Penumbral Brightening Events Observed in AR NOAA 12546

Present-day high resolution observations reveal the dynamic fine scale structure of sunspots created by magnetoconvective interactions (Borrero & Ichimoto 2011; Rempel & Schlichenmaier 2011). Small-scale features in a sunspot (umbral and penumbral) vary in space and time due to a number of different processes. These include oscillations, waves, jets of plasma, and magnetic reconnection. This is believed to lead to flaring events. The latter phenomena involve plasma heating, particle acceleration, and the release of electromagnetic energy from X-rays to radio wavelengths (Shibata & Magara 2011; Benz 2017).

Smaller-scale energy release phenomena detected over and near sunspots' penumbrae have been related to new magnetic elements in emerging flux regions (EFRs, e.g., Guglielmino 2012; Cheung & Isobe 2014) and to magnetic features moving away from the sunspot toward the boundary of the moat region (moving magnetic features (MMFs), see, e.g., Criscuoli et al. 2012; Li et al. 2019, and references therein), which can cancel with preexisting magnetic fields.

In this regard, Kano et al. (2010) analyzed microflares around a well-developed sunspot, by using Hinode (Kosugi et al. 2007) satellite data from the X-ray Telescope (XRT; Golub et al. 2007) and the Narrowband Filter Imager mounted on the Solar Optical Telescope (Tsuneta et al. 2008). They found that half of the observed microflares were caused by magnetic flux cancellation (encounters of opposite polarities in their paper) and when the latter is the main cause, the microflare has one of the X-ray loops connecting the penumbra to the opposite polarity patch of an EFR or an MMF embedded in the moat.

Recently, Bai et al. (2016) reported on a penumbral transient brightening observed with state-of-the-art instruments installed in the Interface Region Imaging Spectrograph (IRIS, De Pontieu et al. 2014) satellite and the 1.6 m New Solar Telescope (Cao et al. 2010; Goode et al. 2010) at the Big Bear Solar Observatory. This penumbral brightening, whose estimated thermal energy was in the range of nanoflares (10²²–10²⁵ erg), displayed signatures from the chromosphere to the corona. The same authors found that an MMF had appeared close to the penumbral boundary and at the same location of one of the footpoints associated with the observed brightening. Bai et al. (2016) attributed the triggering mechanism of the analyzed event to magnetic reconnection occurring in the low corona and explained the brightening seen in the transition region (TR) and chromosphere as due to the local plasma heated up by downward propagating accelerating particles and thermal conduction. However, due to very weak signals of the observations analyzed, those authors could not find any evidence of the chromospheric evaporation that is expected to follow the heating of the chromospheric plasma from reconnection processes in the low corona. This evaporation has been reported from analysis of larger flares (Tian et al. 2014; Graham & Cauzzi 2015), as well as in micro- (Chen & Ding 2010) and nanoflares (Testa et al. 2014) observed outside penumbrae. Since the trigger mechanism of the studied brightening could not be clearly identified by their analysis, Bai et al. (2016) solicited more research on the formation process of brightening events observed in penumbral regions.

Indeed, it is well known that the magnetic reconnection can occur on any spatial or temporal scale in the solar atmosphere (Priest & Forbes 2000). Both the current high resolution observations and magnetohydrodynamical (MHD) numerical simulations indicate that interactions of EFRs with preexisting ambient magnetic fields (Archontis 2012; Cheung & Isobe 2014; Schmieder et al. 2014) play a prominent role in models of large- (e.g., Louis et al. 2015, and references therein) and small-scale eruptive events (e.g., Guglielmino et al. 2010, 2018, 2019, and references therein). In particular, it is expected that the magnetic reconnection can occur at different atmospheric heights depending on the overlying and/or background magnetic fields and on the strength and size of the EFR (Archontis et al. 2004; MacTaggart et al. 2015). Furthermore, Galsgaard et al. (2005, 2007) demonstrated that the magnetic reconnection strongly depends on the relative orientation between the magnetic field components of the emerging flux and the preexisting magnetic field. In fact, only when the two flux systems have almost antiparallel orientation is the magnetic reconnection efficient.

In this study, we analyzed multiwavelength observations of the active region (AR) NOAA 12546 obtained with state-of-the-art instruments to further investigate the trigger mechanism of penumbral brightening events. In particular, we studied the evolution of the penumbral area adjacent to two small-scale EFRs emerged in the AR on 2016 May 20. The analyzed region showed three brightening events detected from the chromosphere to the corona.

The paper is organized as follows: in the next Section we describe the observations and the data processing applied. In Section 3 we feature our analysis and results, which are summarized and discussed in Section 4. Section 5 presents our conclusions.

2.1. Observations

2.2. Data Processing

From 2016 May 15 to 25, ground-based and space-borne solar telescopes detected the on-disk passage of AR NOAA 12546 (AR hereafter), which emerged in the southern solar hemisphere close to the equator. This AR consisted of a leading, large and almost circular sunspot and a trailing, extended plage region. The sunspot, which was one among the largest such structures observed over the solar cycle 24, kept its regular shape nearly unchanged during the disk passage, while the plage region evolved significantly. The studied sunspot also displayed a large number of MMFs, mostly of type II, i.e., unipolar features with the same polarity of spot (e.g., Zuccarello et al. 2009, and references therein), seen to move almost radially away from the sunspot structure toward the boundary of the moat region.

Both the SDO/HMI and IBIS photospheric observations show that the penumbra of AR was formed by bright filaments and dark spines nearly radially aligned over most of the sunspot structure (Murabito et al. 2019), except for an area close to the EFRs observed on 2016 May 20.

3.1. Magnetic Environment in the Photosphere

Figure 1 shows the region analyzed in our study as deduced from SDO/HMI continuum observation performed on 2016 May 19 at 13:34 UT and the region of interest (ROI) discussed in the following and shown in Figures 3, 6, and 7. The bottom panel of Figure 1 displays the LOS magnetogram with overplotted horizontal velocity.

Zoom In Zoom Out Reset image size

Starting from 2016 May 19 12:00 UT the ROI showed in about 24 hr an increase of the LOS magnetic flux of about (2–4) × 10²⁰ Mx for the positive and negative polarity, and a flux decrease afterwards. This flux variation, which is reported in Figure 2, followed the evolution of the region summarized in the panels of Figure 3, which show maps from SDO/HMI observations at 9 representative times from 2016 May 19 08:00 UT to May 21 00:00 UT.

Zoom In Zoom Out Reset image size

Download figure:

Video Standard image High-resolution image

On 2016 May 19 08:00 UT the ROI only included the southern sector of the penumbra and moat area nearby (Figure 3, first row). A few hours later, at 16:00 UT, negative polarity patches (labeled N in Figure 3, panel (b2)) from the new flux of an EFR emerging at X = [20'', 50''], Y = [15'', 40''], had already merged with preexisting network field forming a small pore and a few fainter structures (labeled N₁ and N₂ in Figure 3, panel (b3)). These structures also received the negative polarity flux from a second EFR appearing on 2016 May 20 01:00 UT. Thereafter, positive polarity patches streaming out from the penumbra and from the second EFR merged to form small-scale structures in the southeast section of the ROI (at X = [5'', 30''], Y = [10'', 25'']) that later on, e.g., at 13:48 UT, were only merely discernible.

During the emergence of the second EFR, a bright lane (marked with red arrows in Figure 3, panels (a4)–(a5)) and a few dark patches (marked with the orange arrow in Figure 3, panel (a6)) appeared in the penumbra near the EFR, by creating a gap in the structure (see green arrows in Figure 3, panels (a4)–(a6)) that separated filaments with different orientation.

Later, the filaments closer to N₂ (black oval in Figure 3, panels (b7)–(b8)) broke away from the penumbra (black oval in Figure 3, panel (b8)) when the small-scale negative flux patches from the two EFRs and preexisting fields coalesced and got more aggregated (N₃ in Figure 3, panels (b7)–(b9)).

The animated version in Figure 3 (panels (a) and (b)), available in the online material clearly shows the entire evolution described above and the diverse MMFs discussed below.

The penumbra had a rather homogeneous structure with field inclination of about 60°–90° (Figure 3, third column), except for a few elongated patches with field inclination in the range 40°–60° (marked with the white oval in Figure 3, panels (c3)–(c7)) in the area that broke away. These features were clearly linked to the positive polarity patches of the EFR. The region also displayed two systematic flow patterns, which are shown in Figure 3 (fourth column). The former, directed toward the first EFR, led to the formation of N₃ by the merging of N₁ and N₂, while the second southern-ward pattern contributed to the evolution of the small-scale positive polarity features. Note that the emergence of both EFRs was characterized by strong shear motions with values of the horizontal velocity of the plasma of about 1 km s⁻¹.

The IBIS data provide a close-up view of the evolution of the ROI and of the physical properties therein. In particular, we show in Figure 4 the analyzed area from the IBIS observations taken on 2016 May 20 at 14:00 UT (left panels) and 17:00 UT (right panels), along with results from their inversion. The IBIS data at 14:00 UT indicate that the bright gap in the penumbra (yellow oval) already noticed in the SDO/HMI observations separates filaments with different orientation in the irregularly shaped penumbral sector. The dark structures (red arrow), whose length was of about 10'', decreased in size over time until they were no longer detected, e.g., on 2016 May 20 at 18:00 UT. Noticeably, the plasma LOS velocity in small-scale patches of these structures was opposite with respect to the regular Evershed flows nearby (Figure 4 panel (c)).

Download figure:

Video Standard image High-resolution image

Figure 4 (panels (d)–(e)) also displays filaments at the atmospheric height log τ = −1.0 with average values of the field strength and inclination of about 1.4 kG and about 45°, respectively, in the area that broke away from the penumbra, compared to the values for the same quantities of 0.9 kG and 80° for the homogeneous filaments nearby. Later, e.g., at 17:00 UT, the field pattern in the region was unchanged but for the slightly larger extension of the inhomogeneous patches.

3.2. Signatures from the Chromosphere to the Corona

At the chromospheric heights sampled by the IBIS Ca ii 854.2 nm measurements, the ROI showed a bundle of filaments with intense and repeated brightening at various locations, especially in the area of the two EFRs. These small-scale enhancements were observed in the form of either circular or elongated features (see the online animation of the Ca ii line core images in Figure 4). Moreover, at log τ = −4.6 (Figure 4, panels (f)–(g)) the ROI exhibited average values of the field strength of about 0.9 kG and field inclination ranging from 60° to 100°, except for some small-scale patches in the inhomogeneous area with field inclination lower than 40°.

The transients observed in IBIS Ca ii data were also manifested in the evolution of the average intensity measured over the ROI in the IRIS I1400, SDO/AIA, and Hinode/XRT observations. Figure 5 compares the light curves derived from the above data for the time interval of simultaneous IBIS and IRIS data, along with the variation of the X-ray flux measured by the GOES-15 satellite (Bornmann et al. 1996).

Zoom In Zoom Out Reset image size

All the light curves shown in Figure 5 display a couple of remarkable intensity increases standing out with respect to long-term and smooth variations. In particular, we considered the three abrupt intensity changes marked with vertical dashed lines in Figure 5 and referred to in the following as E1, E2, and E3. Two of these peaks, specifically E2 and E3, had clear counterparts in the variation of the X-ray flux measured by GOES-15 (red curve in Figure 5) and correspond to B-class flares. The above intensity increases, which lasted a few minutes for all the considered data, had different amplitudes depending on the event and wavelength, see, e.g., the peak value for the E2 and E3 events.

In an attempt to localize the source regions and to investigate the possible trigger mechanisms of the above events, Figure 6 displays SDO/HMI, SDO/AIA, and IRIS data taken over the time interval considered in Figure 5. The various panels in Figure 6 show brightening events whose footpoints and emission contours are overplotted on the I2832 observations for ease of comparison. We clearly noticed two homologous small-scale flaring events developed over a sigmoidal region (panels in the first and third rows of Figure 6). These events correspond to E1 occurring at 13:27 UT and E3 recorded by GOES-15 as B-class flares at 14:32 UT, about one hour later than E1. The eastern footpoint of the flaring region was lying in the penumbra, while the other was close to the patch of negative polarity flux N₁. It is worth noting that the latter footpoint can be easily identified in most of the SDO/AIA coronal observations that also show an intensity enhancement in the central part of the sigmoidal flaring area during E3. The latter area is only faintly detected in cotemporal chromosphere and TR, see, e.g., I1400 and A304 maps.

Zoom In Zoom Out Reset image size

Between the above homologous flaring events, we observed the occurrence of E2, a small-scale brightening appeared at about 14:08 UT in the left side of the ROI (panels in the second row of Figure 6) in the form of three elongated small-scale features seen in the A304 and A131 data, and only partly detected in I1400 observations. These bright patches were clearly aligned to penumbral filaments and cospatial to the small-scale positive flux features described in Section 3.1.

We further investigated the differences among E1, E2, and E3 by using I1400 observations, some examples of which are shown in Figure 7. The evolution of E1 was fast (Figure 7, top panels) and with simultaneous signatures seen in the light curves and observations of I1400, A131, and A304. The E2 event lasted about 7 minutes and reached its maximum extension 3 minutes after its start (Figure 7, middle panels) in the form of a bright knot appeared in the penumbra. Three minutes after the first appearance of the bright knot, a thin bright lane connected it to a small bright patch outside the penumbra. This bright lane became brighter with time and reached its maximum extension 4 minutes after its first appearance. E2 produced the strongest signature in the GOES-15 X-ray flux record, but only relatively small changes in the intensity emerging from the low chromosphere to the upper TR compared to the same atmospheric layers during E3. The latter transient produced intensity enhancements whose amplitude increased with the atmospheric height. E3 started at 14:20 UT, lasted about 13 minutes, and reached maximum extension (40'') 11 minutes after its beginning. The maximum of intensity in the I1400 data occurred 2 minutes later than recorded for all the SDO/AIA EUV channels.

Zoom In Zoom Out Reset image size

In Figure 8 we show the Hinode/XRT X-ray intensity maps acquired during the maximum brightening of E1, E2, and E3 events (second column), as well as 1 minute before (first column) and after (third column) the three events. These images display the same plasma behavior as reported in A131 and the strongest coronal brightening observed during E3 too. In particular, the shape of the E2 brightening (panels (b)) indicates the possible connection between opposite polarity patches of the diffuse field and MMF.

Zoom In Zoom Out Reset image size

3.3. IRIS Spectra

The IRIS slit sampled the region at X = [41'']. We show in Figure 9 examples of the line profiles measured during E1 and E3 at the positions marked by the symbols in Figure 7 in both the inhomogeneous and regular penumbra. From a comparison with Figure 6, we can note that the positions marked with triangles correspond to the eastern footpoints of the flaring loop during both E1 and E3. The red curves represent the spectra measured at the time of maximum brightening of each event. Unfortunately, there are no similar data available for E2.

Zoom In Zoom Out Reset image size

During the transients, the C ii and Si iv line profiles, which are formed in the TR, were enhanced by a factor of about 10 with respect to the line profiles measured in the quiet penumbral region (black curves), taken as a reference. At the time of maximum brightening red curves, the double-peaked C ii profiles showed blue lobes stronger than the red ones. In addition, the red wings of the C ii intensity profiles were broader than the blue wings. The C ii profiles showed asymmetric peaks varying in time, specifically the red lobe stronger than the blue one during the initial stages (blue curve) of the brightening, and opposite peak ranking at the time of maximum brightening (red curve). However, during E3 the C ii intensity profile (Figure 9, bottom-left panel) was less asymmetric than reported for E1. The Si iv line profiles during the two events were broadened by a factor of about 3 with respect to that measured in the reference regular penumbra, redshifted, indicating plasma velocities up to about 10 km s⁻¹.

Chromospheric Mg ii h&k lines measurements showed a behavior similar to C ii spectra, since the double-peaked profiles exhibited blue lobes stronger than the red ones at the peak of the brightening events. Moreover, the emission intensity profiles displayed no significant Doppler shift in all the available data. During the transients, the emission of the Mg ii h&k lines was enhanced by a factor 2 with respect to that measured in the regular penumbra. During E1 the profiles were rather symmetric, except for the data taken at maximum brightening (red curve). Conversely, before the E3 transient the profiles displayed red lobes stronger than the blue ones, contrarily to data during the maximum brightening (see blue and red curves in the bottom-right panel of Figure 9).

Noticeably, the Mg ii intensity profiles measured during both E1 and E3 showed emission in the Mg ii 2798.8 Å triplet (marked by the red arrows in Figure 9).

