Small-scale Bright Blobs Ejected from a Sunspot Light Bridge
Fuyu Li, Yajie Chen, Yijun Hou, Hui Tian, Xianyong Bai, Yongliang Song
aa r X i v : . [ a s t r o - ph . S R ] F e b Draft version March 1, 2021
Typeset using L A TEX default style in AASTeX63
Small-scale Bright Blobs from a Sunspot Light Bridge
Fuyu Li, Yajie Chen,
1, 2
Yijun Hou, Hui Tian,
Xianyong Bai, and Yongliang Song School of Earth and Space Sciences, Peking University, 100871 Beijing, China Max-Planck Institute for Solar System Research, Justus-von-Liebig-Weg 3, D-37077 G¨otingen, Germany Key Laboratory of Solar Activity, National Astronomical Observatories, Chinese Academy of Sciences, Beijing 100101, China
Submitted to ApJABSTRACTLight bridges (LBs) are bright lanes that divide an umbra into multiple parts in some sunspots.Persistent oscillatory bright fronts at a temperature of ∼ K are commonly observed above LBsin the 1400/1330 ˚A passbands of the Interface Region Imaging Spectrograph (IRIS). Based on IRISobservations, we report small-scale bright blobs from the oscillating bright front above a light bridge.Some of these blobs reveal a clear acceleration, whereas the others do not. The average speed of theseblobs projected onto the plane of sky is 71 . ± . − , with an initial acceleration of 1 . ± . − . These blobs normally reach a projected distance of 3–7 Mm from their origin sites. From thetransition region images we find an average projected area of 0 . ± .
37 Mm for the blobs. The blobswere also detected in multi-passbands of the Solar Dynamics Observatory, but not in the H α images.These blobs are likely to be plasma ejections, and we investigate their kinematics and energetics.Through emission measure analyses, the typical temperature and electron density of these blobs arefound to be around 10 . K and 10 . cm − , respectively. The estimated kinetic and thermal energiesare on the order of 10 . erg and 10 . erg, respectively. These small-scale blobs appear to show threedifferent types of formation process. They are possibly triggered by induced reconnection or release ofenhanced magnetic tension due to interaction of adjacent shocks, local magnetic reconnection betweenemerging magnetic bipoles on the light bridge and surrounding unipolar umbral fields, and plasmaacceleration or instability caused by upward shocks, respectively. Keywords:
Sun: chromosphere — sunspots — Sun: transition region — Sun: UV radiation INTRODUCTIONBright lanes that divide an umbra into multiple parts, called light bridges (LBs), are among the most prominentbright structures in many sunspots. The magnetic field at LBs is often much weaker and more inclined compared tothat of the umbra (Leka 1997; Shimizu et al. 2009; Toriumi et al. 2015a; Yuan & Walsh 2016; Feng et al. 2020). High-resolution observations have shown dark substructures within light bridges, such as central dark lanes with upflows(Berger & Berdyugina 2003; Rouppe van der Voort et al. 2010) and dark knots with downflows (Zhang et al. 2018).These small-scale structures are likely formed as a result of the magnetoconvection in sunspots (Sch¨ussler & V¨ogler2006). Recent observations and simulations have revealed possible signatures of weakly twisted magnetic field and fluxemergence in LBs (Louis et al. 2015; Tian et al. 2018a; Toriumi et al. 2015a,b; Song et al. 2017; Yang et al. 2019a;Yuan & Walsh 2016; Bai et al. 2019; Lim et al. 2011).The most prominent phenomenon above LBs revealed from chromospheric observations is the recurrent surge-like activity, which was reported as H α surges, light wall oscillations, plasma ejections, or chromospheric jets (Roy Corresponding author: Yajie [email protected] author: Hui [email protected] ∼
100 km s − ) and were found at locations of strong electric current (Louis et al. 2014; Robustini et al. 2016;Yang et al. 2019b; Lim et al. 2020), suggesting that they are likely driven by magnetic reconnection between small-scale magnetic bipoles at LBs and the surrounding unipolar umbral field. Meanwhile, low-speed surges ( ∼
15 km s − )with a stable oscillating period of a few minutes have also been reported, and suggested to be driven by p-modeor magnetoacoustic waves leaking from the underlying photosphere (Yang et al. 2015; Zhang et al. 2017). Based onobservations of the Goode Solar Telescope (GST, Cao et al. 2010) and the Interface Region Imaging Spectrograph(IRIS, De Pontieu et al. 2014), Tian et al. (2018b) found strong evidence of two types of surge-like activity above LBs:persistent up-and-down motions associated with the upward leakage of magnetoacoustic waves from the photosphere,and sporadic high-speed jets triggered by intermittent magnetic reconnection. Based on imaging observations of IRIS,Hou et al. (2017) also reached a similar conclusion.Bright fronts ahead of LB surges are commonly observed in transition region (TR) images (Bharti 2015; Yang et al.2015). It is likely that the bright fronts ahead of the LB surges is heated to TR temperatures either by shocks orthrough compression (Morton 2012; Bharti 2015; Zhang et al. 2017). In this paper, we report small-scale bright blobsat typical TR temperatures from the bright front of surges above a light bridge, which is rarely observed in thepast. The existence of these blobs indicates more complex physical processes above the LB. We perform a detailedinvestigation of their kinematics and energetics, and discuss their possible formation mechanisms. OBSERVATIONSThe observations were performed from 05:17:13 UT to 06:30:18 UT on 2014 Oct 25. The IRIS pointed to a coordinateof (257 ′′ , -318 ′′ ), targeting at the largest sunspot group of solar cycle 24 located in NOAA Active Region (AR) 12192.We used the level 2 IRIS data, which has been corrected through dark current subtraction, flat field, geometrical, andorbital variation corrections. Slit-jaw images (SJIs) of IRIS in the filter of 1400 ˚A were used, and the cadence was 7 . ′′ × ′′ , with a spatial pixel size of 0. ′′ ◦ . Since the slit failed to catch any blobs investigated in this paper, we did not use the spectral data.We also analyzed the data taken simultaneously by the Atmospheric Imaging Assembly (AIA, Lemen et al. 2012) onboard the Solar Dynamics Observatory (SDO) to study the temperature of the blobs. The cadence of AIA observationswas 12 s in the 94, 131, 171, 193, 211, 304, 335 ˚A passbands, and 24 s in the 1600 ˚A passband. The pixel size of theAIA images was ∼ ′′ α core images of NVST were used to examine the chromospheric response ofthe blobs. The bandwidth of the H α filter was 0.25 ˚A. These H α core images have an image scale of 0. ′′
164 pixel − ,a cadence of 12 s , and a FOV of 152 ′′ × ′′ . We coaligned the H α images and SJI 1400 ˚A images by matching thecommonly observed LB and some localized dynamic features. DATA ANALYSIS AND RESULTS3.1.
Kinematics of the ejections
Bharti (2015) and Yang et al. (2015) analyzed this dataset and identified an oscillating bright front above the LBfrom the TR images. The bright front is visible from the 1400 ˚A images shown in Figure 1. It exhibits up-and-downoscillatory motions during the entire observation period. Besides this oscillating bright front, we also identified morethan ten small-scale intermittent bright blobs away from the leading edge.
The blobs are better seen in themovies, and we pointed out each blob by a green arrow in Movie 1. These bright blobs may correspondto plasma ejections, and we cannot rule out the the possibility of heating fronts due to the absence ofspectral data for the bright blobs. Considering both network jets and penumbral microjets, which couldbe caused by heating fronts (De Pontieu et al. 2017; Chen et al. 2019; Rouppe van der Voort & Drews2019; Esteban Pozuelo et al. 2019), usually perform elongated structures, and the bright blobs wereport in this study exhibit circular shape, so they are more likely mass ejections.
Table 1.
Observational details and measured physical parameters of the bright blobs.No. Time
D v a S log EM log T log n e log E k log E t Type(UT) km km s − km s − Mm cm − K cm − erg erg1 05:22:10 - 05:23:09 3980 68.0 0.0 0.85 27.35 5.45 9.69 23.25 23.66 I2 05:33:31 - 05:34:00 2291 78.2 3.4 0.06 27.15 5.49 9.88 21.84 22.16 I3 05:37:55 - 05:39:15 6874 85.3 2.2 0.45 27.17 5.45 9.67 23.02 23.22 I4 05:39:08 - 05:39:45 2533 69.2 3.4 0.24 27.20 5.50 9.75 21.52 22.96 III5 05:41:20 - 05:42:11 3136 61.2 1.1 0.31 - - - - - I6 05:47:11 - 05:48:10 4101 70.0 0.0 0.68 27.15 5.43 9.62 23.06 23.41 II7 05:49:30 - 05:50:07 3136 85.6 2.3 0.36 27.18 5.48 9.70 22.90 23.13 I8 05:51:42 - 05:52:48 6151 93.4 3.4 1.27 27.21 5.47 9.58 23.68 23.82 I9 05:59:45 - 06:00:44 5186 88.6 1.1 1.13 27.50 5.50 9.74 23.71 23.93 I10 06:07:56 - 06:09:01 4704 71.4 3.4 0.88 27.30 5.49 9.66 23.29 23.69 III11 06:12:19 - 06:13:40 3256 40.4 2.3 0.32 27.12 5.46 9.68 22.15 23.01 I12 06:23:11 - 06:23:47 2050 56.0 1.1 0.25 27.13 5.45 9.72 22.30 22.87 III13 06:27:12 - 06:28:55 6633 64.7 1.1 0.59 - - - - - III Note —The second column gives the time ranges during which the blobs appear in SJI 1400 ˚A images. Other columns listthe propagation distance ( D ), velocity ( v ), initial acceleration ( a ), area ( S ), emission measure (log EM), temperature(log T ), electron density (log n e ), kinetic energy (log E k ), and thermal energy (log E t ) for the 13 bright blobs.Since blobs 5 and 13 cannot be unambiguously identified from the AIA channels, their energies and emission measures arenot calculated. The last column shows the types of their possible formation processes. The lifetimes of these blobs are mostly 30-100 s from the 1400 ˚A image sequence. Most of these blobs appear tobe initiated from the bright front, and the rest might be ejected from the LB base. We identified 13 blobs that couldbe unambiguously identified and tracked from the 1400 ˚A images. All of these blobs show only upward propagationand then disappear. No obvious signs of downward motions were found. These bright features were noticed by Bharti(2015), but no detailed analysis was performed. Observational details of these bright blobs are listed in Table 1.There appear to be three types of bright blobs that are possibly associated with different formation processes.Figure 1 shows the IRIS image sequences for three blobs, each falling into one of the three types. In the first example(Figure 1(a1)-(a5), blob 1 in Table 1, type I), we see two small-scale brightenings at the bright front before the brightblobs. These two brightenings subsequently merge into a bigger one (Figure 1(a2)), from which an blob is then clearlyseparated. For most of the eight type I cases, the newly formed features are also brighter.The second example (Figure 1(b1)-(b5), blob 6, type II) appears to have a different formation process. Before theformation of the bright blob, there is an evident brightening (Figure 1(b2)) at the LB (similar to the first case studiedin Hou et al. 2017). Then the emission of the bright front is enhanced at the location right above this brightening,where a faint jet-like structure appeares to connect the LB and the bright front. Later on, a bright blobs is ejectedaway from the bright front. We only found one such case (type II).The third example (Figure 1(c1)-(c5), blob 4, type III) is different from the two examples described above. Beforethe bright blobs forms, neither significant intensity enhancement at the LB nor signature of merging bright pointsat the bright front is observed. However, we noticed that materials arise from a wide range of the LB, and theymerge with the bright front. Then three successive bright blobs appear after 05:37 UT. Figure 2 shows the space-timediagrams along different paths (Figure 1(c2)) of the three successive blobs. It is evident that when the rising materialsfrom the LB merge with the bright front from the south to the north, bright blobs successively occur at the differentmerging locations. There are four such blobs (type III).To measure the velocities ( v ) and initial accelerations ( a ) of these blobs, we produced space-time diagrams alongthe paths of each blob. Figure 3 shows two examples. Blob 11 reveals a constant speed, whereas blob 8 has anobvious acceleration during the initial propagation. We calculated the average speed and acceleration for each case.These blobs have speeds of 71 . ± . − , with accelerations of 1 . ± . − . For the only one type II Figure 1.
Temporal evolution of three bright blobs in SJI 1400 ˚A images. Panels (a1)-(a5), (b1)-(b5), and (c1)-(c5) are forthe blobs 1, 6, and 4 listed in Table 1, respectively. The blue contours in panels (a4), (b5), and (c5) outline the locations of theblobs. The blue dashed lines in panel (c2) mark three cuts that we used to produce space-time diagrams in Figure 2. The greenarrows highlight the blobs and some brightenings related to the origin of these blobs. The images are shown in logarithmicscale. (An animation of this figure is available.) blob, its acceleration is 0 km s − . While for most of the other blobs, we see a clear acceleration. We also calculatedthe propagation distances ( D ) and projected areas ( S ) of the blobs. These blobs normally propagate to a projecteddistance of 3–7 Mm from the bright front in the 1400 ˚A passband. The projected areas, with an average of 0 . ± . , do not reveal a significant change during the propagation. These kinematic parameters of all blobs are listed inTable 1. 3.2. Response in AIA passbands
Most blobs also show clear signatures in images of all or most of the AIA 131, 171, 193, 211, 304, and 335 ˚A filters,and an example is presented in Figure 4. Meanwhile, these blobs do not show any obvious signatures in H α coreimages. The visibility of these blobs in different AIA passbands and SJI 1400 ˚A images suggests either a multi-thermalstructure or a TR temperature. It is difficult to determine the temperature structure through a differential emissionmeasure (DEM) analysis since the low-temperature part of the DEM can not be well constrained by observations inthe AIA passbands (Del Zanna et al. 2011; Testa et al. 2012).Instead, we calculated the EM-loci (emission measure loci) curves for each plasma blob to determine the possibletemperature (e.g., Del Zanna et al. 2002; Winebarger et al. 2013; Tian et al. 2014). Figure 5 shows the EM-loci curvesfor blob 10. The EM-loci curves of different AIA passbands cross a relatively narrow box around the temperatureof 10 . K and emission measure (EM) of 10 . cm − . This temperature and EM are regarded as rough estimatesof the plasma temperature and EM of the blob. Two of the plasma blobs (blobs 5 and 13 in Table 1) do not showobvious signatures in most of the AIA EUV passbands, and we could not perform an EM-loci analysis. The possibletemperature and EM values for the rest of the identified plasma blobs are shown in Table 1. We found that the EMand temperature of the ejected plasma are on average 10 . cm − and 10 . K, respectively. The EM-loci curvesshould just be considered as an upper limit of the true emission measure distribution (Tian et al. 2014).3.3.
Energy estimation D i s t an c e ( Mm ) cut 1 D i s t an c e ( Mm ) cut 2 Time(UT)0123456 D i s t an c e ( Mm ) cut 3 Figure 2.
Space-time diagrams along the cuts shown in Figure 1 (c2). The black dashed lines indicate the rising motion ofsome materials from the LB. The green arrows indicate the apparent origin sites of bright blobs.
The dashed black linesindicate the speeds of the bright edges, that is around Mm s − . Using the EM and temperature estimated from the EM-loci analysis, we can calculate the kinetic and thermal energyfor each blob. Under the assumption that the integration length along the line of sight is similar to the projected size,we estimated the electron density ( n e ) of a blob using Equation 1, EM ≈ n e L (1)where L is the square root of the area of the blob. The electron densities of the blobs were found to be 10 . ± . cm − .Assuming a fully-ionized plasma and a typical coronal composition (e.g., Yang et al. 2020), the mass density can beestimated as the following ρ ≈ . n e m p (2)The kinetic energy ( E k ) and thermal energy ( E t ) can be estimated as E k = 12 ρV v ≈ . n e m p L v (3) E t = 2 n e k B T V ≈ n e k B T L (4)where k B is Boltzmann constant, ρ is the mass density, V is the volume of the blob, and m p is the proton mass. Herewe simply assumed the same length, width and height for the ejected plasma. Considering the projection effect, E k should be considered as a lower limit. We also listed values of E k and E t for all identified blobs in Table 1. Theaverage E k is 10 . ± . erg, and the average E t is 10 . ± . erg. Both are close to the typical nanoflare energy of 10 erg (Parker 1988).Assuming that the blobs are produced due to the release of magnetic energy, and other energies such as gravitationalpotential energy and radiation energy are neglegible, the dissipated magnetic energy E m can be estimated as the sum Ejection 11
Time(UT)0123456 D i s t an c e ( Mm ) Ejection 8
Time(UT)02468 D i s t an c e ( Mm ) Figure 3.
Space-time diagrams along the paths of bright blobs
11 and 8.
Figure 4.
A blob (indicated by the white arrows) in the SJI 1400 ˚A, NVST H α core, and AIA 335, 304, 211, 193, 171 and 131˚A images. An animation of this figure is available online. (An animation of this figure is available.) of E k and E t (Priest 2014; Chen et al. 2015, 2020). Hence, the density of the dissipated magnetic energy can bewritten as E k + E t V ≈ (∆ B ) π , (5) AIA 13106:08:20
AIA 17106:08:23
AIA 19306:08:18
AIA 21106:08:23
AIA 33506:08:26 ( a r cs e c ) ( a r cs e c ) E m i ss i on M ea s u r e ( c m - ) (A) (B) (C)(D) (E) (F) Figure 5. (A)-(E) AIA 131, 171, 193, 211 and 335 ˚A images taken around 06:08:24 UT. The white arrow and red square ineach of panels (A)-(E) indicate a plasma blob and the region for an EM-loci analysis, respectively. (F) The EM-loci curves forthe blob. The small black box shows a region with many crossings of the EM-loci curves. where ∆ B represents the dissipated magnetic field. Among the 10 blobs (except blobs 5, 6, and 13), the absolutevalues of ∆ B are generally 4–6 G. In other words, the magnetic field strength decreases by 4–6 G to trigger theblobs. It should be noted that our estimation based on the assumption of a fully-ionized plasma and a typical coronalcomposition is not applicable to blob 6 (type II), which could be driven by the magnetic energy release in the lowerpartially ionized sunspot atmosphere. 3.4. Possible formation mechanisms
We categorized three types of small-scale bright blobs from the LB in Section 3.1.
Due to the lack of spectralinformation of the blobs, we can only speculate their formation mechanisms from the images.
The firsttype is associated with the merging of two bright features on the bright front, and the newly formed features oftenbecome brighter. The blobs may be triggered by magnetic reconnection between two approaching plasma blobs. Itis also possible that the local magnetic tension strengthens when two brightenings move close to each other, and theenhanced magnetic tension drives these blobs. Both the magnetic reconnection and enhanced magnetic tension couldbe caused by interaction of adjacent shocks, which leads to the merging of different bright features along the brightfronts.The second type appears to be related to localized transient brightenings at the base of the LB. These brighteningsindicate small-scale magnetic reconnection events on the LB, as suggested by Hou et al. (2017) and Tian et al. (2018b).Magnetic reconnection between small-scale emerging magnetic bipoles and the surrounding umbral field could occurintermittently and drive upward propagating hot plasma from the LB. The initial acceleration of the blob, which wecalculated based on the traveling paths above the bright front, is essentially zero. The negligible acceleration alsoimplies that the origin sites of such blobs are below the bright front, thus the main acceleration takes place before theblobs propagate ahead of the bright front.The remaining blobs are related to neither merging of bright features on the bright front nor transient brighteningson the LB. For these events, ascending materials from the LB, likely associated with shock waves, are detected tomerge with the existing bright front. The shocks may accelerate the local plasma on the bright front, leading to blobs.Merging of plasma at the bright front may also cause instabilities, which might trigger the blobs. CONCLUSIONS AND DISCUSSIONWe have performed a detailed study of the kinematics and energetics of small-scale bright blobs above a LB observedin the TR images taken by IRIS. These blobs have a typical lifetime of one minute and an average area of 0 . ± . . All blobs exhibit upward propagation without downward motions. With an average speed of 71 . ± . − ,these blobs could travel to a distance of ∼ but not in the H α image, suggesting that the blobs are heated above the TR temperatures .Using the EM-loci method, we found a typical temperature of 10 . ± . K and EM of 10 . ± . cm − for theejected plasma. The kinetic energy and thermal energy have been estimated to be 10 . ± . erg and 10 . ± . erg,respectively.Some blobs occur after the merging of two brightenings on the bright front, or interaction between rising materialsand the bright front. These blobs may be triggered by magnetic reconnection, release of locally enhanced magnetictension, acceleration of shocks, or instabilities at the bright front as shocks travel through. Obvious acceleration hasbeen found for most of these blobs. One blob appears to travel directly from the LB base, and it might be drivenby intermittent small-scale reconnection on the LB. This blob reveals a constant velocity, and no obvious accelerationwas observed after passing the bright front. Though the observations indicate that the bright blobs abovethe LB are more likely to be mass ejections, it should be noted that we cannot exclude the possibilityof the heating fronts.
Due to the lack of spectral data and magnetic field measurements for these blobs, we cannot provide a conclusivejudgement on the origin of these blobs.
Upcoming 4-m Daniel K. Inouye Solar Telescope (DKIST) wouldperform unprecedented high-resolution observations of the sunspots and the light bridge and providesimultaneous magnetic field measurements from the photosphere to the chromosphere. The SpectralImaging of the Coronal Environment (SPICE) onboard Solar Orbiter could provide the spectra of thelines with a wide range of formation temperatures. These new facilities will likely provide new insightinto the generation mechanisms of these bright blobs.
Advanced three-dimensional numerical simulations ofactive regions, including dynamics above light bridges, will also help us better understand their formation mechanisms.ACKNOWLEDGMENTSThis work is supported by NSFC grants 11825301, 11803002, 11790304(11790300), 11427901, 11873062, 11903050,and 11773039 and the Strategic Priority Research Program of CAS (grant XDA17040507). IRIS is a NASA SmallExplorer mission developed and operated by LMSAL with mission operations executed at NASA Ames Research centerand major contributions to downlink communications funded by ESA and the Norwegian Space Center. SDO is aspace mission in the Living With a Star Program of NASA.REFERENCES