1: \documentclass[preprint]{aastex}
2: \usepackage{graphicx}
3:
4: \begin{document}
5: \title{A Giant Flare from a Soft Gamma Repeater in the Andromeda
6: Galaxy, M31}
7:
8: \author{E.~P.~Mazets\altaffilmark{1}, R.~L.~Aptekar\altaffilmark{1},
9: T.~L.~Cline\altaffilmark{2}, D.~D.~Frederiks\altaffilmark{1},
10: J.~O.~Goldsten\altaffilmark{3}, S.~V.~Golenetskii\altaffilmark{1},
11: K.~Hurley\altaffilmark{4}, A.~von~Kienlin\altaffilmark{5}, and
12: V.~D.~Pal'shin\altaffilmark{1}}
13:
14: %
15: \altaffiltext{1}{Ioffe Physico-Technical Institute of the Russian
16: Academy of Sciences, St. Petersburg, 194021, Russia}
17: %
18: \altaffiltext{2}{Goddard Space Flight Center, NASA, Greenbelt, MD
19: 20771, USA}
20: %
21: \altaffiltext{3}{The Johns Hopkins University Applied Physics
22: Laboratory, MD 20723, USA}
23: %
24: \altaffiltext{4}{Space Sciences Laboratory, University of California
25: at Berkeley, Berkeley, CA 94720-7450, USA}
26: %
27: \altaffiltext{5}{Max-Plank-Institut f\"{u}r extraterrestrische
28: Physik, D-85741 Garching, Germany}
29:
30: \keywords{gamma-ray bursts; soft gamma-ray repeaters}
31:
32: \begin{abstract}
33: The light curve, energy spectra, energetics, and IPN localization of
34: an exceedingly intense short duration hard spectrum burst,
35: GRB~070201, obtained from Konus-Wind, INTEGRAL (SPI-ACS), and
36: MESSENGER data are presented. The total fluence of the burst and the
37: peak flux are $S = 2.00_{-0.26}^{+0.10} \times
38: 10^{-5}$~erg~cm$^{-2}$ and $F_{\mathrm{max}} = 1.61_{-0.50}^{+0.29}
39: \times 10^{-3}$~erg~cm$^{-2}$~s$^{-1}$. The IPN error box has an
40: area of 446 square arcminutes and covers the peripheral part of the
41: M31 galaxy. Assuming that the source of the burst is indeed in M31
42: at a distance of 0.78 Mpc, the measured values of the fluence $S$
43: and maximum flux $F_{\mathrm{max}}$ correspond to a total energy of
44: $Q = 1.5 \times 10^{45}$~erg, and a maximum luminosity $L = 1.2
45: \times 10^{47}$~erg~s$^{-1}$. These data are in good agreement with
46: the corresponding characteristics of the previously observed giant
47: flares from other soft gamma repeaters. The evidence for the
48: identification of this event as a giant flare from a soft gamma
49: repeater in the M31 galaxy is presented.
50: \end{abstract}
51:
52: %***************************************************************
53: \section{INTRODUCTION}
54: Soft gamma repeaters (SGRs) are a special rare class of strongly
55: magnetized neutron stars exhibiting two types of gamma-ray burst
56: emission. Occasionally, SGRs enter an active stage and emit repeated
57: short bursts with spectra which in an energy range above 10--15~keV
58: can be approximated by a soft thermal bremsstrahlung-like function
59: with $kT \approx 20$--30~keV. The bursting activity may last from
60: several days to a year or more, followed by a long quiescent period
61: lasting up to several years.
62:
63: Much more rarely, perhaps once every 30--40 years, an SGR may emit a
64: giant flare (GF) with enormous intensity. The energy released in
65: such a flare in gamma-rays is comparable to the total energy emitted
66: by the Sun over 10$^4$--10$^5$ years or more. GFs display short
67: ($\sim 0.2$--0.5~s) initial pulses of hard gamma rays with a steep
68: leading edge and a more extended trailing edge which as a rule
69: evolves into a soft long decaying tail pulsating with the neutron
70: star rotation period. A GF is frequently preceded by pronounced
71: increase in bursting activity \citep{Frederiks2007a}.
72:
73: The first two SGRs were discovered and localized in March, 1979 by
74: the Konus experiment on Venera 11 and 12. A giant flare from
75: SGR~0526-66 was detected on March 5, 1979 by numerous spacecraft
76: \citep{Mazets1979a, Evans1980, Cline1980}. One
77: of the first results of the interplanetary network (IPN) was the
78: localization of the first GF to an error box inside the N49
79: supernova remnant in the Large Magellanic Cloud (LMC)
80: \citep{Cline1980}. Another important result was the observation and
81: localization by the Konus experiment of $\sim$17 weak soft recurrent
82: bursts emitted by SGR~0526-66 in the years following the GF
83: \citep{Mazets1979a, Golenetskii1984}. Moreover, in 1979 March, Konus
84: detected and localized three similar recurrent bursts from another
85: source, SGR~1900+14 \citep{Mazets1979b}. The third SGR, 1806-20,
86: was discovered in 1983 \citep{Laros1987, Atteia1987,
87: Kouveliotou1987}, and the fourth, SGR~1627-41, in 1998
88: \citep{Hurley1999b, Woods1999}. It has become clear that SGRs are
89: very rare astrophysical objects. The giant flare on 1979 March 5
90: remained a unique event for 20 years, until new giant flares were
91: detected from SGR~1627-41 on 1998 June 18 \citep{Mazets1999a} and
92: from SGR~1900+14 on 1998 August 27 \citep{Hurley1999a, Feroci1999,
93: Mazets1999b}. Finally, a giant flare from SGR~1806-20 on December
94: 27, 2004 was also observed \citep{Hurley2005, Frederiks2007a}. The
95: extremely high intensity of the initial pulses of the giant flares
96: suggested that these events could be detected in galaxies at
97: distances up to 10--30~Mpc.
98:
99: The short hard GRB~051103 was localized by the interplanetary
100: network to the nearby M81 group of interacting galaxies which lies
101: at a distance of 3.6~Mpc \citep{Golenetskii2005}. The possibility
102: of identifying this burst as a GF from an SGR in the M81 group, as
103: opposed to a short GRB in a background galaxy, has been discussed
104: \citep{Frederiks2007b, Hurley2007a}.
105:
106:
107: On 2007 February 1 Konus-Wind recorded a short duration hard
108: spectrum gamma-ray burst which had a higher intensity than any
109: previously observed event. The burst was localized by the IPN
110: \citep{Golenetskii2007}. The final localization (a 446 arcmin$^2$
111: error box) covers the outer arms of the M31 galaxy. The analysis
112: presented below of the time history, spectral characteristics, and
113: energetics argues strongly for the interpretation that this event is
114: actually a GF from a soft gamma repeater in the Andromeda galaxy. In
115: a companion paper, \citet{Abbott2007} discuss the LIGO data taken at
116: the time of this event.
117:
118: %***************************************************************
119: \section{OBSERVATIONS AND LOCALIZATION}
120: The light curve of GRB~070201 (Konus trigger time T$_0$=55390.780~s
121: UT, corresponding to an Earth-crossing time 55390.261~s) recorded by
122: the Konus-Wind S2 detector in the energy range 17--1130~keV is shown
123: in Fig.~\ref{lcSum}. It displays a narrow pulse with a rather steep
124: leading edge ($\sim$20~ms) followed by more prolonged decay to
125: T-T$_0 \approx 180$~ms. The maximum count rate occurs in a $\leq
126: 2$~ms long interval. The burst profile is not smooth: in most cases,
127: the variations evident in the count rate at the top of the pulse and
128: in its decay phase are statistically significant. A weak secondary
129: flash is observed at T-T$_0 \approx 180$--280~ms. Comparison of the
130: light curves in three energy bands G1 (17--70~keV), G2
131: (70--300~keV), and G3 (300--1130~keV) (Fig.~\ref{lc3bands}) and of
132: the hardness ratios G2/G1 and G3/G2 reveals a strong, rapid spectral
133: evolution of the radiation. The most pronounced changes are observed
134: for the high-energy part of the spectrum (energies $E_\gamma >
135: 300$~keV). The duration of the high-energy radiation does not exceed
136: 80 ms. The second, weak rise in intensity at T-T$_0 \approx 180$~ms
137: is present only for the soft part of the spectrum at energies
138: $E_\gamma < 300$~keV. Four multichannel spectra were measured in the
139: course of the burst, each with an accumulation time of 64~ms. They
140: cover the energy range from 17~keV to 14~MeV, but no statistically
141: significant emission is seen above 2~MeV. The boundaries of the
142: successive accumulation intervals N=1, 2, 3, and 4 are shown in
143: Fig.~\ref{lc3bands} by dashed vertical lines. Because of the small
144: number of counts in interval 3, spectra 3 and 4 were combined. The
145: time-integrated spectrum of the entire burst was accumulated over
146: the interval 0--256 ms. The raw count rate spectra were rebinned in
147: order to have at least 20 counts per energy bin and then fitted
148: using XSPEC, version 11.3 \citep{Arnaud1996}. The detector response
149: function was calculated specifically for the $55\fdg9$ burst
150: incidence angle, determined from the IPN localization data.
151:
152: A good fit was obtained for a power-law spectrum with an exponential
153: cutoff, $dN_{ph}/ dE \propto E^{-\alpha} \exp
154: \left[-(2-\alpha)E/E_p\right]$, where $E_p$ is the peak energy in
155: the $EF(E)$ spectrum. We also tested the Band (GRRM) model but no
156: statistically significant high energy power-law tail was found in
157: any fitted spectrum. The deconvolved photon spectra for the
158: intervals 1, 2, and 3+4, and for the time-integrated spectrum of the
159: burst, are shown in Fig.~\ref{spectra}, and the spectral parameters
160: are summarized in Table~\ref{TableSpectra}. Comparison of the $E_p$
161: values shows that the spectrum for interval 1 is the hardest.
162: Returning to Fig.~\ref{lc3bands}, note that the accumulation time of
163: this spectrum is rather long (64~ms) compared to the characteristic
164: evolution time and hence the spectrum is strongly time-averaged.
165: Most of the high-energy photons are accumulated during the first
166: $\sim 20$~ms. This remarkable fact means that the flux of photons
167: with the hardest spectrum is emitted near the peak of the initial
168: pulse. The burst fluence in the 20~keV-–1.2~MeV range is
169: $2.00_{-0.26}^{+0.10} \times 10^{-5}$~erg~cm$^{-2}$. The 2-ms peak
170: flux measured from T$_0$+0.016~s in the same energy band is
171: $1.61_{-0.50}^{+0.29} \times 10^{-3}$~erg~cm$^{-2}$~s$^{-1}$. All
172: the uncertainties are for the 90\% confidence level.
173:
174: GRB~070201 was detected by Wind (Konus) with a time resolution of up
175: to 2~ms, by INTEGRAL (SPI-ACS) with a resolution of 50~ms, by
176: MESSENGER (MErcury Surface, Space ENvironment, GEochemistry, and
177: Ranging , Gamma-Ray and Neutron Spectrometer, \citet{Goldsten2007})
178: with a resolution of 1~s, and by the Swift BAT with 1~s resolution
179: (outside the coded field of view, J. Cummings, private communication
180: 2007). The initial triangulation was given in \citet{Hurley2007b}.
181: The coordinates of the center and vertices of the refined 446
182: arcmin$^2$ $3\sigma$ error box are listed in Table~\ref{TableBox}.
183: The center lies $\sim 1\degr$ away from the center of M31. Assuming
184: that the source of GRB~070201 is situated in M31 at a distance of
185: 0.78~Mpc, the measured values of the fluence and peak flux
186: correspond to an isotropic energy output $Q = 1.5 \times
187: 10^{45}$~erg, and an isotropic peak luminosity $L=1.2 \times
188: 10^{47}$~erg~s$^{-1}$.
189:
190: The relative proximity of M31 makes it worthwhile to search for the
191: afterglow of GRB~070201 in soft gamma rays, i.e., the tail of the
192: possible giant flare. While recording a burst in the trigger mode
193: with high time resolution, Konus also continues to measure the count
194: rates in each of the detectors S1 and S2 in four energy windows G1,
195: G2, G3, and Z (a charged particles channel) in background mode with
196: a time resolution of 2.944~s. The initial pulse of GRB~070201 falls
197: completely in a single 2.944~s interval around T$_0$. We began by
198: selecting a long series of count rate data, excluding this interval.
199: Then, in order to suppress fast statistical fluctuations we formed a
200: new series consisting of a sum over 94.2~s ($32 \times 2.944$s),
201: which can originate at the beginning of any 2.944~s interval. Two
202: such series, in the energy bands G1 and G2, are shown in
203: Figures~\ref{tail}a and b; they are synchronized to the first
204: 2.944~s interval following the excluded one. The position of the
205: excluded interval is marked by a narrow vertical line. The
206: discontinuity at T-T$_0 \sim$240--3800~s is due to the transfer of
207: the accumulated data to the on-board memory. A significant increase
208: in the count rate (4.3$\sigma$ above the average background level)
209: is seen only in the soft energy band G1 and only for the first large
210: interval after T$_0$. Panel~\ref{tail}c shows how the excess counts
211: $\Delta N$ vary if the beginning of the 94.2~s interval is shifted
212: forwards or backwards from T$_0$ step-by-step by 2.944~s. The
213: dependence of $\Delta N$ on the time shift $\Delta$T implies that
214: the gradually decreasing soft gamma-ray flux is indeed present in
215: each 2.944~s interval after T$_0$ but is absent prior to it. This
216: important result confirms the detection of the decaying soft
217: afterglow of GRB~070201. For an OTTB spectrum with $kT \approx
218: 30$~keV, the 94.2~s count excess $\Delta N = 1385 \pm 320$
219: corresponds to a fluence $S \approx 1 \times 10^{-6}$~erg~cm$^{-2}$.
220: Assuming that we are observing emission in the tail of a GF from
221: M31, its energy is $Q_{\mathrm{tail}} \approx 7 \times 10^{43}$~erg.
222:
223: \section{DISCUSSION}
224: The Andromeda galaxy, M31, as the closest massive galaxy, has long
225: been regarded as one of the most likely candidates for searching for
226: observable GFs from distant extragalactic SGRs \citep{Mazets1999b,
227: Bisnovatyi-Kogan1999}. It is well known that one of the dominant
228: features of the structure of M31 is a bright circular ring with a
229: radius of 10~kpc discovered by \citet{Arp1964} in H$\alpha$
230: observations. In agreement with numerous observations, the ring
231: reveals the main star-formation region of the galaxy. In the far
232: infrared, the first \textit{IRAS} images of M31 also revealed a
233: bright circular ring \citep{Habing1984}. The close correspondence
234: between the details of the H$\alpha$ and far infrared images of the
235: galaxy implies that the radiation of massive hot stars
236: \citep{Devereux1994} is the common energy source of these emissions.
237: Recently, high spatial resolution images of M31 have been obtained
238: in the mid- and far infrared at different wavelengths with the
239: \textit{Spitzer} space telescope \citep{Barmby2006, Gordon2006} and
240: in the far and near ultraviolet with \textit{GALEX
241: }\citep{Thilker2005}. These very similar images clearly show the
242: central region, the fragmented spiral arms, the main circular ring
243: with $R \approx 10$~kpc, and a less luminous outer ring with $R
244: \approx 14$~kpc.
245:
246: Figure~\ref{localization} shows the UV image of M31
247: \citep{Thilker2005}. The IPN error box overlaps the northeastern
248: part of the galaxy. It contains segments of both the $R \approx
249: 10$~kpc ring and of the weaker $R \approx 14$~kpc outer ring. The
250: probability of a chance overlapping of the error box and the image
251: is rather low ($\sim 10^{-4}$). Twenty-two sources from an
252: \textit{XMM-Newton} X-ray survey of M31 \citep{Pietsch2005} fall
253: within the error box. Among these, two sources are active galaxy
254: nuclei (AGN) and five are foreground objects. SGRs in the quiescent
255: state are typically sources of pulsating soft X-rays with an average
256: 0.5--10~keV luminosity of $10^{35}$--$10^{36}$~erg~s$^{-1}$ and a
257: spectrum which can be described as a combination of a blackbody and
258: power law \citep{Kulkarni2003, Hurley2000, Kouveliotou2001}.
259: Spectral studies of the unidentified soft X-ray sources in the error
260: box which could be SGRs in M31 would clearly be important.
261:
262: We now examine additional evidence which strengthens the
263: interpretation of the 2007 February~1 event as a GF from an SGR in
264: M31. This comes from a comparison of the energetics, time history,
265: and spectral behavior of GRB~070201 with the corresponding
266: characteristics of the previously observed giant flares on 1979
267: March 5 (SGR~0525-66), 1998 June 18 (SGR~1627-41), 1998 August 27
268: (SGR~1900+14), and 2004 December 27 (SGR 1806-20). Here we are
269: mainly interested in the characteristics of the GF initial pulses,
270: but unfortunately, it is just this information which is the hardest
271: to obtain reliably. When a sensitive gamma-ray detector records an
272: initial pulse with an immense intensity, it is strongly overloaded,
273: which makes it difficult or even impossible to obtain information
274: about the main part of the peak. If a giant pulse is recorded by
275: detectors of low-energy charged particles which have low sensitivity
276: to gamma-rays, there is no saturation problem in most cases.
277: However, severe difficulties are encountered instead when measuring
278: the photon energy, obtaining spectra, and consequently deriving
279: energy estimates. Nevertheless, the entire set of data on initial
280: pulses, obtained both with gamma-ray spectrometers
281: \citep{Mazets1979a, Mazets1982, Hurley1999a, Mazets1999a,
282: Frederiks2007a} and with charged-particle detectors
283: \citep{Hurley2005, Palmer2005, Terasawa2005, Schwartz2005,
284: Tanaka2007}, can be used to establish the general pattern of the
285: initial pulse.
286:
287: We will restrict this comparison of GFs to a limited number of their
288: characteristics, for the following reasons. The time structure of
289: the initial pulse is likely to be very complex. The profile of both
290: the steep leading edge of the pulse, as in SGR~1806-20
291: \citep{Palmer2005}, and its more prolonged decay, as in SGR~1900+14
292: \citep{Tanaka2007}, contain significant intensity variations on both
293: short and long timescales. Among other things, the appearance of the
294: light curve will be strongly affected by the spectral variability of
295: the emission. At high count rates close to saturation level,
296: dead-time and pileup corrections of light curves and energy spectra
297: are difficult, if not impossible. Therefore, we include in
298: Table~\ref{TableSummary} only two fairly reliable characteristics of
299: the initial pulse, $t_R$ and $E_p$; $t_R$, the rise time, is the
300: time between the detection of hard gamma-radiation above the
301: background to the peak of the initial pulse, $E_p$ is the photon
302: energy at the maximum of the time-integrated $EF(E)$ spectrum of the
303: pulse. Table~\ref{TableSummary} also contains data on the peak
304: luminosity $L_{\mathrm{max}}$ and the energy release $Q$ of the
305: initial pulses, as well as the energy in the GF tails
306: $Q_{\mathrm{tail}}$. In keeping with tradition, we suggest the same
307: nomenclature for extragalactic and galactic SGRs, namely SGR
308: followed by the coordinates ($\alpha$, $\delta$) of the center of
309: the error box.
310:
311: Table~\ref{TableSummary} shows that estimates of both the
312: luminosities and energy outputs in GFs are spread over a range of
313: several orders of magnitude. On the other hand, the values of
314: $Q_{\mathrm{tail}}$ are practically equal. Thus it is particularly
315: significant that the afterglow of GRB~070201 is similar to those of
316: the other SGRs. A small spread in SGR $Q_{\mathrm{tail}}$ values
317: implies similar magnetic field strengths, from confinement arguments
318: \citep{Thompson1995}. The only exception so far is the absence of a
319: tail in GF 980618 from SGR~1627-41, which is probably caused by the
320: relative weakness of the magnetic field or its orientation. On the
321: whole, Table 3 clearly demonstrates that both the temporal and
322: energetic characteristics of the event on 2007 February 1 match the
323: general pattern of a giant flare very closely. Beyond a doubt, we
324: can conclude that this event is a GF which originated in
325: SGR~0044+42 in M31.
326:
327:
328: On the Russian side this work was supported by Federal Space Agency
329: of Russia and RFBR grant 06-02-16070. KH is grateful for IPN support
330: under NASA grants NNG06GE69G (the INTEGRAL guest investigator
331: program) and NNX07AR71G (MESSENGER Participating Scientist program).
332:
333:
334: \pagebreak
335: %****************************************************************
336:
337: \begin{thebibliography}{}
338: \bibitem[Abbott et al.(2007)]{Abbott2007}Abbott, B., et al. 2007, ApJ,
339: submitted (astro-ph/0711.1163)
340:
341: %
342: \bibitem[Aptekar et al.(2001)]{Aptekar2001}Aptekar, R. L., et al. 2001, \apjs, 137, 222
343: %
344: \bibitem[Arnaud(1996)]{Arnaud1996}Arnaud, K. A. 1996, in ASP Conf. Ser. 101,
345: Astronomical Data Analysis Software and Systems V, ed. G. Jacoby, \&
346: J. Barnes, (San Francisco: ASP), 17
347: %
348: \bibitem[Arp(1964)]{Arp1964}Arp, H., et al. 1964, \apj, 139, 1045
349: %
350: \bibitem[Atteia et al.(1987)]{Atteia1987}Atteia, J-L., et al. 1987, \apj, 320, L105
351: %
352: \bibitem[Barmby et al.(2006)]{Barmby2006}Barmby, P., et al. 2006, \apj, 650, L45
353: %
354: \bibitem[Bisnovatyi-Kogan(1999)]{Bisnovatyi-Kogan1999}Bisnovatyi-Kogan, G. S. 1999, preprint (astro-ph/9911275)
355: %
356: \bibitem[Cline et al.(1980)]{Cline1980}Cline,T., et. al. 1980, \apj, 237, L1
357: %
358: \bibitem[Devereux et al.(1994)]{Devereux1994}Devereux, N. A., et al. 1994, \aj, 108, 1667
359: %
360: \bibitem[Evans et al.(1980)]{Evans1980}Evans, W. D., et al. 1980, \apj, 237, L7
361: %
362: \bibitem[Feroci et al.(1999)]{Feroci1999}Feroci, M., et al. 1999, \apj, 515, L9
363: %
364: \bibitem[Frederiks et al.(2007a)]{Frederiks2007a}Frederiks, D. D., et al. 2007a, Astronomy Letters, 33, 1
365: %
366: \bibitem[Frederiks et al.(2007b)]{Frederiks2007b}Frederiks, D. D., et al. 2007b, Astronomy Lettrs, 33, 19
367: %
368: \bibitem[Goldsten et al.(2007)]{Goldsten2007}Goldsten, J., et al. 2007, \ssr, 131, 339
369: %
370: \bibitem[Golenetskii et al.(1984)]{Golenetskii1984}Golenetskii, S. V., et al. 1984, \nat, 307, 41
371: %
372: \bibitem[Golenetskii et al.(2005)]{Golenetskii2005}Golenetskii, S. V., et al. 2005, GCN Circular 4197
373: %
374: \bibitem[Golenetskii et al.(2007)]{Golenetskii2007}Golenetskii, S. V., et al. 2007, GCN Circular 6088
375: %
376: \bibitem[Gordon et al.(2006)]{Gordon2006}Gordon, K. D., et al. 2006, \apj, 638, L87
377: %
378: \bibitem[Habing et al.(1984)]{Habing1984}Habing, H. J., et al. 1984, \apj, 278, L59
379: %
380: \bibitem[Hurley et al.(1999a)]{Hurley1999a}Hurley, K., et al. 1999a, \nat, 397, 41
381: %
382: \bibitem[Hurley et al.(1999b)]{Hurley1999b}Hurley, K., et al. 1999b, \apj, 519, L143
383: %
384: \bibitem[Hurley et al.(2000)]{Hurley2000}Hurley, K., et al. 2000, in Gamma-Ray Bursts:
385: 5th Huntsville Symp., ed. R. M. Kippen, R. S., Mallozi, \& G. F.
386: Fishman (Melville: AIP), 763
387: %
388: \bibitem[Hurley et al.(2005)]{Hurley2005}Hurley, K., et al. 2005, \nat, 434, 1098
389: %
390: \bibitem[Hurley et al.(2007a)]{Hurley2007a}Hurley, K., et al. 2007a, in preparation
391: %
392: \bibitem[Hurley et al.(2007b)]{Hurley2007b}Hurley, K., et al. 2007b, GCN Circular 6103
393: %
394: \bibitem[Kouveliotou et al.(1987)]{Kouveliotou1987}Kouveliotou, C., et al. 1987, \apj, 322, L21
395: %
396: \bibitem[Kouveliotou et al.(2001)]{Kouveliotou2001}Kouveliotou, C., et al. 2001, \apj, 558, L47
397: %
398: \bibitem[Kulkarni et al.(2003)]{Kulkarni2003}Kulkarni, S. R., et al. 2003, \apj, 585, 948
399: %
400: \bibitem[Laros et al.(1987)]{Laros1987}Laros, J. L., et al. 1987, \apj, 320, L111
401: %
402: \bibitem[Mazets et al.(1979a)]{Mazets1979a}Mazets, E. P., et al. 1979a, \nat, 282, 587
403: %
404: \bibitem[Mazets et al.(1979b)]{Mazets1979b}Mazets, E. P., et al. 1979b, Sov. Astron. Lett., 5(6), 343
405: %
406: \bibitem[Mazets et al.(1982)]{Mazets1982}Mazets, E. P., Golenetskii, S. V., Guryan,
407: Yu. A., \& Ilyinskii, V. N. 1982, \apss, 84, 173
408: %
409: \bibitem[Mazets et al.(1999a)]{Mazets1999a}Mazets, E. P., et al. 1999a, \apj, 519, L151
410: %
411: \bibitem[Mazets et al.(1999b)]{Mazets1999b}Mazets, E. P., et al. 1999b, Astronomy Letters, 25, 73
412: %
413: \bibitem[Palmer et al.(2005)]{Palmer2005}Palmer, D. M., et al. 2005, \nat, 434, 1107
414: %
415: \bibitem[Pietsch et al. (2005)]{Pietsch2005}Pietsch, W., Freyberg, M., \& Haberl, F. 2005, \aap, 434, 483
416: %
417: \bibitem[Schwartz et al.(2005)]{Schwartz2005}Schwartz, S. J., et al. 2005, \apj, 627, L129
418: %
419: \bibitem[Tanaka et al.(2007)]{Tanaka2007}Tanaka, Y. T., et al. 2007, \apj, 665, L55
420: %
421: \bibitem[Thompson \& Duncan(1995)]{Thompson1995}Thompson, C., \& Duncan, R. 1995, \mnras, 275,
422: 255
423: %
424: \bibitem[Terasawa et al.(2005)]{Terasawa2005}Terasawa, T., et al. 2005, \nat, 434,
425: 1110
426: %
427: \bibitem[Thilker et al.(2005)]{Thilker2005}Thilker, D. A., et al. 2005, \apj, 619, L67
428: %
429: \bibitem[Woods et al.(1999)]{Woods1999}Woods, P. M., et al. 1999, \apj, 519, L139
430: \end{thebibliography}
431:
432:
433:
434:
435: \clearpage
436: %
437: %xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
438: \begin{figure}
439: \centering
440: \includegraphics[]{Fig1.eps}
441: \caption{17--1130 keV Konus-Wind light curve of GRB~070201 (dead
442: time corrected) at the highest available time resolution (2 ms)
443: presented on a linear count rate scale (a), and also on a
444: logarithmic scale (b) to enhance the low-intensity portions.
445: \label{lcSum}}
446: \end{figure}
447:
448: \clearpage
449: %
450: \begin{figure}
451: \centering
452: \includegraphics[]{Fig2.eps}
453: \caption{Konus-Wind light curve in three energy bands: G1 (17--70
454: keV), G2 (70--300 keV), and G3 (300--1130 keV), and the hardness
455: ratios G2/G1 and G3/G2. The vertical dashed lines indicate the four
456: successive 64-ms intervals where the energy spectra were measured.
457: \label{lc3bands}}
458: \end{figure}
459:
460: \clearpage
461: %
462: \begin{figure}
463: \centering
464: \includegraphics[]{Fig3.eps}
465: \caption{Deconvolved photon spectra of the burst accumulated over
466: the (T-T$_0$) intervals 0-–64 ms (a), 64-–128 ms (b), 128-–256 ms
467: (c), and the time-integrated spectrum, 0-–256 ms (d).
468: \label{spectra}}
469: \end{figure}
470:
471: \clearpage
472: %
473: \begin{figure}
474: \centering
475: \includegraphics[]{Fig4.eps}
476: \caption{Waiting mode count rates $N$ in the energy bands G1 (a) and
477: G2 (b) averaged over 94.208~s intervals (excluding the burst). For
478: G1, the first interval after T$_0$ exhibits a count excess $\Delta
479: N$ of 4.3$\sigma$ above the average background level. The bottom
480: panel (c) shows how this excess $\Delta N$ varies if the beginning
481: of the 94.208-interval is shifted by an interval $\Delta$T forwards
482: or backwards from T$_0$ step by step, with $\Delta$T=2.944~s.
483: \label{tail}}
484: \end{figure}
485:
486: \clearpage
487: %
488: \begin{figure}
489: \centering
490: \includegraphics[width=0.7\textwidth]{Fig5.eps}
491: %\plotone{Fig5.eps}
492: \caption{UV image of the M31 galaxy
493: \citep{Thilker2005} and the 3$\sigma$ IPN error box of GRB~070201.
494: \label{localization}}
495: \end{figure}
496:
497:
498: \clearpage
499: %
500: \begin{deluxetable}{rccc}
501: %\tabletypesize{\scriptsize}
502: \tablewidth{0pt} \tablecaption{Summary of spectral fits
503: \label{TableSpectra}} \tablehead{ \colhead{Time interval} &
504: \colhead{$\alpha$} &
505: \colhead{$E_p$} & \colhead{$\chi^2$/dof}\\
506: \colhead{(s)} & \colhead{} & \colhead{(keV)} & \colhead{}}
507: %
508: \startdata
509: %
510: 0--0.064 & $0.52_{-0.15}^{+0.13}$ & $360_{-38}^{+44}$ & 31.6/35\\
511: 0.064--0.128 & $0.56_{-0.42}^{+0.38}$ & $128_{-16}^{+24}$ & 11.3/15\\
512: 0.128--0.256 & $1.06_{-0.52}^{+0.42}$ & $123_{-25}^{+54}$ & 19.3/16\\
513: 0--0.256 & $0.98_{-0.11}^{+0.10}$ & $296_{-32}^{+38}$ & 39.7/40\\
514: %
515: \enddata
516: \end{deluxetable}
517: %
518: \begin{deluxetable}{ccc}
519: %\tabletypesize{\scriptsize}
520: \tablewidth{0pt} \tablecaption{IPN error box of GRB~070201
521: \label{TableBox}} \tablehead{ \colhead{} & \colhead{RA(2000)} &
522: \colhead{Dec(2000)}}
523: %
524: %
525: \startdata
526: %
527: Center & $00^{\mathrm h} 44^{\mathrm m} 32^{\mathrm s}$ &
528: $+42\arcdeg 14\arcmin 21\arcsec$\\
529: Vertices & & \\
530: 1 & $00^{\mathrm h} 46^{\mathrm m} 18^{\mathrm s}$ & $+41\arcdeg
531: 56\arcmin 42\arcsec$\\
532: 2 & $00^{\mathrm h} 41^{\mathrm m} 51^{\mathrm s}$ & $+42\arcdeg
533: 52\arcmin 08\arcsec$\\
534: 3 & $00^{\mathrm h} 42^{\mathrm m} 47^{\mathrm s}$ & $+42\arcdeg
535: 31\arcmin 41\arcsec$\\
536: 4 & $00^{\mathrm h} 47^{\mathrm m} 14^{\mathrm s}$ & $+41\arcdeg
537: 35\arcmin 35\arcsec$\\
538: \enddata
539: \end{deluxetable}
540:
541:
542: \begin{deluxetable}{lcccccc}
543: \tabletypesize{\scriptsize} \rotate \tablewidth{0pt}
544: %
545: \tablecaption{Some characteristics of known giant
546: flares\label{TableSummary}} \tablehead{\colhead{} & \colhead{SGR
547: 0526-66\tablenotemark{a} } & \colhead{SGR 1627-41\tablenotemark{b}}
548: & \colhead{SGR 1900+14\tablenotemark{c}} & \colhead{SGR
549: 1806-20\tablenotemark{d}} & \colhead{SGR 0952+69\tablenotemark{e}} &
550: \colhead{SGR 0044+42\tablenotemark{f}}\\
551: \colhead{} & \colhead{(LMC)} & \colhead{} & \colhead{} & \colhead{}
552: & \colhead{(M81)} & \colhead{(M31)}}
553: %
554: %
555: \startdata
556: %
557: Data, YYMMDD & 790305 & 980618 & 980827 & 041227 & 051103 &
558: 070201\\
559: %
560: Distance, kpc & 50 & 10 & 15 & 15 & 3600 & 780\\
561: %
562: Rise time, $t_R$, ms & $\sim 25$ & $\sim 15$ & $\sim 15$ & $\sim 25$
563: & $\sim 6$ & $\sim 20$\\
564: %
565: Peak energy, $E_p$, keV & $\sim 500$ & $\sim 150$ & $> 250$ & $\sim
566: 850$ & $\sim 900$ & $\sim 300$\\
567: %
568: Luminosity, $L_{\mathrm{max}}$, erg~s$^{-1}$ & $6.5 \times 10^{45}$
569: & $3.4 \times 10^{44}$ & $2.3 \times 10^{46}$ & $3.5 \times 10^{47}$
570: & $4.3
571: \times 10^{48}$ & $1.2 \times 10^{47}$\\
572: %
573: Energy release, $Q$, erg & $7 \times 10^{44}$ & $1 \times 10^{43}$ &
574: $4.3 \times 10^{44}$ & $2.3 \times 10^{46} $ & $7 \times 10^{46}$ &
575: $1.5
576: \times 10^{45}$\\
577: %
578: Tail energy, $Q_{\mathrm{tail}}$, erg & $3.6 \times 10^{44}$ &
579: absent & $1.2 \times 10^{44}$ & $2.1 \times 10^{44}$ & not detected
580: & $7 \times
581: 10^{43}$\\
582: \enddata
583: \tablenotetext{a}{The previous dead-time correction for the event
584: \citep{Mazets1979a, Mazets1982} has been revised here, assuming that
585: the initial pulse displays the expected strong hard-to-soft spectral
586: evolution. This resulted in a reduction in $t_R$ and an increase in
587: $L_{\mathrm{max}}$ and $Q$.}
588: %
589: \tablenotetext{b}{\citet{Mazets1999a, Aptekar2001}.}
590: %
591: \tablenotetext{c}{\citet{Tanaka2007, Hurley1999a, Mazets1999b}.}
592: %
593: \tablenotetext{d}{\citet{Palmer2005, Frederiks2007a}.}
594: %
595: \tablenotetext{e}{\citet{Frederiks2007b}.}
596: %
597: \tablenotetext{f}{This work.}
598: %
599: \end{deluxetable}
600: %
601: \end{document}
602: