Triggering Process of the X1.0 Three-ribbon Flare in the Great Active Region NOAA 12192

In this study, we basically used the analysis method developed in Bamba et al. (2013). The analysis method was originally developed for Hinode/SOT data, but Bamba et al. (2014) already examined the applicability of the analysis method to the SDO data sets. Here, we briefly summarize the procedures of the analysis method.

We used HMI level 1.5 line-of-sight (LOS) magnetograms (hmi.M_45s series) and AIA level 1.0 data (aia.lev1_euv_12s and aia.lev1_uv_24s series). We first calibrated all the HMI filter magnetograms and all the AIA images using the aia_prep procedure in the Solar Soft-Ware (SSW) package. By this process, spatial fluctuations were reduced, and the images were rotated so that the solar EW and NS axes are aligned with the horizontal and vertical axes of the image, respectively. Moreover, LOS magnetograms and AIA images were resampled to the same size because the pixel scales are different between HMI and AIA. Thus, the positions of the LOS magnetograms and AIA images were aligned. Next, we chose an HMI filter magnetogram and an AIA image closest in time, and these two images were superimposed on each other. We drew the PILs and the strong emission contours in the AIA image on the LOS magnetogram at each time. The PILs were drawn on the AIA images at each time as well.

The SP scan data were calibrated using the sp_prep procedure (Lites & Ichimoto 2013) in the SSW package assuming a Milne–Eddington atmosphere. The inversion code MEKSY (developed by Dr. Takaaki Yokoyama) was adopted, and the 180° ambiguity in the vector magnetograms is resolved using the AZAM utility (Lites et al. 1995). Note that the observed "LOS" and "transverse" magnetic field vectors (i.e., image-plane magnetic field) were converted to the heliographic magnetic field B_x, B_y, and B_z using Equation (1) of Gary & Hagyard (1990).

We investigated the distribution of the magnetic shear over the AR before the flare onset, using the vector magnetogram obtained from the SP scan data. We first calculated the potential field using the fff procedure in the nlfff package (developed by Dr. Yuhong Fan) in SSW. Then we measured the angles between the potential field vector ${{\boldsymbol{B}}}_{p}$ and horizontal field vector ${{\boldsymbol{B}}}_{h}=\sqrt{{{B}_{x}}^{2}+{{B}_{y}}^{2}}$ in each pixel, and we defined these angles as the relative shear angle χ. The relative shear angles are defined between $\pm 180^\circ$, where 0° means vectors ${{\boldsymbol{B}}}_{p}$ and ${{\boldsymbol{B}}}_{h}$ are oriented in the same direction. The direction of the vector ${{\boldsymbol{B}}}_{h}$ deviates from that of the vector ${{\boldsymbol{B}}}_{p}$ counterclockwise (clockwise) when the magnetic helicity is positive (negative) and the value of χ is positive (negative). We colorize the relative shear angles, as shown in Section 4.3.

The temporal evolution of the brightenings and flare ribbons in AIA 1600 Å, which is sensitive to the emission from the upper chromosphere and the transition region, is shown in Figure 2. We can see the contours of those brightenings with the magnetic field structure in Figure 3, where white/black indicates the positive/negative polarity of the LOS magnetic field. The green lines and the red contours denote the PIL (line of 0 G) and the significant brightenings (2000 DN) in AIA 1600 Å, respectively. The green lines in Figure 2 are the same as those in Figure 3. The positive polarity that is intruding on the negative region N1 (henceforth IPP: intruding positive polarity) and the weak negative region N2 is located between the major bipole P1 and N1, as can be seen in Figure 3(a). We can also see the small brightenings B1 and B2, which are indicated by the yellow arrows in panel (a) of both Figures 2 and 3. These brightenings are seen intermittently at the west side of the IPP at 14:48–14:58 UT. The other brightening B3 is also seen in the IPP in panel (a). The flare ribbons of the C5.1 flare appear about five minutes later, as shown in panel (b). The positive ribbons CR1 and CR2 are clearly seen in the P1 and IPP regions. There are two negative ribbons in the N1 region, but these brightenings are connected to each other in the AIA 304 Å image (see Figure 4(b) and the following description). We then identify these brightenings in the N1 region as the negative ribbon CR3, as seen in Figure 2(b). B1 and B2 disappear in the last minute before the C5.1 flare, while B3 enhances as the flare ribbon CR2 in panel (b). It is inferred that B1 and B2 are caused by a local magnetic reconnection that occurred at the west side of the IPP before the C5.1 flare onset, and that the B1 and B2 brightenings appear at the footpoints of the small magnetic loops connecting the IPP and N2. It is also suggested that B3 is located at the footpoint of the magnetic loop that connects the IPP and N1 and which corresponds to the C5.1 flare. Once the three ribbons CR1, CR2, and CR3 of the C5.1 flare disappear by 15:15 UT, the C9.7 flare ribbons appear from 15:40 UT as seen in panel (c). The C9.7 flare also shows three flare ribbons CR1, CR2, and CR3 at the same location as the ribbons of the previous C5.1 flare. The three ribbons of the C5.1 and C9.7 flares almost never propagate with time. A faint ribbon-like brightening remained at the region where CR1 is seen even after the three ribbons disappeared, as seen in Figure 2(d). Panel (e) shows the initial flare ribbons of the X1.0 flare that appear about one hour after the C9.7 flare. In this phase, XR1 (positive ribbon) and XR2 (negative ribbon) at the west side/middle are more clearly seen than XR3 (negative ribbon) at the east side. These three ribbons slowly grow  between 16:35–17:10 UT as shown in panel (f). In particular, XR1 and XR3 get longer in the southward and northward directions, respectively. However, they almost never propagate to the outer sides (i.e., westward and eastward) with time. XR2 disappears first in the late phase, while XR1 and XR3 remain more than one hour after the flare onset.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Figure 4 shows the AIA 304 Å images with the same FOV and almost the same time as Figures 2 and 3. The PILs are overplotted only in panel (a). The intermittent brightenings B1, B2, and B3 are indicated by the yellow arrows in panel (a), and these are seen at the west side of the IPP as well as shown in the AIA 1600 Å image of Figure 2(a). The brightenings B1 and B2 are connected to each other via a small loop striding over the local PIL at the west side of the IPP. It is clearly seen at 14:48–15:00 UT although it persisted from 13:00 UT to just before the onset of the C5.1 flare. It also suggests the existence of small magnetic loops that connect the IPP and N2 on the west side of the IPP. On the other hand, the brightening B3 is connected with N1, where CR3 will appear, as clearly seen in Figures 4(a) and (b). The brightenings, which will be the flare ribbons CR1 and CR2, are already seen from 13:00 UT, and  CR3 also appears from 14:00 UT. The three ribbons start becoming enhanced from 14:30 UT, and these will be the three ribbons of the C5.1 flare in panel (b). There is a filament on the PIL between P1 and N2, and some bright points start to move along the filament just before the C5.1 flare (from 14:58 UT). The three ribbons CR1, CR2, and CR3 remain even after the onset of the C5.1 flare, and the bright point motion along the filament continues and becomes intense. These bright structures flow down along the filament to the footpoints in the P1 and N1 regions, and then the three ribbons CR1, CR2, and CR3 of the C9.7 flare are enhanced from 15:47 UT (see panel (c)). Note that the strong brightening, which looks like another flare ribbon between CR1 and CR2 in panel (c), is not a flare ribbon, but a bright point moving along the filament like the tether-cutting reconnection (see Figure 6 and the following description). The flare ribbons remain clearer than those seen in Figure 2(c), especially CR1 and CR3, which are clearly seen in Figure 4(d). The remaining ribbons are enhanced and a new flare ribbon along the filament gradually appeared in the N2 region from 16:30 UT, becoming XR2 in panel (e). This is the X1.0 flare, and the flare ribbons XR1 and XR3 correspond to CR1 and CR3, respectively. CR2 also remains, but it is not seen in the AIA 1600 Å images (see Figures 2(e) and (f)). In the initial phase of the X1.0 flare (until 16:55 UT), the ex-CR2 brightening and the negative ribbon XR3 are connected, as seen in Figure 4(e). It corresponds to the first peak of the X1.0 flare in the GOES soft X-ray light curve (cf. Figure 1). Then, the bright bridge connecting ex-CR2 and XR3 weakened, and XR3 becomes longer to the north as seen in panel (f). XR1 and XR2 are also enhanced at that time, and the post-flare loop connecting XR1 and XR2 appeared. XR2 is weakened first while XR1 and XR3 likewise remain in the AIA 1600 Å images.

The coronal loops in the AIA 131 Å images are shown in Figure 5, with the PILs only shown in panel (a). In the early phase (panel (a)), the bright loops L1 and L2 connect IPP–N1 and P1–N1, respectively. The IPP has continuously emerged from October 19, and it has intruded onto the negative sunspot (N1). Therefore, it is suggested that a strong electric current layer is formed in between L1/L2 and the overlying magnetic arcade that connects P1–N1 (it is located higher/more outside than L2). In this layer, coronal loops are heated by the magnetic reconnection between L1/L2 and the overlying magnetic arcade, even though the field directions of the magnetic loops do not differ strongly. Note that there is another faint loop, L3, indicated in panel(a), and it might connect P1–N2 (and also P1–N1). The footpoints of L1 and L2 brighten as seen in Figure 4(a). These heated loops L1 and L2 are enhanced at the onset of the C5.1 flare as seen in Figure 5(b). Once the intensity of the coronal loops decrease at 15:15–15:35 UT, the loops L1 and L2 become bright again, and the three ribbons of the C9.7 flare appear as indicated by the yellow broken lines in panel (c). Then, the intensity of the coronal loops continuously increases until the onset time of the X1.0 flare (panel (d)). The coronal loops connecting XR1 and XR2 enhance (panel (e)), XR3 appears, and the faint coronal loops connecting XR1 and XR3 can be seen in panel (f).

Zoom In Zoom Out Reset image size

Interestingly, the tether-cutting magnetic reconnection process (Moore et al. 2001) is clearly seen between the C5.1 and C9.7 flares. Figure 6 shows the AIA 171 Å images with a similar FOV to Figures 2–5. The PILs are overplotted only in panel (a), and there is a filament as indicated by the red arrow in panel (b). We can see the faint loop structures that intersect at point O and along the filament. Some small brightenings frequently move along these faint loops as illustrated by the blue arrows in panel (b). These bright structures are seen from 13:00 UT, and these are enhanced from 15:00 UT, just before the C5.1 flare. The motion along the loops becomes faster, and the loops slightly expand between 15:40 and 16:00 UT (panel (c)). The flare ribbons CR1 and CR3 of the C9.7 flare appear at the footpoints of the faint loops, as illustrated by the red broken lines in panel (d), and we can see the three ribbons in the chromosphere at that time (see Figure 4(b)). These are clearly seen in Movie 1. A similar motion of the bright structure along the filament is also seen in the AIA 304 Å images just after the C5.1 flare to the onset of the C9.7 flare. It suggests that the tether-cutting magnetic reconnection occurs in the C5.1 and C9.7 flares at point O.

The bright points B1, B2, and B3; the tether-cutting reconnection; and the bright coronal loops connecting XR1 and XR2 are also seen in the AIA 193 Å images (Figure 7) with much less saturation. In panel (a), the brightenings B1, B2, and B3 are clearly seen, and they are also seen in the AIA 1600 and 304 Å images. The brightening B3 corresponds to the positive footpoint of the coronal loop L1 that connects the IPP and N1 (cf. Figure 5(a)). The positive footpoint of L2 is also seen as a bright point in the P1 region, as seen in Figure 7(a). The tether-cutting reconnection is also seen in panel (b), although it is fainter than that seen in the AIA 304 and 171 Å images. After the tether-cutting reconnection, from 16:30 UT, the faint coronal loops rooted in XR1 and XR2 drastically expand to the southwest, as illustrated by the blue arrow in panel (c). This expanding motion of the faint coronal loops correspond to the enhancement of the coronal loops seen in Figures 5(d)–(f). Then, the flare ribbons XR1 and XR3 elongate southward and northward, respectively.

Zoom In Zoom Out Reset image size

The flare ribbons of the two C-class flares (the C5.1 and C9.7 flares) appear at the same location, as we described in the previous section. Therefore, here we compare the locations and evolution between the C9.7 and X1.0 flare ribbons. Figure 8 shows the distribution of the C9.7 and X1.0 flare ribbons. The background images are the HMI LOS magnetograms at the onset time of the X1.0 flare (16:31 UT), and the green lines indicate the PILs. The red contour outlines the brightenings in AIA 1600 Å, such as the flare ribbon at 15:50 UT (just after the C9.7 flare onset), in both panels (a) and (b). The blue contours outline the brightenings at 16:45 UT (before the X1.0 flare onset) in panel (a) and at 17:06 UT (after the X1.0 flare onset) in panel (b), respectively. The initial flare ribbons of the C9.7 and X1.0 flares are seen in panel (a). XR1 and XR3 of the X1.0 flare are located slightly to the outer side of CR1 and CR3 of the C9.7 flare. XR1, XR3, CR1, and CR3 slightly propagate to the outer side (i.e., westward and eastward) with time, although XR1 and XR3 clearly grow southward and northward, as noted in Section 4.1. Therefore, it is suggested that the C5.1, C9.7, and X1.0 flares occur serially, corresponding to almost the same magnetic structures. In other words, the X1.0 flare is an extension of the C-class flares, and the X1.0 flare is driven by the reconnection of the magnetic field anchored in the outer region of the C-class flare ribbons.

Zoom In Zoom Out Reset image size

The vector magnetic field in the central part of the AR was observed by Hinode/SP between 11:00 and 11:33 UT, as shown in Figure 9. In panel (a), the white/black background indicates the positive/negative polarity of B_z. Green lines indicate the PILs (line of ${B}_{z}=0\,{\rm{G}}$). The horizontal magnetic field vector ${{\boldsymbol{B}}}_{h}$ (red arrows) is overplotted on the B_z map. The horizontal field is strongly sheared along the PIL located between the P1 and N2 regions. It suggests that the major magnetic helicity along the PIL of the AR is negative. On the other hand, the vectors at the west side of the IPP stride over the PIL locally toward the northwest. It is consistent with the shape of the small bright loop connecting B1 and B2 in Figure 4(a). Therefore, the local magnetic field at the west side of the IPP has a positive magnetic helicity, which is opposite that of the major magnetic helicity along the PIL. It is more clearly seen in panel (b), which shows the relative shear angle χ, defined as the angle between the potential field ${{\boldsymbol{B}}}_{p}$ and the horizontal field ${{\boldsymbol{B}}}_{h}$ at each point. The black lines indicate the PILs, which is the same as the green lines in panel (a). The relative shear angle χ along the flaring PIL (between P1 and N2) and that on the west side of the IPP correspond to the shear angle ${\theta }_{0}$ of the AR and the azimuth ${\varphi }_{e}$ of the small magnetic disturbance in Kusano et al. (2012), respectively. Obviously, χ is around $-90^\circ$ (blue) along the PIL but χ is 90° (red) on the west side of the IPP. Therefore, the distribution of the relative shear angle χ suggests that the west side of the IPP satisfies the RS-type flare-trigger condition proposed by Kusano et al. (2012). They proposed that shear cancellation between the global magnetic field and the RS-type small magnetic field structure can trigger a flare, and significant brightenings should be observed in the solar atmosphere during the shear cancellation over the RS-type structure. In our case, significant brightenings, B1 and B2, are observed at the west side of the IPP where the magnetic shear is opposite the shear of the global magnetic field along the flaring PIL. It is consistent with the theoretical prediction. Therefore, the west side of the IPP could be the location where the flares were triggered in the AR.

Zoom In Zoom Out Reset image size

In this study, we extrapolated the magnetic field in the corona, and compared it to the observed features. We derived the coronal magnetic field from the vector magnetic field data taken by HMI at October 25, 15:00 UT, using the NLFFF extrapolation method developed by Inoue et al. (2014). Figure 10 shows the coronal magnetic field lines anchored to the flare ribbons of the C9.7 and the X1.0 flares. The grayscale is an HMI LOS magnetogram at October 25, 15:00 UT. The red/blue contour outlines the brightenings (700 DN) in AIA 1600 Å, such as the flare ribbons observed at 15:50 UT/17:03 UT. The blue and orange tubes indicate the coronal magnetic field lines. The orange tubes are anchored to the ribbons of the C9.7 flare (CR1, CR2, and CR3), as shown in panels (a) and (b). On the other hand, the sky-blue tubes are anchored to the three ribbons of the X1.0 flare (XR1, XR2, and XR3), as seen in panels (c) and (d). The small magenta tubes indicated by the magenta arrow are the local magnetic field lines at the west side of the IPP. Obviously, the sky-blue magnetic loops connecting XR1 and XR3 are located at the outer edge of the orange loops connecting CR1 and CR3. These are consistent with the observed features that the flare ribbons of the X-class flare are on the outer side of the ribbons of the C-class flares, as shown in Figure 8.

Zoom In Zoom Out Reset image size

Here, we discuss the flare-trigger scenario of consecutive C- and X-class flares by comparing with the theoretical model proposed by Chen & Shibata (2000). In this region, there are four kinds of possible magnetic connectivity: P1–N1, IPP–N1, IPP–N2, and P1–N2. Moreover, there is a filament trapped by the P1–N2 loops as seen in the AIA images (Figures 4, 6, and 7). If we focus on the connectivities IPP–N1 and P1–N2, the topology of the magnetic field is consistent with case B of Chen & Shibata (2000), in which the emerging flux appears on the outer edge of the filament channel. The IPP–N1 and P1–N2 connectivities have the same orientation relative to each other's main flaring PIL, and IPP–N1 can be considered as a flux emerging on the east edge of the filament channel P1–N2. Magnetic reconnection of the IPP–N1 flux, with the flux passing over the main flaring PIL (between P1 and N2), will weaken the latter flux. It can destabilize the current-carrying flux in the core region above the main flaring PIL, and the reconnection also produces short loops at the PIL between IPP–N2 and long loops connecting P1–N1.

In this scenario, to trigger an X-class flare, it is required that the flux rope exists on the PIL and that the flux rope erupts. The former is likely because a multitude of flares including three X-class flares had occurred in the same PIL between P1–N2. However, the latter is not consistent with the observed features that the filament along the PIL between P1–N2 does not move through the two C-class and X-class flare processes (cf. Figure 4). It indicates that the eruption of the flux rope, such as that required in the case B scenario, is unlikely. Moreover, the IPP is intruding onto N1, i.e., the magnetic loop connecting IPP–N1 moves away from P1–N2, as the IPP emerges; this is  described in Section 4.1. This motion is in contrast to the scenario expected in case B of Chen & Shibata (2000), which requires that IPP–N1 approach P1–N2. Therefore, the flare trigger by the case B scenario may be difficult, although the magnetic field topology is consistent with this case.

Another flare-trigger scenario is the RS type, which is proposed by Kusano et al. (2012). In this scenario, the key structure is the small magenta loops connecting IPP–N2, which are indicated by the magenta arrow in Figure 10. We found that the west side of the IPP satisfied the condition of the RS-type flare-trigger field based on the distribution of the relative shear angle (cf. Figure 9(b)). The long orange loops connecting P1–N1 are topologically equivalent to the overlying blue field line in Figures 4(a)–(c) of Kusano et al. (2012). Therefore, the following scenario is conceivable.

(1) Step-1 (trigger for the C-class flares). Magnetic shear cancellation proceeds between the IPP–N2 (magenta) loops and the P1–N1 (orange) loops at the west side of the IPP. The shear-canceling reconnection locally heats the atmosphere and causes chromospheric evaporation above the west side of the IPP. These local heating and evaporation are observed as the brightenings B1, B2, and B3 in the AIA images (Figures 2–4). The reconnected fluxes are transferred into the small loops connecting IPP–N1 and P1–N2, and the magnetic pressure above the west side of the IPP must decrease. The loops IPP–N1, P1–N1, and P1–N2 are heated by the reconnection. These are observed as the bright coronal loops L1, L2, and L3 in AIA 131 Å (Figure 5), although the location of the negative footpoint of L3 is fuzzy. The schematics of the magnetic field line before and after step-1 reconnection are shown in Figures 11(a) and (b). The orange and magenta loops are equivalent to the P1–N1 (orange loops) and IPP–N2 (magenta loops) loops in Figures 10(a) and (b). The small green loops in Figure 11(b) represent the small loops formed by the magnetic reconnection between the RS-type trigger field (magenta loops) and the overlying sheared field (orange loops).

Zoom In Zoom Out Reset image size

(2) Step-2 (reconnection of the C-class flares). The sheared loop P1–N1 collapses inward because an inflow (such as that studied by Ugai & Shimizu 1996) drags flux into the region where magnetic pressure was decreased by the step-1 reconnection, i.e., above the west side of the IPP. Then, reconnection occurs in the collapsed loops (inside the orange loops connecting P1–N1 in Figures 10, 11(b)), and it shows the three ribbons CR1, CR2, and CR3 of the C5.1 and C9.7 flares. The reconnection is observed as the tether-cutting reconnection in the AIA 171 Å (Figure 6 and Movie 1) and 193 Å (Figure 7) images. Note that the CR2 in the IPP was the brightening B3 in the pre-flare phase, and that the brightening in the IPP represents the intensive electric current layer formed by the step-1 reconnection (see the Appendix).

(3) Step-3 (trigger for the X1.0 flare). It is inferred that the long twisted flux rope connecting P1–N1 might be formed by the tether-cutting reconnection, as illustrated by the thick orange line in Figure 11(c), according to Moore et al. (2001) and Kusano et al. (2012). There  are other sheared loops (sky-blue loops connecting P1–N1 and P1–N2) above the orange loops and the long twisted flux rope. The flux rope gradually rises upwards with the overlying sky-blue loops, then the magnetic fluxes are transferred upward, and the magnetic pressure over the PIL between the P1 and N1 regions decreases. After the magnetic pressure is decreased sufficiently, magnetic reconnection between the sky-blue loops occurs. It causes the X1.0 flare and produces the three ribbons (XR1, XR2, and XR3) at the footpoints of the sky-blue loops. The reconnection might produce a further long twisted flux rope connecting P1–N1, then the flare ribbons XR1 and XR3 are elongated to the south and north as magnetic reconnection with different field lines.

Therefore, the C5.1 and C9.7 flares can be originally triggered by the RS-type structure (the magenta loops connecting the IPP–N2) at the west side of the IPP, and the rise of the flux rope formed by the C-class flares triggers the X1.0 flare, in a three-step process. In Kusano et al. (2012), they put a small bipole that satisfies the RS-type condition just above the highly sheared PIL. Then, shear cancellation (step-1 reconnection) between the trigger field and the pre-existing sheared field occurs on the PIL. However, in the current AR, the IPP–N2 magenta loops are not located on the PIL between the major flare ribbons XR1 and XR2, i.e., the flare-trigger field is located slightly off the main flaring PIL. In the Appendix, we discuss the applicability of the RS-type scenario in the case of a trigger field that exists far from the main flaring PIL. According to an extra simulation, an RS-type field can work as a trigger of the flare even if it is located away from the highly sheared PIL. It is suggested that the RS-type flare-trigger process has a certain flexibility with the distance of the flare-trigger field from the highly sheared PIL.

Figure 12(a) is the side view of Figure 10(f) in which the vertical cross-section displays the decay index n between 0 and 1.5. The decay index is defined as $n=-d\mathrm{ln}| {{\boldsymbol{B}}}_{p,h}| /d\mathrm{ln}| z|$, where ${{\boldsymbol{B}}}_{p,h}$ and z are the horizontal component of the potential magnetic field calculated from the observed B_z and the height, respectively (e.g., Shafranov 1966; Kliem & Török 2006). Obviously, the coronal magnetic loops anchored to both the C9.7 and the X1.0 flare ribbons have not reached the isosurface of n = 1.5, and the flux tube remained stable because the overlying magnetic field above the flux tube did not reach the critical decay index. Therefore, the highly sheared magnetic loops, which are related to the flares, have not satisfied a condition of the torus instability, although they satisfy the flare-trigger condition summarized in Inoue et al. (2015). In other words, the closed loops (red loops in panel (b)), which are over the orange and the sky-blue loops, might prohibit the twisted flux rope, which is created in the central part of the AR via tether-cutting reconnection with the C- and X-class flares, from erupting (Inoue et al. 2016).

Zoom In Zoom Out Reset image size

According to Kusano et al. (2012), the causality between the flux rope eruption, which may create CMEs, and the magnetic reconnection of flares is different between the two types of flare-trigger processes. The RS-type flare-trigger process is a "reconnection-induced eruption" process where the tether-cutting reconnection of the flare is triggered before the eruption of the flux rope. It means that the RS-type flare does not necessarily cause a large flux rope eruption such as a CME. Therefore, it is also consistent that the flares triggered by the RS-type magnetic configuration did not have an associated CME in the AR.