Soft X-ray Fluxes of Major Flares Far Behind the Limb as Estimated Using STEREO EUV Images
N.V. Nitta, M.J. Aschwanden, P.F. Boerner, S.L. Freeland, J.R. Lemen, J.-P. Wuelser
aa r X i v : . [ a s t r o - ph . S R ] A p r Solar PhysicsDOI: 10.1007/ ••••• - ••• - ••• - •••• - • Soft X-ray Fluxes of Major Flares Far Behind theLimb as Estimated Using STEREO EUV Images
N.V. Nitta · M.J. Aschwanden · P.F. Boerner · S.L. Freeland · J.R. Lemen · J.-P. Wuelser c (cid:13) Springer ••••
Abstract
With increasing solar activity since 2010, many flares from the backside ofthe Sun have been observed by the
Extreme Ultraviolet Imager (EUVI) on eitherof the twin STEREO spacecraft. Our objective is to estimate their X-ray peakfluxes from EUVI data by finding a relation of the EUVI with GOES X-rayfluxes. Because of the presence of the Fe xxiv line at 192 ˚A, the response of theEUVI 195 ˚A channel has a secondary broad peak around 15 MK, and its fluxesclosely trace X-ray fluxes during the rise phase of flares. If the flare plasma isisothermal, the EUVI flux should be directly proportional to the GOES flux.In reality, the multithermal nature of the flare and other factors complicate theestimation of the X-ray fluxes from EUVI observations. We discuss the uncer-tainties, by comparing GOES fluxes with the high cadence EUV data from the
Atmospheric Imaging Assembly (AIA) on board the
Solar Dynamics Observatory (SDO). We conclude that the EUVI 195 ˚A data can provide estimates of theX-ray peak fluxes of intense flares ( e.g. , above M4 in the GOES scale) withuncertainties of a factor of a few. Lastly we show examples of intense flares fromregions far behind the limb, some of which show eruptive signatures in AIAimages.
Keywords:
Extreme ultraviolet · Flares · Photometry · SDO · Soft X-rays · STEREO
1. Introduction
The
Solar Terrestrial Relations Observatory (STEREO; Kaiser et al. , 2008) waslaunched in October 2006, and continues to deliver unique views of the Sunnot accessible from any Earth-bound instruments. STEREO consists of theAhead (A) and Behind (B) spacecraft, which drift about 22 ◦ a year in opposite Lockheed Martin Solar and Astrophysics Laboratory,A021S, Bldg 252, 3251 Hanover Street, Palo Alto, CA 94304USA email: [email protected],[email protected],[email protected],[email protected],[email protected],[email protected]
SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 1 .V. Nitta directions from the Sun-Earth line. Both are equipped with nearly identicalcopies of the
Extreme Ultraviolet Imager (EUVI; Wuelser et al. , 2004) as partof the
Sun Earth Connection Coronal and Heliospheric Investigation (SECCHI;Howard et al. , 2008). In combination with images from the
Atmospheric ImagingAssembly (AIA; Lemen et al. , 2012) on board the
Solar Dynamics Observatory (SDO; Pesnell, Thompson, and Chamberlin, 2012) we have observed the fulllongitudes of the solar corona in EUV since February 2011.With the increasing separation of STEREO from the Sun-Earth line togetherwith increasing solar activity since 2010, the EUVI has observed many flares farbehind the limb from Earth view. Their X-ray emission is therefore completelyocculted by the limb, and one of the frequently asked questions is “what wouldbe the magnitude of such a flare in soft X-rays?” Data from the GOES
X-raySensor (XRS) have played a major role in our understanding of the physics ofsolar flares as they have been widely used to derive plasma parameters ( e.g. ,Feldman et al. , 1996; Battaglia, Grigis, and Benz, 2005; Ryan et al. , 2012). Theflare class defined by the peak flux in the 1 – 8 ˚A band is generally thoughtto be a good measure of the released energy. In particular, the so-called X-class ( F − ∼ > − W m − ) flares are often treated with special interests( e.g. , Sudol and Harvey, 2005; LaBonte, Georgoulis, and Rust, 2007), becausethey release enormous amount of energy, and some of them may have a higherpotential to perturb the heliosphere.In this article, we explore the possibility to use EUVI data to estimate theGOES X-ray peak fluxes of flares far behind the limb. The response of the EUVIwith temperature is vastly different from that of the GOES XRS. However,during flares, one of the four channels, namely the one that encompasses theFe xii complex at 195 ˚A ( ≈ >
10 MK) plasma (see, for example, Dere and Cook (1979) who derived a typicaldifferential emission measure of flares) due to the Fe xxiv line at 192 ˚A, whichhas the peak temperature of 15 – 20 MK, depending on the assumed ionizationequilibrium. Therefore we focus on this EUVI channel to estimate the X-rayflux. In the next section, the temperature response of the EUVI is compared tothat of the AIA. In Section 3, we use AIA data to study the origin and extent ofthe uncertainties that limit the accuracy of our work. This is followed in Section4 by our main objective of finding an empirical relation of the EUVI with theGOES XRS fluxes. In Section 5 we show examples of flares from the backsidethat are estimated to be intense. Some of them leave observable signatures inAIA data. We summarize the work in Section 6, which also lists items that maypossibly improve this work.
2. Temperature Response
The EUVI observes the Sun in four channels of EUV passbands that are similarto those of the
Extreme-ultraviolet Imaging Telescope (EIT; Delaboudini`ere et al. ,1995) on the
Solar and Heliospheric Observatory (SOHO). They are 171 ˚A(Fe ix , Fe x ), 195 ˚A (Fe xii , Fe xxiv ), 284 ˚A (Fe xv ) and 304 ˚A (He ii ). Theirtemperature responses are plotted in Figure 1 as solid lines. Those for the AIA SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 2 oft X-ray Fluxes of Flares Far Behind the Limb
Response with Temperature (EUVI, AIA and GOES) DN s − ( E M = c m − ) −9 −8 −7 −6 −5 −4 GO ES F l u x ( W m − ) f o r E M = c m −
171 Å193 (195) Å211 (284) Å304 Å94 Å131 Å335 ÅGOES1−8Å0.5−4Å
Figure 1.
Temperature response of EUVI on STEREO-A (solid lines), AIA (dashed lines)and GOES-14 (dotted lines). They are calculated using the CHIANTI atomic database version7.0 (Dere et al. , 1997; Landi et al. , 2012). are plotted as dashed lines. In order to gain more comprehensive understandingof the structures and dynamics of the corona and transition region, the AIAhas three more EUV channels, namely 94 ˚A (Fe xviii ), 131 ˚A (Fe viii , Fe xxi ),and 335 ˚A (Fe xvi ). Additionally, the AIA replaces the 284 ˚A channel used byprevious missions with the 211 ˚A channel that contains the Fe xiv line, but thepeak temperature is similar.The two bands of the GOES XRS (plotted as dotted lines) have monotonicallyincreasing response with temperature. Although none of these EUV channels hasa similar temperature response, some of them have either primary or secondarypeaks at high temperatures that characterize flares (Dere and Cook, 1979). Al-though AIA’s 131 ˚A channel (peaking at ≈
10 MK) may be most often correlatedwith the XRS as indicated in the next section, it is clear that the 195 ˚A is theonly EUVI channel that has a distinct peak of response above 10 MK. Thisbroad peak around 15 MK is due to the Fe xxiv line at 192 ˚A. Therefore weuse this channel exclusively for estimating the X-ray peak fluxes of flares notobserved by GOES.The high temperature response of the 195 ˚A channel is only secondary tothe main peak at ≈ SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 3 .V. Nitta
EUVI and AIA DN s − ( N o r m a li z ed a t t he . M K M a i n P ea k ) GOES/EUVI, GOES/AIA GO ES / E U V I, GO ES / A I A ( N o r m a li z ed a t M K ) (a) (b)4 50 AIA 193 ÅEUVI 195 Å
Figure 2. (a) Comparison of EUVI’s 195 ˚A channel with AIA’s 193 ˚A channel, normalizedat the respective primary peaks at ≈ xii . Because AIA’s passband is shiftedto shorter wavelength, the secondary peak around 15 MK is more pronounced. (b) Althoughthe GOES response monotonically increases with temperature (Figure 1), the ratio of GOESwith AIA and EUVI has a dip due to the secondary peak of the latter instruments. at 195 ˚A, and thus has a higher response at the wavelength of the Fe xxiv line.Figure 6 of Howard et al. (2008) and Figure 8 of Boerner et al. (2012) give theeffective areas of the EUVI and AIA, respectively. Specifically, the response to15 MK plasma relative to the main response to 1.5 MK plasma is about a factorof two smaller for EUVI than for AIA (Figure 2(a)). The small difference of theeffective area also affects the flux at a given temperature with respect to theGOES XRS flux. The GOES/EUVI and GOES/AIA ratios both decrease from10 MK to 15 MK and increase above 15 MK, but reach the level of 10 MK atdifferent temperatures as will be shown later in Figures 5 and 7. According toFigure 2(b), the GOES/AIA ratio at 35 MK is still small than that at 10 MK,whereas the GOES/EUVI ratio at 30 MK is already comparable to that at10 MK.
3. AIA-GOES Relation
In order to understand various uncertainties that may sometimes be critical inour objective of estimating the GOES X-ray peak flux from EUVI 195 ˚A data,we first compare light curves of flares from the AIA with those from the GOESXRS. The AIA takes images every 12 seconds in all the EUV channels, whereasthe typical cadence of EUVI 195 ˚A images is 5 minutes. We select all flares abovethe GOES C3 ( F − = 3 × − W m − ) level between June 2010 and September2012. In order to be free of the uncertainty from occultation by the limb, we limit SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 4 oft X-ray Fluxes of Flares Far Behind the Limb N o r m a li z ed F l u x N o r m a li z ed F l u x −0.020.000.020.04 N o r m a li z ed F l u x N o r m a li z ed F l u x N o r m a li z ed F l u x N o r m a li z ed F l u x N o r m a li z ed F l u x
94 Å Max: 5.066e+07 Min (pre−peak): 1.307e+07 Min: 1.307e+07131 Å Max: 2.423e+08 Min (pre−peak): 5.719e+07 Min: 5.719e+07171 Å Max: 2.162e+09 Min (pre−peak): 2.074e+09 Min: 2.029e+09193 Å Max: 2.884e+09 Min (pre−peak): 2.339e+09 Min: 2.339e+09211 Å Max: 8.120e+08 Min (pre−peak): 7.208e+08 Min: 7.208e+08304 Å Max: 3.895e+08 Min (pre−peak): 3.100e+08 Min: 3.100e+08335 Å Max: 5.685e+07 Min (pre−peak): 3.511e+07 Min: 3.511e+07 N o r m a li z ed F l u x N o r m a li z ed F l u x N o r m a li z ed F l u x N o r m a li z ed F l u x N o r m a li z ed F l u x N o r m a li z ed F l u x N o r m a li z ed F l u x
94 Å Max: 5.388e+07 Min (pre−peak): 1.648e+07 Min: 1.532e+07131 Å Max: 2.301e+08 Min (pre−peak): 6.708e+07 Min: 6.471e+07171 Å Max: 2.281e+09 Min (pre−peak): 2.138e+09 Min: 2.124e+09193 Å Max: 3.114e+09 Min (pre−peak): 2.680e+09 Min: 2.680e+09211 Å Max: 9.998e+08 Min (pre−peak): 9.261e+08 Min: 9.261e+08304 Å Max: 4.093e+08 Min (pre−peak): 2.946e+08 Min: 2.909e+08335 Å Max: 9.416e+07 Min (pre−peak): 6.894e+07 Min: 6.755e+07
Figure 3.
Time variations of fluxes in AIA images. The channels plotted are (from top tobottom): 94 ˚A, 131 ˚A, 171 ˚A, 193 ˚A, 211 ˚A, 304 ˚A, and 335 ˚A. The preflare minimum issubtracted from each time profile. The GOES 1 – 8 ˚A light curves is overplotted in gray, scaledto [0,maximum] of each EUV light curve. to only those flares whose 2-D radial distance is less than 0.997 r ⊙ . We used thethe Heliophysics Event Registry (Hurlburt et al. , 2012) for the flare locations.Light curves are produced from the headers of individual level-1 full-diskimages, which contain the average pixel values. To isolate the emission fromthe flare, we subtract the minimum flux during the 30 minute period precedingthe GOES peak time. In Figure 3, we show two examples of the light curves.They are X-class flares (SOL2011-02-15T01:56 and SOL2011-03-09T23:23) . TheGOES XRS light curves are also shown normalized to the positive scale of theEUV flux. In both examples, time variations of the 193 ˚A flux (in the fourthrow) is found to be well correlated with those of the X-ray flux up to the GOESpeak. After the peak, the 193 ˚A flux does not decay as fast as the GOES fluxpresumably because of contributions from cooler plasma. The first flare was quiteeruptive and associated with an energetic coronal mass ejection (CME) ( e.g. ,Schrijver et al. , 2011), whereas the second one was confined without a CME.Whether the flare is associated with a CME may affect our purpose because ofthe associated coronal dimming that is pronounced at temperatures of 1 – 2 MK.Figure 3 indicates that the effect of dimming in the 193 ˚A channel may not besubstantial, at least for intense flares. For SOL identification convention, see
Solar Phys. , pp.1 – 2, 2010.
SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 5 .V. Nitta
AIA 193 Å E U V F l u x ( DN s − ) AIA 131 Å E U V F l u x ( DN s − ) GOES 1−8 ÅAIA 193 Å FDAIA 193 Å Partial GOES 1−8 ÅAIA 131 Å FD (a) (b)
Figure 4.
Light curves of AIA images in (a) 193 ˚A and (b) 131 ˚A channels for a M1.1 flarein comparison with the GOES soft X-ray time variations. In 193 ˚A, there are variations thatdo not track the GOES, although a good match is seen in 131 ˚A. Even if the field of view isrestricted to a sub-field around the flare (see the line in red) as opposed to full-disk (FD), thevariations observed in 193 ˚A remain, suggestive of increase of low-temperature material in theflare that is not observed by GOES.
Concerning other channels, the EUV peak is delayed at 94 ˚A and 335 ˚A withrespect to the GOES probably because plasma that cools from >
10 MK to therespective response peaks (see Figure 1) also contributes to the peaks in the lightcurves. At low temperatures, most notably at 304 ˚A but also at 171 ˚A and evenat 211 ˚A, the EUV peak precedes the XRS peak, reflecting the impulsive phase inwhich non-thermal electrons collisionally thermalize plasma at loop foot-points( e.g. , Fletcher et al. , 2011). The dimming is most pronounced at 171 ˚A, but itsrelation with the CME is not clear. In the EUV wavelengths, the flux tends todecay much more slowly than in soft X-rays, and there are even secondary peaks.These tendencies seem to be least prominent in the 193 ˚A channel as far as theexamples in Figure 3 are concerned.For less intense flares, there are more complications in the 193 ˚A data dueprimarily to the flux of cool plasma that contributes to the main response around1.5 MK. In full-disk 193 ˚A images, the flare contribution is typically only ≈ F − = (3 – 4) × − W m − ). Figure 4(a) shows that thelight curves are not well correlated between the GOES and 193 ˚A and that thisdoes not change substantially if we use the flux within a sub-field (300 ′′ × ′′ ).Furthermore, we could not find a sub-field in which the first peak that corre- SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 6 oft X-ray Fluxes of Flares Far Behind the Limb
Flares Observed by AIA AIA 193 Å (DN s −1 )10 −6 −5 −4 −3 GO ES − Å ( W m − ) CMX GO ES F l a r e C l a ss i f i c a t i on Figure 5.
Scatter plot of the GOES 1 – 8 ˚A and AIA 193 ˚A background-subtracted fluxes of540 flares. In each flare, the GOES-AIA pair is made around the X-ray peak time. The fivelines show how the relation of the GOES and AIA fluxes depends on the temperature of theflare. The order reflects the AIA response relative to the GOES response (see Figure 2(b)).The size of the symbols represents the flare temperature as determined by the GOES 0.5 – 4 ˚Aand 1 – 8 ˚A ratio. sponds to the GOES peak is higher than the second peak around 17:03 UT. Thisexample thus indicates that some flares can have a higher emission measure at1.5 MK relative to 20 MK than found by Dere and Cook (1979). For this M1.1flare (SOL2011-07-27T16:07), the 131 ˚A flux is still well-correlated with theGOES (Figure 4(b)). One of the reasons that smaller flares correlate better withthe 131 ˚A than with the 193 ˚A may be that they are cooler, and thus reachtemperatures of Fe xxi but not those of Fe xxiv .Coronal dimming can also play a major role in less intense flares than thoseshown in Figure 3. Out of the 609 flares above C3 and observed in AIA’s normalmode, 69 flares, mostly C-class, show less 193 ˚A flux at the GOES peak thanduring the preflare interval. The most intense flare in this categoy is M2.5. Thefrequency of a smaller flux at the flare peak is not not much different if weinclude only a small area around the flare because of deeper dimming closer tothe flare.Cooling plasma from previous flares can contribute to the main response ofthe 193 ˚A channel, making it difficult to determine the background level. Thisis especially true when the increase of flux due to the flare is only a few percent
SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 7 .V. Nitta
Rise Time vs Peak Flux (GOES Flares 2007/03/01 − 2012/09/30) −7 −6 −5 −4 −3 Soft X−ray (1−8 Å) Peak Flux (W m −2 )1101001000 R i s e T i m e ( P ea k − O n s e t ) i n M i nu t e s Figure 6.
The distribution of the flare rise time from the NOAA flare list. of less. Therefore, flares that occur one after another can pose a serious problemfor comparing EUV and X-ray fluxes.In Figure 5, we show a scatter plot of the AIA 193 ˚A and GOES XRS 1 – 8 ˚Afluxes around the GOES peak times. Both fluxes reflect pre-flare backgroundsubtraction. This includes all the 540 flares that have positive 193 ˚A fluxes. Thetime difference between the AIA and GOES fluxes is only up to the three-secondsampling rate of the latter. Different symbols are used to indicate the GOES XRSchannel ratio temperature (White et al. , 2005). Pearson’s correlation coefficientfor the fluxes in the logarithmic scales is 0.81 for the entire population and 0.92for the 51 flares above M3. The lines represent the expected relations betweenthe GOES and AIA fluxes for flares characterized by single temperatures (seeFigure 2(b)). Most points above the M3 level fall between T =15 MK and T =35MK lines. The GOES channel ratio temperature is generally above 15 MK forthese flares. The scatter in this plot could be made smaller first by limitingthe area and then by conducting pixel-to-pixel differential measure analyses andextracting only those pixels that contain hot plasma to be observed by GOES. Itis beyond the scope of this article, although such an effort could be instructive.
4. GOES EUVI Relation
Now we apply the same analysis on EUVI 195 ˚A data. The most significantdifference of these images from AIA images is that they are taken much less
SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 8 oft X-ray Fluxes of Flares Far Behind the Limb
Flares Observed by STEREO EUVI EUVI 195 Å (DN s −1 )10 −6 −5 −4 −3 GO ES − Å ( W m − ) CMX GO ES F l a r e C l a ss i f i c a t i on Figure 7.
Scatter plot of the GOES 1 – 8 ˚A and EUVI 195 ˚A fluxes of ≈
450 flares. Only thoseflares observed by AIA and not occulted from STEREO’s view are plotted. The same formatas Figure 5. The order of the lines representing five temperatures reflects the EUVI responserelative to the GOES response (see Figure 2(b)). frequently, typically once every five minutes. This may mean a poorer correlationwith the GOES XRS flux because the next image after the flare onset can beafter the flare peak, by which time the excess in EUV flux may already start.In Figure 6, for all the flares included in the NOAA flare list, we plot the timedifference between the soft X-ray peak and onset as a function of the X-ray peakflux. Most flares above the M4 level have the rise time longer than 4 minutes,so it is likely that for most intense flares the rising phase is included where weexpect a good correlation between GOES and EUVI 195 ˚A. There were intenseflares in 2007 up to M8 (Aschwanden et al. , 2009; Nitta et al. , 2013), but wecannot include them because their rise times are typically less than ten minutes,and 195 ˚A images were taken only once every ten minutes. Until August 2009,EUVI’s “primary wavelength” was 171 ˚A.From the same set of flares that are plotted in Figure 5, we select only thosethat were observed on disk by one or both STEREO spacecraft. This is achievedwithin SolarSoft (Freeland and Handy, 1998) by combining the flare locationson flares that come from the Heliophysics Event Registry (Hurlburt et al. , 2012)with the STEREO orbital information. We also drop those flares in which thetime difference of the EUVI image from the GOES peak time is more than fiveminutes. This leaves a total of 450 flares with Ahead and Behind combined. No
SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 9 .V. Nitta
Ratio of Observed and Expected GOES Fluxes for T =20 MK −5 −4 −3 GOES 1−8 Å (W m −2 )0.11.010.0 R a t i o Figure 8.
The ratio of the observed GOES flux with the expected flux from EUVI 195 ˚A fluxassuming an isothermal plasma of 20 MK. distinction is made between the EUVI flux from Ahead and Behind, since theresponse of the 195 ˚A channel to 10 MK differs by only ≈ T =15 MK line ( e.g. , the right-most point, which is the X6.9 flare of SOL2011-08-09T08:05 (Asai et al. , 2012)),suggesting the need for fine tuning of the EUVI absolute flux calibration.Irrespective of the absolute calibration, we can now use the points of intenseflares to estimate the GOES peak X-ray flux. In Figure 8 we show how much thescatter is from what is expected by assuming that flares have a single temper-ature of 20 MK. Here the error bars reflect a crude estimate of the uncertaintyof the EUVI flux, where it increases from 3% at F EUVI(195) = 10 to 50% at F EUVI(195) = 10 . This is based on an inspection of temporal variations of theEUVI flux in comparison with the XRS flux for all the events included here.Figure 8 suggests that we can determine the GOES peak X-ray flux for flaresmore intense than M4 to within the factor of three. The best fit we find is F GOES = 1 . · − F EUVI(195) (1)The uncertainty is larger for less intense flares as is clear from the largerscatter. Nevertheless, it is not much larger than an order of magnitude, so for
SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 10 oft X-ray Fluxes of Flares Far Behind the Limb
Table 1.
List of major events since 2010 that come from regions > ◦ behind the limb.These are either candidates of X-class flares or multi-spacecraft SEP events. The fourevents that are shown in Figure 9 are marked with asterisks.Date Time A or B Location Est. F GOES
Range Refs2010/01/17 03:55:40 B S25 E128 6.4 · − M3.4 – M9.6 1, 22010/08/31 20:55:53 A S22 W146 1.7 · − M8.4 – X2.52010/09/01 21:50:53 A S22 W162 1.1 · − M5.4 – X1.62011/03/21 02:10:49 A N20 W128 3.1 · − M1.3 – X1.3 32011/06/04 07:05:58 A N15 W140 1.0 · − M5.2 – X1.62011/06/04 * 21:50:58 A N17 W148 8.1 · − X4.0 – X122011/10/23 23:15:44 A N19 W151 1.1 · − X5.3 – X1.62011/11/03 * 22:40:44 B N08 E156 9.4 · − M4.7 – X1.4 42012/03/26 22:16:02 B N18 E123 1.6 · − M8.2 – X2.52012/04/29 12:45:53 B, A N12 E163 1.7 · − M8.3 – X2.52012/07/23 * 02:30:56 A S15 W133 1.5 · − M8.2 – X2.52012/08/21 20:10:54 B, A S22 E158 1.4 · − M6.8 – X2.02012/09/11 07:55:50 B, A S22 E178 2.6 · − X1.3 – X3.92012/09/19 11:15:49 B, A S15 E171 1.8 · − M9.1 – X2.72012/09/20 * 15:00:48 B, A S15 E153 1.2 · − X5.8 – X182012/09/22 03:05:48 B S15 E134 1.7 · − M8.4 – X2.5References: 1. Veronig et al. (2010), 2. Dresing et al. (2012), 3. Rouillard et al. (2012), 4.Mewaldt et al. (2012) convenience we will put the range of 0.5 – 1.5 of Equation (1) for flares with F EUVI(195) > · DN s − and 0.3 – 3 for flares with F EUVI(195) . · DNs − .
5. Major Events on the Farside
We have surveyed the mission-long EUVI archive and found a number of can-didates of X-class flares. Table 1 lists them with their locations and estimatedranges of the GOES peak X-ray flux. It also includes a few less intense flaresthat have been discussed largely in the context of multi-spacecraft observations ofsolar energetic particle (SEP) events (Dresing et al. , 2012; Rouillard et al. , 2012;Mewaldt et al. , 2012). It turns out that these events (17 January 2010, 21 March2011, and 3 November 2011) are not particularly intense in the GOES scale. Butat least the first two events stand out in terms of one of the attributes of theCME, namely an EUV wave (Veronig et al. , 2010; Rouillard et al. , 2012).We have a few notes concerning the list in Table 1. First, the frequency ofcandidates of X-class flares behind the limb mimics that of the X-class flaresactually observed on the visible side. Single active regions may produce multipleX-class flares over a wide range of longitude. Although we had to wait until 15February 2011 for the first X-class flare in solar cycle 24, the flare on 31 August2010 was probably the first one.EUVI data of these behind-the-limb flares (from Earth view) show a varieddegree of association with CMEs. While some are confined and not associated
SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 11 .V. Nitta (a) 2011/06/04 (b) 2011/11/03(c) 2012/07/23 (d) 2012/09/20
Figure 9.
Running ratio images of four events that occurred far behind the limb (see Table 1for their locations), and were observed directly by AIA because of high cadence and sensitivity. with a major CME in white-light coronagraph data, others are highly eruptive,also accompanying extended dimming. Note that dimming is usually observedonly after a sharp increase of flux, indicating that the peak flux is well captured.Some of the highly eruptive events, despite deep occultation by the limb, emergein AIA images either as an eruption itself or in the form of waves or propagatingfronts. Figure 9 gives snapshots from the tri-color running ratio movies that areaccessible at http://aia.lmsal.com/occulted events.
We also point out that theevent on 23 July 2012 is more eruptive than that on 20 September 2012, whichis probably the most intense flare in solar cycle 24 as of the end of 2012.
6. Summary and Discussion
We have correlated GOES X-ray and EUV (195 ˚A) fluxes of 450 flares as ob-served by STEREO, and found an empirical relation, which lets us estimate the
SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 12 oft X-ray Fluxes of Flares Far Behind the Limb
GOES peak X-ray flux of intense events, whose X-ray emission is occulted bythe limb. There are more than a dozen candidates of X-class flares as shownin Table 1. Many of the events included here to obtain the EUVI-XRS fluxrelation have impulsive time profiles, and in such events it is possible that wemay underestimate the soft X-ray flux by missing the real peak because of theEUVI five-minute cadence limitation.Using AIA observations we study the origins and extents of uncertainties inthe estimation of the peak GOES X-ray flux from EUV data. The main difficultylies in the fact that the Fe xxiv line represents only a small secondary responsepeak in the 195 ˚A (193 ˚A) channel. The main response peak not only picks upcooling or heating plasma from the present and previous flares in the 1 – 2 MKrange but also shows decrease because of coronal dimming when the flare isassociated with a CME. It appears that the effect of coronal dimming is notsubstantial until the soft X-ray peak in flares above M3 or so, but it can besignificant in less intense flares. In extreme cases, the 193 ˚A flux is found todecrease in the rising phase of the flare. The primary response of AIA’s 193 ˚Aand EUVI’s 195 ˚A channels is around 1.5 MK, and in order to correlate thefluxes in these channels with the GOES flux, we should in principle extract thosepixels that contain ∼ >
10 MK plasma, by conducting differential emission measureanalysis. One caveat is that one needs to monitor full-disk images for sympatheticflaring at a remote region, which may also contribute to the spatially-integratedGOES flux.Other related issues include the absolute calibration of the involved instru-ments, including the GOES XRS. We used calibration data for the XRS onGOES-14, but most flares analyzed in this work were observed by the XRSon GOES-15, whose response data have not been available (S. White, personalcommunication, 2012) . Also needed is a fine tuning as to the absolute calibrationof EUVI and AIA. Apart from the calibration of the instruments, we have notconsidered the possible effect of non-thermal flare emission (especially on GOES)and ionization equilibrium of hot plasma.Lastly, we may ask how significant the GOES X-ray flare classification is forthe interpretation of energetic events. We have conventionally and convenientlyused the GOES X-ray class as if it represents the magnitude of the flare. However,it is argued that the energy radiated in soft X-rays is only a small fractionof the total energy released in flares ( e.g. , Kretzschmar et al. , 2010). For thetotal energy budget, we should include presumably larger energy that goes intoSun’s lower atmosphere. The fact that even C-class flares produce white-lightemission ( e.g. , Matthews et al. , 2003) indicates that the X-ray flux may not becorrelated with the total energy. Here we consider the GOES X-ray flux to be stillan important reference, however, thanks partly to its existence for more thanthree solar cycles. As demonstrated by the recent work of Ryan et al. (2012),the GOES XRS measurement is useful for collectively deriving flare plasmaparameters, even though it is also possible that we may find in SDO data a The response of the XRS on GOES-15 became available in March 2013, around the time ofthe second referee report, but the difference from that of the XRS on GOES-14 seems to betoo small to affect this work in a significant way.
SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 13 .V. Nitta better parameter more intimately correlated with the total energy that includesenergy flow both into the heliosphere and lower atmosphere.
Acknowledgements
This work has been supported by the NASA STEREO mission underNRL Contract No. N00173-02-C-2035, and by the AIA contract NNG04EA00C to LMSAL.We are greateful to the referee for valuable comments that clarified the motivations of thiswork.
References
Aschwanden, M.J., Wuelser, J.-P., Nitta, N.V., Lemen, J.R. : 2009,
Solar Phys. , 3. doi:10.1007/s11207-009-9347-4Asai, A., Ishii, T.T., Isobe, H., Kitai, R., Ichimoto, K., UeNo, S., Nagata, S., Morita, S.,Nishida, K., Shiota, D., Oi, A., Akioka, M., Shibata, K.: 2012,
Astrophys. J. Lett. ,L18. doi: 10.1088/2041-8205/745/2/L18Battaglia, M., Grigis, P.C., Benz, A.O.: 2005,
Astron. Astrophys. , 737. doi: 10.1051/0004-6361:20053027Boerner, P., Edwards, C., Lemen, J., Rausch, A. Schrijver, C., Shine, R., Shing, L. Stern, R.,Tarbell, T., Title, A., Wolfson, C. J., Soufli, R., Spiller, E., Gullikson, E., McKenzie, D.,Windt, D., Golub, L. Podgorski, W., Testa, P., Weber, M.: 2012,
Solar Phys. , 41. doi:10.1007/s11207-011-9804-8Delaboudini`ere, J.-P., Artzner, G.E., Brunaud, J., Gabriel, A.H., Hochedez, J.F., Millier, F.,Song, X.Y., Au, B., Dere, K.P., Howard, R.A., Kreplin, R., Michels, D.J., Moses, J.D.,Defise, J.M., Jamar, C., Rochus, P., Chauvineau, J.P., Marioge, J.P., Catura, R.C., Lemen,J.R., Shing, L., Stern, R.A., Gurman, J.B., Neupert, W.M., Maucherat, A., Clette, F.,Cugnon, P., van Dessel, E.L.: 1995,
Solar Phys. , 291. doi: 10.1007/BF00733432Dere, K.P., Cook, J.W.: 1979,
Astrophys. J. , 772. doi:10.1086/157013Dere, K.P., Landi, E., Mason, H.E., Monsignori Fossi, B.C., Young, P.R.: 1997,
Astron.Astrophys. Suppl. , 149. doi: 10.1051/aas:1997368Dresing, N., G´omez-Herrero, R., Klassen, A., Heber, B., Kartavukh, Y., Dr¨oge, W.: 2012,
SolarPhys. , 281. doi: 10.1007/s11207-012-0049-yFeldman, U., Doscheck, G.A., Behring, W.E., Phillips, K.J.H.: 1996,
Astrophys. J. , 1034.doi: 10.1086/177030Fletcher, L., Dennis, B.R., Hudson, H.S., Krucker, S., Phillips, K., Veronig, A., Battaglia, M.,Bone, L., Caspi, A., Chen, Q., Gallagher, P., Grigis, P.T., Ji, H., Liu, W., Milligan, R.O.,Temmer, M.: 2011,
Space Sci. Rev. , 19. doi: 10.1007/s11214-010-9701-8Freeland, S.L., Handy, B.N.: 1998,
Solar Phys. , 497. doi: 10.1023/A:1005038224881Howard, R.A., Moses, J.D., Vourlidas, A., Newmark, J.S., Socker, D.G., Plunkett, S.P., Ko-rendyke, C.M., Cook, J.W., Hurley, A., Davila, J.M., Thompson, W.T., St Cyr, O.C.,Mentzell, E., Mehalick,K., Lemen, J.R., Wuelser, J.-P., Duncan, D.W., Tarbell, T.D.,Wolfson, C.J., Moore, A., Harrison, R.A., Waltham, N.R., Lang, J., Davis, C.J., Eyles,C.J., Mapson-Menard, H., Simnett, G.M., Halain, J.P., Defise, J.M., Mazy, E., Rochus,P., Mercier, R., Ravet, M.F., Delmotte, F., Auch`ere, F., Delaboudini`ere, J. P., Bothmer,V., Deutsch, W., Wang, D., Rich, N., Cooper, S., Stephens, V., Maahs, G., Baugh, R.,McMullin, D., Carter, T.: 2008,
Space Sci. Rev. , 67. doi: 10.1007/s11214-008-9341-4Hurlburt, N., Cheung, M., Schrijver, C., Chang, L., Freeland, S., Green, S., Heck, C., Jaffey,A., Kobashi, A., Schiff, D., Serafin, J., Seguin, R., Slater, G., Somani, A., Rimmons, R.:2012,
Solar Phys. , 67. doi: 10.1007/s11207-010-9624-2Kaiser, M.L., Kucera, T.A., Davila, J.M., St.Cyr, O.C., Guhathakurta, M., Christian, E.: 2008,
Space Sci. Rev. , 5. doi: 10.1007/s11214-007-9277-0Kretzschmar, M., Dudok de Wit, T., Schumutz, W., Mekaoui, S., Hochedez, J.-F., Dewitte,S.: 2010,
Nature Physics , 690. doi: 10.1038/nphys1741LaBonte, B.J., Georgoulis, M.K., Rust, D.M.: 2007, Astrophys. J. , 955. doi:10.1086/522682Landi, E., Del Zanna, G., Young, P.R. Dere, K.P., Mason, H.E.: 2012,
Astrophys. J. , 99.doi: 10.1088/0004-637X/744/2/99Lemen, J.R., Title, A.M., Akin, D.J., Boerner, P.F., Chou, C., Drake, J.F., Duncan, D.W.,Edwards, C.G., Friedlaender, F.M., Heyman, G.F., Hurlburt, N.E., Katz, N.L., Kushner,
SOLA: ms_20130406_2.tex; 16 October 2018; 6:22; p. 14 oft X-ray Fluxes of Flares Far Behind the LimbG.D., Levay, M., Lindgren, R. W., Mathur, D. P., McFeaters, E.L., Mitchell, S., Rehse,R.A., Schrijver, C.J., Springer, L.A., Stern, R.A., Tarbell, T.D., Wuelser, J.-P., Wolfson,C.J., Yanari, C., Bookbinder, J.A., Cheimets, P.N., Caldwell, D., DeLuca, E.E., Gates, R.,Golub, L., Park, S., Podgorski, W.A., Bush, R.I., Scherrer, P.H., Gummin, M.A., Smith,P., Auker, G., Jerram, P., Pool, P., Soufli, R., Windt, D.L., Beardsley, S., Clapp, M., Lang,J., Waltham, N.: 2012,
Solar Phys. , 17. doi: 10.1007/s11207-011-9776-8Matthews, S.A., van Driel-Gesztelyi, L., Hudson, H.S., Nitta, N.V.: 2003,
Astron. Astrophys. , 1107. doi: 10.1051/0004-6361:20031187Mewaldt, R.A., Cohen, C.M.S., Mason, G.M., von Rosenvinge, T.T., Leske, R.A., Luhmann,J.G., Odstrcil, D., Vourlidas, A.: 2012, In: Solar Wind 13 Conference Proceedings.Nitta, N.V., Aschwanden, M.J., Freeland, S.L., Lemen, J.R., Wuelser, J.-P., Zarro, D.M.: 2013,
Solar Phys. submitted.Pesnell, W.D., Thompson, B.J., Chamberlin, P.C.: 2012,
Solar Phys. , 3. doi:10.1007/s11207-011-9841-3Rouillard, A.P., Sheeley, N.R., Jr., Tylka, A., Vourlidas, A., Ng, C.K., Rakowski, C., Cohen,C.M.S., Mewaldt, R.A., Mason, G.M., Reames, D., Savani, N.P., St. Cyr, O.C., Szabo, A.:2012,
Astrophys. J. , 44. doi: 10.1088/0004-637X/752/1/44Ryan, D.F., Milligan, R.O., Gallagher, P.T., Dennis, B.R., Tolbert, A.K., Schwartz, R.A.,Young, C.A.: 2012,
Astrophys. J. Suppl. , 11. doi: 10.1088/0067-0049/202/2/11Schrijver, C.J., Aulanier, G., Title, A.M., Pariat, E., Delann´ee: 2011,
Astrophys. J. , 167.doi: 10.1088/0004-637X/738/2/167Sudol, J.J., Harvey, J.W.: 2005,
Astrophys. J. , 647. doi: 10.1086/497361Veronig, A.M., Muhr, N., Kienreich, I.W., Temmer, M., Vrˇsnak, B.: 2010,
Astrophys. J. Lett. , L57. doi: 10.1088/2041-8205/716/1/L57White, S.M., Thomas, R.J., Schwartz, R.A.: 2005,
Solar Phys. , 231. doi: 10.1007/s11207-005-2445-zWuelser, J.-P., Lemen, J.R., Tarbell, T.D., Wolfson, C.J., Cannon, J.C., Carpenter, B.A.,Duncan, D.W., Gradwohl, G.S., Meyer, S.B., Moore, A.S., Navarro, R.L., Pearson, J.D.,Rossi, G.R., Springer, L.A., Howard, R.A., Moses, J.D., Newmark, J.S., Delaboudini`ere, J.-P., Artzner, G.E., Auch`ere, F., Bougnet, M., Bouyries, P., Bridou, F., Clotaire, J.-Y., Colas,G., Delmotte, F., Jerome, A., Lamare, M., Mercier, R., Mullot, M., Ravet, M.-F., Song, X.,Bothmer, V., Deutsch, W.: 2004, In: Fineschi, S., Gummin, M.A. (eds.),
Telescopes andInstrumentation for Solar Astrophysics, Proc. SPIE , 111. doi: 10.1117/12.506877, 111. doi: 10.1117/12.506877