We presented our analysis of brightening events observed in the penumbra of AR NOAA 12546. This region included one among the largest sunspots of the solar cycle 24. The data analyzed in our study consist of multiwavelength and multi-instrument measurements obtained with state-of-the-art facilities, specifically with the IBIS, SDO/HMI, SDO/AIA, IRIS, and Hinode/XRT instruments. The bright filaments and dark spines forming the penumbra were nearly radially aligned over most of the sunspot structure and during its entire disk passage, except for a small area at a given time. We analyzed the phenomena occurring in this small area at representative times. We found that:

• 1.  
  A first EFR emerged close to the penumbral field, with a flux content of about 3 × 10²⁰ Mx.
• 2.  
  The newly emerged flux coalesced to preexisting network field and formed a small pore of opposite polarity flux with respect to the one in the umbra.
• 3.  
  Another small EFR occurred in the sunspot moat with a flux content of about 5 × 10¹⁹ Mx. The negative flux patches from this EFR merged with the pore formed from the previous EFR.
• 4.  
  During the emergence of the second EFR, a bright gap and some dark patches appeared inside the penumbra, in which the positive polarity of the second EFR was embedded. The bright gap separated filaments with different orientation than the one of the filaments nearby.
• 5.  
  The above photospheric features were preceded by the formation of elongated patches in the penumbra with magnetic field inclination ranging from 40° to 60°, different with respect to the inclination of the field in the filaments nearby, of about 60°–90°. These elongated patches also showed plasma LOS velocity opposite with respect to the regular Evershed flows nearby.
• 6.  
  The horizontal plasma flow exhibited two systematic velocity patterns directed from the penumbra toward the first EFR and to the southern side of the ROI. Strong diverging motions with values of the horizontal plasma velocity of about 1 km s⁻¹ characterized the area of the two EFRs.
• 7.  
  Three brightening events occurred during the above evolution of the photospheric magnetic environment, with clear signatures detected from the chromosphere to the corona. All the events had a footpoint located in the penumbra.
• 8.  
  Two of these events were recorded by GOES as B-class flares. One of these (E3) can be also classified as a homologous flare (e.g., Romano et al. 2015), since it showed the same configuration as a previous flaring event (E1). The first homologous flaring event lasted less than the second one. It left simultaneous signatures at all the chromospheric, TR, and coronal analyzed data. The second homologous event lasted about 13 minutes. The flaring region of E3 was about 40'' wide (29 Mm). It is worth mentioning that another homologous flaring event was seen in the SDO/AIA filtergrams after the end of the IRIS observations, at 16:39 UT.
• 9.  
  IRIS spectra relevant to the penumbral footpoints during the two homologous flares reveal enhanced emission and asymmetries in both the C ii and Mg ii h&k line profiles, and weak emission at the Mg ii 2798.8 Å triplet. In particular, at the time of maximum brightening, the blue lobe profiles are stronger than the red ones.

Observations of AR NOAA 12546 show a large number of short-living MMFs flowing away from the sunspot, some of them occurring in the ROI.

Figure 10 displays the evolution of a few MMFs relevant to the transient brightening events described above. To ease comparison, we overplotted the intensity contours of the studied events on all of the maps in Figure 10. Close to each footpoint of E1, we note type I MMFs (zoomed in the inserts in the left and right corners in Figure 10, panels (a), (b), (c), and (d)).

Zoom In Zoom Out Reset image size

The MMF at the eastern side streamed out from the penumbra. Eventually, at 13:24 UT, i.e., at the time the E1 occurred, a unipolar feature remained. The MMF at the western footpoint likely underwent a partial magnetic cancellation of the positive flux (see panel (d) in Figure 10), slightly before E1. After E1, we note two couples of type I MMFs (see insert in the left corner of Figure 10, panel (e)) that disappeared at the time of E3. At the western footpoint, the negative flux patch from previous MMF merging was approached by a positive flux feature leading to a partial magnetic flux cancellation event.

The photospheric structures that appeared in the penumbra facing the second EFR had vertical magnetic fields. In the same region we found a bright gap and diverging filaments. We consider these features to be the counterpart of the positive polarity of the EFR, which broke up the uncombed penumbra. This is likely due to the interaction of rising magnetic flux elements with the inclined magnetic field of the penumbra (Ichimoto et al. 2007; Su et al. 2010).

The same regions also exhibited LOS velocity opposite to the Evershed flow observed in the penumbral filaments nearby. We consider that these opposite flows are a further consequence of the above change in the field inclination. Counter Evershed flows were also found in umbral filaments (e.g., Kleint & Sainz Dalda 2013), during the formation of penumbrae (Schlichenmaier et al. 2012; Romano et al. 2014; Murabito et al. 2016, 2018), and over stable penumbrae (Louis et al. 2014; Siu-Tapia et al. 2017).

Lim et al. (2011) presented observations of two granular-like features (GLFs), about 1'' wide, at the tip of penumbral filaments and associated with MMFs. Both the observed GLFs were preceded by elongated darkening and one of these showed fine structure consisting of thin, dark, and bright threads. The two GLFs revealed significant chromospheric transients, specifically brightening in Hα and SDO/AIA 1600 Å images and jets. The relatively long-lasting, dark patches we observed in the penumbra were five times wider than reported by the above authors and showed no threads. Besides, the GLFs found by Lim et al. (2011) had opposite polarity to the field in the sunspot umbra, whereas the dark structure observed in our region had the same polarity as the umbral field.

The brightening observed by Bai et al. (2016), about 10'' wide, crossed a penumbra. It started at the border of the penumbra and then elongated toward the umbra, by leaving signatures detected from the chromosphere to the corona. The transient events analyzed in this work differ from the one reported by Bai et al. (2016) mostly for their larger extension and activation over an area that included both the penumbra and moat, as well as for their longer lifetime. Moreover, they do not show any expansion motion.

IRIS spectroscopic information is relevant to the eastern footpoints of events E1 and E3, embedded in the penumbra. According to Rathore et al. (2015), IRIS C ii intensity profiles are a powerful discriminant of upper chromospheric structures, and their asymmetries and line shifts are a robust velocity diagnostic. They showed that the C ii spectral profiles can have a single peak or two or more peaks, depending on how the source function varies with the atmospheric height. They attributed the asymmetry of intensity profiles to velocity gradients between the formation height of the intensity peaks and of the line core. In particular, by comparing the intensity profiles measured in active and quiet Sun regions, Rathore et al. (2015) reported line profiles with the blue peak stronger/weaker than the red one and attributed this profile feature to plasma with downflow/upflow above the peak formation height. In our data, we found: (i) the intensity profiles measured during the initial stages of both events with red peaks stronger than the blue ones, thus revealing plasma upflow at the time of those measurements; (ii) at maximum brightening, intensity profiles with blue peaks stronger than red ones, thus hinting at plasma downflow; (iii) E1 showed a stronger asymmetry of the C ii intensity profiles than that derived from E3; (iv) the intensity profiles measured after E3 display blue peaks stronger than red ones, whereas the ones measured after E1 do not clearly show the same feature.

The C ii spectra of the brightening event reported by Bai et al. (2016) are asymmetric with the red peak stronger than the blue one, in contrast with the observations analyzed in our study.

The intensity profiles of the chromospheric Mg ii h&k lines measured during E1 and E3 are also double-peaked but rather symmetric except for the data taken at maximum brightening which have blue peaks stronger than red ones as for plasma downflow at the formation height of the intensity peaks (Leenaarts et al. 2013). Finally, the width of the intensity profiles of the Mg ii h&k lines does not change significantly during the studied events as for a temperature increase associated with the brightening not occurring in the lower chromosphere (Pereira et al. 2015). It is worth noting that the Mg ii 2796.35 Å intensity profile presented by Bai et al. (2016) is symmetric as expected from quiet plasma at the formation height of the analyzed intensity peak in contrast to our data.

Indeed, we notice that the blue peaks are stronger than the red ones for both the C ii and Mg ii h&k line data, especially during event E3, which is the stronger brightening. This asymmetry suggests the occurrence of chromospheric evaporation at the penumbral footpoints following a magnetic reconnection process occurring at higher atmospheric heights, as hypothesized by Bai et al. (2016).

Furthermore, the Mg ii intensity profiles we measured during both E1 and E3 show a weak emission of Mg ii 2798.8 Å triplet. Pereira et al. (2015) attributed this rare emission to a steep temperature increase in the lower chromosphere. The Mg ii 2798.8 Å emission can involve either the far wings or core of the line, depending on the range of interested column masses. In particular, the emission seen in the wings of the intensity profile points to plasma heating occurring deeper down the chromosphere than that recorded at the line core, which is suggestive of pure chromospheric heating (Hansteen et al. 2017). Our data show Mg ii 2798.8 Å emission in the line core, thus revealing that E1 and E3 produced a steep temperature increase at chromospheric heights.

We also considered the results presented above with respect to the findings of MHD numerical simulations of coronal and chromospheric microflares performed by Jiang et al. (2012). In these numerical models the microflares leave different atmospheric signatures and have different size depending on the atmospheric height of the magnetic reconnection episode responsible for the flaring. In particular, in the case of the simulated coronal microflares, the size of the brightening region ranges from 15'' to 22''. The upper atmosphere response to the event simulated by Jiang et al. (2012) includes both hot plasma jets (≈1.8 × 10⁶ K) observed in the EUV/X-ray bands and cold jets (≈10⁴ K) revealed as Hα/Ca brightening events. Moreover, the numerical studies indicate a time delay between the signatures of the reconnection process seen in the upper and lower atmosphere. In particular, the enhanced EUV/X-ray emission was found to appear before the Hα/Ca brightening. Jiang et al. (2012) roughly derived an estimate for this time delay. By taking into account the temperature response at 1500 km (the height formation of Hα, Vernazza et al. 1981) those authors reported a time delay of about 3–5 minutes. The spatial scale of the studied brightening events in their signatures in the various layers of atmosphere agree with those reported by the simulation results of Jiang et al. (2012) for coronal microflares.

We have seen that the magnetic configuration analyzed in Figure 10 for the three penumbral brightenings suggest a clear link between the EFR, MMFs, and the preexisting penumbral field. In regard to the trigger mechanism of the observed transient events, E1 and E3 seem due to the interaction between the penumbral fields and the EFR. In particular, the outer negative footpoint of the EFR interacts with the MMFs departing from the penumbra, as suggested in the model proposed by Kano et al. (2010; flux emergence case). Moreover, the positive footpoint of the EFR inside the penumbra also comes into contact with negative polarities around the penumbra originated by either MMFs or flux return patches. This interaction may activate magnetic reconnection events. On the other hand, E2 seems due to the interplay between MMFs and the preexisting diffuse field to the south of the sunspot, as suggested by the crossing enhanced connections visible in the Hinode/XRT images. This is reminiscent of the model introduced by Kano et al. (2010; MMFs interaction case). Therefore, for the three studied events there is a tight distinctive interplay among EFR, MMFs, and preexisting magnetic flux systems.

In this paper, we analyzed multiwavelength observations of the AR NOAA 12546 obtained with state-of-the-art instruments to further investigate the trigger mechanism of penumbral brightening events. In particular, we studied the evolution of the AR penumbral area adjacent to two small-scale EFRs that emerged in the AR on 2016 May 20. The analyzed region showed three brightening events detected from the chromosphere to the corona. From analysis of IBIS, SDO/HMI, SDO/AIA, IRIS, and Hinode observations we found that two of the studied events likely derived from reconnection processes occurring at different atmospheric heights that were activated by the interaction of the preexisting fields with either the EFRs or MMFs.

Penumbral transient brightening events have been attributed to magnetic reconnection processes occurring in the low corona (Bai et al. 2016). The spatial scale of the events analyzed in our study in their signatures in the atmosphere agree with those reported by the simulation results of Jiang et al. (2012) for coronal microflares. Moreover, our observations confirm the model for nanoflare events sketched in Figure 12 of the paper by Kano et al. (2010), both for the flux emergence and MMFs interaction cases. They also complement the results presented by Bai et al. (2016) showing the signatures of the chromospheric evaporation following the flaring events not found by those authors.

We plan to further analyze penumbral brightening events to understand the role of the less energetic occurrences in the heating of the chromosphere and upper atmosphere. Spectropolarimetric observations to accurately infer the vector magnetic field at different heights in the solar atmosphere, and simultaneous images and spectra from the photosphere outward, such as those expected by the upcoming new generation solar observatories (DKIST, Tritschler et al. 2015; Solar Orbiter, Müller et al. 2013) are essential to perform these studies.

The authors wish to thank the anonymous referee for helpful comments. The authors are grateful to Doug Gilliam for the IBIS observations analyzed in this study. S.L.G. and M.M. wish to thank Peter Young for useful discussion. This work has received funding from the European Union's Horizon 2020 research and Innovation program under grant agreements No. 739500 (PRE-EST) and No. 824135 (SOLARNET). S.L.G. acknowledges support from the Università degli Studi di Catania (Piano per la Ricerca Università di Catania 2016–2018—Linea di intervento 1 "Chance"; Linea di intervento 2 "Ricerca di Ateneo—Piano per la Ricerca 2016/2018"). This research has made use of NASA's Astrophysics Data System. IRIS is a NASA small explorer mission developed and operated by LMSAL with mission operations executed at NASA Ames Research center and major contributions to downlink communications funded by the Norwegian Space Center (NSC, Norway) through an ESA PRODEX contract.