Explosive chromospheric evaporation driven by nonthermal electrons around one footpoint of a solar flare loop
aa r X i v : . [ a s t r o - ph . S R ] M a y Explosive chromospheric evaporation driven by nonthermalelectrons around one footpoint of a solar flare loop
D. Li , , , Z. J. Ning , Y. Huang , and Q. M. Zhang , Key Laboratory of Dark Matter and Space Astronomy, Purple Mountain Observatory, CAS, Nanjing210008, China CAS Key Laboratory of Solar Activity, National Astronomical Observatories, Beijing 100012, China Key Laboratory of Modern Astronomy and Astrophysics (Nanjing University), Ministry of Education,Nanjing 210023, China
ABSTRACT
We explore the temporal relationship between microwave/HXR emission and Doppler velocityduring the impulsive phase of a solar flare on 2014 October 27 (SOL2014-10-27), which displaysa pulse on the light curves in microwave (34 GHz) and hard X-ray (HXR, 25 −
50 keV) bandsbefore the flare maximum. Imaging observation shows that this pulse mainly comes from onefootpoint of a solar flare loop. The slit of
Interface Region Imaging Spectrograph ( IRIS ) staysat this footpoint during this solar flare. The Doppler velocities of Fe
XXI IV XXI T ∼ IV T ∼ Subject headings:
Sun: flares — Sun: UV radiation — Sun: radio radiation — Sun: X-rays, gamma rays— line: profiles
1. Introduction
In a typical solar flare, the released energy isabout 10 erg within dozens of minutes. Suchhuge energy is thought to be transferred fromthe magnetic energy via reconnection, which isthought to heat the plasma and to accelerate thebi-directional nonthermal electrons through thesolar chromosphere, transition region and corona.This is well known as the standard solar flaremodel. These accelerated electrons are guided bythe newly reconnected magnetic field lines. Someof the accelerated electrons travel up to the inter-planetary space, and the others propagate down-ward to the lower corona and upper chromosphere, Correspondence should be sent to: [email protected]. where they lose their energies through Coulombcollisions with the denser plasma. Subsequently,two footpoint sources are often produced along theflare loop legs in the hard X-ray (HXR) and mi-crowave bands (Brown 1971; Asai et al. 2006), andthe double ribbons are brightened in H α , ultravi-olet (UV), and extreme-ultraviolet (EUV) wave-lengths (Czaykowska 1999; Del Zanna et al. 2006;Milligan 2015). Meanwhile, the local chromo-spheric material is rapidly heated up to ∼
10 MK(Antonucci et al. 1982), resulting in a higher pres-sure, which drives the chromospheric materialupward into the corona along the solar flareloops. Thus the hot plasma fills the solar flareloops in a process called ‘chromospheric evapora-tion’ (Brown 1973; Fisher et al. 1985a,b; Liu et al.1006; Ning & Cao 2010; Brosius & Daw 2015;Tian et al. 2015; Young et al. 2015; Zhang et al.2016; Lee et al. 2017). At the same time, theoverpressure also drives the denser plasma down-ward into the lower chromosphere in a processcalled ‘chromospheric condensation’ (Fisher et al.1985a,b; Teriaca et al. 2006; Li et al. 2015).Chromospheric evaporation is observed in mul-tiple wavelengths, i.e., HXR, EUV/UV, and mi-crowave bands. Imaging observations show thatthe HXR emission tends to rise with the doublefootpoint sources along the loop legs, eventuallymerging them into a single source at the looptop (e.g., Liu et al. 2006; Ning & Cao 2010; Ning2011; Nitta et al. 2012). This process is thought tobe the HXR signature of chromospheric evapora-tion. On the other hand, the heated materials riseupward to the corona and disturb the local plasma.This process causes the microwave emission sud-denly cut-off on the dynamic spectra. Observa-tionally, a high-frequency cut-off drifts to low-frequency is the radio signature of chromosphericevaporation (Aschwanden & Benz 1995; Karlicky1998; Ning et al. 2009). Joint observations fromspectra and images show that the blue shift of hotlines in corona tends to appear at the outside offlare ribbon (Czaykowska 1999; Li & Ding 2004;Li et al. 2015).Chromospheric evaporation is divided into twotypes, i.e., ‘explosive’ and ‘gentle’. Explosiveevaporation takes place if the input energy fluxexceeds a critical value of ∼ erg cm − s − (Fisher et al. 1985a,b). In spectroscopic observa-tions, explosive evaporation is usually accompa-nied by blue shift with high speed ( ∼ −
400 km s − )in the hot lines from corona, and by red shiftwith low speed ( ∼ −
40 km s − ) in the coollines from upper chromosphere and transitionregion (Feldman et al. 1980; Ding et al. 1996;Del Zanna et al. 2006; Teriaca et al. 2006; Milligan & Dennis2009; Veronig et al. 2010; Chen & Ding 2010;Brosius 2013; Doschek et al. 2013; Tian et al.2014; Brosius & Daw 2015; Brosius et al. 2016;Zhang et al. 2016). Observationally, the speed ofblue shift is an order of magnitude larger thanthat of red shift, which is due to the fact that thedensity of the overlying corona is less than that ofthe underlying lower chromosphere (Fisher et al.1985a,b; Milligan & Dennis 2009; Doschek et al.2013). Gentle evaporation occurs when the in- put energy flux is smaller than the critical value,and all the emission lines display blue shift fromchromosphere through transition region to corona(Milligan et al. 2006; Brosius 2009; Li & Ding2011).In the literatures, there are three mechanismsto drive chromospheric evaporation. The firstone is electron-driven, which emphases that thenonthermal electrons accelerated by magneticreconnection play an important role in drivingchromospheric evaporation (Fisher et al. 1985b;Brosius 2003; Milligan & Dennis 2009; Tian et al.2014, 2015; Li et al. 2015; Zhang et al. 2016).The second one focusses on the thermal energy(Fisher et al. 1985a; Falewicz et al. 2009). Whilethe third one is driven by the energy depositionthrough Alfv´en waves (Reep & Russell 2016). Thecorrelation between Fe XXI
Interface Region Imaging Spectrograph ( IRIS ,De Pontieu et al. 2014),
NoRH (Hanaoka et al.1994),
Reuven Ramaty High Energy Solar Spec-troscopic Imager ( RHESSI , Lin et al. 2002), At-mospheric Imaging Assembly (AIA, Lemen et al.2012) and Helioseismic and Magnetic Imager(HMI, Schou et al. 2012) aboard
Solar Dynam-ics Observatory ( SDO ), we study the temporalrelationship between the microwave/HXR emis-sion and Doppler velocity of the SOL2014-10-27event to prove the nonthermal electrons drivingchromospheric evaporation around one footpointof a solar flare loop.
2. Data Analysis and Results
The studied event takes place in NOAA AR12192on 2014 October 27. It is an M7.1 flare in
GOES light curve. It displays typical features of solarflares, such as two footpoint sources in microwaveand HXR images. The slit of
IRIS stays at onefootpoint during the impulsive phase of the so-lar flare, which gives us a good chance to studythe temporal evolution of the Doppler velocity incoronal and transition region lines, i.e., Fe
XXI IV GOES
SXR fluxes in 1.0 − − GOES
SXR light curves. Mean-while, the microwave 34 GHz (purple) and HXR25 −
50 keV (orange) emissions display a series ofregular and periodic peaks during the impulsivephase of this solar flare, which result in severalnonthermal pulses (e.g., Li et al. 2017), as markedby the Roman Capitals ‘I’, ‘II’. The first pulse (‘I’)is studied in this paper.Figure 2 shows the multi-wavelength imagesfrom several instruments during the first pulse.They have the same field-of-view (FOV) of 90 ′′ × ′′ .Panel (a) displays the integral intensity im-age ( ∼ ′′ /pixel) in microwave 34 GHz from00:09 UT to 00:10 UT. The double footpointsources are marked by the white boxes. Amongthem, one is bright (ft1), while the other one isfaint (ft2). This is confirmed by RHESSI
CLEANimage ( ∼ ′′ /pixel) at 25 −
50 keV, as shown inpanel (b). The contour of ft2 is from the mi-crowave emission. Panel (c) gives the line-of-sight(LOS) magnetogram ( ∼ ′′ /pixel) overlaid withthe AIA 1600 ˚A contours. The bottom pan-els show the EUV/UV snapshots in SDO /AIA131 ˚A and 1600 ˚A ( ∼ ′′ /pixel), IRIS /SJI 1330 ˚A( ∼ ′′ /pixel). The AIA 1600 ˚A image is ap-plied to co-align with the SJI 1330 ˚A image(Cheng et al. 2015; Tian et al. 2015; Li et al.2015), because they both contain continuum emis-sion around the temperature minimum which isdominant in many bright features. The flare rib-bons in AIA 1600 ˚A and SJI 1330 ˚A are connectedby the hot loops in AIA 131 ˚A, as indicated bythe dashed lines. Consistent with the standardsolar flare model, panel (e) shows that the doublefootpoint sources overlap on the two flare ribbons.The slit of IRIS crosses one flare ribbon and fixeson the edge of one footpoint (ft1), as shown by thelong blue line along the 45 degree to the North-South direction.The hot line of Fe
XXI T ∼ IV T ∼ IRIS observation at spectral win-dows of ‘Fe
XXI ’ (a, c, e) and ‘Si IV ’ (b, d, f)at three times. The Y-axis is along the slit direc-tion. They have been pre-processed with the rou- tines of ‘iris orbitval corr l2.pro’ (Tian et al. 2014;Cheng et al. 2015) and ‘iris prep despike.pro’(De Pontieu et al. 2014) in Solar Soft Ware (SSW).The absolute wavelength calibration is also man-ually performed with the relatively strong neutrallines, i.e., ‘O I ’ 1355.5977 ˚A and ‘S I ’ 1401.5136 ˚A(e.g., Tian et al. 2015). The overplotted black pro-file shows the spectrum at the position of about38.6 ′′ , which is marked by the short purple lineon the left-hand side. This position is located atthe edge of a flare ribbon in the northwest labeledwith the short blue line in Figure 2 (e). The multi-Gaussian functions (purple profile) superimposedon a linear background (green line) are used to fitthe IRIS spectrum at ‘Fe
XXI ’ window to extractthe flare line of Fe
XXI
XXI I line at1354.29 ˚A, Fe II lines at 1353.02 ˚A, 1354.01 ˚A, and1354.75 ˚A, Si II lines at 1352.64 ˚A and 1353.72 ˚A,some unidentified lines at 1353.32 ˚A and 1353.39 ˚A(Polito et al. 2015; Young et al. 2015; Tian et al.2015). Here, we fixed these blended line posi-tions, constrained their widths, and tied their in-tensities to the lines in other spectral windows(e.g., Li et al. 2015, 2016). Finally, we can obtainthe line profile of Fe XXI IV XXI IV XXI IV ′′ (short purple linein Figure 3). The hot line of Fe XXI IV − and 36 km s − during the firstpulse between 00:08 − − NoRH , RHESSI and
IRIS have the cadences of1 s, 4 s and 16.2 s, respectively. In this paper, weuse all 9 points from the Doppler velocity aroundthe pulse (between 00:08:30 UT and 00:11:00 UT),and their nearby points of microwave and HXRemissions. Thus, 9 points are extracted from allthe light curves, as marked by the purple symbols(‘+’ or ‘ × ’).Figure 4 (b) shows the Doppler velocities ofFe XXI IV × ’) dependence on microwave 34 GHz (black)and HXR 25 −
50 keV (purple) emissions dur-ing the pulse interval, i.e., between 00:08:30 UTand 00:11:00 UT. As expected from the modelof electron-driven chromospheric evaporation, wefind a negative correlation (-0.96/-0.87) betweenthe microwave/HXR emission and the Dopplervelocity of hot line (Fe
XXI
XXI IV IV ∼ ∼ ∼ ′′ . It clearlyshows the evaporation upflows from double foot-points through legs to loop top, as indicated bythe purple arrows during 00:08:30 − ∼
112 km s − and ∼
99 km s − without con-sidering the projection effect.
3. Conclusions and Discussions
Based on the observations from
IRIS , NoRH ,and
RHESSI , we explore the temporal relationshipbetween Doppler velocity of Fe
XXI IV −
50 keVduring a pulse of solar flare on 2014 October27. The completely new observational resultis that the explosive chromospheric evaporationdriven by nonthermal electrons originates fromone footpoint of a solar flare loop. Meanwhile,the evaporation upflows are also observed fromthe double footpoints to loop top in AIA 131 ˚Aimage. This is consistent well with the stan-dard solar flare model. Our results agree wellwith previous findings about the temporal cor-relation between the Doppler velocity and HXRflux which represents the deposition rate (Li et al.2015; Tian et al. 2015; Lee et al. 2017), and arealso consistent well with the studies of the spa-tial correlation between the upflows/dowflows andHXR sources in solar flares (Brosius & Holman2009; Milligan & Dennis 2009; Veronig et al. 2010;Brosius et al. 2016; Zhang et al. 2016). It hasbeen recently demonstrated that the explosiveevaporation can also be driven by the dissipationthrough Alfv´en waves, and this mechanism canproduce the HXR bursts (Reep & Russell 2016).Therefore, only the presence of an HXR burstis impossible to rule out the contribution fromthe heating of Alfv´en waves. On the other hand,a microwave pulse is observed around the peaktime of
IRIS
Doppler velocity at one footpoint ofa solar flare loop, implying that the nonthermalelectrons are injected into the chromosphere alongthe flare loop in the Sun. Thus, the observations4n this case support an electron-driven chromo-spheric evaporation.Figure 2 (e) and (f) shows that the slit of
IRIS crosses one of flare ribbons. The studied positionof Doppler velocity is just at the edge of flare rib-bon, indicating that the explosive chromosphericevaporation appears at the outside of flare rib-bon, which is consistent well with previous find-ings (Czaykowska 1999; Li & Ding 2004; Li et al.2015; Tian et al. 2015). On the other hand, Fig-ure 2 (a) exhibits that the slit of
IRIS does notappear in the footpoint center, which may be dueto the lower spatial resolution of microwave image( ∼ ′′ ). Future work needs higher resolution ob-servations in microwave or HXR bands, includingthe time and spatial resolutions.In this letter, only the first pulse (‘I’) in mi-crowave and HXR emissions is used to investi-gate the explosive chromospheric evaporation. Wedid not find any correspondence between the mi-crowave (or HXR) emission and Doppler veloc-ity during the other pulses, i.e., the second mi-crowave pulse at about 00:13 UT (‘II’). This maybe because the upflows/downfows of the other mi-crowave pulses are very complex. The upflows arefrom the new heated plasma which moves upwardalong the loop, while the downflows from the lastevaporated material may eventually cool and pre-cipitate back down along the loop (e.g., Brosius2003; Milligan & Dennis 2009). It could also bebecause the location of energy deposition is notcovered by the slit of IRIS . The good correlationbetween Doppler velocity and microwave/HXRemission may be found when shifting the slit po-sition (e.g., Li et al. 2015).The authors would appreciate the anonymousreferee for his/her valuable comments and sugges-tions to improve the manuscript. We thank theteams of
IRIS , GOES , NoRH , RHESSI , SDO /AIAand it SDO/HMI for their open data use policy.The authors also thank Y. Gong for reading ourmanuscript. This study is supported by NSFCunder grants 11603077, 11573072, 11473071,11333009, KLSA201708, and Laboratory No.2010DP173032. Dr. D. Li is also supported bythe Youth Fund of Jiangsu No. BK20161095, andDr. Q. M. Zhang is supported by the SurfaceProject of Jiangsu No. BK 20161618 and theYouth Innovation Promotion Association CAS.
REFERENCES
Antonucci, E., Gabriel, A. H., Acton, L. W., et al.1982, Sol. Phys., 78, 107Aschwanden, M. J., & Benz, A. O. 1995, ApJ,438, 997Asai, A., Nakajima, H., Shimojo, M., et al. 2006,PASJ, 58, L1Brosius, J. W. 2003, ApJ, 586, 1417Brosius, J. W. 2009, ApJ, 701, 1209Brosius, J. W., & Holman, G. D. 2009, ApJ, 692,492Brosius, J. W. 2013, ApJ, 762, 133Brosius, J. W., & Daw, A. N. 2015, ApJ, 810, 45Brosius, J. W., Daw, A. N., & Inglis, A. R. 2016,ApJ, 830, 101Brown, J. C. 1971, Sol. Phys., 18, 489Brown, J. C. 1973, Sol. Phys., 31, 143Chen, F., & Ding, M. D. 2010, ApJ, 724, 640Cheng, X., Ding, M. D., & Fang, C. 2015, ApJ,804, 82Czaykowska, A., & De Pontieu, B., & Alexander,D., & Rank, G. 1999, ApJ, 521, L75De Pontieu, B., & Title, A. M., & Lemen, J. R.,et al. 2014, Sol. Phys., 289, 2733Del Zanna, G., & Schmieder, B., & Mason, H., &Berlicki, A., & Bradshaw, S. 2006, Sol. Phys.,234, 95Ding, M. D., & Watanabe, T., & Shibata, K., etal. 1996, ApJ, 458, 391Doschek, G. A., & Warren, H. P., & Young, P. R.2013, ApJ, 767, 55Falewicz, R., & Rudawy, P., & Siarkowski, M.2009, A&A, 508, 971Feldman, U., & Doschek, G. A., & Kreplin, R. W.,& Mariska, J. T. 1980, ApJ, 241, 1175Fisher, G. H, & Canfield, R. C., & McClymont,A. N. 1985a, ApJ, 289, 4145isher, G. H, & Canfield, R. C., & McClymont,A. N. 1985b, ApJ, 289, 425Hanaoka, Y., Shibasaki, K., Nishio, M., et al.1994, Proceedings of Kofu Symposium, 35Karlicky, M. 1998, A&A, 338, 1084Lee, K.-S., Imada, S., Watanabe, K., Bamba, Y.,& Brooks, D. H. 2017, ApJ, 836, 150Lemen, J. R., & Title, A. M., & Akin, D. J., et al.2012, Sol. Phys., 275, 17Li, Y., & Ding, M. D. 2011, ApJ, 727, 98Li, L. P., Peter, H., Chen, F., & Zhang, J. 2014,A&A, 570, A93Li, D., Ning, Z. J., & Zhang, Q. M. 2015, ApJ,813, 59Li, D., Innes, D. E., & Ning, Z. J. 2016, A&A,587, A11Li, D., Zhang, Q. M., Huang, Y., Ning, Z. J., &Su, Y. N. 2017, A&A, 597, L4Li, J. P., & Ding, M. D. 2004, ApJ, 606, 583Lin, R. P., & Dennis, B. R., & Hurford, G. J., etal. 2002, Sol. Phys., 210, 3Liu, W., & Liu, S. M., & Jiang, Y. W., et al. 2006,ApJ, 649, 1124Milligan, R. O., & Gallagher, P. T., & Math-ioudakis, M., et al. 2006, ApJ, 642, L169Milligan, R. O., & Dennis, B. R. 2009, ApJ, 699,968Milligan, R. O. 2015, Sol. Phys., 290, 3399Ning, Z. J., & Cao, W. D., & Huang, J., et al.2009, ApJ, 699, 15Ning, Z. J., & Cao, W. D. 2010, ApJ, 717, 1232Ning, Z. J. 2011 Sol. Phys., 273, 81Nitta, S., Imada, S., & Yamamoto, T. T. 2012,Sol. Phys., 276, 183Polito, V., Reeves, K. K., Del Zanna, G., Golub,L., & Mason, H. E. 2015, ApJ, 803, 84 Schou, J., Scherrer, P. H., Bush, R. I., et al. 2012,Sol. Phys., 275, 229Reep, J. W., & Russell, A. J. B. 2016, ApJ, 818,L20Teriaca, L., & Falchi, A., & Falciani, R., & Cauzzi,G., & Maltagliati, L. 2006, A&A, 455, 1123Tian, H., & Li, G., & Reeves, K. E., et al. 2014,ApJ, 797, L14Tian, H., Young, P. R., Reeves, K. K., et al. 2015,ApJ, 811, 139Veronig, A. M., & Ryb´ak, J., & G¨om¨ory, P., et al.2010, ApJ, 719, 655Young, P. R., Tian, H., & Jaeggli, S. 2015, ApJ,799, 218Zhang, Q. M., Li, D., Ning, Z. J., et al. 2016, ApJ,827, 27
This 2-column preprint was prepared with the AAS L A TEXmacros v5.2. − − −
50 keV (orange) from the whole flare regionduring the same time intervals.Fig. 2.— The multi-wavelength images with aFOV of 90 ′′ × ′′ on the 2014 October 27 flare. Thepurple contours are from the microwave of 34 GHz,and the white boxes indicate the sources of twofootpoints. The green contours represent the AIA1600 ˚A intensities at the scale of 200 DN s − . Thered dashed lines in AIA 131 ˚A outline the possibleflare loops. The long blue line represents the slitof IRIS and the short blue line marks the regionstudied here. Fig. 3.—
IRIS spectra at ‘Fe
XXI ’ (left) and ‘Si IV ’ (right) windows on 2014 October 27. Theblack profile is the line spectrum across the slitposition marked by a purple line on the left-handside of each image, the overplotted purple profilerepresents the fitting results. The turquoise lineis Fe XXI I NoRP
34 GHz at ft1 (solid black) and ft2 (dashedblack) during 00:06 − RHESSI −
50 keV (green) is from the wholeflare at the same time. The temporal evolutionof Doppler velocities at Fe
XXI IV IRIS . Panel (b):Scatter plots of Doppler velocities dependence onmicrowave and HXR emissions during the firstpulse. The correlation coefficients (cc) are labeled. Fig. 5.— Upper: The difference map in AIA131 ˚A, two solid lines (purple) outline a flare loop.Bottom: The time-distance image along flare loopin AIA 131 ˚A. The start point (d=0) is fromft2. The arrows indicate the upflow directions.Two dotted lines mark the upflow time between00:08:30 −−