1: \documentclass[12pt,preprint]{aastex}

2: 

3: %\usepackage {amssymb}

4: %\usepackage {amsmath}

5: %\usepackage {graphicx}

6: 

7: \bibliographystyle{aa}

8: 

9: \newcommand{\oh}[1]{\omit\hidewidth #1\hidewidth}

10: 

11: \begin{document}

12: 

13: \title{Initial Observations of Sunspot Oscillations Excited by Solar Flare}

14: \author{A. G. Kosovichev}

15: \affil{W.W.Hansen Experimental Physics Laboratory, Stanford

16: University, Stanford, CA 94305, USA}

17:  \author{T. Sekii}

18:  \affil{National Astronomical Observatory of Japan,

19: Mitaka, Tokyo 181-8588, Japan}

20: %\date{\today}

21: %\maketitle

22: %\Large

23: 

24: \begin{abstract}

25: Observations of a large solar flare of December 13, 2006, using

26: Solar Optical Telescope (SOT) on Hinode spacecraft revealed

27: high-frequency oscillations excited by the flare in the sunspot

28: chromosphere. These oscillations are observed in the region of

29: strong magnetic field of the sunspot umbra, and may provide a new

30: diagnostic tool for probing the structure of sunspots and

31: understanding physical processes in solar flares.

32: \end{abstract}

33: \keywords{Sun: flares -- Sun: oscillations}

34: \section{Introduction}

35: Solar flares represent a process of conversion of magnetic energy

36: into heat, kinetic energy of plasma eruptions and high-energy

37: particles. Solar flares may excite various types of oscillations and

38: waves in various layers of the Sun, from "sunquakes" in the interior

39: \citep{Kosovichev1998} to coronal Morton waves \citep{Moreton1960}

40: and coronal loop oscillations \citep{Aschwanden1999}. The mechanisms

41: of these oscillations and waves are not fully understood yet, but

42: obviously related to the energy release and transport properties.

43: For instance, the seismic response ("sunquakes") are believed to be

44: related to the hydrodynamic impact on the lower atmosphere and

45: photosphere by shocks generated in the area heated by high-energy

46: electrons. The spatial-temporal properties of the seismic wave

47: source are closely related to the properties of hard X-ray source

48: \citep{Kosovichev2006a,Kosovichev2006b}. The sunquakes represent

49: packets of high-frequency acoustic waves traveling through the Sun's

50: interior. The waves propagate through the active regions and

51: sunspots with strong magnetic field without significant distortion

52: of the wave front and large changes in the travel times. The

53: magnetic field effects in the sunquake waves, which to some extent

54: are obviously present, have not been detected.

55: 

56: Here, we report on observations from Hinode spacecraft of a

57: different type of oscillations excited by a solar flare. These

58: oscillations are observed in the chromosphere of the sunspot umbra

59: and inner penumbra, and, thus, probably represent some kind of

60: magnetohydrodynamic oscillatory modes. Unfortunately, the relatively

61: low cadence of the observations did not allow us to identify the

62: specific mode of the oscillations. Nevertheless, these oscillations

63: carry potentially interesting information about the flare energy

64: release and transport and properties of sunspots, and deserve

65: further observational and theoretical studies.

66: 

67: \section{HINODE Observations of December 13, 2006, Flare}

68: 

69: Images of the solar flare of December 13, 2006, were obtained by the

70: Solar Optical Telescope (SOT) onboard the solar space mission Hinode

71: in two spectral filters, the molecular line G-band and Ca~II~H. In

72: each filter the images were obtained every 2 minutes with the

73: spatial resolution of 0.2 acrsec. The pixel size of the $2048\times

74: 1024$ images was 0.109 arcsec. The first observations of sunspot

75: oscillations from Hinode by \citet{Nagashima2007} have shown that

76: the G-band signal is suppressed in the sunspot umbra, but the

77: Ca~II~H data reveal high-frequency oscillations with a peak around 6

78: mHz, originally discovered by \citet{Beckers1969}. We also use the

79: Ca~II~H images in this study (Fig.~\ref{fig1}a-d. The images were

80: carefully tracked to remove the proper motion of the large sunspot.

81: 

82: A first signature of the flare appeared in the Ca~II~H images at

83: about 2:08 UT between two sunspots of the opposite magnetic polarity

84: (Fig.~\ref{fig1}a). The energy release was probably triggered by

85: strong shearing flows around the following (smaller in size) sunspot

86: \citep{Zhang2007}. Strong shearing flows have been observed also in

87: other flares \citep[e.g.][]{Deng2006}. The flare emission quickly

88: extended during the next 10 min, forming a two-ribbon structure

89: (Fig.~\ref{fig1}b). The ribbons expanded in length and separated

90: from from each other moving into the sunspot areas

91: (Fig.~\ref{fig1}c). The two-ribbon structure is also clearly visible

92: in the G-band images (Fig.~\ref{fig1}g) indicating that the

93: perturbation extends into the photosphere. At this stage a bright

94: arcade of connecting loops with the footpoints in the ribbons

95: appeared in the Ca~II~H images (Fig.~\ref{fig1}d). A movie of the

96: image differences revealed a characteristic oscillatory pattern in

97: the umbra of the larger sunspot after the start of the flare. The

98: oscillations appear to form traveling waves with a more-or-less

99: regular wave front.

100: %Figure~\ref{fig2} shows 12

101: %consecutive frames of this movie. The round-shaped structure in the

102: %middle is the sunspot umbra (best visible in the first image at

103: %2:18:37). It is surrounded by filamentary radial structures of

104: %penumbra. The bright and black structure at the bottom of the

105: %subsequent images is the flare ribbon moving in the sunspot umbra.

106: %The large-scale circular structures best visible at the umbra

107: %boundary in the bottom row (02:34;37-02:40:38) are the sunspot

108: %oscillations excited by the flare. The oscillations propagated to

109: %the umbra-penumbra boundary, and the signal was lost in the

110: %penumbra, which have much stronger background oscillations than the

111: %umbra.

112: 

113: To illustrate the temporal behavior of the oscillations we plot the

114: relative variations of the Ca~II~H intensity in 12 points (indicated

115: by their position numbers in Fig.~\ref{fig2}) at different distances

116: from the final location of the flare ribbon, in Fig.~\ref{fig3}, and

117: compare with the variations of the soft X-ray flux from GOES-12 and

118: hard X-ray in 50-100 keV energy range from RHESSI (the first few

119: minutes of the flare are missing in RHESSI observations, which

120: started at 2:28). Evidently, the oscillations in the umbra started

121: immediately after the hard X-ray peak and before the soft X-ray

122: maximum. The oscillation amplitude of the flare-excited oscillations

123: exceeds the amplitude of preflare oscillation in the umbra by a

124: factor 2-4. The RMS of the relative intensity variations

125: (Fig.~\ref{fig4}) shows that the relative oscillations in the

126: central part of the umbra changes from 0.01-0.03 to 0.04-0.06. The

127: relative amplitude increase is smaller in the outer umbra. The flare

128: oscillation signal is essentially lost in the penumbra.

129: 

130: It can be seen from Fig.~\ref{fig3} that the oscillations excited by

131: the flare have higher frequency than the oscillations prior the

132: flare. The characteristic period of these oscillations is probably

133: shorter than 3 min. Unfortunately, the 2-min cadence of the Hinode

134: observations did not allow us to investigate the spectral and

135: dispersion properties of these oscillations. Their amplitude may be

136: significantly underestimated because of the poor sampling.

137: 

138: After the flare, the observations seem to show intrusions of hot

139: flare material into the umbra area in addition to the flare ribbons.

140: These cause variations of the Ca~II~H intensity in the umbra after

141: the flare, thus masking and distorting the oscillation signal. Thus,

142: it is rather difficult to follow the evolution of the oscillations

143: and estimate the lifetime. Figure~\ref{fig5} shows the Ca~II~H

144: intensity for longer than 12 hours in a $10\times 10$-pixel area

145: located in one of the least disturbed regions of the umbra. It seems

146: that the oscillation amplitude returned to the pre-flare level after

147: 6 hours. However, the lifetime is not yet determined.

148: 

149: An interesting question is whether the observed oscillations

150: represent traveling or standing waves. Unfortunately, the low

151: cadence of this observing run on Hinode did not allow us to

152: convincingly answer this. Nevertheless, there is a hint that the

153: waves were traveling North-ward from the flare ribbon during a short

154: period after the flare impulse, and then the oscillations became

155: more random. The traveling wave feature can be seen in

156: Fig.~\ref{fig3} as a slight positive slope (from vertical) in the

157: wave signals at points 1--6 between 2:30 and 3:00. To get an

158: estimate of the wave speed, these signals were re-sampled to a

159: 30-sec cadence and then cross-correlated with each other. The time

160: lags were measured by fitting second-degree polynomials to the

161: cross-correlation functions, and determining their maxima. This

162: procedure was repeated for 10 North-South columns along the

163: locations, 1--6, numbered in Fig.~\ref{fig2}, and the mean and

164: standard deviation values of these measurements are shown in

165: Fig.~\ref{fig6}. It seems that the wave speed is of the order of

166: 50-100 km/s. This is much higher than the sound speed in the

167: chromosphere. Thus, these waves must be of an MHD type. In general,

168: the physics of sunspot oscillations is quite complicated.

169: Understanding of this requires numerical modeling

170: \citep[e.g.][]{Khomenko2006}.

171: 

172: \section{Discussion}

173: 

174: The observations of the solar flare of December 13, 2006, from

175: Hinode reveal a new type of flare-excited oscillations. The

176: oscillations observed in Ca~II~H images appeared in the

177: chromospheric layers of the sunspot umbra immediately after the

178: impulsive phase. They had the amplitude 2--4 times larger than the

179: pre-flare oscillations in the umbra. Also, their frequency seemed to

180: be higher. There is a weak evidence that during the first 30-40 min

181: the oscillations represent waves traveling through the umbra in the

182: direction away from the flare ribbon with a speed of 50--100 km/s.

183: Then, the oscillation become more irregular with some occasional

184: wave packets. The lifetime of these oscillations is probable about 8

185: hours. The uncertainties in the data analysis and interpretation are

186: caused by the low cadence of the Hinode observing program during the

187: flare. The images were taken every 2 min while the characteristic

188: oscillation period is about 3 min. Most of the oscillation power is

189: probably even at the shorter periods. Thus, for detailed studies of

190: these oscillations it will be important to increase the image

191: cadence.

192: 

193: The image cadence should be sufficiently high to capture the initial

194: waves traveling with a speed of $\sim 100$ km/s according to our

195: preliminary estimates. This speed indicates that the waves are of an

196: MHD type, and if their speed is of the order of magnitude of the

197: Alfven speed then they should propagate rather low in the sunspot

198: chromosphere. Thus, simultaneous observations in photospheric lines

199: would be interesting. The oscillation amplitude was several times

200: higher than the amplitude of preflare umbral oscillations, which can

201: be as high as  5--6 km/s \citep{Yoon1995}, and thus may reach

202: supersonic velocities of 10--20 km/s.

203: 

204: Sunspot oscillations have been studied intensively for many years

205: (for a review see \citet{Staude1999}) but these Hinode observations

206: seem to be first that show enhanced oscillations in the umbra,

207: associated with a solar flare. Further investigations of these

208: oscillations are of great interest for understanding the processes

209: in solar flares and sunspots.

210: 

211: 

212: \acknowledgements

213: 

214:  Hinode is a Japanese mission developed and

215: launched by ISAS/JAXA, with NAOJ as domestic partner and NASA and

216: STFC (UK) as international partners. It is operated by these

217: agencies in co-operation with ESA and NSC (Norway). This work was

218: partly carried out at the NAOJ Hinode Science Center, which is

219: supported by the Grant-in-Aid for Creative Scientific Research "The

220: Basic Study of Space Weather Prediction" from MEXT, Japan (Head

221: Investigator: K. Shibata), generous donations from Sun Microsystems,

222: and NAOJ internal funding.

223: 

224: 

225: \begin{thebibliography}{}

226: 

227: \bibitem[Aschwanden et al.(1999)]{Aschwanden1999} Aschwanden, M.~J.,

228: Fletcher, L., Schrijver, C.~J., \& Alexander, D.\ 1999, \apj, 520,

229: 880

230: 

231: \bibitem[Beckers \& Tallant(1969)]{Beckers1969} Beckers, J.~M., \&

232: Tallant, P.~E.\ 1969, \solphys, 7, 351

233: 

234: \bibitem[Deng et al.(2006)]{Deng2006} Deng, N., Xu, Y., Yang,

235: G., Cao, W., Liu, C., Rimmele, T.~R., Wang, H., \& Denker, C.\ 2006,

236: \apj, 644, 1278

237: 

238: 

239: \bibitem[Khomenko \& Collados(2006)]{Khomenko2006} Khomenko, E., \&

240: Collados, M.\ 2006, \apj, 653, 739

241: 

242: 

243: 

244: \bibitem[Kosovichev \& Zharkova(1998)]{Kosovichev1998} Kosovichev, A. G. {\&} Zharkova, V. V., 1998,

245: Nature, 393, 317

246: 

247: 

248: \bibitem[Kosovichev(2006a)]{Kosovichev2006a} Kosovichev, A.~G.\ 2006a,

249: \solphys, 238, 1

250: 

251: \bibitem[Kosovichev(2006b)]{Kosovichev2006b} Kosovichev, A.~G.\ 2006b,

252: Sunquake sources and wave propagation, in: Proceedings of SOHO

253: 18/GONG 2006/HELAS I, Beyond the spherical Sun, ESA SP-624, p.134.1

254: 

255: 

256: \bibitem[Moreton(1960)]{Moreton1960} Moreton, G.~E.\ 1960, \aj, 65,

257: 494

258: 

259: \bibitem[Nagashima et al.(2007)]{Nagashima2007} Nagashima, K., et

260: al.\ 2007, ArXiv e-prints, 709, arXiv:0709.0569

261: 

262: 

263: \bibitem[Staude(1999)]{Staude1999} Staude, J.\ 1999, Third

264: Advances in Solar Physics Euroconference: Magnetic Fields and

265: Oscillations, 184, 113

266: 

267: \bibitem[Yoon et al.(1995)]{Yoon1995} Yoon, T.-S., Yun, H.~S.,

268: \& Kim, J.-H.\ 1995, Journal of Korean Astronomical Society, 28, 245

269: 

270: 

271: \bibitem[Zhang et al.(2007)]{Zhang2007} Zhang, J., Li, L., \&

272: Song, Q.\ 2007, \apjl, 662, L35

273: 

274: 

275: 

276: 

277: 

278: \end{thebibliography}

279: 

280: \clearpage

281: 

282: \begin{figure}

283: \centerline{\includegraphics[scale=0.8]{f1.eps}} \caption{A sample

284: of Hinode/SOT images of the solar flare of December 13, 2006, taken

285: in Ca~II~H line (a-d) and in the G-band (e-h). To reveal better the

286: flare signal in the G-band, the preflare image (panel e) was

287: subtracted in panels f-g.} \label{fig1}

288: \end{figure}

289: 

290: %\begin{figure}

291: %\centerline{\includegraphics[scale=0.8]{f2_new.eps}} \caption{ A

292: %series of the image differences in the Ca~II H intensity of the

293: %larger sunspot (located close to the top boundary of Fig.~\ref{fig1}

294: %images). } \label{fig2}

295: %\end{figure}

296: 

297: \begin{figure}

298: \centerline{\includegraphics[scale=1]{f2.eps}} \caption{The image of

299: the sunspot umbra with a portion of the flare ribbon at 04:08:39

300: near the end of the solar flare. The numbers from 1 to 12 show the

301: position of the intensity variation plots in Fig.~\ref{fig3}. The

302: distance between the points is approximately 1.6 Mm. The small

303: rectangular box is used to show long-term intensity variations in

304: Fig.~\ref{fig5}.} \label{fig2}

305: \end{figure}

306: 

307: \begin{figure}

308: \centerline{\includegraphics[scale=0.8]{f3.eps}} \caption{ a) The

309: relative intensity variations at 12 position within the sunspot

310: umbra and penumbra indicated in Fig.~\ref{fig2}; b) The soft x-ray

311: flux measured by GOES-10 (thick smooth curve) and hard X-ray flux in

312: the range 25-50 keV, measured by RHESSI (thin sharp curve).

313: }\label{fig3}\end{figure}

314: 

315: \begin{figure}

316: \centerline{\includegraphics[scale=0.7]{f4.eps}} \caption{RMS of the

317: relative intensity oscillation signals 2 hours before the flare,

318: 0:00-2:00 UT, (triangles) and 2 hours after the flare, 3:00-5:00 UT

319: (stars).}\label{fig4}\end{figure}

320: 

321: \begin{figure}

322: \centerline{\includegraphics[scale=0.7]{f5.eps}} \caption{a) Ca~II~H

323: intensity variations in a small umbra area shown by a rectangular

324: box in Fig.~\ref{fig2}. b) Soft X-ray flux from GOES-12 satellite.

325: }\label{fig5}\end{figure}

326: 

327: \begin{figure}

328: \centerline{\includegraphics[scale=0.7]{f6.eps}} \caption{Estimates

329: of the mean wave speed between 2:28 and 3:28~UT at 5 positions in

330: the sunspot umbra, shown in Fig.~\ref{fig2}.

331: }\label{fig6}\end{figure}

332: 

333: 

334: \clearpage

335: 

336: 

337: \end{document}

338: