1: %\documentstyle[12pt,aasms4]{article}
2: \documentclass[12pt,preprint]{aastex}
3:
4: %\documentclass[preprint2]{emulateapj}
5: %\documentclass[10pt,preprint2]{aastex}
6: % ---------------------------------------------------
7:
8: \def\simg{\hspace{1ex} ^{>} \hspace{-2.5mm}_{\sim} \hspace{1ex}}
9: \def\siml{\hspace{1ex} ^{<} \hspace{-2.5mm}_{\sim} \hspace{1ex}}
10: \def\reference{\par\noindent\hangindent=1cm\hangafter=1}
11: \newcommand{\ins}[2]{\makebox [#1]{#2}}
12: \newcommand{\inc}[1]{\ {\mbox{#1}}\ }
13: \newcommand{\mt}{{{m_1}\over {m_1 + m_2}}}
14: \newcommand{\mb}{{m_2}\over {m_1 + m_2}}
15: \newcommand{\mtb}{{m_1 m_2}\over {m_1 + m_2}}
16: \newcommand{\eq}{\begin{equation}}
17: \newcommand{\ee}{\end{equation}}
18: \newcommand{\bra}[1]{\left( #1 \right)}
19: \newcommand{\brasq}[1]{\left[ #1 \right]}
20: \newcommand{\pdrv}[2]{{{\partial #1}\over {\partial #2}}}
21: \newcommand{\pddrv}[2]{{{\partial^2 #1}\over {\partial #2^2}}}
22: \newcommand{\pderiv}[1]{{{\partial}\over {\partial #1}}}
23: \newcommand{\pdderiv}[1]{{{\partial^2}\over {\partial #1 ^2}}}
24: \newcommand{\drv}[2]{{{d #1}\over {d #2}}}
25: \newcommand{\deriv}[1]{{d\over {d #1}}}
26: \newcommand{\half}{{1\over 2}}
27: \newcommand{\thalf}{{3\over 2}}
28: \newcommand{\bxi}{\mbox{\boldmath$ \xi $}}
29: \newcommand{\bgk}[1]{\mbox{\boldmath$ #1 $}}
30: \def\t0{\theta_{\circ}}
31: \def\muo{\mu_{\circ}}
32: \def\sd{\partial}
33: \def\be{\begin{equation}}
34: \def\en{\end{equation}}
35: \def\bv{\bf v}
36: \def\bvo{\bf v_{\circ}}
37: \def\ro{r_{\circ}}
38: \def\rhoo{\rho_{\circ}}
39: \def\etal{et al.\ }
40: \def\msun{M_{\sun}}
41: \def\rsun{R_{\sun}}
42: \def\lsun{L_{\sun}}
43: \def\msunyr{M_{\sun} yr^{-1}}
44: \def\kms{\rm \, km \, s^{-1}}
45: \def\md{\dot{M}}
46: \def\Md{\dot{M}}
47: \def\tper{\tau_{\rm per}}
48: \def\tcore{\tau_{\rm core}}
49: \def\tgrow{\tau_{\rm grow}}
50: \def\tnu{\tau_{\rm \nu}}
51: \def\tmig{\tau_{\rm ma}}
52: \def\ttmig{\tau_{\Delta}}
53: \def\texc{\tau_{\rm exc}}
54: \def\x{b_{\rm R}}
55: \def\Delres{\Delta_{\rm res}}
56: \def\gapp{\ \lower 3pt\hbox{${\buildrel > \over \sim}$}\ }
57: \def\lapp{\ \lower 3pt\hbox{${\buildrel < \over \sim}$}\ }
58: %-------------------------------------------
59:
60: \begin{document}
61:
62:
63: \title{Multi-color Shallow Decay and Chromatic Breaks in the
64: GRB\,050319 Optical Afterglow}
65:
66: \author{
67: K.Y. \textsc{Huang}\altaffilmark{1},
68: Y. \textsc{Urata}\altaffilmark{2,3},
69: P.H. \textsc{Kuo}\altaffilmark{1},
70: W.H. \textsc{Ip}\altaffilmark{1},
71: K. \textsc{Ioka}\altaffilmark{4},
72: T. \textsc{Aoki}\altaffilmark{5},
73: C.W. \textsc{Chen}\altaffilmark{1},
74: W.P. \textsc{Chen}\altaffilmark{1},
75: M. \textsc{Isogai}\altaffilmark{5},
76: H.C. \textsc{Lin}\altaffilmark{1},
77: K. \textsc{Makishima}\altaffilmark{3,6},
78: H. \textsc{Mito}\altaffilmark{5},
79: T. \textsc{Miyata}\altaffilmark{5},
80: Y. \textsc{Nakada}\altaffilmark{5},
81: S. \textsc{Nishiura}\altaffilmark{7},
82: K. \textsc{Onda}\altaffilmark{2},
83: Y. \textsc{Qiu}\altaffilmark{8},
84: T. \textsc{Soyano}\altaffilmark{5},
85: T. \textsc{Tamagawa}\altaffilmark{3},
86: K. \textsc{Tarusawa}\altaffilmark{5},
87: M. \textsc{Tashiro}\altaffilmark{2},
88: and
89: T. \textsc{Yoshioka}\altaffilmark{9}
90: }
91:
92: \altaffiltext{1}{Institute of Astronomy, National Central University, Chung-Li 32054, Taiwan, Republic of China. (Huang email: d919003@astro.ncu.edu.tw)}
93: \altaffiltext{2}{Department of Physics, Saitama University, Shimo-Okubo, Sakura, Saitama, 338-8570, Japan.}
94: \altaffiltext{3}{RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan.}
95: \altaffiltext{4}{Department of Physics, Kyoto University, Kyoto 606-8602, Japan}
96: \altaffiltext{5}{Kiso Observatory, Institute of Astronomy, The University of Tokyo, Mitake-mura, Kiso-gun, Nagano 397-0101, Japan.}
97: \altaffiltext{6}{Department of Physics, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan.}
98: \altaffiltext{7}{Department of Astronomy and Earth Sciences, Tokyo Gakugei University, Koganei, Tokyo 184-8501, Japan.}
99: \altaffiltext{8}{National Astronomical Observatories, Chinese Academy of Sciences, Beijing 100012, China, PR.}
100: \altaffiltext{9}{Department of Physics, Nagoya University, Furo-cho, Chikusa, Nagoya, Aichi 464-8602, Japan.}
101:
102: %\authoremail
103:
104: \begin{abstract}
105:
106: Multi-wavelength $B$, $V$, $R$, $I$ observations of the optical
107: afterglow of GRB\,050319 were performed by the 1.05-m telescope at
108: Kiso Observatory and the 1.0-m telescope at Lulin Observatory from
109: 1.31 hours to 9.92 hours after the burst. Our $R$ band lightcurves,
110: combined with other published data, can be described by the smooth
111: broken power-law function, with $\alpha_1$ = $-$0.84 $\pm$0.02 to
112: $\alpha_2$ = $-$0.48$\pm$0.03, 0.04 days after the GRB. The optical
113: lightcurves are characterized by shallow decays--- as was also
114: observed in the X-rays--- which may have a similar origin, related to
115: energy injection. However, our observations indicate that there is
116: still a puzzle concerning the chromatic breaks in the $R$ band
117: lightcurve (at 0.04~days) and the X-ray lightcurve (at 0.004~days)
118: that remains to be solved.
119: \end{abstract}
120:
121: \keywords{gamma-ray: burst : afterglow}
122:
123: \section{Introduction}
124:
125: The GRB afterglow as perceived in the X-ray, optical and radio
126: wavelengths is now understood to be the result of the collision
127: between relativistic ejecta from the gamma-ray bursts and the
128: interstellar medium (ISM). A comparison of afterglow lightcurves
129: obtained at different wavelengths gives important information about
130: the surrounding ISM environment and the interaction processes. Such
131: analyses can also provide essential input for theoretical
132: models. Recently, the pace of this type of activity has quickened
133: significantly, stimulated by the capabilities of the quick response
134: and accurate localization of the GRB by the {\it Swift} Satellite
135: (Gehrels et al. 2004). This has meant that the number of GRB optical
136: afterglow detections in the first several hours by ground-based
137: telescopes has recently increased significantly. It is interesting to
138: note that the observations by {\it Swift} of the early X-ray emissions
139: from a number of GRBs reveal a canonical behavior. The X-ray
140: lightcurves can be divided into three distinct power law segments
141: (Nousek et al. 2006). Some X-ray and optical observations show that
142: the evolution of both lightcurves changes at the same time (Blustin et
143: al. 2006; Rykoff et al. 2006), however chromatic breaks were also
144: found in some cases (Fan \& Piran 2006; Panaitescu et al. 2006). The
145: nature of the afterglow early breaks in the lightcurves is thus
146: uncertain. A detailed comparison of changes in the evolution of the
147: optical, radio and X-ray lightcurves should therefore be very
148: interesting. This kind of physical study demands both a
149: well-coordinated observational program and careful data analysis. We
150: use GRB\,050319 which has comprehensive observational coverage in both
151: the X-ray and optical wavelengths may be used as just such an example.
152:
153: GRB\,050319 was detected by the Burst Alert Telescope (BAT)
154: instrument onboard the {\it Swift} satellite on March 19, 2005 at
155: 09:31:18.44 UT (Krimm et al. 2005). However, a re-analysis of the BAT
156: data showed two flares, which indicated that GRB\,050319 had already
157: started 137 sec before the trigger. The 15--350~keV fluence for the
158: entire burst duration of $T_{90}= 149.6\pm0.7$ sec has been estimated
159: to be 1.6 $\times 10^{-6}~{\rm erg\, cm}^{-2}$ (Cusumano et
160: al. 2006). The X-ray emission of GRB\,050319 after the burst was
161: monitored by the XRT from 225 sec to 28 days\footnote[1] { The burst
162: time in the article is 09:29:01.44 UT, 137~s before the BAT
163: trigger}. Two breaks in the emission curves were found (Cusumano et
164: al. 2006). The initial sharp decline can be described by a power-law
165: with an index of $\alpha_{1}= -5.53\pm 0.67$ to be followed by
166: $\alpha_{2}= -0.54\pm 0.04$ after 0.004 days since the burst. The
167: unusually flat decline in the second part might have been caused by
168: continuous energy injection (Cusumano et al. 2006). At about 0.313
169: days after the burst, the power-law index changed to $\alpha_{3}=
170: -1.14\pm 0.2$ which can be readily explained as a jet break or a
171: reduction in the energy injection (Cusumano et al. 2006; Zhang et
172: al. 2006) .
173:
174: The early optical afterglow emission at 230~s was observed by the UVOT
175: telescope on {\it Swift} (Mason et al. 2006) and by two ground-based
176: robotic telescopes, {\it ROTSE-III} (Quimby et al., 2006) and {\it
177: RAPTOR} (Wo\'zniak et al. 2005). The best single power-law fit of
178: unfiltered data from {\it ROTSE-III} and {\it RAPTOR} indicates that
179: $\alpha = -0.854\pm 0.014$. A number of optical observatories have
180: joined the follow-up observations (see Yoshioka et al. 2005; Torii et
181: al. 2005; Sharapov et al. 2005a,b; George et al. 2006; Misra et
182: al. 2005; Kiziloglu et al. 2005; Greco et al. 2005). The spectral
183: measurements of the afterglow by the Nordic Optical Telescope (NOT)
184: indicate that was redshift $z=3.24$ of this event (Fynbo et al. 2005).
185:
186: \section{Observations and Analysis}
187:
188: After receiving the GRB alert message from {\it Swift} and the
189: afterglow position was reported by Rykoff et al. (2005), the
190: Target-of-Opportunity procedures of the East-Asia GRB Follow-up
191: Observation Network (EAFON\footnote[2]{EAFON web-page:
192: http://cosmic.riken.jp/grb/eafon/}, Urata et al. 2005) were
193: immediately carried out. A series of multi-band follow-up observations
194: were successfully performed by the 1.05-m Schmidt telescope of the
195: Kiso Observatory in Japan and the Lulin One-meter Telescope (LOT) in
196: Taiwan. Photometric $B$ and $R$ images were obtained at the Kiso site
197: with a 2k $\times$ 2k CCD camera (Urata et al. 2005) between 0.055 and
198: 0.326 days after the burst. A number of parallel $B$, $V$, $R$ and $I$
199: images were obtained by LOT with a PI1300 CCD camera (Kinoshita et
200: al. 2005) from 0.080 to 0.413 days after the burst.
201:
202: The standard routine including bias subtraction and dark
203: subtraction; flat-field corrections were employed with the appropriate
204: calibration data needed to process the data using IRAF. The afterglow
205: can be clearly seen in the images. The signal-to-noise ratio was
206: improved by combing the LOT $B$ band data with median filtering. The
207: DAOPHOT package (Stetson 1989) was then used to perform point-spread
208: function (PSF) fitting for the GRB images. Four field stars were used
209: to create a PSF model which was applied to the optical afterglow of
210: each GRB image. For absolute photometric calibration, we used
211: calibrated data of the GRB field obtained by the USNOFS 1.0-m
212: telescope (Henden 2005). The photometric error and the systematic
213: calibration error were included in the magnitude error
214: estimation\footnote[3]{The errors in this article were quoted for
215: 68$\%$ (1-$\sigma$) confidence level}.
216: \section{Results}
217:
218: \subsection{Lightcurve}
219: Figure\,1 shows the multi-band lightcurves of the GRB\,050319
220: afterglow. Besides our $B$, $V$, $R$ and $I$ band data (Table\,1) we
221: also included the $R$ band measurements from {\it ROTSE-III} (Quimby
222: et al. 2006), {\it RAPTOR} (Wo\'zniak et al. 2005) and several GCN
223: reports (Greco et al. 2005; Kiziloglu et al. 2005; Misra et al. 2005;
224: Sharapov et al. 2005a,b). In addition, we also made use of several $B$
225: and $V$ band measurements taken by the {\it Swift} UVOT (Mason et
226: al. 2006). The GCN $R$ band points were re-calibrated using the
227: GRB\,050319 field stars reported by Henden (2005) so they could be
228: plotted on the same magnitude scale. The magnitude differences between
229: photometric field stars in Henden (2005) and in USNO-A2.0, USNO-B1.0
230: stars are $+0.18$ and $-0.22$ mag, respectively. We re-measured the
231: reference stars from Greco et al. (2005) from the LOT $R$ band images
232: and obtained the average magnitudes and rms errors.
233:
234: After fitting the $B$, $V$, $R$ and $I$ band lightcurves to a single
235: power-law expression $F \propto t^{\alpha}$, where $\alpha$ is the
236: index and $t$ is the time after the burst, we get $\alpha = -0.56 \pm
237: 0.06$ ($\chi^{2}/\nu$= 2.90 for $\nu=19$) for the $B$ band, $\alpha =
238: -0.65 \pm 0.03$ ($\chi^{2}/\nu$= 2.60 for $\nu=27$) for the $V$ band,
239: $\alpha = -0.59 \pm 0.01$ ($\chi^{2}/\nu$= 5.3 for $\nu=97$) for the
240: $R$ band, and $\alpha = -0.52 \pm 0.15$ ($\chi^{2}/\nu$= 7.7 for
241: $\nu=9$) for the $I$ band. This single power-law fitting indicates
242: that these lightcurves, obtained with different filters have similar
243: power-law decay even though the reduced chi-square values are
244: relatively large.
245:
246: Since the data sets of the $V$ and $R$ measurements are more complete,
247: it is possible with following expression to attempt the fitting of the
248: corresponding lightcurves with a smoothly broken power-law function:
249:
250: \begin{equation}
251: F(\nu,t) = {2^{1/k}F_{\nu,b} \over [(t/t_{\rm b})^{-k\alpha_1}+(t/t_{\rm
252: b})^{-k\alpha_2}]^{(1/k)}},
253: \end{equation}
254: where $t_{\rm b}$ is the break time, $\alpha_1$ and $\alpha_2$ are the
255: power-law indices before and after $t_{\rm b}$, $F_{\nu,b}$ is flux at
256: break $t_{\rm b}$, and $k$ is a smoothness factor. For the $V$ band,
257: we obtain $\alpha_1 = -0.87 \pm 0.21$, $\alpha_2 = -0.49 \pm 0.05$,
258: $t_{\rm b}$ = 0.042 $\pm$ 0.058 days and $k= -30$ ($\chi^{2}/\nu$=
259: 1.48 for $\nu=24$). For the $R$ band, we obtain $\alpha_1 = -0.84 \pm
260: 0.02$, $\alpha_2 = -0.48 \pm 0.03$, $t_{\rm b} = 0.046 \pm 0.008$
261: days, and $k = -21$ ($\chi^{2}/\nu$= 2.24 for $\nu=90$). This result
262: implies a mild break in both the $V$ and the $R$ band lightcurves at
263: around 0.04 days after the occurrence of the GRB.
264:
265: Taking $t_{\rm b}$ = 0.04 days, we fit the data in the $B$ and $I$
266: bands to a respective power-law before and after the break. In this
267: manner, we find $\alpha_1$ = $-$0.79 $\pm$ 0.09 ($\chi^2/\nu=1.09$ for
268: $\nu=7)$; $\alpha_2$ = $-$0.36 $\pm$ 0.05 ($\chi^2/\nu=1.23$ for
269: $\nu=9)$ for the $B$ band and $\alpha_2$ = $-$0.52 $\pm$ 0.15
270: ($\chi^2/\nu=7.7$ for $\nu=9)$ for the $I$ band. The best-fit
271: parameters for the $B$, $V$, $R$, and $I$ bands are summarized in
272: Table 2. Our results show not only the clear presence of mild breaks
273: in the $V$, $R$ band lightcurves but a flattening trend after the
274: break. Furthermore, our $R$ band slope before the break ($\alpha \sim
275: -0.84$) is in agreement with the corresponding value derived by Quimby
276: et al. (2006) for the interval between 0.0019 and 0.05 days after the
277: burst.
278:
279: \subsection{Color and spectral flux distribution}
280:
281: Our multi-wavelength observations indicate that median colors between
282: 0.07 and 0.35 days are $V-R = 0.45 \pm 0.11$, $R-I = 0.46 \pm 0.10 $,
283: and $B-V = 0.84 \pm 0.14 $. These values have been corrected for
284: foreground reddening of E($B-V) = $ 0.011 mag (Schlegel et
285: al. 1998). The $V-R$ and $R-I$ colors so derived are consistent with
286: those of the typical long GRBs (Simon et al. 2001), but the $B-V$
287: color is slightly redder than those of the typical long GRBs ($B-V =
288: 0.47 \pm 0.17$). The larger $B-V$ value may imply a certain absorption
289: effect because the redshift of GRB\,050319 was determined to be 3.24
290: (Fynbo et al. 2005).
291:
292: The $B$, $V$, $R$ and $I$ magnitudes have been further converted to
293: fluxes using the effective wavelengths and normalizations of Fukugita
294: et al. (1995). The effect of the Galactic interstellar extinction has
295: been corrected. Figure 2 shows two samples of spectral energy
296: distribution obtained by LOT at 0.13 and 0.21 days after the
297: occurrence of GRB\,050319. A drop in the $B$ band flux at about
298: $4380\AA$ can be clearly seen. We subsequently fitted the flux
299: distribution of $V$, $R$ and $I$ bands with a power-law function $
300: F(\nu,t) \propto \nu^{\beta}$; here $F(\nu,t)$ is the flux at
301: frequency $\nu$ with a certain $t$ and $\beta$ is the spectral
302: index. We find that $\beta = -1.08 \pm 0.05$ ($\chi^{2}/\nu$= 0.05 for
303: $\nu=1$) at 0.13 days and that $\beta = -1.08 \pm 0.32$
304: ($\chi^{2}/\nu$ = 2.3 for $\nu=1$) at 0.21 days. Our result ($\beta =
305: -1.08$ with a rms error 0.23) is consistent with the X-ray fitting
306: value ($\beta = -0.69 \pm 0.06$) in a 3-$\sigma$ level.
307:
308: With a redshift of 3.24, the Ly${\alpha}$ absorption feature would
309: shift into the $B$ bandpass, causing reduction of the afterglow flux
310: in the $B$ band. To correct for this absorption effect, we used the
311: formulation derived by Yoshii et al. (1994), in which the optical
312: depth is a function of the observed wavelength and source
313: redshift. With the computed optical depth in the $B$ band, and a
314: spectral slope of $\beta = -1.08$, we found the expected $B$ band
315: magnitude after Ly${\alpha}$ absorption to be $21.33 \pm 0.05$ at 0.13
316: days. This value compares very well with our observed value of $B$ =
317: 21.26 $\pm$ 0.17 at 0.13 days after correction for Galactic
318: extinction. The drop at the B band is hence fully produced by the
319: Ly${\alpha}$ absorption and no spectral breaks should have taken place
320: during our observation.
321:
322: \section{Discussion and Summary}
323:
324: It is important to note that {\it Swift} found two breaks at 0.004 and
325: 0.313~days after the burst in the X-ray afterglow observations
326: (Cusumano et al. 2006), but we only found a single break in our $V$
327: and $R$ lightcurves (see Figure 1). In the following since there are
328: more data points available for the $R$ band data we will focus on
329: this. It is useful to remember that $\alpha_1$ = $-0.84$ and
330: $\alpha_2=-0.48$ at the break time of $t_{\rm b}$ = 0.04~days.
331:
332: \subsection{Before the optical break ( t $<$ 0.04 days)}
333:
334: The slope $\alpha_1$ (=$-$0.84) is consistent with the typical range
335: of $\alpha = -0.62$ to $-2.3$ for many well observed GRBs. According
336: to the standard afterglow model relating the power-law index
337: ($\alpha$) to the power-law index ($\it{p}$) of an electron spectrum
338: (Zhang \& M\'esz\'aros, 2004; Dai \& Cheng, 2001), the corresponding
339: value for $\alpha_1=-0.84$ is $\it{p}$ = 2.1, which is in agreement
340: with the constant-density ISM model with slow cooling in which $\it{p}
341: >$ 2 for $\nu_{\rm m}< \nu_{\rm opt} < \nu_{\rm c}$ ($\nu_{\rm m}$ is
342: the typical frequency; $\nu_{\rm opt}$ is the optical frequency;
343: $\nu_{\rm c}$ is the cooling frequency). In light of the XRT
344: observations, the first break was likely caused by the transition from
345: the tail end of the low energy prompt emission to the afterglow phase
346: (Zhang et al. 2006). However, it is important to note that the X-ray
347: break at 0.004 days (where the steeper slope becomes shallow) is not
348: accompanied by an $R$ band break. At the same time, the power-law
349: decay slope in the X-ray ($\sim -$ 5.53) and the $R$ band ($\sim -$
350: 0.84) are quite different. This is an indication not only that the
351: behaviors of the X-ray afterglow and optical afterglow of the
352: GRB\,050319 event are different, but also suggested that the afterglow
353: phase already dominated the optical bands when the optical emission
354: was first detected.
355:
356:
357: \subsection{Shallow Decay}
358:
359: The power law index becomes shallow after the break ($t_{\rm b}$ =
360: 0.04 days). Neither the jet (Rhoads 1999) nor the break frequencies
361: across the optical wavelength (Sari et al., 1998) suitably explain the
362: break they see in the GRB\,050319 lightcurves. As discussed before,
363: the X-ray lightcurve between the two breaks 0.004 and 0.313~days after
364: the burst is also characterized by shallow decay. Zhang et al. (2006)
365: suggested that such behavior is related to continuous energy injection
366: into the ISM. For a long-lasting central engine, the energy injection
367: rate is $\dot{E}(t) \propto t^{-q}$ and with $q < 1$ (Zhang \&
368: M\'esz\'aros 2001). For slow cooling in the ISM, the temporal index
369: can be expressed as : $\alpha = [(2p-6)+(p+3)q] / 4 = [(q-1) +
370: {(2+q)\beta] / 2},$\, when\, $\nu_{\rm m} < \nu < \nu_{\rm
371: c}$. Using this formulation, Zhang et al. (2006) obtained $q$ = 0.6
372: and $p$ = 2.4 from the X-ray observations. With $\alpha$ = $-$0.48 and
373: $\beta$ = $-$1.08 from the $R$ band observations, we find that $q$ =
374: 0.72 and $p$ = 2.12. The results not only indicate that the electron
375: spectrum power law index is the same before and after the break, they
376: also compare well with the results of Zhang et al. (2006). These
377: results indicate that the shallow decays evidenced by both X-ray and
378: optical afterglows could be of a similar origin, related to a
379: continuous energy injection mechanism.
380:
381: According to the energy injection model, would also expect an X-ray
382: break at the time of the optical flattening break, because the onset
383: of the energy injection should also alter the X-ray temporal
384: index. However, such an X-ray break is not observed, which suggests
385: that some modifications to the injection model may be needed. As
386: mentioned in $\S$ 4.1, the X-ray break at 0.004 days was not
387: accompanied by a break in the $R$ lightcurve. Although a chromatic
388: break in the X-ray was found at 0.004 days and in the optical region
389: at 0.04 days, the lightcurves at both wavelengths indeed showed
390: shallow decay after the breaks, which can be explained by the energy
391: injection model. However, it is difficult for energy injection from
392: 0.004 days to 0.04 days to affect only high energies. This difficulty
393: indicates that energy injection is an imperfect mechanism for
394: explaining the shallow optical or X-ray phase associated with the
395: GRB\,050319 event.
396:
397: Several models have recently been proposed to explain the shallow
398: decay effect. Using the multiple-subjet model (Nakamura, 2000), Toma
399: et al. (2006) invoked the superposition of afterglows from many
400: off-axis subjets. Eichler \& Granot (2006) favored a combination of
401: the tail of prompt emission model with the afterglow emissions
402: observed from a viewing angle outside the edge of the jet. These
403: arguments hence suggest that the multiple-subjet model and the
404: patchy-shell model (Kumar \& Piran, 2000) might provide a theoretical
405: basis for explaining the observed shallow decays in the X-ray and
406: optical lightcurves. It is interesting to note that in order to
407: sustain the shallow decay process these models all require high
408: gamma-ray efficiency (75-90$\%$); additional mechanisms such as prior
409: activity (Ioka et al. 2005) and time-dependent shock generation (Fan
410: \& Piran, 2006) have been also proposed. Comprehensive
411: multi-wavelength observations, such as those reported here, provide
412: important keys to improve these models.
413:
414: Finally, the second break in the X-ray emissions ($\sim$ 0.313~days
415: after the burst), has been interpreted as being due to an unusual flat
416: jet break (Cusumano et al. 2006), Zhang et al. (2006) however provided
417: an alternate explanation, a sudden cessation of the energy
418: injection. In both interpretations, the corresponding break should
419: appear in both the X-ray and optical lightcurves. This effect cannot
420: be clearly identified in our measurements until 0.413 days after the
421: burst. The lack of data for the subsequent time interval could lead to
422: uncertainty in the power-law fitting. We thus cannot fully exclude the
423: existence of a second break in the optical lightcurves. However,
424: Panaitescu et al. (2006) have studied several afterglows. They found
425: the shallow power-law decay evidenced by the X-ray emissions to
426: steepen about 0.04$-$0.17 days after the burst, although there was no
427: accompanying break found in the optical range. They suggest that such
428: chromatic X-ray breaks may be common. The chromatic breaks (e.g. the
429: shallow X-ray phase becomes steeper, with no accompanying optical
430: break) may be caused by differences in the X-ray and optical outflow
431: (Panaitescu et al. 2006), or by changes in the typical electron energy
432: parameters (the so-called microphysical parameters) at the end of
433: energy injection (Panaitescu 2006).
434:
435: In summary, our analysis of the optical multi-wavelength observations
436: of GRB\,050319 made at Kiso and Lulin compared with the X-ray
437: observations from {\it Swift} found the following major results:
438:
439: \begin{itemize}
440:
441: \item The $B$, $V$, $R$, and $I$ band lightcurves displayed unusual
442: shallow decays.
443:
444: \item The $R$ lightcurve can be described by a smooth broken power-law
445: function. $\alpha_1 \sim -0.84$ becomes shallow ($\alpha_2 \sim
446: -0.48$) 0.04 days after the occurrence of the GRB.
447:
448: \item The shallow decay observed in the X-ray and optical lightcurves
449: may have a similar origin related to energy injection. However, our
450: observations indicate that there is still a major puzzle remaining
451: concerning the chromatic breaks in the $R$ band lightcurve (at
452: 0.04~days) and the X-ray lightcurve (at 0.004~days).
453:
454: \item Our calculations revealed that the drop in spectral energy
455: distribution was fully caused by a shift in the Ly${\alpha}$
456: absorption to the $B$ bandpass at $z$ = 3.24.
457:
458: \end{itemize}
459:
460: \acknowledgments
461:
462: We thank the referee for his/her valuable advice. This work is
463: supported by NSC\,95-2752-M-008-001-PAE, NSC\,95-2112-M-008-021, the
464: Japan Society for the Promotion of Science (JSPS) Grant-in-Aid for
465: Young Scientists (B) No.\,18740147 and JSPS Research Fellowships for
466: Young Scientists (Y.U.).
467:
468: \clearpage
469:
470:
471: \begin{thebibliography}{}
472:
473: \bibitem[Blustin (2006)]{blustin} Blustin, E. et al. 2006 \apj, 637, 901
474: \bibitem[Cusumano (2006)]{cusumano} Cusumano, G. et al. 2006, \apj, 639, 316
475: \bibitem[Dai \& Cheng (2001)]{dai2} Dai, Z. G., \& Cheng, K. S. 2001, \apj, 558, 109
476: \bibitem[Eichler et al.(2006)]{eichler} Eichler, D., \& Granot, J. 2006, \apj, 641, L5
477: \bibitem[Fan (2006)]{fan} Fan, Y., \& Piran, T. 2006, \mnras, 369, 197
478: \bibitem[Fynbo (2005)]{fynbo} Fynbo, J. P. U. et al. 2005, GCN Circ. 3136
479: \bibitem[Fukugita 1995]{fukugita} Fukugita, M., Shimasaku, K., \& Ichikawa, T. 1995, PASP, 107, 945
480: \bibitem[Gehrels et al. (2004)]{gehrels} Gehrels, N. et al. 2004 \apj, 611, 1005
481: \bibitem[George et al. (2006)]{George} George, K., Banerjee, D. P. K., Chandrasekhar, T., \& Ashok, N. M. 2006 \apj, 640, L13
482: \bibitem[Greco et al. (2005)]{greco} Greco, G., Bartolini, C., Guarnieri, A., Piccioni,A., Ferrero, P. \& Bruni, I. 2005, GCN Circ. 3142
483: \bibitem[Henden (2005)]{Henden}Henden, A. 2005, GCN Circ. 3454
484: \bibitem[Ioka (2005)]{Ioka} Ioka, K., Toma, K., Yamazaki, R., \& Nakamura, T. 2005, A\&A in press (astro-ph/0511749)
485: \bibitem[Kinoshita et al.(2005)]{kinoshita} Kinoshita, D., et al. 2005, ChJAA, 5, 315
486: \bibitem[Kiziloglu et al. (2005)]{kiziloglu et al.} Kiziloglu, U. et al. 2005, GCN Circ. 3139
487: \bibitem[Krimm et al. (2005)]{Krimm05a} Krimm, H. et al. 2005, GCN Circ. 3119
488: \bibitem[Kumar 2000]{kumar} Kumar, P. \& Piran, T. 2000, \apj, 535, 152
489: \bibitem[Mason (2006)]{mason} Mason, K. O. et al. 2006, \apj, 639, 311
490: \bibitem[Misra et al. (2005)]{misra} Misra, K., Kamble, A. P. \& Pandey, S.B. 2005, GCN Circ. 3130
491: \bibitem[Nakamura 2000]{nakamura} Nakamura, T. 2000, \apj, 534, L159
492: \bibitem[Nousek (2006)]{nousek} Nousek, J. A. et al. 2006, \apj, 642, 389
493: \bibitem[Panaitescu et al. (2006)]{panaitescu} Panaitescu, A., M\'esz\'aros, P., Burrows, D., Nousek, J., O'Brien, P. \& Willingale, R. 2006, MNRAS, 369, 2059
494: \bibitem[Panaitescu (2006)]{panaitescu2} Panaitescu, A., 2006, preprint (astro-ph/0607396)
495: \bibitem[Quimby et al. (2006)]{quimby} Quimby, R. M. et al. 2006, \apj, 640, 402
496: \bibitem[Rhoads et al. (1999)]{quimby} Rhoads, J. E. 1999, \apj, 525, 737
497:
498: \bibitem[Rykoff et al.(2005)]{rykoff} Rykoff, E., Schaefer, B., \& Quimby, R. 2005, GCN Cir. 3116
499: \bibitem[Rykoff et al.(2006)]{rykoff06} Rykoff, E. et al. 2006, \apj, 638, L5
500: \bibitem[Sari (2005)]{sari} Sari, R., Piran, T., \& Narayan, R. 1998, \apj, 497, L17
501: \bibitem[Schlegel, Finkbeiner, \& Davis (1998)]{schlegel} Schlegel, D. J., Finkbeiner, D. P., \& Davis, M. 1998, ApJ, 500, 525
502: \bibitem[Sharapov et al. (2005a)]{sharapov05a} Sharapov, D. et al. 2005a, GCN Circ. 3124
503: \bibitem[Sharapov et al. (2005b)]{sharapov05b} Sharapov, D. et al. 2005b, GCN Circ. 3140
504: \bibitem[Simon (2001)]{simon} Simon, V., Hudec, R., Pizzichini, G., \& Masetti, N. 2001, A\&A, 377, 450
505: \bibitem[Stetson (1987)]{Stetson87} Stetson, P. B. 1987, PASP, 99, 191
506: \bibitem[Toma et al.(2006)]{toma} Toma, K., Ioka, K., Yamazaki, R. \& Nakamura, T. 2006, \apj 640, L139
507: \bibitem[Torii et al. (2005)]{torii} Torii, K. 2005, GCN Circ. 3121
508: \bibitem[Urata et al. (2005)]{urata} Urata, Y. et al. 2005, Nuovo Cimento, 28, 775
509: \bibitem[Wo\'zniak et al.(2005)]{wozniak} Wo\'zniak, P. R., Vestrand, W. T., Wren, J. A., White, R. R., Evans, S. M., \& Casperson, D. 2005, ApJ, 627, L13
510: \bibitem[Yoshii(1994)]{yoshii} Yoshii, Y. \& Peterson, B. A. 1994, \apj, 436, 551
511: \bibitem[Yoshioka et al. (2005)]{yoshioka} Yoshioka, T. et al. 2005, GCN Circ. 3120
512: \bibitem[Zhang \& M\'esz\'aros (2001) ]{zhang3} Zhang, B., \& M\'esz\'aros, P. 2001, \apj, 552, L35
513: \bibitem[Zhang \& M\'esz\'aros (2004) ]{zhang1} Zhang, B., \& M\'esz\'aros, P. 2004, Int. J. Mod. Phys. A., 19, 2385
514: \bibitem[Zhang (2006) ]{zhang2} Zhang, B. et al. 2006, \apj, 642, 354
515: \end{thebibliography}
516:
517: \clearpage
518:
519: \begin{table}
520: \begin{center}
521: \caption{GRB\,050319 Optical Afterglow Photometry\tablenotemark{a}}
522: \begin{tabular}{cccc}
523: \tableline \tableline
524: Days after GRB\tablenotemark{b} & Filter & Magnitude\tablenotemark{c} & Site \\
525: \tableline
526: 0.05443 & R & 19.01$\pm$0.08 & Kiso\\
527: 0.06023 & B & 20.42$\pm$0.13 & Kiso\\
528: 0.08037 & V & 19.75$\pm$0.07 & Lulin \\
529: 0.10439 & I & 19.09$\pm$0.07 & Lulin\\
530: 0.27821 & B & 20.92$\pm$0.12 & Kiso\\
531: 0.33424 & I & 19.91$\pm$0.09 & Lulin\\
532: 0.40097 & V & 20.68$\pm$0.12 & Lulin\\
533: 0.41346 & R & 20.20$\pm$0.07 & Lulin\\
534: \tableline
535: \tablenotetext{a}{Table 1 is published in its entirety in the electronic edition of the $\it{Astrophysical\,Journal}$.}
536: \tablenotetext{b}{The burst time is 2005 March 19, UT 09:29:01.44}
537: \tablenotetext{c}{The magnitudes are not corrected for Galactic extinction.}
538: \end{tabular}
539: \end{center}
540: \end{table}
541:
542: \clearpage
543:
544: \begin{table}
545: \begin{center}
546: \caption{Fitting results of the GRB\,050319 lightcurves : The $B$ and
547: $I$ band data were fitted by a respective power-law model [$F \propto
548: t^{\alpha}$] before and after the break ($t_b =0.04\,{\rm days}$). On
549: the other hand, the $V$ and $R$ data were fitted by a smoothly broken
550: power-law [equation (1)].}
551: \begin{tabular}{cccc}
552: \tableline \tableline
553: Filter & $\alpha_1$ & $\alpha_2$ & $t_b ($\rm day$)$\\
554: \tableline
555: B & -0.79$\pm$0.09 & -0.36$\pm$0.05 & -\\
556: V & -0.87$\pm$0.21 & -0.49$\pm$0.05 & 0.042$\pm$0.058\\
557: R & -0.84$\pm$0.02 & -0.48$\pm$0.03 & 0.046$\pm$0.008\\
558: I & - & -0.52$\pm$0.15 & -\\
559: \tableline
560:
561: \end{tabular}
562: \end{center}
563: \end{table}
564:
565: \clearpage
566:
567: \begin{figure}
568: \begin{center}
569: \includegraphics[width=0.8\textwidth]{f1.ps}
570: \caption{Optical lightcurves of GRB\,050319 : The solid lines present
571: the best fit by the single power-law model [$F \propto t^{\alpha}$]
572: for the $B$ band ($\alpha = -0.56 \pm 0.06$) and the $I$ band ($\alpha
573: = -0.52 \pm 0.15$). The dashed line indicates the best fit by the
574: smooth broken power-law [equation (1)] with the $V$ band ($\alpha_1 =
575: -0.87 \pm 0.21$, $\alpha_2 = -0.49 \pm 0.05$, $t_{\rm b}$ = 0.042
576: $\pm$ 0.058 days) and the $R$ band ($\alpha_1 = -0.84 \pm 0.02$,
577: $\alpha_2 = -0.48 \pm 0.03$, $t_{\rm b}$ = 0.046 $\pm$ 0.008
578: days). These observations are based on our EAFON data and Quimby et
579: al. (2006), Wo\'zniak et al. (2005), Mason et al. (2005), Greco et
580: al. (2005), Kiziloglu et al. (2005), Misra et al. (2005), and Sharapov
581: et al. (2005a,b). The dotted lines represent the break times of X-ray
582: afterglows at 0.004 and 0.31 days after the burst. The two breaks were
583: not found in X-ray observations, but a mild break seems to exist in
584: the $V$ and $R$ band lightcurves. }
585: \end{center}
586: \end{figure}
587:
588: \clearpage
589:
590: \begin{figure}
591: \begin{center}
592: \includegraphics[width=0.8\textwidth]{f2.ps}
593: \caption{Spectral energy distribution of GRB\,050319 between the $V$,
594: $R$ and $I$ bands at 0.13 and 0.21 days after the burst (corrected for
595: Galactic extinction). The solid and dashed lines indicate the best fit
596: by the power-law model [$F(\nu) \propto \nu^{\beta}$], where $\beta =
597: -1.08 \pm 0.05$ for 0.13 days and $\beta = -1.08 \pm 0.32$ for 0.21
598: days, respectively.}
599: \end{center}
600: \end{figure}
601:
602: \end{document}
603:
604:
605:
606:
607:
608:
609:
610:
611:
612: