1: %%
2: %% GRB 990712 ApJL paper
3: %%
4:
5: \documentclass[11pt,aasms4]{aastex}
6: %\usepackage{emulateapj5}
7:
8: \usepackage[dvips]{epsfig}
9: %\usepackage{psfig,natbib}
10:
11: %\citestyle{aa}
12:
13: %\documentstyle[12pt,aasms4]{article}
14: %\documentstyle[11pt,myaaspp4,flushrt,psfig]{article}
15: %\documentstyle[aas2pp4]{article}
16:
17: %\documentstyle[11pt,eqsecnum,aaspp4]{article}
18:
19: % Authors are permitted to use the fonts provided by the American Mathematical
20: % Society, if they are available to them on their local system. These fonts
21: % are not part of the AASTeX macro package or the regular TeX distribution.
22:
23: %\documentstyle[12pt,amssym,aasms4]{article}
24:
25: % Here's some slug-line data. The receipt and acceptance dates will be
26: % filled in by the editorial staff with the appropriate dates. Rules will
27: % appear on the title page of the manuscript until these are uncommented
28: % out by the editorial staff.
29:
30: %\received{4 August 1988}
31: %\accepted{23 September 1988}
32: %\journalid{337}{15 January 1989}
33: %\articleid{11}{14}
34:
35: %\slugcomment{}
36:
37: % Authors may supply running head information, if they wish to do so, although
38: % this may be modified by the editorial offices. The left head contains a
39: % list of authors, usually three allowed---otherwise use et al. The right
40: % head is a modified title of up to roughly 44 characters. Running heads
41: % are not printed.
42:
43: %\shorttitle{The Polarization Variability of GRB 990712}
44: %\shortauthors{Bj\"ornsson \& Lindfors}
45:
46: %\doublespace
47:
48: \begin{document}
49:
50: \title{The Polarization Variability in the Optical Afterglow of GRB~990712}
51:
52: \author{Gunnlaugur~Bj\"ornsson
53: and Elina~J.\ Lindfors\\
54: Science Institute, University of Iceland\\
55: Dunhaga~3, IS--107 Reykjavik, Iceland\\
56: e-mail:gulli@raunvis.hi.is}
57:
58: \begin{abstract}
59: In a recent paper, Rol and colleagues present evidence for a variable
60: polarization in the optical afterglow following the gamma-ray burst
61: GRB~990712. The variation is highly significant, but the position angle
62: appears time independent. Contrary to their conclusion, we point out that
63: this can in fact be explained with existing afterglow models, namely that
64: of a laterally expanding jet.
65: \end{abstract}
66:
67: % The different journals have different requirements for keywords. The
68: % keywords.apj file, found on aas.org in the pubs/aastex-misc directory,
69: % contains a list of keywords used with the ApJ and Letters. These are
70: % usually assigned by the editor, but authors may include them in their
71: % manuscripts if they wish.
72:
73: \keywords{gamma rays: bursts --- radiation mechanisms: synchrotron --- polarization}
74:
75: \section{Introduction}
76: \label{sec:intro}
77:
78: It is generally accepted that the optical emission from gamma-ray burst
79: afterglows is synchrotron radiation from relativistic electrons (e.g.\
80: M\'esz\'aros \& Rees 1997).
81: Models of optical afterglows based on synchrotron emission, and either
82: spherical or collimated outflow geometry have successfully been applied to
83: a number of sources (e.g.\ Galama et~al.\ 1998; Bloom et~al.\ 1998;
84: Holland et~al.\ 2000). As synchrotron radiation under favorable
85: conditions can be up to 70\% polarized, polarization measurements of
86: optical afterglows have recently been added to the tool box of
87: afterglow researchers.
88:
89: The first attempt, by Hjorth et~al.\ (1999), resulted in an upper limit
90: of 2.3\% for the polarization of GRB~990123 about 18.3 h after the burst.
91: The polarization level of GRB~990510 was successfully measured by
92: Covino et~al.\ (1999) about 18.5 h after the burst, and Wijers et~al.\ (1999)
93: about 2 h later. These latter measurements were obtained using the same
94: instruments on the same telescope and the polarization remained constant
95: at 1.7\% during the 2 hour interval. Wijers et~al.\ (1999) obtained an
96: additional measurement at burst age of 43.3 h, but the polarization level
97: at that time was marginally detectable at similar level, mainly due to the
98: faintness of the source and worse observing conditions.
99:
100: In a recent preprint, Rol et~al.\ (2000), present polarization measurements
101: of GRB~990712 at three different burst ages, 10.6, 16.7 and 34.7 h after the
102: gamma ray event. The polarization level varied between the three measurements
103: from $2.9\%\pm 0.4$\% to $1.2\%\pm 0.4$\% and $2.2\%\pm 0.7$\%, respectively.
104: An interesting part of the result is that the position angle does {\em not}
105: seem to vary over the 24 h period from the first to the last data point.
106: Rol et~al.\ (2000) conclude, based on the constant position angle they find,
107: that none of the existing models can successfully explain their result.
108: The purpose of this {\em Letter} is to point out that it is in fact possible
109: to obtain varying degree of polarization {\em and} a constant position angle
110: in beamed models.
111:
112: \section{Variable Polarization From a Collimated Outflow}
113: \label{sec:obs}
114:
115: Several models have been put forward to explain how a polarized emission
116: may arise in an optical afterglow, despite the fact that the magnetic field
117: generated is expected to be highly tangled with no preferred direction and
118: therefore no net polarization. Examples include the spherically symmetric model
119: of Gruzinov \& Waxman (1999), and the polarization scintillation model of
120: Medvedev \& Loeb (1999). Recently, Sari (1999; hereafter referred to as S99)
121: and Ghisellini \& Lazzati (1999; hereafter GL99), independently and
122: simultaneously, showed that a non-zero and variable polarization can arise
123: from an almost totally tangled magnetic field that has some degree of
124: alignment, if the emission arises in a collimated outflow, and the observers line
125: of sight is located off the outflow axis but within the collimated beam.
126: This polarization variability is essentially a geometrical effect, the most
127: important points being the following (we will assume here that all geometries
128: are conical and we refer the reader to Fig.\ 2 in GL99 and Figs.\ 2 and 3 in S99):
129: Initially, when the expansion is highly relativistic, the observer only receives
130: radiation from within the relativistic cone of angular size $1/\Gamma$, where
131: $\Gamma$ is the bulk Lorentz factor. This cone is centered on, and symmetric with
132: respect to, the line of sight and therefore no polarization is observed. As the
133: expansion slows down, the edge of the relativistic cone reaches the edge of the
134: collimated beam and thereafter looks asymmetric to the observer. A net polarization
135: arises, that reaches a maximum as the relativistic cone expands and looks more and
136: more asymmetric, and drops to zero again when the emitting areas contributing to
137: the polarization in two different directions (``vertical and horizontal'')
138: become equal. The polarization then rises again with increasing asymmetry
139: between the areas emitting the two possible polarization directions, but with
140: the position angle rotated by $90^\circ$. The polarization finally drops to
141: zero again when $\Gamma \rightarrow 1$ (GL99), or exhibits a third
142: maximum if the jet is spreading (S99).
143:
144: In Figure 1a we show a typical evolution of the polarization for a conical
145: beam of fixed opening angle, $\theta_c=5^\circ$, with the observers line of
146: sight making an angle $\theta_0=f\theta_c$, ($f<1)$, with the cone symmetry
147: axis. We used the approach of GL99 to construct the figure and therefore only
148: exhibit two maxima. The evolution is shown as a function of the inverse bulk
149: Lorentz factor for two different values of $f$. The Lorentz factor can be
150: converted to time, using the relation $\Gamma=\Gamma_0 (t/t_0)^{-3/8}$.
151: For this figure we have used $\Gamma_0=100$ for the initial value of the
152: Lorentz factor and $t_0=50$ s. Note that lowering $f$ for a fixed $\theta_c$,
153: brings the observers line of sight closer to the symmetry axis and therefore
154: decreases the net polarization. It also shifts the first maximum and the
155: minimum to later times (lower $\Gamma$), while the second maximum occurs
156: almost at the same time ($\Gamma\approx 5$). Note also, that the first maximum
157: occurs typically less than an hour after the burst, the minimum less than 10 h
158: after the burst and the second maximum from 1-2 days after the burst.
159: The polarization level is modestly affected by the radiation spectral index.
160: We assume a power law spectral distribution, and take the spectral index
161: to be $\beta=0.6$ as e.g.\ observed for GRB~990712 (Sahu et~al.\ 2000).
162:
163: A crucial effect that is not discussed in detail by GL99 and for which S99
164: considers only one particular example in his toy model, is the {\em evolution}
165: of $f$, the ratio of the angle the line of sight makes with the jet axis to
166: that of the collimated beam. The jet axis is most likely defined by the angular
167: momentum of the burst progenitor system, and is presumably fixed in space.
168: The angle between the jet axis and the line of sight should therefore be
169: constant, unless the jet is precessing for which there is no evidence.
170: If the jet is expanding laterally, the ratio
171: $f=\theta_0/\theta_c$, {\em decreases} with time. Generating a sequence of
172: polarization curves as in Fig.\ 1a, varying (increasing) only the jet opening
173: angle, shows a decreasing magnitude of both maxima and a shift of the first
174: maximum and the minimum to the right (towards lower $\Gamma$ or later times).
175: The evolution of the second maximum is particularly interesting as it takes
176: place entirely under the polarization curve defined by the initial value of $f$,
177: with the maximum occurring at almost constant value of $\Gamma$
178: (see also S99 and GL99).
179:
180: We show an example in Figure 1b, with the data points of GRB~990712 superimposed.
181: As the first data point is obtained about 11 h after the burst, we assume that
182: the polarization at that time has already evolved into the region of second
183: maximum and therefore that the position angle has already changed by 90$^\circ$.
184: With $\Gamma_0=100$, and $t_0=50$~s, we find that an opening angle of
185: $\theta_c=5.1^\circ\pm 0.1^\circ$, with $f=0.9$, fits the first point. We then
186: let $\theta_c$ increase to $6.0^\circ\pm 0.2^\circ$, over the next 6 h, giving
187: $f=0.77$, and finally a modest increase to $6.2^\circ\pm 0.5^\circ$ fits the
188: last point 18 h later. The last two data points are also consistent with being
189: on the same polarization light curve. In that case we would be observing a
190: widening of the collimation angle by $1^\circ$ over a 6 h period between
191: the first and the second point, and approximately a constant opening angle
192: thereafter, or a slower rate of lateral expansion that may be due to density
193: irregularities in the local environment. We emphasize that the
194: above is obtained by taking ``snapshots'' of the evolving polarization light
195: curve, where only the jet opening angle has been changed between each shot.
196: The ``error estimates'' on the opening angle are determined by searching for
197: values of $\theta_c$ that give polarization within the error of the measured
198: polarization points at the appropriate time. It is interesting that a modest
199: variation in the jet opening angle (about 20\%), can easily change the
200: polarization by a factor of two to three. An important consequence of the
201: variable polarization being due to temporal evolution of the second maximum
202: is a constant position angle, naturally explaining the observations of the
203: GRB~990712 afterglow.
204:
205: The above analysis is a simple extension of the GL99 model, and complements the
206: approach of S99 that assumed that the jet opening angle evolved as $1/\Gamma$,
207: once $\Gamma$ had decreased below the inverse of the initial jet opening angle.
208: The initial polarization evolution, i.e.\ the first maximum and minimum, is
209: therefore similar in both approaches, differences arising when the jet opening
210: angle starts spreading, but the polarization light curve at that time is already
211: in the region of second maximum. Despite differences in details of GL99 and S99,
212: the location of the the second maximum in both cases occurs on similar time scales,
213: about 1-2 days after the burst. We have shown here that a modest variation in the
214: jet opening angle is sufficient to explain the polarization measurements of
215: Rol et~al.\ (2000). Numerical simulations using more realistic models are needed
216: to follow the detailed temporal evolution of the polarization light curve,
217: in particular the evolution of the second maximum.
218:
219: The optical light curve of GRB~990712 decayed as a power law with an index of
220: $\alpha\approx-1.0$ (Sahu et~al.\ 2000, Hjorth et~al.\ 2000). This is similar
221: to the decay index of GRB~990123 and somewhat steeper than that of GRB~990510,
222: before the break in their light curves. In the latter two cases the light curve
223: steepened about 1-2 days after the burst (e.g.\ Kulkarni et~al.\ 1999;
224: Harrison et~al.\ 1999; Stanek et~al.\ 1999; Israel et~al.\ 1999; Holland et~al.\ 2000).
225: A model of a collimated outflow predicts a steepening of the light curve when
226: $1/\Gamma \approx\theta_c$, the sharpness of the break depending on the rate
227: of lateral expansion (e.g.\ Rhoads 1999), and the break in the light curves of
228: GRB~990123 and GRB~990510 indicates a jet opening angle of about $5^\circ$.
229: Interpreting the polarization data for GRB~990712 with a spreading jet, therefore
230: implies that the optical light curve should show a break after about 1-2 days.
231: The modest increase in the opening angle implied by the polarization measurements
232: requires the break to have been rather sharply defined in time.
233: No such break has been reported (Sahu et~al.\ 2000; Hjorth et~al.\ 2000), perhaps
234: because the host galaxy is bright and was already affecting the magnitudes of the
235: optical transient 10 h after the burst (second data point on the optical light curve).
236:
237: The polarization measurement of GRB~990123 at 18.3 h and of GRB~990510 at
238: 18.5 h, 20.7 h and 43.3 h, are likely to have
239: been taken during the second maximum. It is crucial that polarization
240: measurements be attempted as soon as possible after the discovery of an
241: optical afterglow. In particular, to demonstrate the 90$^\circ$ change in the
242: position angle, a positive detection well before a burst age of 10 h is needed
243: and would nicely confirm the applicability of collimated models. A well sampled
244: polarization light curve is a powerful tool in exploring the properties of
245: burst afterglows and their surroundings and can potentially provide more
246: detailed information than the optical light curve alone.
247:
248: \acknowledgments
249:
250: \acknowledgements
251: This work was supported by the Icelandic Research Council
252: and the University of Iceland Research Fund. We thank the
253: anonymous referee for useful suggestions.
254:
255: \begin{thebibliography}{}
256: \bibitem[2000]{}
257: Bloom, J.\ S., et~al.\ 1998, ApJ, 508, L21
258:
259: \bibitem[2000]{}
260: Covino, S., et~al.\ 1999, A\&A, 348, L1
261:
262: \bibitem[2000]{}
263: Galama, T.\ J., Wijers, R.\ A.\ M.\ J., Bremer, M., Groot, P.\ J.,
264: Strom, R.\ G., Kouveliotou, C.\ \& van Paradijs, J.\ 1998,
265: ApJ, 500, L97
266:
267: \bibitem[2000]{}
268: Ghisellini, G.\ \& Lazzati, D.\ 1999, MNRAS, 309, L7 (GL99)
269:
270: \bibitem[2000]{}
271: Gruzinov, A.\ \& Waxman, E.\ 1999, ApJ, 511, 852
272:
273: \bibitem[2000]{}
274: Harrison F.\ A., et~al.\ 1999, ApJ, 523, L121
275:
276: \bibitem[2000]{}
277: Hjorth, J., et~al.\ 1999, Science, 283, 2073
278:
279: \bibitem[2000]{}
280: Hjorth, J., Holland, S., Courbin, F., Dar, A., Olsen, L.\ F.\ \&
281: Scodeggio, M.\ 2000, ApJ, 534, L147
282:
283: \bibitem[2000]{}
284: Holland, S., Bj\"ornsson, G., Hjorth, J.\ \& Thomsen, B.\ 2000, A\&A, in press
285:
286: \bibitem[2000]{}
287: Israel, G., L., et~al.\ 1999, A\&A, 348, L5
288:
289: \bibitem[2000]{}
290: Kulkarni, S.\ R., et~al.\ 1999, Nature, 398, 389
291:
292: \bibitem[2000]{}
293: Medvedev, M.\ V.\ \& Loeb, A.\ 1999, ApJ, 526, 697
294:
295: \bibitem[2000]{}
296: M\'esz\'aros, P.\ \& Rees, M.\ J.\ 1997, ApJ, 476, 232
297:
298: \bibitem[2000]{}
299: Rhoads, J., 1999, ApJ, 575, 737
300:
301: \bibitem[2000]{}
302: Rol, E., et~al.\ ApJ, in press
303: ({\tt astro-ph/0007015})
304:
305: \bibitem[2000]{}
306: Sahu, K.\ C., et~al.\ 2000, ApJ, 540, 74
307:
308: \bibitem[2000]{}
309: Sari, R.\ 1999, ApJ, 524, L43 (S99)
310:
311: \bibitem[2000]{}
312: Stanek K.\ Z., Garnavich, P.\ M., Kaluzny, J., Pych, W.\ \& Thompson, I.\ 1999, ApJ, 522, L39
313:
314: \bibitem[2000]{}
315: Wijers, R.\ A.\ M.\ J., et~al.\ 1999, ApJ, 523, L33
316: \end{thebibliography}
317:
318: %\clearpage
319:
320: %\figcaption[]{}
321:
322: \begin{figure}[t]
323: \epsscale{0.3}
324: \centerline{\epsfig{figure=fig1.ps,height=6cm,clip=}\epsfig{figure=fig2.ps,height=6cm,clip=}}
325: %\plottwo{fig1.ps}{fig2.ps}
326: \caption{a) Typical evolution of the polarization light curve for constant
327: $\theta_c=5^\circ$, and $f=0.9$ (solid curve) and $f=0.67$ (dotted curve).
328: Note that the first maximum and the minimum shifts to the right as $f$ is
329: decreased, while the second maximum only decreases in amplitude.
330: b) Polarized light curves of a collimated outflow with a varying opening
331: angle and a fixed angle between the jet symmetry axis and the line of
332: sight. If $\theta_c$ is increased, $f$ decreases. The solid curve has
333: $\theta_c=5.1^\circ$ and $f=0.9$, the dashed has $\theta_c=6.0^\circ$ and
334: $f=0.77$, and the dotted has $\theta_c=6.2^\circ$ and $f=0.74$. As in panel
335: a), the first maximum and the minimum move to the right while the second
336: maximum shifts to the right by a factor of 2 in time but does not cross the
337: initial curve (solid). The polarization values for GRB~990712
338: are superimposed. We have assumed that the maximum polarization from
339: synchrotron radiation is $P_0=70\%$, in both a) and b).
340: }
341: \end{figure}
342:
343: \end{document}
344: