1: \documentclass[12pt,preprint]{aastex}
2:
3: \newcommand{\vdag}{(v)^\dagger}
4: \newcommand{\myemail}{skywalker@galaxy.far.far.away}
5:
6: \shorttitle{Two types of softening}
7:
8: \shortauthors{Qin et al.}
9:
10: \begin{document}
11:
12: \title{Two types of softening detected in X-ray
13: afterglows of Swift bursts: internal and external shock origins?}
14:
15: \author{Y.-P. Qin\altaffilmark{1,2}, A. C. Gupta\altaffilmark{3,1},
16: J. H. Fan\altaffilmark{1}, R.-J. Lu\altaffilmark{2}}
17:
18: \altaffiltext{1}{Center for Astrophysics, Guangzhou University,
19: Guangzhou 510006, P. R. China; ypqin@gzhu.edu.cn}
20:
21: \altaffiltext{2}{Physics Department, Guangxi University, Nanning
22: 530004, P. R. China}
23:
24: \altaffiltext{3}{Aryabhatta Research Institute of Observational
25: Sciences (ARIES), Manora Peak, Nainital - 263129, India}
26:
27: \begin{abstract}
28:
29: The softening process observed in the steep decay phase of early
30: X-ray afterglows of Swift bursts has remained a puzzle since its
31: discovery. The softening process can also be observed in the later
32: phase of the bursts and its cause has also been unknown. Recently,
33: it was suggested that, influenced by the curvature effect, emission
34: from high latitudes would shift the Band function spectrum from
35: higher energy band to lower band, and this would give rise to the
36: observed softening process accompanied by a steep decay of the flux
37: density. The curvature effect scenario predicts that the terminating
38: time of the softening process would be correlated with the duration
39: of the process. In this paper, based on the data from the UNLV GRB
40: group web-site, we found an obvious correlation between the two
41: quantities. In addition, we found that the softening process can be
42: divided into two classes: the early type softening ($t_{s,max}\leq
43: ``4000"s$) and the late type softening ($t_{s,max} > ``4000"s$). The
44: two types of softening show different behaviors in the duration vs.
45: terminating time plot. In the relation between the variation rates
46: of the flux density and spectral index during the softening process,
47: a discrepancy between the two types of softening is also observed.
48: According to their time scales and the discrepancy between them, we
49: propose that the two types are of different origins: the early type
50: is of internal shock origin and the late type is of external shock
51: origin. The early softening is referred to the steep decay just
52: following the prompt emission, whereas the late decay typically
53: conceives the transition from flat decay to late afterglow decay. We
54: suspect that there might be a great difference of the Lorentz factor
55: in two classes which is responsible for the observed discrepancy.
56:
57: \end{abstract}
58:
59: \keywords{gamma-rays: bursts --- gamma-rays: theory ---
60: relativity}
61:
62:
63: \section{Introduction}
64:
65: The newly discovered phenomenon, the {\it spectral softening} of
66: gamma-ray bursts (GRBs) observed by the Swift instruments in their
67: early X-ray afterglows [1$-$5] has not been predicted by any of the
68: existing GRB models. Credit of this discovery goes to the Swift
69: instruments, which can monitor X-ray emission of GRBs at quite early
70: time after the trigger events. Soon after this discovery, authors of
71: Ref. [6] performed a systematic analysis on a selected sample of
72: Swift bursts and found that $\sim 75\%$ of the bursts (33 out of 44)
73: show an obvious spectral softening process. Besides their papers,
74: the UNLV (University of Nevada, Las Vegas) GRB group also presented
75: the temporal and spectral data of Swift bursts on their
76: web-site\footnote{http://swift.physics.unlv.edu/.}, which are
77: continued to be accumulated as the number of Swift bursts keeps
78: increasing. Their data show, besides the steep decay phase, the
79: softening can be observed in later phases as well (see our analysis
80: below).
81:
82: Most softening processes are detected in the steep decay phase in
83: the early X-ray afterglows of GRBs. Since the steep decay phase
84: promptly follows, and is smoothly connected to, the prompt emission
85: phase, it is regarded as the prompt emission tail [7$-$10]. It is
86: generally believed that this phenomenon is due to the high latitude
87: emission of fireballs, in which the so-called curvature effect must
88: play a role [6, 9, 11$-$18]. However, the steep decay tails are
89: expected from the curvature effect but the softening process is not,
90: which puzzled astronomers.
91:
92: The curvature effect arises from the emission from the surface of an
93: relativistically expanding fireball, where the delay of time, the
94: shifting of the intrinsic spectrum due to high latitude emission
95: areas, as well as other relevant factors of expanding fireballs must
96: be taken into account (for detail explanation and analysis, see [17,
97: 19$-$21]). Due to the great amount of energy release,
98: relativistically expanding fireballs would be produced at very early
99: epoch of GRBs [22, 23]. The curvature effect is thus expected in the
100: prompt gamma-ray emission phase. Investigations on the profile of
101: the light curves of pulses, the spectral lags, the power-law
102: relation between the pulse width and the energy, the evolution of
103: the hardness ratio and the evolution of the peak energy in the
104: prompt emission phase have been performed by various groups in the
105: last decade [19$-$21, 24$-$34]. It was shown that the effect can
106: also play an important role in the early afterglow period [11, 12,
107: 17, 35].
108:
109: There are only few attempts of interpreting the softening phenomenon
110: which indicate that the phenomenon is beyond the expectation of
111: current or underlying models. The few attempts of interpretation
112: include: cooling of the internal-shocked region might be responsible
113: for strong softening [6]; at least in some bursts, the softening
114: might be accounted for by the central engine, which is assumed to
115: produce a soft and decaying afterglow emission [1, 6, 36]; both the
116: temporal behavior and the spectral softening of bursts might be a
117: consequence of the cannonball model of GRBs [37]; a hard-to-soft
118: behavior lasting to the latest phases of the afterglow can be
119: expected base on the ``fireshell'' model with a ``canonical GRB''
120: light curve containing two sharply different components [38, 39].
121: One of the most remarkable investigation on this issue is performed
122: by authors of Ref. [16]. They concluded that the early emission in
123: $>90\%$ of early afterglows has a curved $\nu f_{\nu}$ spectrum and
124: that $E_{peak}$ (peak energy) likely evolves from the $\gamma$-rays
125: through the soft X-ray bands on the timescales of $10^2 - 10^4$ s
126: after the GRBs. Along with this is the discovery of Ref. [5]: the
127: $E_{peak}$ of GRB 060614, which is one of the members of Ref. [6]'s
128: sample, decreases to as low as $\sim$ 8 keV at the beginning of the
129: XRT observations. The same phenomenon was revealed in literature as
130: early as in 2000 by the analysis of BeppoSAX data: the peak energy
131: was found to evolve from the prompt to the afterglow phase of GRBs,
132: decreasing from $>700 keV$ to $<3 keV$ for some bursts [40].
133:
134: Motivated by Ref. [16]'s finding, very recently, the author of [35]
135: has shown that the curvature effect alone can produce both the
136: softening and the decaying behavior observed in the early X-ray
137: afterglow of the Swift bursts. It is due the shifting of the Band
138: function spectrum [41] which gives rise to the softening along with
139: the temporal decaying. Two factors of the curvature effect, the time
140: delay and the variation of the Doppler effect of higher latitude
141: emission from the fireball surface, cause the shifting of the Band
142: function spectrum.
143:
144: According to the curvature effect scenario, the start time of the
145: softening is much smaller than the terminating time of the process,
146: and then it is expected that the softening duration must be
147: correlated with its terminating time [35]. We will investigate
148: statistically in the following if the two quantities are correlated
149: or not. At the same time, some other statistical properties will
150: also be explored. We will not limit our analysis on the steep decay
151: phase, but instead, any softening detected in the XRT light curve
152: will be considered.
153:
154: The paper is organized as follows: in Section 2, we describe our
155: sample of Swift data and the data reduction; in Section 3, we
156: discuss the statistical properties obtained from our analysis;
157: conclusions of the present work are reported in the last section.
158:
159: \section{Data}
160:
161: The data (up to May 23, 2008) employed in our analysis on the
162: relation between the duration and terminating time of the softening
163: process are taken from the UNLV GRB group web-site (see footnote 1),
164: where Swift/XRT time-resolved spectra of selected bursts are
165: available (see Ref. [6] for selection criteria of the bursts). The
166: softening is observed in some of the bursts and we selected only
167: those bursts which have noticeable X-ray softening, i.e., those
168: bursts which should contain at least three data points starting from
169: a smaller spectral index and ending at a larger one (refer to the
170: time intervals presented in Table 1 and the corresponding spectral
171: evolution figures presented in the UNLV GRB group web-site).
172:
173: The start time ($t_{s,min}$) and the terminating time ($t_{s,max}$),
174: together with the corresponding values of the spectral index
175: ($\beta_{min}$ and $\beta_{max}$), of the softening process of the
176: selected bursts are listed in Table 1. Here we divide the softening
177: process into two distinct classes according to the corresponding
178: terminating time: for class 1, $t_{s,max}\leq ``4000"s$ (called the
179: early type softening); and for class 2, $t_{s,max} > ``4000"s$
180: (called the late type softening). Note that, for some bursts, there
181: might exist both types of softening (see Table 1).
182:
183: \begin{figure}[tbp]
184: \begin{center}
185: \includegraphics[width=5in,angle=0]{f1.eps}
186: \end{center}
187: \caption{The spectral evolution (the lower panel) and light curve
188: (the upper panel) of GRB 070520B. The data are taken from
189: http://swift.physics.unlv.edu/ (see the spectra.txt and the lc.txt
190: files of GRB 070520B). The dash lines in the lower panel denote the
191: time positions of $t_{s,min}$ and $t_{s,max}$, and those in the
192: upper panel represent that of $t_{f,min}$ and $t_{f,max}$.}
193: \label{Fig. 1}
194: \end{figure}
195:
196: An examples of selecting the softening as well as the corresponding
197: start time and terminating time is displayed in the lower panel of
198: Fig. 1 (note that we employ index $\beta$ instead $\Gamma$), where
199: the spectral evolution of GRB 070520B is shown. The light curve of
200: this burst is displayed in the upper panel of the figure. The
201: spectral and light curve data are taken from
202: http://swift.physics.unlv.edu/ (spectra.txt and lc.txt files). We
203: first selected $t_{s,min}$ and $t_{s,max}$ by viewing the $\beta$
204: vs. $t$ plot (see the lower panel of Fig. 1 and the dash lines
205: there), and then located them in the spectra.txt file of the burst,
206: and then read and calculated (see the explanation below) their
207: values as well as their uncertainties from this file. The provided
208: data have been analyzed by the UNLV GRB group. The details of the
209: analysis are described in Ref. [6]. The UNLV GRB group have
210: developed a time filter for the time-resolved spectral analysis
211: which can be automatically performed. Time intervals for analyzing
212: the spectral index are determined by two criteria raised by them,
213: and hence they are different from each other. For example, for GRB
214: 070520B, the lower limit of the time interval associated with its
215: $\beta_{min}=1.196\pm 0.054$ is $t_1 = 125 s$ and the upper limit of
216: this interval is $t_2 = 195 s$ (see Fig. 1 and Table 1, and also
217: the spectra.txt file). This gives rise to a time interval of $70 s$.
218: According to the data format file, the mean time of this interval is
219: $t=(t_1+t_2)/2=160 s$ and its error is $\sigma_t=(t_2-t_1)/2=35 s$
220: (see Table 1). However, the lower limit of the time interval
221: associated with $\beta_{max}=1.92\pm 0.21$ of this burst is $t_1 =
222: 308 s$, and the upper limit of this interval is $t_2 = 391 s$. That
223: measures the time interval as $83 s$. The mean time of this interval
224: is $t=(t_1+t_2)/2=349.5 s$ and its error is
225: $\sigma_t=(t_2-t_1)/2=41.5 s$ (see Table 1). The UNLV GRB group use
226: different NH for different sources. They used the XSPEC spectral
227: fitting model: a simple power law combined with the absorptions of
228: both our Galaxy and the GRB host galaxy, wabs$^{Gal} \times$
229: zwabs$^{host} \times$ power law (for bursts with known redshifts) or
230: wabs$^{Gal} \times$ wabs$^{host} \times$ power law (for bursts whose
231: redshifts are unknown) (see Ref. [6]).
232:
233:
234: \section{Results}
235:
236: \subsection{Relation between the duration and terminating time of the softening process}
237:
238: \begin{figure}[tbp]
239: \begin{center}
240: \includegraphics[width=5in,angle=0]{f2.eps}
241: \end{center}
242: \caption{Relation between the duration $\Delta t_s$ and the
243: terminating time $t_{s,max}$ of the softening process observed in the
244: X-ray afterglows of the selected bursts. The upper and lower solid
245: lines are drawn by $\Delta t_s = t_{s,max}$ and $\Delta t_s =
246: t_{s,max}/2$, respectively. Open circles and open circles with dots
247: represent the early type softening and the late type softening,
248: respectively.} \label{Fig. 1}
249: \end{figure}
250:
251: The duration of the softening is calculated by $\Delta t_{s} =
252: t_{s,max} - t_{s,min}$. The relation between $\Delta t_s$ and
253: $t_{s,max}$ for these bursts is displayed in Fig. 2. We find that
254: the data of ($t_{s,max}$, $\Delta t_s$) for the selected bursts are
255: distributed mainly within the area confined by the lines represented
256: by $\Delta t_s = t_{s,max}$ and $\Delta t_s = t_{s,max}/2$. We
257: observed a difference between the two types of softening: the
258: ($t_{s,max}$, $\Delta t_s$) data for the late type softening are
259: well within the mentioned area, while some of the data for the early
260: type softening dropped out the mentioned area (especially when
261: $t_{s,max}$ is relatively smaller).
262:
263: We performed a Spearman's correlation analysis and obtained: for the
264: early type softening, log $\Delta t_s = (1.172 \pm 0.051)$ log
265: $t_{s,max} - (0.69 \pm 0.14)$, and the fitting parameters are
266: $R=0.951$ (the correlation coefficient), $N=57$ (the number of data
267: points), and $P=8.64\times 10^{-30}$ (the chance probability); for
268: the late type softening, log $\Delta t_s = (1.026 \pm 0.040)$ log
269: $t_{s,max} - (0.19 \pm 0.18)$, with $R=0.982$, $N=26$, and
270: $P=7.51\times 10^{-19}$. This is well in agreement with the prediction made by
271: the curvature effect [35]. In addition, the results show that the
272: correlation between the two quantities for the late type softening
273: well follows the trend of the identical curve, while in some extent
274: it betrays the identical curve for the early type softening (see
275: also Fig. 1). The cause of this difference is currently not known.
276:
277:
278: \subsection{Two types of softening distinguished in other aspects}
279:
280: In addition to the duration and terminating time of the softening
281: process, we also measured the variations of the spectral index and
282: the flux density during this period.
283:
284: The period of the softening is confined by the lower limit ($t_1$)
285: of the interval that measures $\beta_{min}$ and the upper limit
286: ($t_2$) of the interval that measures $\beta_{max}$. For example,
287: for GRB 070520B, the lower limit of the time interval associated
288: with its $\beta_{min}=1.196\pm 0.054$ is $t_1 = 125 s$, and the
289: upper limit of the time interval associated with its
290: $\beta_{max}=1.92\pm 0.21$ is $t_2 = 391 s$ (see the last section),
291: and thus the softening period of this burst is the time interval
292: from $125 s$ to $391 s$. Within this period, we search the minimum
293: and maximum of the flux $f_{\nu}$ from the lc.txt file of GRB
294: 070520B. The two extreme values of the flux are denoted by
295: $f_{\nu,min}$ and $f_{\nu,max}$ respectively, and the corresponding
296: times are denoted by $t_{f,min}$ and $t_{f,max}$ respectively. The
297: values of $t_{f,min}$ and $t_{f,max}$ and their uncertainties are
298: estimated in the same way adopted in measuring $t_{s,min}$ and
299: $t_{s,max}$ and their uncertainties (see the last section). For GRB
300: 070520B, we found $f_{\nu,max}=(52.7\pm 3.9)\times 10^{-10} erg\cdot
301: cm^{-2}\cdot s^{-1}$, $f_{\nu,min}=(10.5\pm 6.1)\times 10^{-11}
302: erg\cdot cm^{-2}\cdot s^{-1}$, $t_{f,max}=181.6\pm 1.0 s$, and
303: $t_{f,min}=353.6\pm 1.0 s$ (see Table 2). One might observe that for
304: this burst $t_{f,max}$ is different from $t_{s,min}$ and $t_{f,min}$
305: is different from $t_{s,max}$. Two facts cause this difference. The
306: first is that time intervals for measuring the flux are generally
307: smaller than those for measuring the spectral index (see Fig. 1,
308: where the number of data points of the flux within the softening
309: period is much larger than that of the spectral index). The second
310: is that in the softening process some bursts undergo a rise phase
311: and then a decay phase in its light curve, and close to the lower
312: limit of the softening we measure the peak of the flux instead of
313: the flux at the very beginning of the softening process (see also
314: Fig. 1). (Note that we identified a softening process according to
315: the spectral index but not the flux.)
316:
317: In Table 2, we listed the maximum and minimum values of the flux
318: density detected during the softening process. We calculated the
319: variation rates of the two quantities by $\Delta \beta / \Delta t_s
320: = (\beta_{max}-\beta_{min}) / \Delta t_s$ and $\Delta \rm{log}
321: f_{\nu} / \Delta t_f = (\rm{log} f_{\nu,max} - \rm{log} f_{\nu,min})
322: / \Delta t_f$, where $\Delta t_f = t_{f,min}-t_{f,max}$ is the time
323: interval between $f_{\nu,max}$ and $f_{\nu,min}$, which can be
324: different from $\Delta t_s$ (see Table 2).
325:
326: \begin{figure}[tbp]
327: \begin{center}
328: \includegraphics[width=5in,angle=0]{f3.eps}
329: \end{center}
330: \caption{Relation between the variation rates of the flux density
331: and spectral index, $\Delta \rm{log} f_{\nu} / \Delta t_f$ and
332: $\Delta \beta / \Delta t_s$, during the softening process observed
333: in the X-ray afterglows of the selected bursts. Solid lines from top
334: to the bottom represent the $\Delta \rm{log} f_{\nu} / \Delta t_f =
335: k \Delta \beta / \Delta t_s$ lines with $k=6$, 2, 1, and 0.5,
336: respectively. The open circles represent the early type softening
337: while the open circles with dots represent the late type softening.}
338: \label{Fig. 2}
339: \end{figure}
340:
341: In Fig. 3, we presented the relation between the two variation
342: rates. It shows that the two types of softening do have distinct
343: behaviors in the variation rate. Data of the late type softening are
344: distributed within the $\Delta \rm{log} f_{\nu} / \Delta t_f = 2
345: \Delta \beta / \Delta t_s$ and $\Delta \rm{log} f_{\nu} / \Delta t_f
346: = 6 \Delta \beta / \Delta t_s$ curves, which well follow the trend
347: of the identical curve. For the early type softening, the data are
348: scattered in a wider area confined by the $\Delta \rm{log} f_{\nu} /
349: \Delta t_f = 0.5 \Delta \beta / \Delta t_s$ and $\Delta \rm{log}
350: f_{\nu} / \Delta t_f = 6 \Delta \beta / \Delta t_s$ curves. The
351: trend of the early type is obviously deviated from that of the
352: identical curve.
353:
354: A Spearman's correlation analysis between the two variation rates
355: was also performed. For the early type softening we got $\rm{log}
356: \Delta \rm{log} f_{\nu} / \Delta t_f = (0.556 \pm 0.045) \rm{log}
357: \Delta \beta / \Delta t_s - (0.85 \pm 0.11)$, with $R=0.857$,
358: $N=57$, and $P=1.84\times 10^{-17}$, and for the late type softening
359: we obtained $\rm{log} \Delta \rm{log} f_{\nu} / \Delta t_f = (0.987
360: \pm 0.071) \rm{log} \Delta \beta / \Delta t_s + (0.51 \pm 0.33)$,
361: with $R=0.943$, $N=26$, and $P=5.49\times 10^{-13}$. The correlation
362: must owe to the fact that the softening scope (represented by
363: $\Delta \beta$) and the decaying scale (described by $\Delta
364: \rm{log} f_{\nu}$) vary mildly for different sources, but the time
365: interval of the process differ significantly. The two types of
366: softening occupy distinct areas in the plot due to the large
367: variance of the softening duration between them. But why the trends
368: of the relation for the two types are so different remains unclear.
369:
370:
371: \section{Discussion and conclusions}
372:
373: In the present work, we studied the relation between the duration
374: and terminating time of the softening process observed in the X-ray
375: afterglows of the Swift bursts. We found that these two quantities
376: are obviously correlated, as expected by the curvature effect. The
377: analysis reveals that the softening can be divided into two classes
378: merely on the basis of the corresponding terminating time: the early
379: type softening ($t_{s,max}\leq ``4000"s$) and the late type
380: softening ($t_{s,max} > ``4000"s$). The two types of softening show
381: different behaviors in their duration and the terminating time plot.
382: We also investigated the relation between the variation rates of the
383: flux density and the spectral index during the softening process. In
384: this aspect, more obvious discrepancy is observed between the two
385: types of softening.
386:
387: As revealed in Ref. [35], the duration of the softening can be
388: affected by three parameters: the Lorentz factor, the radius, and
389: the intrinsic radiative peak energy concerned. As shown in Ref. [35]
390: Fig. 6, for the same Lorentz factor $\Gamma$ and radius $R_c$, a
391: smaller intrinsic peak energy $E_{0,p}$ can lead to a smaller
392: duration of the process, while the corresponding terminating time
393: will be unchanged. It would give rise to the betray observed in Fig.
394: 1. Why this occurs in the early type softening but not in the late
395: type softening? We suspect, probably the early type softening is of
396: the internal shock origin while the late type softening is of the
397: external shock origin. In the former case the Lorentz factor is
398: large and then the curvature effect is sensitive to the fireball
399: parameters while in the latter case the Lorentz factor is small and
400: hence the curvature effect is less sensitive to the fireball
401: parameters. Is it due to the selection effect? This is unlikely,
402: because if the earlier softening tends to have a relatively smaller
403: duration due to the overlapping of its start time by other
404: components of emission then it will affect both variation rates of
405: the flux density and spectral index in the same way and then the
406: possible influence will be canceled. But Fig. 2 clearly shows that
407: the early softening does have a different behavior relative to the
408: rest.
409:
410: The fact that the GRBs with lowest $t_{s,max}$ tend to deviate from
411: the $\Delta t_s \sim t_{s,max}$ law may be partially due to the fact
412: that $\Delta t_s$ is defined as the difference of the linear
413: quantities $t_{s,max}$ and $t_{s,min}$ and we take the logarithm of
414: $\Delta t_s$ and correlate it with the logarithm of $t_{s,max}$. Is
415: it better that we use the ratio between $t_{s,max}$ and $t_{s,min}$
416: rather than use $\Delta t_s$? The ratio, when passing to logarithms,
417: becomes the difference between $log t_{s,max}$ and $log t_{s,min}$.
418: Is it a more suitable quantity to be correlated with $log
419: t_{s,max}$? We studied this issue by replacing $\Delta t_s$ with
420: $t_{s,max}/t_{s,min}$ in Fig. 2. We found a weak correlation between
421: $log (t_{s,max}/t_{s,min})$ and $log t_{s,max}$ (the plot is
422: omitted). The data are quite scattered in the $log
423: (t_{s,max}/t_{s,min})$ vs. $log t_{s,max}$ plane. The analysis does
424: not provide any information of the causes of the mentioned
425: deviation. This must be due to the fact that $log (t_{s,max} -
426: t_{s,min})$ is closer to $log t_{s,max}$ than $log t_{s,max} - log
427: t_{s,min}$ is, as long as the discrepancy between $t_{s,max}$ and
428: $t_{s,min}$ is large enough (say, when $t_{s,max}$ is one order of
429: magnitude larger than $t_{s,min}$).
430:
431: The strongest reason in favor of our suggestion of two types of
432: origin might be that the two types of softening occur at very
433: different time scales (the former appears much earlier and the
434: latter emerges very late). In addition we found that the early type
435: softening is observed in the steep decay phase which is believed to
436: be due to the high latitude emission of the prompt phase [6, 9,
437: 11$-$18] and the late type softening is found in the normal
438: afterglow phase which was believed to be due to the external shocks
439: [42$-$44].
440:
441: \begin{figure}[tbp]
442: \begin{center}
443: \includegraphics[width=5in,angle=0]{f4.eps}
444: \end{center}
445: \caption{Relation between $\Delta t_s$ and $t_{s,max}$ for the newly
446: defined two types of the softening process. Filled circles in the
447: upper panel and open circles in the lower panel represent the
448: process with $t_{s,min}>500s$ (the newly defined late type softening
449: process). Filled circles in the lower panel and open circles in the
450: upper panel stand for the process with $t_{s,min}\le 500s$ (the
451: newly defined early type softening process). The two solid lines are
452: the same as they are in Fig. 2.} \label{Fig. 1}
453: \end{figure}
454:
455: It should be pointed out that the softening process is divided into
456: two classes empirically. If the two kinds of softening are
457: associated with internal and external shock mechanisms, it might be
458: more natural to divide them according to the start time of the
459: process. By examining the data in Tables 1 and 2, we roughly
460: redivided the softening process into two classes according to
461: $t_{s,min}$ being less or larger than $500 s$. Shown in Fig. 4 are
462: the relations between the duration and the terminating time of the
463: softening process for the newly defined early and late types. As
464: expected, they do not dramatically affect the result. The two types
465: are distributed in two distinguishable domains in the $\Delta t_s$
466: vs. $t_{s,max}$ plane. However, a slight overlapping between the two
467: distributions is observed. We do not know if this overlapping is due
468: to the overlapping of physical parameters or merely the statistical
469: fluctuation. One might notice that few of the new early type occupy
470: the late type domain near the identical curve and few of the late
471: type are located in the early type domain near the $\Delta t_s =
472: t_{s,max}/2$ curve. Both seem to be a result of the extension of the
473: two types. We therefore insist that this possibility cannot be ruled
474: out with the current data. However, as an empirical analysis, we
475: prefer the former division, i.e. dividing them according to
476: $t_{s,max}$, since no overlapping is observed in this division
477: scenario.
478:
479: \begin{figure}[tbp]
480: \begin{center}
481: \includegraphics[width=5in,angle=0]{f5.eps}
482: \end{center}
483: \caption{Relation between the variations of the flux density and
484: spectral index, $\Delta \rm{log} f_{\nu}$ and $\Delta \beta$, during
485: the softening process observed in the X-ray afterglows of the
486: selected bursts. Filled circles stand for the late type of softening
487: and open circles for the early type.} \label{Fig. 2}
488: \end{figure}
489:
490: Studied in Fig. 3 is the relation between the variation rates of the
491: flux density and spectral index during the softening process. How it
492: would be when we simply study the relation between $\Delta \rm{log}
493: f_{\nu}$ and $\Delta \beta$? Shown in Fig. 5 is the result. One
494: finds that the two types seem to have different distributions of
495: $\Delta \rm{log} f_{\nu}$ and $\Delta \beta$. The late type
496: softening tends to have smaller $\Delta \beta$ and slightly larger
497: $\Delta \rm{log} f_{\nu}$. However, the overlapping is so heavy that
498: we cannot tell the type of a softening merely according to its
499: location in the $\Delta \rm{log} f_{\nu}$ vs. $\Delta \beta$ plane.
500: This difference, if confirmed statistically later, might become a
501: hint in searching the physical difference between the two softening
502: processes.
503:
504:
505: \acknowledgments
506:
507: Our special thanks are given to the anonymous referee for his or her
508: comments and suggestions which have improved the paper greatly. This
509: work is supported in part by the National Natural Scientific
510: Foundation of China (10573005, 10633010, 10747001) and the 973
511: project (No. 2007CB815405). We also thank the financial support from
512: the Guangzhou Education Bureau and Guangzhou Science and Technology
513: Bureau.
514:
515: %\begin{thebibliography}{}
516: %\bibliography{}
517:
518: \section{References}
519:
520: [1]{} ~S. Campana et al., 2006, {\it Nature}, 442, 1008
521:
522: [2]{} ~G. Ghisellini et al., 2006, {\it Mon. Not. R. Astron. Soc}, 372, 1699
523:
524: [3]{} ~N. Gehrels et al., 2006, {\it Nature}, 444, 1044
525:
526: [4]{} ~B. Zhang et al., 2007, {\it Astrophys. J.}, 655, L25
527:
528: [5]{} ~V. Mangano et al., 2007, {\it Astron. Astrophys.}, 470, 105
529:
530: [6]{} ~B. B. Zhang, E. W. Liang, and B. Zhang, 2007, {\it Astrophys. J.}, 666, 1002
531:
532: [7]{} ~G. Tagliaferri et al., 2005, {\it Nature}, 436, 985
533:
534: [8]{} ~S. D. Barthelmy et al., 2005, {\it Astrophys. J.}, 635, L133
535:
536: [9]{} ~E. W. Liang et al., 2006, {\it Astrophys. J.}, 646, 351
537:
538: [10]{} ~T. Sakamoto et al., 2007, {\it Astrophys. J.}, 669, 1115
539:
540: [11]{} ~P. Kumar, and A. Panaitescu, 2000, {\it Astrophys. J.}, 541, L51
541:
542: [12]{} ~C. D. Dermer, 2004, {\it Astrophys. J.}, 614, 284
543:
544: [13]{} ~J. Dyks et al., 2005, astro-ph/0511699
545:
546: [14]{} ~A. Panaitescu et al., 2006, {\it Mon. Not. R. Astron. Soc}, 366, 1357
547:
548: [15]{} ~B. Zhang et al., 2006, {\it Astrophys. J.}, 642, 354
549:
550: [16]{} ~N. R. Butler, and D. Kocevski, 2007, {\it Astrophys. J.}, 663, 407
551:
552: [17]{} ~Y.-P. Qin, 2008, {\it Astrophys. J.}, 683, 900
553:
554: [18]{} ~R. L. C. Starling et al., 2008, {\it Mon. Not. R. Astron. Soc}, 384, 504
555:
556: [19]{} ~Y.-P. Qin, 2002, {\it Astron. Astrophys.}, 396, 705
557:
558: [20]{} ~Y.-P. Qin et al., 2004, {\it Astrophys. J.}, 617, 439
559:
560: [21]{} ~Y.-P. Qin et al., 2006, {\it Phys. Rev. D} 74, 063005
561:
562: [22]{} ~J. Goodman, 1986, {\it Astrophys. J.}, 308, L47
563:
564: [23]{} ~B. Paczynski, 1986, {\it Astrophys. J.}, 308, L43
565:
566: [24]{} ~E. E. Fenimore et al., 1996, {\it Astrophys. J.}, 473, 998
567:
568: [25]{} ~R. Sari, and T. Piran, 1997, {\it Astrophys. J.}, 485, 270
569:
570: [26]{} ~F. Ryde, and V. Petrosian, 2002, {\it Astrophys. J.}, 578, 290
571:
572: [27]{} ~D. Kocevski, F. Ryde, and E.-W. Liang, 2003, {\it Astrophys. J.}, 596, 389
573:
574: [28]{} ~Y.-P. Qin, and R.-J. Lu, 2005, {\it Mon. Not. R. Astron. Soc}, 362, 1085
575:
576: [29]{} ~R.-F. Shen et al., 2005, {\it Mon. Not. R. Astron. Soc}, 362, 59
577:
578: [30]{} ~R.-J. Lu, Y.-P. Qin, Z.-B. Zhang, and T.-F. Yi, 2006,
579: {\it Mon. Not. R. Astron. Soc}, 367, 275
580:
581: [31]{} ~R.-J. Lu, Z.-Y. Peng, and W. Dong, 2007, {\it Astrophys. J.}, 663, 1110
582:
583: [32]{} ~Z.-Y. Peng et al., 2006, {\it Mon. Not. R. Astron. Soc}, 368, 1351
584:
585: [33]{} ~Y.-P. Qin et al., 2005, {\it Astrophys. J.}, 632, 1008
586:
587: [34]{} ~L.-W. Jia, 2008, {\it Chinese J. Astron. Astrophys.}, 4, 451
588:
589: [35]{} ~Y.-P. Qin, 2008, ApJ, accepted [arXiv:0806.3339v1]
590:
591: [36]{} ~Y. Z. Fan et al., 2006, {\it JCAP}, 9, 13
592:
593: [37]{} ~S. Dado, A. Dar, and A. D. Rujula, 2008, {\it Astrophys.
594: J.}, 681, 1408
595:
596: [38]{} ~R. Ruffini et al., 2006, {\it Astrophys. J.}, 645, L109
597:
598: [39]{} ~C. L. Bianco et al., 2008, {\it AIPC}, 966, 12
599:
600: [40]{} ~F. Frontera et al., 2000, {\it Astrophys. J. Suppl. Ser.},
601: 127, 59
602:
603: [41]{} ~D. Band et al. 1993, {\it Astrophys. J.}, 413, 281
604:
605: [42]{} ~P. Meszaros, and M. J. Rees, 1997, {\it Astrophys. J.}, 476,
606: 232
607:
608: [43]{} ~R. Sari, T. Piran, and R. Narayan, 1998, {\it Astrophys.
609: J.}, 497, L17
610:
611: [44]{} ~R. A. Chevalier, and Z.-Y. Li, 2000, {\it Astrophys. J.},
612: 536, 195
613:
614: %\end{thebibliography}
615:
616: \begin{deluxetable}{llllll}
617:
618: \tablewidth{0pt} \tablecaption{The minimum and maximum values of the
619: spectral index ($\beta_{min}$ and $\beta_{max}$) and the
620: corresponding times ($t_{s,min}$ and $t_{s,max}$), of the softening
621: process of the selected bursts.} \tablehead{ \colhead{type} &
622: \colhead{burst} & \colhead{$t_{s,min} (s)$} &
623: \colhead{$\beta_{min}$} & \colhead{$t_{s,max} (s)$} &
624: \colhead{$\beta_{max}$} } \startdata
625:
626: early & 050315 & 91.0 $\pm$ 7.0 & 1.02 $\pm$
627: 0.27 & 620 $\pm$ 230 & 1.58 $\pm$
628: 0.45 \\
629: & 050421 & 124.0 $\pm$ 9.0 & -0.23 $\pm$
630: 0.32 & 247 $\pm$ 44 & 1.35 $\pm$
631: 0.36 \\
632: & 050502B & 340 $\pm$ 270 & 0.972 $\pm$
633: 0.053 & 1200 $\pm$ 150 & 2.29 $\pm$
634: 0.26 \\
635: & 050714B & 188 $\pm$ 31 & 4.80 $\pm$
636: 0.20 & 474 $\pm$ 73 & 6.06 $\pm$
637: 0.40 \\
638: & 050716 & 114 $\pm$ 10 & -0.13 $\pm$
639: 0.12 & 396 $\pm$ 88 & 1.019 $\pm$
640: 0.084 \\
641: & 050717 & 94.0 $\pm$ 3.0 & 0.08 $\pm$
642: 0.24 & 247 $\pm$ 32 & 0.90 $\pm$
643: 0.20 \\
644: & 050724 & 87.9 $\pm$ 4.3 & 0.38 $\pm$
645: 0.10 & 318 $\pm$ 16 & 2.24 $\pm$
646: 0.31 \\
647: & 050726 & 366 $\pm$ 32 & 0.72 $\pm$
648: 0.26 & 689 $\pm$ 58 & 1.02 $\pm$
649: 0.26 \\
650: & 050730 & 145 $\pm$ 12 & 0.245 $\pm$
651: 0.089 & 729 $\pm$ 66 & 0.962 $\pm$
652: 0.062 \\
653: & 050814 & 173.1 $\pm$ 8.1 & 0.885 $\pm$
654: 0.091 & 358 $\pm$ 26 & 1.88 $\pm$
655: 0.25 \\
656: & 050904 & 182 $\pm$ 13 & 0.043 $\pm$
657: 0.072 & 533 $\pm$ 48 & 0.977 $\pm$
658: 0.093 \\
659: & 050922B & 673 $\pm$ 17 & 0.93 $\pm$
660: 0.29 & 1460 $\pm$ 220 & 2.21 $\pm$
661: 0.41 \\
662: & 051117A & 153 $\pm$ 40 & 0.722 $\pm$
663: 0.028 & 1560 $\pm$ 170 & 1.268 $\pm$
664: 0.027 \\
665: & 051227 & 112 $\pm$ 11 & 0.21 $\pm$
666: 0.20 & 340 $\pm$ 140 & 0.91 $\pm$
667: 0.31 \\
668: & 060115 & 126.7 $\pm$ 4.5 & 0.71 $\pm$
669: 0.14 & 810 $\pm$ 290 & 1.60 $\pm$
670: 0.31 \\
671: & 060124 & 559 $\pm$ 31 & 0.032 $\pm$
672: 0.023 & 871 $\pm$ 49 & 1.591 $\pm$
673: 0.087 \\
674: & 060210 & 113.0 $\pm$ 9.0 & 0.472 $\pm$
675: 0.093 & 500 $\pm$ 120 & 1.503 $\pm$
676: 0.049 \\
677: & 060211A & 202 $\pm$ 16 & 0.689 $\pm$
678: 0.058 & 351 $\pm$ 29 & 1.30 $\pm$
679: 0.20 \\
680: & 060218 & 1013 $\pm$ 13 & 0.424 $\pm$
681: 0.087 & 2122 $\pm$ 27 & 1.057 $\pm$
682: 0.072 \\
683: & 060413 & 132 $\pm$ 11 & 0.626 $\pm$
684: 0.099 & 278 $\pm$ 31 & 1.25 $\pm$
685: 0.21 \\
686: & 060510B & 291 $\pm$ 11 & 0.214 $\pm$
687: 0.060 & 417 $\pm$ 16 & 1.32 $\pm$
688: 0.18 \\
689: & 060522 & 160 $\pm$ 11 & 0.40 $\pm$
690: 0.24 & 390 $\pm$ 190 & 0.98 $\pm$
691: 0.16 \\
692: & 060526 & 285 $\pm$ 30 & 0.640 $\pm$
693: 0.037 & 429 $\pm$ 44 & 1.89 $\pm$
694: 0.13 \\
695: & 060607A & 97 $\pm$ 10 & 0.428 $\pm$
696: 0.073 & 149 $\pm$ 16 & 1.12 $\pm$
697: 0.11 \\
698: & 060607A & 229 $\pm$ 24 & 0.560 $\pm$
699: 0.061 & 353 $\pm$ 38 & 1.114 $\pm$
700: 0.088 \\
701: & 060614 & 104.3 $\pm$ 5.5 & 0.070 $\pm$
702: 0.042 & 451 $\pm$ 25 & 2.09 $\pm$
703: 0.13 \\
704: & 060707 & 213 $\pm$ 49 & 0.72 $\pm$
705: 0.15 & 970 $\pm$ 470 & 0.96 $\pm$
706: 0.14 \\
707: & 060714 & 116.5 $\pm$ 7.5 & 0.361 $\pm$
708: 0.083 & 223 $\pm$ 28 & 2.13 $\pm$
709: 0.17 \\
710: & 060729 & 132.09 $\pm$ 0.50 & 0.94 $\pm$
711: 0.20 & 286.8 $\pm$ 7.5 & 4.90 $\pm$
712: 0.47 \\
713: & 060814 & 83.3 $\pm$ 4.8 & 0.241 $\pm$
714: 0.074 & 377 $\pm$ 36 & 1.51 $\pm$
715: 0.12 \\
716: & 060904A & 76.2 $\pm$ 3.0 & 0.036 $\pm$
717: 0.079 & 226 $\pm$ 14 & 2.19 $\pm$
718: 0.21 \\
719: & 061007 & 106 $\pm$ 18 & 0.847 $\pm$
720: 0.032 & 252 $\pm$ 57 & 0.967 $\pm$
721: 0.029 \\
722: & 061110A & 80.4 $\pm$ 3.0 & 1.57 $\pm$
723: 0.14 & 210 $\pm$ 15 & 3.23 $\pm$
724: 0.21 \\
725: & 061121 & 71.7 $\pm$ 2.0 & -0.455 $\pm$
726: 0.062 & 155.8 $\pm$ 5.3 & 1.68 $\pm$
727: 0.31 \\
728: & 061222A & 108.8 $\pm$ 2.3 & 0.80 $\pm$
729: 0.20 & 177.8 $\pm$ 9.6 & 2.05 $\pm$
730: 0.35 \\
731: & 070110 & 111 $\pm$ 10 & 0.79 $\pm$
732: 0.11 & 245 $\pm$ 25 & 1.32 $\pm$
733: 0.26 \\
734: & 070129 & 299 $\pm$ 46 & 0.250 $\pm$
735: 0.030 & 1030 $\pm$ 150 & 2.36 $\pm$
736: 0.44 \\
737: & 070223 & 121.0 $\pm$ 3.0 & -0.22 $\pm$
738: 0.26 & 218 $\pm$ 22 & 1.17 $\pm$
739: 0.13 \\
740: & 070318 & 174 $\pm$ 22 & 0.19 $\pm$
741: 0.12 & 397 $\pm$ 89 & 0.924 $\pm$
742: 0.069 \\
743: & 070330 & 141 $\pm$ 54 & 0.43 $\pm$
744: 0.27 & 540 $\pm$ 230 & 0.76 $\pm$
745: 0.20 \\
746: & 070419B & 106 $\pm$ 17 & 0.569 $\pm$
747: 0.047 & 333 $\pm$ 25 & 1.144 $\pm$
748: 0.076 \\
749: & 070518 & 101 $\pm$ 22 & 1.147 $\pm$
750: 0.093 & 226 $\pm$ 31 & 1.99 $\pm$
751: 0.15 \\
752: & 070520B & 160 $\pm$ 35 & 1.196 $\pm$
753: 0.054 & 350 $\pm$ 42 & 1.92 $\pm$
754: 0.21 \\
755: & 070616 & 204 $\pm$ 65 & -0.060 $\pm$
756: 0.028 & 1100 $\pm$ 100 & 0.72 $\pm$
757: 0.15 \\
758: & 070621 & 122.0 $\pm$ 2.0 & 1.18 $\pm$
759: 0.21 & 235 $\pm$ 39 & 1.56 $\pm$
760: 0.16 \\
761: & 070704 & 252 $\pm$ 57 & 0.385 $\pm$
762: 0.084 & 511 $\pm$ 26 & 1.07 $\pm$
763: 0.25 \\
764: & 070714B & 74.0 $\pm$ 4.0 & -0.11 $\pm$
765: 0.22 & 396 $\pm$ 88 & 1.13 $\pm$
766: 0.33 \\
767: & 070721B & 265 $\pm$ 44 & 0.08 $\pm$
768: 0.16 & 930 $\pm$ 160 & 0.50 $\pm$
769: 0.17 \\
770: & 071031 & 118.0 $\pm$ 6.0 & 0.320 $\pm$
771: 0.078 & 252 $\pm$ 57 & 1.201 $\pm$
772: 0.035 \\
773: & 071112C & 624 $\pm$ 140 & 0.43 $\pm$
774: 0.11 & 1410 $\pm$ 210 & 0.65 $\pm$
775: 0.16 \\
776: & 080123 & 124 $\pm$ 13 & 0.57 $\pm$
777: 0.10 & 700 $\pm$ 450 & 1.35 $\pm$
778: 0.30 \\
779: & 080319B & 77 $\pm$ 10 & 0.671 $\pm$
780: 0.021 & 133 $\pm$ 18 & 0.761 $\pm$
781: 0.022 \\
782: & 080325 & 173 $\pm$ 12 & 0.682 $\pm$
783: 0.070 & 407 $\pm$ 62 & 2.188 $\pm$
784: 0.068 \\
785: & 080430 & 196 $\pm$ 54 & 0.56 $\pm$
786: 0.27 & 910 $\pm$ 270 & 0.91 $\pm$
787: 0.26 \\
788: & 080503 & 118 $\pm$ 18 & 0.240 $\pm$
789: 0.057 & 407 $\pm$ 62 & 1.55 $\pm$
790: 0.35 \\
791: & 080506 & 167 $\pm$ 18 & 0.224 $\pm$
792: 0.051 & 288 $\pm$ 35 & 1.03 $\pm$
793: 0.21 \\
794: & 080523 & 118 $\pm$ 18 & 1.287 $\pm$
795: 0.063 & 299 $\pm$ 45 & 2.15 $\pm$
796: 0.17 \\\hline
797: late & 050315 & 40000 $\pm$ 18000 & 0.99 $\pm$
798: 0.10 & 400000 $\pm$ 130000 & 1.31 $\pm$
799: 0.23 \\
800: & 050721 & 830 $\pm$ 97 & 0.55 $\pm$
801: 0.22 & 20400 $\pm$ 4100 & 1.16 $\pm$
802: 0.50 \\
803: & 050726 & 4850 $\pm$ 900 & 0.95 $\pm$
804: 0.11 & 19300 $\pm$ 4100 & 1.25 $\pm$
805: 0.30 \\
806: & 050730 & 4580 $\pm$ 570 & 0.434 $\pm$
807: 0.049 & 22600 $\pm$ 1300 & 0.714 $\pm$
808: 0.080 \\
809: & 050803 & 9800 $\pm$ 3300 & 0.879 $\pm$
810: 0.088 & 60000 $\pm$ 30000 & 1.29 $\pm$
811: 0.20 \\
812: & 050826 & 210 $\pm$ 95 & 0.63 $\pm$
813: 0.30 & 84000 $\pm$ 54000 & 1.64 $\pm$
814: 0.54 \\
815: & 050904 & 17800 $\pm$ 1300 & 0.730 $\pm$
816: 0.063 & 45000 $\pm$ 15000 & 1.28 $\pm$
817: 0.12 \\
818: & 051109A & 4070 $\pm$ 570 & 0.720 $\pm$
819: 0.093 & 56800 $\pm$ 6200 & 1.11 $\pm$
820: 0.18 \\
821: & 060105 & 4820 $\pm$ 180 & 0.92 $\pm$
822: 0.15 & 81000 $\pm$ 24000 & 1.44 $\pm$
823: 0.19 \\
824: & 060204B & 530 $\pm$ 140 & 1.03 $\pm$
825: 0.26 & 13800 $\pm$ 4100 & 1.63 $\pm$
826: 0.16 \\
827: & 060306 & 135 $\pm$ 38 & 1.00 $\pm$
828: 0.16 & 12500 $\pm$ 6500 & 1.39 $\pm$
829: 0.14 \\
830: & 060313 & 178 $\pm$ 65 & 0.42 $\pm$
831: 0.16 & 7100 $\pm$ 2900 & 1.01 $\pm$
832: 0.14 \\
833: & 060714 & 980 $\pm$ 220 & 0.78 $\pm$
834: 0.19 & 6750 $\pm$ 650 & 1.19 $\pm$
835: 0.21 \\
836: & 060807 & 4960 $\pm$ 480 & 1.09 $\pm$
837: 0.15 & 17400 $\pm$ 1300 & 1.39 $\pm$
838: 0.20 \\
839: & 070220 & 397 $\pm$ 87 & 0.23 $\pm$
840: 0.20 & 12100 $\pm$ 1300 & 0.80 $\pm$
841: 0.20 \\
842: & 070318 & 990 $\pm$ 220 & 0.55 $\pm$
843: 0.11 & 12480 $\pm$ 890 & 1.01 $\pm$
844: 0.27 \\
845: & 070419B & 5970 $\pm$ 180 & 0.50 $\pm$
846: 0.15 & 28100 $\pm$ 1200 & 0.82 $\pm$
847: 0.11 \\
848: & 070508 & 1600 $\pm$ 270 & 0.55 $\pm$
849: 0.12 & 57000 $\pm$ 23000 & 1.31 $\pm$
850: 0.19 \\
851: & 070721B & 5020 $\pm$ 770 & 0.43 $\pm$
852: 0.11 & 10610 $\pm$ 530 & 0.68 $\pm$
853: 0.27 \\
854: & 071020 & 630 $\pm$ 140 & 0.53 $\pm$
855: 0.11 & 17000 $\pm$ 1500 & 1.38 $\pm$
856: 0.69 \\
857: & 071025 & 174 $\pm$ 22 & 0.412 $\pm$
858: 0.049 & 13400 $\pm$ 2100 & 1.39 $\pm$
859: 0.30 \\
860: & 080207 & 5260 $\pm$ 530 & 1.69 $\pm$
861: 0.23 & 17500 $\pm$ 1400 & 2.03 $\pm$
862: 0.42 \\
863: & 080319C & 650 $\pm$ 190 & 0.535 $\pm$
864: 0.079 & 30190 $\pm$ 970 & 1.30 $\pm$
865: 0.31 \\
866: & 080328 & 5590 $\pm$ 200 & 0.57 $\pm$
867: 0.26 & 18200 $\pm$ 1200 & 1.33 $\pm$
868: 0.20 \\
869: & 080413B & 7000 $\pm$ 1000 & 0.79 $\pm$
870: 0.10 & 18800 $\pm$ 1300 & 1.18 $\pm$
871: 0.18 \\
872: & 080430 & 6490 $\pm$ 910 & 0.75 $\pm$
873: 0.16 & 26000 $\pm$ 4000 & 0.90 $\pm$
874: 0.15 \\
875:
876: \enddata
877: \end{deluxetable}
878:
879:
880:
881: \begin{deluxetable}{llllll}
882:
883: \tablewidth{0pt} \tablecaption{The maximum and minimum values of the
884: flux density ($f_{\nu, max}$ and $f_{\nu, min}$) and the
885: corresponding times ($t_{f,max}$ and $t_{f,min}$), detected during
886: the softening process.} \tablehead{ \colhead{type} & \colhead{burst}
887: & \colhead{$t_{f,max} (s)$} & \colhead{$f_{\nu, max}^a$} &
888: \colhead{$t_{f,min} (s)$} & \colhead{$f_{\nu, min}^a$} } \startdata
889:
890: early
891: & 050315 & 88.0 $\pm$ 1.5 & (8.6 $\pm$
892: 1.4)E-10 & 620 $\pm$ 77 & (12.8 $\pm$
893: 2.3)E-12\\
894: & 050421 & 130.4 $\pm$ 1.0 & (17.1 $\pm$
895: 3.4)E-10 & 285.0 $\pm$ 4.7 & (9.6 $\pm$
896: 4.5)E-11\\
897: & 050502B & 743.2 $\pm$ 1.0 & (66.6 $\pm$
898: 4.6)E-10& 1277 $\pm$ 26 & (29.5 $\pm$
899: 9.8)E-12\\
900: & 050714B & 162.4 $\pm$ 1.0 & (19.2 $\pm$
901: 2.6)E-08 & 273.2 $\pm$ 3.5 & (2.2 $\pm$
902: 1.5)E-09\\
903: & 050716 & 109.0 $\pm$ 1.0 & (25.1 $\pm$
904: 3.1)E-10& 457.0 $\pm$ 1.0 & (9.0 $\pm$
905: 4.5)E-11\\
906: & 050717 & 96.0 $\pm$ 1.0 & (32.2 $\pm$
907: 3.9)E-10& 202.0 $\pm$ 1.0 & (15.1 $\pm$
908: 9.2)E-11\\
909: & 050724 & 79.94 $\pm$ 0.25 & (18.6 $\pm$
910: 2.1)E-09 & 312.0 $\pm$ 1.1 & (7.6 $\pm$
911: 2.4)E-10\\
912: & 050726 & 382.3 $\pm$ 6.9 & (17.3 $\pm$
913: 2.9)E-11 & 617.8 $\pm$ 8.6 & (5.9 $\pm$
914: 1.4)E-11\\
915: & 050730 & 139.6 $\pm$ 1.0 & (20.5 $\pm$
916: 2.4)E-10 & 603.6 $\pm$ 1.0 & (22.4 $\pm$
917: 6.7)E-11\\
918: & 050814 & 167.09 $\pm$ 0.30 & (24.6 $\pm$
919: 4.1)E-10 & 363.7 $\pm$ 2.9 & (17.0 $\pm$
920: 3.2)E-11\\
921: & 050904 & 178.0 $\pm$ 1.0 & (34.6 $\pm$
922: 3.4)E-10 & 566.0 $\pm$ 1.0 & (14.8 $\pm$
923: 5.6)E-11\\
924: & 050922B & 755.6 $\pm$ 1.6 & (5.7 $\pm$
925: 1.0)E-09 & 1426 $\pm$ 12 & (19.1 $\pm$
926: 8.6)E-12\\
927: & 051117A & 134.0 $\pm$ 1.0 & (87.3 $\pm$
928: 5.7)E-10& 1257.7 $\pm$ 1.1 & (41.5 $\pm$
929: 9.3)E-11\\
930: & 051227 & 114.2 $\pm$ 1.0 & (11.7 $\pm$
931: 2.1)E-10 & 460 $\pm$ 10 & (10.5 $\pm$
932: 7.4)E-12\\
933: & 060115 & 124.45 $\pm$ 0.25 & (32.9 $\pm$
934: 6.2)E-10 & 741 $\pm$ 32 & (19.4 $\pm$
935: 4.6)E-12\\
936: & 060124 & 572.14 $\pm$ 0.25 & (84.1 $\pm$
937: 5.8)E-09 & 849.50 $\pm$ 0.70 & (10.8 $\pm$
938: 2.1)E-10\\
939: & 060210 & 106.8 $\pm$ 1.0 & (64.9 $\pm$
940: 5.6)E-10 & 302.8 $\pm$ 1.0 & (13.6 $\pm$
941: 6.8)E-11\\
942: & 060211A & 187.0 $\pm$ 1.0 & (47.1 $\pm$
943: 3.6)E-10 & 379.0 $\pm$ 1.0 & (10.9 $\pm$
944: 5.5)E-11\\
945: & 060218 & 1258.04 $\pm$ 0.25 & (18.0 $\pm$
946: 4.0)E-09 & 1975.54 $\pm$ 0.25 & (39.3 $\pm$
947: 9.3)E-10\\
948: & 060413 & 122.0 $\pm$ 1.0 & (12.5 $\pm$
949: 1.0)E-09 & 302.8 $\pm$ 1.8 & (7.7 $\pm$
950: 2.2)E-10\\
951: & 060510B & 305.53 $\pm$ 0.25 & (12.1 $\pm$
952: 1.5)E-09 & 442.59 $\pm$ 0.77 & (29.8 $\pm$
953: 9.4)E-11\\
954: & 060522 & 158.2 $\pm$ 1.0 & (5.3 $\pm$
955: 1.2)E-10 & 461.4 $\pm$ 7.2 & (10.3 $\pm$
956: 7.7)E-12\\
957: & 060526 & 255.78 $\pm$ 0.61 & (14.8 $\pm$
958: 1.2)E-09 & 463.4 $\pm$ 1.0 & (14.6 $\pm$
959: 5.5)E-11\\
960: & 060607A & 99.8 $\pm$ 1.0 & (84.9 $\pm$
961: 6.4)E-10 & 153.8 $\pm$ 1.0 & (50.5 $\pm$
962: 9.7)E-11\\
963: & 060607A & 267.8 $\pm$ 1.0 & (38.9 $\pm$
964: 2.8)E-10 & 387.8 $\pm$ 1.0 & (14.0 $\pm$
965: 4.9)E-11\\
966: & 060614 & 106.03 $\pm$ 0.25 & (75.2 $\pm$
967: 5.6)E-09 & 465.2 $\pm$ 3.6 & (43.6 $\pm$
968: 5.2)E-11\\
969: & 060707 & 188.3 $\pm$ 3.5 & (17.0 $\pm$
970: 3.3)E-11 & 1301 $\pm$ 28 & (4.1 $\pm$
971: 1.8)E-12\\
972: & 060714 & 139.6 $\pm$ 1.0 & (72.7 $\pm$
973: 4.5)E-10 & 223.6 $\pm$ 1.0 & (13.9 $\pm$
974: 7.3)E-11\\
975: & 060729 & 132.34 $\pm$ 0.25 & (11.8 $\pm$
976: 1.1)E-08 & 276.48 $\pm$ 0.51 & (21.9 $\pm$
977: 6.9)E-10\\
978: & 060814 & 79.84 $\pm$ 0.25 & (45.3 $\pm$
979: 4.6)E-09 & 365.8 $\pm$ 1.5 & (9.9 $\pm$
980: 1.6)E-10\\
981: & 060904A & 74.44 $\pm$ 0.25 & (37.1 $\pm$
982: 3.8)E-09 & 214.86 $\pm$ 0.68 & (4.1 $\pm$
983: 1.4)E-10\\
984: & 061007 & 91.4 $\pm$ 1.0 & (39.5 $\pm$
985: 1.7)E-09 & 306.96 $\pm$ 0.58 & (37.2 $\pm$
986: 4.8)E-10\\
987: & 061110A & 86.64 $\pm$ 0.25 & (9.9 $\pm$
988: 1.3)E-09& 198.37 $\pm$ 0.30 & (15.3 $\pm$
989: 5.1)E-10\\
990: & 061121 & 74.91 $\pm$ 0.25 & (19.2 $\pm$
991: 1.1)E-08 & 154.29 $\pm$ 0.75 & (12.5 $\pm$
992: 2.2)E-10\\
993: & 061222A & 109.82 $\pm$ 0.25 & (13.0 $\pm$
994: 1.9)E-09 & 175.7 $\pm$ 1.1 & (19.7 $\pm$
995: 3.6)E-10\\
996: & 070110 & 102.0 $\pm$ 1.0 & (13.8 $\pm$
997: 1.7)E-10 & 249.1 $\pm$ 4.2 & (9.9 $\pm$
998: 2.1)E-11\\
999: & 070129 & 365.0 $\pm$ 1.0 & (26.5 $\pm$
1000: 1.5)E-09 & 1054 $\pm$ 10 & (2.4 $\pm$
1001: 1.1)E-11\\
1002: & 070223 & 118.9 $\pm$ 1.0 & (40.3 $\pm$
1003: 4.9)E-10 & 231.0 $\pm$ 1.0 & (6.6 $\pm$
1004: 1.9)E-10\\
1005: & 070318 & 274.8 $\pm$ 1.0 & (17.3 $\pm$
1006: 2.2)E-10 & 482.8 $\pm$ 1.0 & (12.3 $\pm$
1007: 5.5)E-11\\
1008: & 070330 & 221.9 $\pm$ 2.9 & (29.5 $\pm$
1009: 6.1)E-11 & 616 $\pm$ 16 & (13.5 $\pm$
1010: 5.1)E-12\\
1011: & 070419B & 104.0 $\pm$ 1.0 & (15.8 $\pm$
1012: 1.0)E-09 & 352.0 $\pm$ 1.0 & (16.1 $\pm$
1013: 2.1)E-10\\
1014: & 070518 & 104.0 $\pm$ 1.0 & (12.9 $\pm$
1015: 1.6)E-10 & 256.0 $\pm$ 1.0 & (4.0 $\pm$
1016: 2.9)E-11\\
1017: & 070520B & 181.6 $\pm$ 1.0 & (52.7 $\pm$
1018: 3.9)E-10 & 353.6 $\pm$ 1.0 & (10.5 $\pm$
1019: 6.1)E-11\\
1020: & 070616 & 486.0 $\pm$ 1.0 & (20.0 $\pm$
1021: 1.2)E-09 & 1174.8 $\pm$ 5.4 & (15.7 $\pm$
1022: 4.0)E-11\\
1023: & 070621 & 124.8 $\pm$ 1.0 & (61.9 $\pm$
1024: 5.1)E-10 & 272.8 $\pm$ 1.0 & (11.7 $\pm$
1025: 8.3)E-11\\
1026: & 070704 & 315.2 $\pm$ 1.0 & (106.5 $\pm$
1027: 7.9)E-10 & 535.2 $\pm$ 1.0 & (2.7 $\pm$
1028: 1.6)E-10\\
1029: & 070714B & 73.5 $\pm$ 1.0 & (24.3 $\pm$
1030: 3.2)E-10 & 466.1 $\pm$ 6.0 & (1.5 $\pm$
1031: 1.0)E-11\\
1032: & 070721B & 312.0 $\pm$ 1.0 & (25.8 $\pm$
1033: 2.7)E-10 & 935.6 $\pm$ 5.4 & (3.1 $\pm$
1034: 1.9)E-11\\
1035: & 071031 & 121.4 $\pm$ 1.0 & (58.2 $\pm$
1036: 4.1)E-10 & 305.4 $\pm$ 1.0 & (5.5 $\pm$
1037: 1.0)E-10\\
1038: & 071112C & 569.1 $\pm$ 2.4 & (27.9 $\pm$
1039: 6.8)E-11 & 1413.8 $\pm$ 7.2 & (2.2 $\pm$
1040: 1.1)E-11\\
1041: & 080123 & 122.0 $\pm$ 1.0 & (17.7 $\pm$
1042: 2.1)E-10 & 921 $\pm$ 77 & (6.9 $\pm$
1043: 4.9)E-13\\
1044: & 080319B & 68.2 $\pm$ 1.0 & (135.9 $\pm$
1045: 3.4)E-09 & 150.09 $\pm$ 0.91 & (45.7 $\pm$
1046: 2.0)E-09\\
1047: & 080325 & 222.2 $\pm$ 1.0 & (108.3 $\pm$
1048: 7.4)E-10 & 460.2 $\pm$ 1.0 & (5.0 $\pm$
1049: 1.6)E-10\\
1050: & 080430 & 172.3 $\pm$ 6.0 & (21.1 $\pm$
1051: 5.3)E-11 & 603 $\pm$ 15 & (2.4 $\pm$
1052: 1.1)E-11\\
1053: & 080503 & 105.2 $\pm$ 1.0 & (48.4 $\pm$
1054: 3.9)E-10 & 409.3 $\pm$ 5.0 & (3.1 $\pm$
1055: 1.4)E-11\\
1056: & 080506 & 156.4 $\pm$ 1.0 & (57.4 $\pm$
1057: 4.4)E-10 & 320.4 $\pm$ 1.0 & (5.3 $\pm$
1058: 3.8)E-11\\
1059: & 080523 & 104.8 $\pm$ 1.0 & (19.9 $\pm$
1060: 2.0)E-10 & 306.8 $\pm$ 1.0 & (10.2 $\pm$
1061: 4.6)E-11\\
1062: \hline
1063: late
1064: & 050315 & 23810 $\pm$ 270 & (9.7 $\pm$
1065: 1.1)E-12 & 432400 $\pm$ 9400 & (40.8 $\pm$
1066: 7.4)E-14\\
1067: & 050721 & 777.0 $\pm$ 5.0 & (14.1 $\pm$
1068: 3.4)E-11 & 23530 $\pm$ 190 & (8.8 $\pm$
1069: 6.2)E-13\\
1070: & 050726 & 4001 $\pm$ 53 & (27.3 $\pm$
1071: 6.6)E-12 & 23010 $\pm$ 380 & (10.7 $\pm$
1072: 2.9)E-13\\
1073: & 050730 & 4370.1 $\pm$ 3.7 & (5.2 $\pm$
1074: 1.0)E-10 & 23912 $\pm$ 28 & (7.2 $\pm$
1075: 2.9)E-12\\
1076: & 050803 & 10502 $\pm$ 53 & (5.7 $\pm$
1077: 4.0)E-11 & 83100 $\pm$ 5500 & (9.7 $\pm$
1078: 1.4)E-13\\
1079: & 050826 & 169.7 $\pm$ 5.0 & (20.8 $\pm$
1080: 5.1)E-11 & 111000 $\pm$ 12000 & (10.4 $\pm$
1081: 6.8)E-14\\
1082: & 050904 & 18582.8 $\pm$ 8.9 & (8.6 $\pm$
1083: 1.8)E-11 & 47020 $\pm$ 120 & (4.9 $\pm$
1084: 3.4)E-13\\
1085: & 051109A & 4185 $\pm$ 32 & (61.0 $\pm$
1086: 8.3)E-12 & 57940 $\pm$ 670 & (27.2 $\pm$
1087: 5.3)E-13\\
1088: & 060105 & 5571.2 $\pm$ 3.4 & (24.4 $\pm$
1089: 5.7)E-11 & 87600 $\pm$ 300 & (12.6 $\pm$
1090: 4.8)E-13\\
1091: & 060204B & 411 $\pm$ 17 & (11.6 $\pm$
1092: 1.9)E-11 & 16720 $\pm$ 110 & (2.1 $\pm$
1093: 1.1)E-12\\
1094: & 060306 & 102.4 $\pm$ 1.7 & (10.3 $\pm$
1095: 1.5)E-10 & 16970 $\pm$ 220 & (40.3 $\pm$
1096: 9.5)E-13\\
1097: & 060313 & 149.4 $\pm$ 2.4 & (27.2 $\pm$
1098: 6.1)E-11 & 4500 $\pm$ 17 & (3.0 $\pm$
1099: 2.1)E-12\\
1100: & 060714 & 880 $\pm$ 10 & (8.4 $\pm$
1101: 1.8)E-11 & 6860 $\pm$ 28 & (3.7 $\pm$
1102: 2.1)E-12\\
1103: & 060807 & 4743 $\pm$ 53 & (36.6 $\pm$
1104: 4.9)E-12 & 17850 $\pm$ 250 & (62.4 $\pm$
1105: 9.5)E-13\\
1106: & 070220 & 542.5 $\pm$ 3.5 & (41.0 $\pm$
1107: 8.0)E-11 & 12786 $\pm$ 53 & (4.1 $\pm$
1108: 2.5)E-12\\
1109: & 070318 & 789.1 $\pm$ 4.1 & (15.8 $\pm$
1110: 3.8)E-11 & 7584 $\pm$ 39 & (2.8 $\pm$
1111: 2.0)E-12\\
1112: & 070419B & 5807.5 $\pm$ 3.8 & (17.3 $\pm$
1113: 4.3)E-11 & 29245 $\pm$ 19 & (6.1 $\pm$
1114: 4.6)E-12\\
1115: & 070508 & 1583.3 $\pm$ 2.4 & (8.0 $\pm$
1116: 2.1)E-10 & 53632 $\pm$ 269 & (11.7 $\pm$
1117: 8.3)E-13\\
1118: & 070721B & 4438 $\pm$ 15 & (40.4 $\pm$
1119: 9.5)E-12 & 10501 $\pm$ 38 & (5.1 $\pm$
1120: 2.4)E-12\\
1121: & 071020 & 566.6 $\pm$ 2.9 & (31.4 $\pm$
1122: 6.8)E-11 & 17000 $\pm$ 1500 & (11.4 $\pm$
1123: 3.6)E-13\\
1124: & 071025 & 159.0 $\pm$ 1.0 & (100.8 $\pm$
1125: 7.5)E-10 & 10866 $\pm$ 39 & (2.3 $\pm$
1126: 1.3)E-12\\
1127: & 080207 & 4751 $\pm$ 19 & (17.3 $\pm$
1128: 4.1)E-11 & 18230 $\pm$ 100 & (9.8 $\pm$
1129: 4.6)E-12\\
1130: & 080319C & 505.1 $\pm$ 7.7 & (44.9 $\pm$
1131: 6.0)E-11 & 30511 $\pm$ 35 & (1.8 $\pm$
1132: 1.3)E-12\\
1133: & 080328 & 5517 $\pm$ 14 & (15.0 $\pm$
1134: 3.4)E-11 & 18298 $\pm$ 43 & (4.9 $\pm$
1135: 3.0)E-12\\
1136: & 080413B & 8541 $\pm$ 53 & (3.5 $\pm$
1137: 1.4)E-11 & 12367 $\pm$ 53 & (2.1 $\pm$
1138: 1.5)E-12\\
1139: & 080430 & 5965 $\pm$ 35 & (14.2 $\pm$
1140: 3.5)E-12 & 29943 $\pm$ 57 & (2.2 $\pm$
1141: 1.1)E-12\\
1142:
1143: \enddata
1144: \tablenotetext{a}{in units of $erg \cdot cm^{-2} s^{-1}$.}
1145: \end{deluxetable}
1146:
1147: \end{document}
1148: