A study on the dynamic spectral indices for SEP events on 2000 July 14 and 2005 January 20
aa r X i v : . [ a s t r o - ph . S R ] O c t Research in Astronomy and Astrophysics manuscript no.(L A TEX: ms2019-0144.tex; printed on October 17, 2019; 0:40)
A study on the dynamic spectral indices for SEP events on 2000 July14 and 2005 January 20
Ming-Xian Zhao , Gui-Ming Le Key Laboratory of Space Weather, National Center for Space Weather, China MeteorologicalAdministration, Beijing, 100081, China; (
[email protected] ) Received 20** June **; accepted 20** July **
Abstract
We have studied the dynamic proton spectra for the two solar energetic particle(SEP) events on 2000 July 14 (hereafter GLE59) and 2005 January 20 (hereafter GLE69).The source locations of GLE59 and GLE69 are N22W07 and N12W58 respectively. Protonfluxes >
30 MeV have been used to compute the dynamic spectral indices of the two SEPevents. The results show that spectral indices of the two SEP events increased more swiftlyat early times, suggesting that the proton fluxes >
30 MeV might be accelerated particularlyby the concurrent flares at early times for the two SEP events. For the GLE69 with sourcelocation at N12W58, both flare site and shock nose are well connected with the Earth at theearliest time. However, only the particles accelerated by the shock driven by eastern flank ofthe CME can propagate along the interplanetary magnetic field line to the Earth after the flare.For the GLE59 with source location at N22W07, only the particles accelerated by the shockdriven by western flank of the associated CME can reach the Earth after the flare. Resultsshow that there was slightly more than one hour during which the proton spectra for GLE69are softer than that for GLE59 after the flares, suggesting that the shock driven by easternflank of the CME associated with GLE69 is weaker than the shock driven by the westernflank of the CME associated with GLE59. The results support that quasi-perpendicular shockhas stronger potential in accelerating particles than the quasi-parallel shock. The results alsosuggest that only a small part of the shock driven by western flank of the CME associatedwith the GLE59 is quasi-perpendicular.
Key words:
Sun: flares — Sun: coronal mass ejections (CMEs) — Sun: particle emission— Sun: solar-terrestrial relations
Large gradual solar energetic particle (SEP) events are often accompanied by both flare and fast coronalmass ejection (CME). Both the flare and CME-driven shock may contribute to the productions of SEPs.
M.-X. Zhao & G.-M. Le
However, whether a gradual flare can accelerate protons to high energy and even to relativistic energy isstill an open question. For example, Reames (1999) suggested that only CME-driven shock can accelerateprotons to high energy in large gradual SEP events. However, some researchers argued that flares maydominate in the acceleration of particles at the early phase of large gradual SEP events (e.g., Cane et al.2003). When relativistic solar protons (RSPs) reach the atmosphere of the Earth, the interaction betweenthe RSPs and the particles of the Earth’s atmosphere causes the atmospheric cascade.At times only a small fraction of the RSPs accelerated to the energy of ≥ ynamic protons energy spectra comparison 3 SUNEarthSUNEarth C M E S h o c k A B C M E S h o c k Fig. 1: The two positions on the shock surface connect with the Earth at two different momentsof interplanetary shock. Figure 1 shows the variation of the points on the shock surface connected with theEarth. Because shock intensity at different points on shock surface is different, hence the energy spectralindex of particles observed by GOES should change continuously as CME moves away from the Sun andpropagates in interplanetary space.To investigate possible source for the earliest particles accelerated by associated solar flares and checkwhether perpendicular shock is more effective in accelerating protons than parallel shock, dynamic energyspectral indices of protons for GLE59 and GLE69 are to be computed and compared in this study. Theenergy spectral indices of double power law for SEP events associated with GLE59 and GLE69 have beendetermined by a few researchers (Mewaldt et al. 2012; Wu & Qin 2018). However, the double power lawis the event integrated differential spectra, which cannot reflect the variation of the energy spectral indexwith time. This article is arranged as follows. Data analysis is presented in section 2. Discussion is given insection 3. Summary and conclusion are noted in the final section.
Solar active region(SAR) 9077, which is located at N22W07, produced a X5.7 flare. The flare started at10:03 UT and peaked at 10:24 UT on 2000 July 14 and then a CME associated with the flare firstly enteredSolar and Heliospheric Observatory (SOHO)/Large Angle and Spectrometric Coronagraph (LASCO)-C2field of view (2.2-6 R s ) at 10:54 UT. The projected speed of the CME was 1674 km/s ( https://cdaw.gsfc.nasa.gov/CME_list/ ; e.g., Yashiro et al. 2004). A large gradual SEP event accompanied withthe flare and CME was observed by ACE and GOES 8, which is shown in Figure 2. M . - X . Z h a o & G . - M . L e -5 -4 GO ES X r a y - A -4 -2 P r o t on f l u x ( A C E and GO ES ) ACE P1ACE P2ACE P3ACE P4ACE P5ACE P6ACE P7ACE P8GOES P1GOES P2GOES P3GOES P4GOES P5GOES P6GOES P7GOES P8GOES P9GOES P10GOES P11
Fig. 2: Fluxes of solar soft X-rays and proton particles with different energies during 13-18, July 2000. The upper panel shows the flux of soft X-ray in 1-8 ˚A observed by GOES.The lower panel shows ACE EPAM/LEMS120 (Gold et al. 1998) ion fluxes with energy (P1)0.047-0.068 MeV, (P2)0.068-0.115 MeV, (P3)0.115-0.195 MeV, (P4)0.195-0.321MeV, (P5)0.310-0.580 MeV, (P6)0.587-1.060 MeV, (P7)1.060-1.900 MeV and (P8)1.900-4.800 MeV. GOES EPS corrected proton flux with energy (P1)0.6-4.0 MeV, (P2)4.0-9.0MeV, (P3)9.0-15.0 MeV, (P4)15.0-44.0 MeV, (P5)40.0-80.0, (P6)80.0-165.0 and (P7)165.0-500.0 MeV. GOES HEPAD proton flux with energy (P8)350.0-420 MeV, (P9)420-510MeV, (P10)510-700 MeV, and (P11) >
700 MeV. All data are of 5 min resolution. yn a m i c p r o t on s e n e r gy s p ec t r ac o m p a r i s on5 -7 -6 -5 -4 GO ES X r a y - A -6 -4 -2 P r o t on f l u x ( A C E and GO ES ) ACE P1ACE P2ACE P3ACE P4ACE P5ACE P6ACE P7ACE P8GOES P1GOES P2GOES P3GOES P4GOES P5GOES P6GOES P7GOES P8GOES P9GOES P10GOES P11
Fig. 3: Fluxes of solar soft X-rays and proton particles with different energies during 20-24, January 2005.
M.-X. Zhao & G.-M. Le
SAR 10720 located at N12W58 produced a X7.1 flare. The flare started at 06:36 UT and peaked at 07:01 UTon 2005 January 20. A CME associated with the flare with a projected speed 882 km/s was first observed bySOHO/LASCO C2 at 06:54 UT on 2005 January 20( https://cdaw.gsfc.nasa.gov/CME_list/ ;e.g., Yashiro et al. 2004). A large gradual SEP event accompanied with the flare and CME was observed byACE and GOES 11, which is shown in Figure 3. >
100 MeV for two SEP events
The fluxes of protons with different energies usually increase swiftly after the eruptions of the associatedflares and CMEs. The flux of energy (E) >
100 MeV proton usually reached its peak flux no longer afterthe eruptions of the associated flare and CME, suggesting that the strongest acceleration for E >
100 MeVproton takes place in the Sun or in the interplanetary space near the Sun. The fluxes of E >
100 MeV protonsfor GLE59 and GLE69 are shown in Figure 4. It is seen that the flux of E >
100 MeV proton of the GLE69event reached its peak flux faster than that of GLE59. The peak fluxes of E >
100 MeV protons of theGLE69 and GLE59 events are 698 pfu and 408 pfu, respectively (Le et al. 2016, 2017). [1 proton flux unit(pfu)= cm − sr − s − ].It is evident that peak flux of E >
100 MeV proton of the GLE69 is much stronger than that of theGLE59. However, the flux of E >
100 MeV proton of the GLE69 decayed much faster than that of theGLE59 after their peak fluxes. The source location of the GLE69 is N12W58, which is well connectedwith the Earth, because the location is far away from the solar center (Swalwell 2017). However, the sourcelocation N22W07 of GLE59 is not well connected with the Earth, because the location is close to the solarcenter. This may be the reason why the flux of E >
100 MeV of the GLE69 reached its peak flux faster thanthat of the GLE59.The shock nose driven by the GLE69-associated CME is well connected with the Earth at the earliesttime, and then the eastern flank of the shock is connected with the Earth and the shock intensity declinedgradually as the CME moved away from the Sun. On the contrary, the particles accelerated by the westernflank of the shock associated with the GLE59 can reach the Earth and the shock intensity also changedcontinuously as the CME moves away from the Sun. One can understand from Figure 4 that the flux ofE >
100 MeV proton of the GLE59 is stronger than that of the GLE69 no longer after their peak fluxes,suggesting that the intensity of the western flank shock associated with the GLE59 may be stronger thanthat of eastern flank shock associated with the GLE69.
Double power laws were used to study the energy spectra of GLEs that occurred during solar cycle 23. Theresults showed that the breaking energies for GLE59 and GLE69 are 24.2 MeV and 8.18 MeV respectively(Mewaldt et al. 2012). The breaking energies for both GLEs are lower than 30 MeV. Hence, E >
30 MeVprotons observed by GOES are used to calculate the dynamic energy spectral indices for the two SEP event. ynamic protons energy spectra comparison 7 -10 0 10 20 30 40 50
Hours -2 -1 > M e V P r o t on F l u x Fig. 4: Comparison between fluxes of E >
100 MeV protons of two major SEP events f ( E ) ∝ E − γ is used to calculate the dynamic spectral index of the two SEP events. Time resolution ofprotons observed by GOESs is 5-minutes. The SEP data observed by GOES 8 and GOES 11 are used tocalculate the dynamic energy spectral indices for GLE59 and GLE69 respectively. The start times of twoflares are all toggled to zero time. 7 differential channels (channels from P5 to P11, energy ranging from40 to >
700 MeV), and 4 integral channels ( > > >
60, and >
100 MeV, described in Mewaldt et al.2005) observed by GOES are used to calculate energy spectral indices.
The dynamic spectral indices calculated for the GLE59 and GLE69 are shown in Figure 5, which exposesthat the spectral indices for the two GLEs increased faster and reached peak value promptly. The decayphases of the spectral indices for the two events differ a lot. The decay phase of the spectral index forGLE69 declines much more promptly than that for the GLE59. In fact, the decay phase of the spectralindex for GLE59 declines abruptly. In this regard, Firoz et al. (2019a) observed that the GLE69-associatedDH-type II burst ended about 112 min earlier than the flare, implying that the CME shock did not operateover the decay phase of the GLE69 particle event, whereas CME shock operated over the decay phase ofthe GLE59 particle event.As mentioned in the earlier (Figure 1), the source location for GLE69 is well connected with the Earththat the particles accelerated by the flare and latter reached the CME shock nose can directly propagateto the Earth along the interplanetary magnetic field line at the earliest time. However, only the particlesaccelerated by eastern flank shock can reached the Earth. For GLE59, only the particles accelerated bywestern flank shock can reach the Earth. We can also see from Figure 5 that there was slightly more than1 hour during which the energy spectral index for GLE59 is higher than that for GLE69, suggesting thatwestern flank shock associated with GLE59 is stronger than eastern flank shock associated with GLE69during this period. The shock on the eastern flank is quasi-parallel shock, while the shock in the western
M.-X. Zhao & G.-M. Le -6 -5 -4 -3 GO ES X -r a y - A W / m X M γ Fig. 5: Dynamic energy spectral indices for the two SEP events. The upper and lower panels indicate theflux of solar X-ray in 1-8 ˚A and dynamic energy spectral indices of the two SEP events respectively. Thevertical dashed line indicates the peak time of X7.1 that occurred on 2005 January 20flank is quasi-perpendicular shock (Reames 1999). In this context, quasi-perpendicular shock is strongerthan quasi-parallel shock.
The energy spectral indices for GLE69 increased quickly and then reached its peak value at 06:50 UTon 2005 January 20. It is evident that the hardest proton spectrum occurred during the rising phase of theassociated flare. Higher energy protons have closer association with the associated flare, while lower energyprotons have closer association with associated CME-driven shock (Le & Zhang 2017). In this context, thephenomenon that spectral index increased quickly at early times indicates that E >
30 MeV protons in thetwo GLEs should be mainly accelerated by the concurrent flares.The flux of E >
100 MeV proton for GLE69 reached its peak flux faster than that for GLE59. The peakflux of E >
100 MeV proton for GLE69 is much stronger than that for GLE59. The proton spectra for GLE69is harder than that for GLE59 at early times (Figure 5), suggesting that solar eruptions associated withGLE69 have stronger potential to accelerate protons to E >
100 MeV than that associated with GLE59 atearly times. Gopalswamy et al. (2005) proposed that the speed of the CME associated with GLE69 is 3242km/s, which is much faster than the CME projected speed 882 km/s. If the speed of the CME associatedwith GLE69 is really 3242 km/s or even close to this value, the E >
30 MeV protons may be accelerated byboth concurrent flare and CME-driven shock at early times. However, E >
30 MeV protons may still mainlybe accelerated by concurrent flare at early times because proton spectrum became harder at early times. ynamic protons energy spectra comparison 9
Figure 5 shows that there was only slightly more than 1 hour during which the energy spectral index forGLE59 is higher than that for GLE69, suggesting that quasi-perpendicular shock associated with GLE59 isstronger than quasi-parallel shock associated with GLE69, which is consistent with the simulation resultsobtained by Qin et al. (2018). To be noticed, only small part of the western flank shock associated withGLE59 is quasi-perpendicular shock.The results of the present study support the results obtained by Firoz et al. (2019) that GeV protons areaccelerated by concurrent flare. In this context, the result that GeV particles were accelerated by associatedflare obtained in the paper of Zhao et al. (2018) is reasonable. Now the question is for GLE59, how theRSPs accelerated by the associated flare with source location at N22W07 propagated to the Earth? Thesimulation of the results made by Zhang & Zhao (2017) showed that if the perpendicular diffusion is about10% of what is derived from the random walk of field lines at the rate of supergranular diffusion, particlesinjected at the compact solar flare site can spread to a wide range of longitude and latitude, very similar tothe behavior of particles injected at a large CME shock. The results of present study that E >
30 MeV protonmay be mainly accelerated by concurrent flare associated with GLE59 at early times give an evidence thatparticles accelerated by associated flare can spread to a wide range of longitude and latitude, very similarto the behavior of particles injected at a large CME shock (Zhang & Zhao 2017).
We have analyzed solar proton fluxes of E >
30 MeV and studied the spectral indices for the SEP events on2000 July 14 (GLE59) and 2005 January 20 (GLE69). Important results are summarized as follows.1. Solar acceleration processes during the GLE69 event has stronger potential to accelerate the protonsto GeV energetics than those during the GLE59 event. E >
30 MeV protons for both the GLE59 and GLE69seemed to have been accelerated mainly by the flares at early times. Our analysis has been illustrated by thesimulation study of Zhang & Zhao (2017) that the particles injected from the flare site can spread to a widerange of longitude and latitude, which is very similar to the behavior of particles injected at a large CMEshock.The results of our study also support the viewpoints proposed by Firoz et al. (2019a) that the MeVenergetic protons are initiated over the flare initial phases and then accelerated to GeV energetic over theflare prompt phases associated with the coronal shocks manifested in metric-type II burst.2. The spectral index for GLE59 is higher than that for GLE69 for about 1 hour over the flare decayphases where coronal shocks manifested in DH-type II bursts played much stronger roles for the GLE59(e.g., see Firoz et al. 2019a,b). The results suggest that quasi-perpendicular shock associated with GLE59 isstronger than quasi-parallel shock associated with GLE69, and only a small part of the western flank shockassociated with GLE59 is quasi-perpendicular shock.
Acknowledgements
We are grateful to SOHO/LASCO,and CDAW, GOES data for making their dataavailable online. This work is funded by the National Natural Science Foundation of China (Grant No.41674166)
References
Aschwanden M. J., & Freeland S. L., 2012, ApJ, 754, 112 2Cane H. V., Reames D. V., & Rosenvinge T. T. VONR, 1988, J. Geophys. Res., 93(A9), 9555 2Cane H. V., von Rosenvinge T. T., Cohen C. M. S., & Mewaldt R. A., 2003, Geophys. Res. Lett., 30(12),8017 2Firoz K. A., Cho K.-S., Hwang J., et al., 2010, J. Geophys. Res., 115, A09105 2Firoz K. A., Moon Y.-J., Park S.-H., et al., 2011, ApJ, 743, 190 2Firoz K. A., Gan W. Q., Li Y. P., et al., 2019a, ApJ, 872, 178 7, 9Firoz K. A., Gan W. Q., Moon Y. J., et al., 2019b, ApJ, 883, 91 2, 9Firoz K. A., Gan W. Q., Moon Y.-J., & Li, C., 2012, ApJ, 758, 119 2Gold R. E., Krimigis S. M., Hawkins S. E., et al., 1998, Space Sci. Rev., 86, 541 4Grechnev V. V., Kurt V.G., Chertok I. M., et al., 2008, Sol. Phys., 252, 149 2Gopalswamy N., Xie H., Yashiro S., Usoskin I., 2005, in Proceedings of 29th International Cosmic RayConference, Vol. 1, ed., B. Sripathi Acharya, S. Gupta, P. Jagadeesan, A. Jain, S. Karthikeyan, S. Morris,S. Tonwar (Tata Institute of Fundamental Research, Mumbai), 169 8Jokipii J. R.,1987, ApJ, 313, 842 2Kahler S. W., 2016, ApJ, 819, 105 2Kallenrode M.-B., 2001, J. Geophys. Res., 106, 24989 2Klein K.-L., Trottet G., Lantos P., & Delaboudini´ere J.-P., 2001, A&A, 373, 1073 2Klein K.-L., Masson S., Bouratzis C., et al., 2014, A&A, 572, A4 2Le G.-M., Tang Y.-H., & Han Y.-B., 2006, ChJAA (Chin. J. Astron. Astrophys.), 6(6), 751 2Le G.-M., Li C., Tang Y.-H., et al., 2016, Research in Astron. Astrophys. (RAA), 16, 14 6Le G.-M., Li C., Zhang X.-F., 2017, Research in Astron. Astrophys. (RAA), 17(7), 73 6Le G.-M., & Zhang X.-F., 2017, Research in Astron. Astrophys. (RAA), 17(12), 123 2, 8Li G., Zank G. P., & Rice W. K. M., 2003, J. Geophys. Res., 108(A2), 1082 2Li G., Zank G. P., & Rice W. K. M., 2005, J. Geophys. Res., 110, A06104 2Li C., Tang Y. H., Dai Y., Zong W. G., & Fang C., 2007, A&A, 461, 1115 2Masson S., Klein K.-L., B¨utikofer R., et al., 2009, Sol. Phys., 257, 305 2McCracken K. G., Moraal H., & Stoker P. H., 2008, J. Geophys. Res., 113, A12101 2Mewaldt R. A., Cohen C. M. S., Labrador A. W., et al., 2005, J. Geophys. Res., 110, A09S18 7Mewaldt R. A., Looper M. D., Cohen C. M. S., et al., 2012, Space Sci. Rev., 171(1-4), 97 2, 3, 6Qin G., Kong F.-J., & Zhang L.-H., 2018, ApJ, 860:3 2, 9Reames D. V., 1999, Space Sci. Rev., 90, 413 2, 8Simnett G. M., 2006, A&A, 445, 715 2Simnett G. M., 2007, A&A, 472, 309 2Swalwell B., Dalla S., & Walsh R.W., 2017, Sol. Phys., 292, 173 6Wu S.-S., & Qin G., 2018, J. Geophys. Res.:Space Physics, 123, 76 2, 3Yashiro S., Gopalswamy N., Michalek G., et al., 2004, J. Geophys. Res., 109, A07105 3, 6Zank G. P., Li G., Florinski V., et al., 2006, J. Geophys. Res., 111, A06108 2 ynamic protons energy spectra comparison 11ynamic protons energy spectra comparison 11