EUV wave is likely a 3D structure propagating along the solar surface with a fast-mode speed. The wavefronts observed in the 193 Å and 171 Å passbands are the line-of-sight integrations projected on the solar disk at the temperature of 1.5 MK and 1 MK, respectively. As reported by Warmuth & Mann (2005) and Downs et al. (2021), the fast-mode speed increases with height and leads to the wavefront at larger heights having a faster apparent motion on the solar surface. Our result that the wavefront seen at 193 Å propagates faster than the one in 171 Å is consistent with this when we consider the effect that temperature has on the observability of the wavefronts. The densities of plasma at various temperatures show a different fall-off with height due to the different hydrostatic scale height. Thus, emission of the cooler plasma at 171 Å is more concentrated at lower heights than emission at 193 Å  and the same holds for the observed wavefronts. Therefore, the different propagation speeds determined from the locations of the wavefronts in 193 Å and 171 Å might well be a result of this hydrostatic weighting bias. For 304 Å, the presence of emission lines with different formation temperatures in the 304 Å passband makes it hard to determine the dominated emission for the wave signatures. However, we found that the wavefront has a chromospheric height and is preceded by and lower than the 193 Å wavefront, indicating that the 304 Å wavefront is likely dominated by the He ii line. For the chromospheric counterpart Moreton-Ramsey wave, the wavefront signature is dominated by the chromospheric H$\alpha$ line and also clearly preceded by the 193 Å wavefront. Then, the difference of the propagating speeds between the coronal and chromospheric/transition region wavefronts would be also an effect of temperature.

Shen & Liu (2012a) reported an EUV wave simultaneously observed in the 1600/1700 Å, H$\alpha$, 193 Å, and 211 Å filters, covering a broad temperature range from the photosphere to the low corona. They found that the EUV wave had similar initial propagating velocities in the photospheric and coronal passbands, and suggested that the coronal wave and its photospheric/chromospheric counterparts had a common origin. We also observed a chromospheric counterpart of the EUV wave in the H$\alpha$ images, indicating that the EUV wave is accompanied by a Moreton-Ramsey wave. However, no wave signature appeared in the 1600/1700 Å images from SDO/AIA. This implies that the EUV wave in our study may have affected the upper chromosphere but not deeper in the atmosphere.

Several previous observations have shown that EUV waves could be detected in the 304 Å passband (Liu et al., 2010; Long et al., 2011; Shen & Liu, 2012a, b; Ying et al., 2018; Long et al., 2019). In quiet-Sun regions and ARs, the AIA 304 Å passband is dominated by two He ii 303.8 Å lines, which are formed in the transition region at a temperature of log(T/K)$\approx$4.7, but a coronal line of Si xi 303.33 Å at the temperature of log(T/K)$\approx$6.2 could also make an important contribution to the 304 Å passband in some cases (O’Dwyer et al., 2010). Shen & Liu (2012a) found that the propagation of an EUV wave in the 304 Å images was similar to that in the 193 Å passband and suggested that the wavefront in the 304 Å passband was the manifestation of the contribution of the coronal Si xi line rather than the transition region He ii line. However, in our event, we found that the 304 Å wavefront with a chromospheric height is preceded by and lower than the 193/195 Å wavefront, suggesting that the 304 Å wavefront is likely dominated by the He ii line. Also, the 304 Å wavefront clearly precedes the H$\alpha$ wavefront, suggesting that the 304 Å passband may be used to reveal the link between the coronal EUV wave and the chromospheric Moreton-Ramsey wave.

For the EUV waves, dome-like wavefronts have been reported by some authors (Veronig et al., 2010; Li et al., 2012; Liu et al., 2012; Selwa et al., 2012; Shen et al., 2014a; Zhou et al., 2021a). Veronig et al. (2010), Li et al. (2012), and Shen et al. (2014a) found that the upward expansions of their studied wavefronts were faster than the lateral expansions of the wavefronts. With AIA observations, Liu et al. (2012) analyzed an off-limb EUV wave driven by a dome-shaped CME front, and suggested that its wavefront propagated forwardly inclined to the solar surface (Liu et al., 2018). Similarly, in our event, the propagating tracks in the warm passbands (e.g., 193 Å and 195 Å) precede those in the cool passbands (e.g., 304 Å) by $\sim$36.4 Mm. Meanwhile, we found the wavefront in the 195 Å passband $\sim$67.4 Mm higher than that in the 304 Å passband from dual-perspective and multi-passband observations. Based on these findings, we suggest that the propagating wavefront appears to be a dome-like structure, propagating forwardly inclined toward the solar surface with an averaged tilt angle of $\sim$53.2^∘ in the time period of 15:32–15:38 UT. This is consistent with the finding of Liu et al. (2012), in which the slit angle of the wavefront near the flare site appeared to decrease with decreasing hight, ranging from 73^∘ at 72 Mm to 48^∘ at the coronal base. Also, it has been reported that wavefronts of the EUV waves tend to get increasingly tilted during their propagation (Hudson et al., 2003; Patsourakos & Vourlidas, 2009; Kienreich et al., 2009; Liu et al., 2012) due to the increase of fast-mode speed with height in the low corona (Warmuth & Mann, 2005).

In this study, we have presented analysis results of a remarkable global EUV wave and its chromospheric counterpart ‘Moreton-Ramsey wave’ with stereoscopic observations from SDO/AIA, STEREO-Ahead/EUVI, FY-3E/X-EUVI, and GONG. The EUV wave is associated with an X1.0 flare and a halo CME, and appears in the H$\alpha$ and EUV passbands with different characteristic temperatures (e.g., 304 Å, 171 Å, and 193/195 Å).

The EUV wave becomes evident at a distance of $\sim$150 Mm away from the eruption center and propagates circularly outward with an initial velocity of 600–720 km s^-1 and a significant deceleration of 110–320 m s^-2. During the propagation of the EUV wave, the wavefront propagates forwardly inclined to the solar surface with a tilt angle of $\sim$53.2^∘. This implies that the propagating wavefront is likely a dome-like structure, which could interact with the solar atmosphere. The plasma in the low corona is heated from log(T/K)$\approx$5.9 to log(T/K)$\approx$6.2 on the propagation path of the wavefront. The 304 Å and H$\alpha$ passbands also reveal signatures of the wave, indicating that the EUV wave could affect the low atmosphere. This event is also accompanied by a dimming region, which just follows the wavefront and evolves from a small region to a ring-shaped region stopping at a distance of $<$150 Mm away from the eruption center. Our results suggest that the observed EUV wave is likely a fast-mode MHD wave or shock driven by the expansion of its associated CME.