0808.4115/ms.tex
1: \documentclass[12pt,preprint]{aastex}
2: \usepackage{epsfig,emulateapj5,apjfonts}
3: %\usepackage{epsfig}
4: %%%%%%%%%%%%%%%%%%%%% macro definitions %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
5: \newcommand       \Angstrom     {\,{\rm \AA}}
6: \newcommand       \AU           {\,{\rm AU}}
7: \newcommand       \cm           {\,{\rm cm}}
8: \newcommand       \mm           {\,{\rm mm}}
9: \newcommand       \erg          {\,{\rm erg}}
10: \newcommand       \eV           {\,{\rm eV}}
11: \newcommand       \g        {\,{\rm g}}
12: \newcommand       \K            {\,{\rm K}}
13: \newcommand       \pc       {\,{\rm pc}}
14: \newcommand       \kpc      {\,{\rm kpc}}
15: \newcommand       \s        {\,{\rm s}}
16: \newcommand       \sr           {\,{\rm sr}}
17: \newcommand       \yr       {\,{\rm yr}}
18: \newcommand       \yrs      {\,{\rm yrs}}
19: \newcommand       \Myr      {\,{\rm Myr}}
20: \newcommand       \Myrs     {\,{\rm Myrs}}
21: \newcommand       \Gyr      {\,{\rm Gyr}}
22: \newcommand       \nH           {n_{\rm H}}
23: \newcommand       \simlt        {\lesssim}
24: \newcommand       \simgt        {\gtrsim}
25: \newcommand       \gtsim        {\gtrsim}
26: \newcommand       \ltsim        {\lesssim}
27: \newcommand       \um           {\mu{\rm m}}
28: \newcommand       \mum          {\,{\rm \mu m}}
29: \newcommand       \ppm          {\,{\rm ppm}}
30: \newcommand       \Teff         {T_{\rm eff}}
31: \newcommand       \Msun         {\,{M_\odot}}
32: \newcommand       \Lsun         {\,{L_\odot}}
33: \newcommand       \mH       {m_{\rm H}}
34: \newcommand       \NH       {N_{\rm H}}
35: \newcommand       \simali   {\sim\,}
36: \newcommand       \Alambda  {A_\lambda}
37: \newcommand       \AV       {A_V}
38: \newcommand       \Fnu      {F_\nu}
39: \newcommand       \Fo       {F_{\rm o}}
40: \newcommand       \dof      {{\rm d.o.f.}}
41: \newcommand       \magni    {\,{\rm mag}}
42: \def	\beq	{\begin{equation}}
43: \def	\eeq	{\end{equation}}
44: \def	\beqa	{\begin{eqnarray}}
45: \def	\eeqa	{\end{eqnarray}}
46: %
47: %\newcommand{\figwidth}{4.0in}
48: \newcommand{\figwidth}{6.0in}
49: \newcommand{\btdnote}[1]{{\bf[#1]}}
50: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
51: %------- delete following for submission to ApJ --------
52: %\pagestyle{myheadings}
53: % today's date
54: %use number register 200 for "decade"
55: %\countdef\decade=200
56: %\decade=0
57: %\advance\decade by \year
58: %\advance\decade by -2000   %to suppress two leading digits of yearb
59: %\countdef\hours=201
60: %\hours=0
61: %\advance\hours by \time
62: %\divide\hours by 60
63: %\countdef\mins=202
64: %\mins=0
65: %\advance\mins by \hours
66: %\multiply\mins by 60
67: %\multiply\hours by 100
68: %\countdef\miltime=203
69: %\miltime=0
70: %\advance\miltime by \hours
71: %\advance\miltime by \time
72: %\advance\miltime by -\mins
73: %\def\today{\number\decade.\number\month.\number\day.\number\miltime}
74: %\markright{\today: DRAFT}
75: 
76: \shorttitle{Dust Extinction of GRB Host Galaxies}
77: 
78: %\received{2003 September 18}
79: \begin{document}
80: \title{
81: %------------- enable for labelling preprint ---------------------------
82: % \vspace*{-2.0em}
83: %  {\normalsize\rm submitted to {\it The Astrophysical Journal Letters}}\\
84: % \vspace*{1.0em}
85: On Dust Extinction of Gamma-ray Burst Host Galaxies
86: %\\{\small DRAFT: \today ~~}
87:      }
88: \author{Aigen Li\altaffilmark{1},
89:         S.~L. Liang\altaffilmark{1},
90:         D.~A. Kann\altaffilmark{2},
91:         D.~M. Wei\altaffilmark{3},
92:         S. Klose\altaffilmark{2},
93:         and Y.~J. Wang\altaffilmark{4}}
94: \altaffiltext{1}{Department of Physics and Astronomy, University of
95:                  Missouri, Columbia, MO 65211; {\sf lia@missouri.edu}}
96: \altaffiltext{2}{Th\"{u}ringer Landessternwarte Tautenburg, 
97:                  D-07778 Tautenburg, Germany}
98: \altaffiltext{3}{Purple Mountain Observatory, Chinese Academy of
99:                  Sciences, Nanjing 210008, China}
100: 
101: \altaffiltext{4}{Department of Physics, Hunan Normal University,
102:                  Changsha 410071, China} 
103: 
104: \begin{abstract}
105: Although it is well recognized that gamma-ray burst (GRB) 
106: afterglows are obscured and reddened by dust in their host 
107: galaxies, the wavelength-dependence and quantity of dust 
108: extinction are still poorly known. 
109: %
110: Current studies on this mostly rely on fitting
111: the afterglow spectral energy distributions (SEDs)
112: with template extinction models.
113: %
114: The inferred extinction (both quantity and 
115: wavelength-dependence) and dust-to-gas ratios
116: are often in disagreement with that obtained from 
117: dust depletion and X-ray spectroscopy studies.
118: %
119: We argue that this discrepancy could result from
120: the prior assumption of a template extinction law.
121: %
122: We propose an analytical formula to approximate
123: the GRB host extinction law.
124: %
125: With the template extinction laws self-contained,
126: and the capability of revealing extinction laws 
127: differing from the conventional ones, 
128: it is shown that this is a powerful approach 
129: in modeling the afterglow SEDs to derive GRB host extinction.  
130: %
131: \end{abstract}
132: 
133: \keywords{dust, extinction --- gamma rays: bursts}
134: 
135: \section{Introduction\label{sec:intro}}
136: %
137: In addition to the Galactic foreground extinction,
138: GRBs and their afterglows are 
139: subject to extinction caused by the dust within 
140: their host galaxies. Evidence for this includes ---
141: \begin{itemize}
142: \item ``{\it Dark bursts}'' -- an appreciable fraction
143:       of GRBs with X-ray and/or radio afterglows
144:       lack an optical afterglow 
145:       (Jakobsson et al.\ 2004).\footnote{%
146:         Prior to the launch of {\it Swift}, 
147:         nearly $\simali$60\% of the X-ray afterglows
148:         reportedly had no optical counterparts.
149:         Despite rapid and deep searches in the {\it Swift} era, 
150:         it was found that $\simali$1/3 GRBs with bright X-ray 
151:         afterglows remain undetected at optical wavelengths
152:         (Fiore et al.\ 2007, Schady et al.\ 2007).
153:        }
154:       %
155:       A natural explanation for dark bursts is that they
156:       lie behind significant obscuring dust columns
157:       in their host galaxies
158:       which effectively suppresses the optical light 
159:       [although some dark bursts may be intrinsically 
160:         faint or occur at high redshifts (say, $z\gtsim 5$)
161:         where the Ly$\alpha$ break has moved through 
162:         the optical bands, leading to absorption of 
163:         the optical light by the Ly$\alpha$ forest].
164:       %Indeed, Schady et al.\ (2007) found that the host galaxies 
165:       %of dark GRBs are intrinsically dustier than those of GRBs 
166:       %with detected optical counterparts.
167:       Indeed, Schady et al.\ (2007) found that the X-ray afterglows 
168:       of GRBs not detected by UVOT were more affected by extinction 
169:       than those of GRBs with detected UVOT counterparts.
170:       %
171:       The recent detection of the near infrared (IR) afterglows
172:       of some GRBs (which would have been considered as ``dark bursts''
173:       since their afterglows were not detected in any bluer bands)
174:       provides another piece of evidence for dust obscuration
175:       (e.g. see Jaunsen et al.\ 2008, Tanvir et al.\ 2008).
176: %
177: \item {\it Reddening} -- some GRB afterglows with low redshifts
178:     appear very red, due to effects of extinction -- ultraviolet
179:     (UV)/visible light is extinguished more by dust than red light 
180:     (e.g. see Klose et al.\ 2000, Levan et al.\ 2006).
181:     Dust reddening is also indicated by the significant deviation
182:     of the optical/near-IR spectral energy distributions (SEDs) of 
183:     many afterglows from that expected from standard models.
184:     Also because of dust reddening, the Balmer line ratios in 
185:     the spectra of some GRB host galaxies 
186:     (e.g. see Djorgovski et al.\ 1998),
187:     known as the {\it Balmer decrement},
188:     deviate from the expected ratios for 
189:     the standard Case B recombination,
190:     which are fairly independent of physical conditions
191:    (Osterbrock \& Ferland 2006).
192: %
193: \item {\it Depletion} -- dust-forming heavy elements 
194: such as Si and Fe were found to be substantially 
195: depleted from the gas phase in some host galaxies 
196: (e.g. see Savaglio et al.\ 2003). This indirectly
197: shows the presence of dust in GRB host galaxies
198: since the missing heavy elements must have
199: been locked up in dust grains.
200: %
201: \item {\it Connection between long GRBs and massive stars} 
202: -- there are multiple strong lines of evidence 
203: that long-duration ($\simgt 2\s$) GRBs 
204: are associated with the death of massive stars, 
205: occurring in regions of active star formation 
206: embedded in dense clouds of dust and gas
207: (see Woosley \& Bloom 2006).
208: %
209: \end{itemize}
210: 
211: A precise knowledge of the extinction 
212: (quantity, wavelength-dependence) and 
213: the nature (size, composition, and quantity) 
214: of the dust in GRB host galaxies is crucial for 
215: %
216: \begin{itemize}
217: \item Correcting for the extinction of afterglows 
218:     from X-ray to near-IR wavelengths to derive their
219:     intrinsic luminosities --
220:       this is particularly important
221:       for studying the luminosity distribution of 
222:       GRB afterglows and their intrinsic SEDs
223:       (e.g. see Kann et al.\ 2008);
224: %
225: \item Constraining the nature of the GRB progenitors 
226:     (i.e. collapsing massive stars 
227:     or merging neutron stars) --
228:       if long-duration GRBs are indeed linked to
229:       the collapse of massive stars, 
230:       it is most likely that
231:       their optical and near-IR afterglows will 
232:       suffer from significant attenuation in the 
233:       star-forming molecular clouds heavily 
234:       enshrouded by dust
235:       -- the birth place of these short-lived
236:       ($\simali 10^{6}\yrs$) massive stars;
237: %
238: \item Tracing the physical conditions of 
239:     (and processes occurring in) the environments 
240:     where GRBs occur which hold clues for understanding 
241:     the mechanism for making a burst, e.g.,
242:       a flat or gray extinction law for GRB host galaxies
243:       would imply a dense circumburst environment where 
244:       dust undergoes coagulational-growth 
245:       or a preferential destruction of small grains; and 
246: % 
247: \item Probing the interstellar medium (ISM) 
248:     of high-redshift galaxies 
249:     and the cosmic star formation history --
250:       because of their intense luminosity
251:       which allows their detection at cosmological distances,
252:       GRBs are a powerful tool to study the star formation 
253:       history up to very high redshifts;
254:       e.g., the dust and extinction properties 
255:       of GRB hosts would help understand the nature 
256:       of dark bursts and the dark burst fraction
257:       which would place important constraints
258:       on the fraction of obscured star formation 
259:       in the universe 
260:       (e.g. see Djorgovski et al. 2001, Ramirez-Ruiz et al.\ 2002).
261: %
262: \end{itemize}
263: 
264: However, our current understanding of the dust extinction 
265: in GRB host galaxies is still very poor. 
266: Existing studies on this often draw conclusions
267: in conflict with each other 
268: (see \S\ref{sec:status} for details).
269: We argue that this could be caused by the prior
270: adoption of a {\it template} extinction law
271: in fitting the observed GRB afterglow spectra
272: to derive dust extinction (\S\ref{sec:status}).
273: We propose in this work an alternative, robust method 
274: based on an analytical formula which can restore the 
275: widely adopted template extinction laws (\S\ref{sec:approach}).
276: For illustration, we apply this approach to 
277: GRB\,000301C and GRB\,021004 (\S\ref{sec:test}).
278: We demonstrate in \S\ref{sec:discussion} the uniqueness 
279: of the derived extinction laws.
280: The robustness of this approach will be discussed in
281: a separate paper (Liang \& Li 2008a) in which the afterglow
282: SEDs of $>$\,50 GRBs of a wide range of properties 
283: are successfully modeled and for which the inferred
284: extinction curves are diverse, with some differing 
285: substantially from any of the template extinction curves. 
286: 
287: \section{Current Status\label{sec:status}}
288: %
289: At present, the amount of extinction (usually the rest-frame 
290: visual extinction $A_{V_r}$) and the wavelength-dependence of
291: the extinction (``extinction curve'' or ``extinction law''; 
292: $A_\lambda/A_V$ or $A_\nu/A_V$ if expressed in frequency) 
293: are commonly derived by fitting the UV, optical, and near-IR 
294: afterglow photometry 
295: ($F_\nu$; with the Galactic extinction corrected) 
296: with a power-law model ($\propto \nu^{-\beta}$; 
297: approximating their intrinsic spectra) 
298: reddened by an {\it assumed}, 
299: template extinction law $A_\nu/A_V$
300: %
301: \begin{equation}
302: \label{eq:Fnu}
303: \Fnu = \Fo\,\left(\nu/{\rm Hz}\right)^{-\beta} 
304: \exp\left[-\frac{A_{V_r}}{1.086} \frac{A_{(1+z)\nu}}{A_{V_r}}\right]~~,
305: \end{equation}
306: where $\beta$ is the intrinsic power-law slope 
307: of the afterglow, %\footnote{%
308: %  $\beta$ is usually estimated based on the standard
309: %  fireball model (Sari et al.\ 1998).
310: %  Let $\beta_X$ be the X-ray afterglow spectral index,
311: %  and $p$ be the electron energy distribution index.
312: %  One can first estimate $p$ from $p$\,=\,2\,$\beta_X$ 
313: %  by assuming that the X-ray frequencies $\nu_X$ are 
314: %  above the cooling frequency $\nu_c$ (Berger et al.\ 2003),
315: %  and then estimate the UV-optical-near-IR spectral index
316: %  $\beta$ from $\beta$\,=\,$(p-1)/2$ for $\nu$\,$<$\,$\nu_c$,
317: %  or $\beta$\,=\,$p/2$ for $\nu$\,$>$\,$\nu_c$.
318: %  Alternatively, one can also estimate $p$ from
319: %  the afterglow power-law decay index 
320: %  $\alpha$ ($F_\nu$\,$\sim$\,$t^{-\alpha}$; see Sari et al.\ 1998):
321: %  $p$\,=\,4\,$\alpha/3$\,$+$\,1 in the slow-cooling regime,
322: %  or $p$\,=\,4\,$\alpha/3$\,$+$\,2/3 in the fast-cooling regime.
323: %  } 
324: $\Fo$ is a normalization constant 
325: (normalized to the overall afterglow flux level), 
326: $A_{(1+z)\nu}$ is the rest-frame extinction,
327: and $z$ is the GRB redshift.
328: The factor of ``1.086'' in eq.(\ref{eq:Fnu})
329: arises from the conversion of extinction (in magnitude)
330: to optical depth.
331: %
332: As a priori, six {\it template} extinction laws have 
333: been widely adopted in the literature to derive the dust 
334: extinction of GRB hosts:
335: (1) a simple power-law  
336:       $A_\lambda/A_V$\,$\sim$\,$\lambda^{-\gamma}$
337:       %(e.g. see Savaglio \& Fall 2004),
338:       or even just a linear function of inverse wavelength
339:       $A_\lambda/A_V$\,$\sim$\,$\lambda^{-1}$
340:       (``Linear'' thereafter); 
341:       %(e.g. see Galama \& Wijers 2001).
342: (2) the Milky Way (MW) extinction curve
343:     (with a prominent bump at 2175$\Angstrom$)
344:     characterized by $R_V$, the total-to-selective
345:     extinction ratio (the Galactic average value is 
346:     $R_V\approx 3.1$);
347: (3) the featureless Small Magellanic Cloud (SMC) extinction curve 
348:      which steeply rises with inverse wavelength 
349:       from near-IR to far-UV
350:       ($A_\lambda/A_V \sim \lambda^{-1.2}$);      
351: (4) the Large Magellanic Cloud (LMC) curve being 
352:     intermediate between that of the MW and the SMC;
353: (5) the featureless ``Calzetti'' attenuation law 
354:     for the dust in local starburst galaxies 
355:     (Calzetti et al.\ 1994);\footnote{%
356:       We should note that recent Spitzer observations 
357:       in the near- and mid-IR argue against GRB hosts 
358:       being strongly starbursting galaxies
359:       (Le Floc'h et al.\ 2006),
360:       although their morphological and average radio/submillimeter 
361:       properties suggest that they are likely massive 
362:       and actively star-forming galaxies
363:       (Berger et al.\ 2003; Conselice et al. 2005).
364:       }
365:       and
366: (6) the relatively flat ``Maiolino'' extinction law for the dust 
367:     in the dense circumnuclear region of AGNs (Maiolino et al.\ 2001) 
368:     where the dust size distribution is skewed 
369:     toward large grains (see Fig.\,1).
370: %
371: 
372: To our knowledge, exceptions to the ``template''
373: extinction approach described here are that of 
374: Chen et al.\ (2006) and Li et al.\ (2008), 
375: both of which were based on the fireball model. 
376: The latter approach is limited to bursts of which 
377: the X-ray and optical decay indices are the same.
378: %
379: In most studies (which assume an extinction template)
380: a SMC-type extinction curve is preferred.
381: This is probably because the 2175$\Angstrom$ extinction
382: feature (which is prominent in the MW and LMC curves)  
383: is rarely seen in the afterglow spectra of GRBs.
384: So far, its possible detection is only reported in
385: four bursts: GRB\,970508 (Stratta et al.\ 2004),
386: GRB\,991216 (Kann et al.\ 2006, Vreeswijk et al.\ 2006), 
387: GRB\,050802 (Schady et al.\ 2007),
388: and more definitely GRB\,070802 
389: (Kr\"uhler et al.\ 2008; 
390: \'A. El\'iasd\'ottir et al.\ 2008, in preparation).   
391:    
392: However, some studies favour a much flatter or even gray 
393: extinction curve (e.g. see Savaglio et al.\ 2003, 
394: Savaglio \& Fall 2004, Stratta et al.\ 2005, 
395: Chen et al.\ 2006, Li et al.\ 2008, Perley et al.\ 2008).
396: %
397: With a SMC-type curve, the amount of visual extinction 
398: $A_V$ or reddening\footnote{%
399:   Reddening is usually expressed as 
400:   $E(B-V)\equiv A_B-A_V \equiv A_V/R_V$,
401:   where $A_B$ is the extinction at 
402:   the $B$ band ($\lambda_B\approx 4400\Angstrom$).
403:   %and $R_V$ is the total-to-selective extinction ratio.  
404:   By definition, gray dust (for which the extinction
405:   is just weakly dependent on $\lambda$) is characterized
406:   by small reddening $E(B-V)$ and large $R_V$.
407:   Apparently, for gray dust, a small {\it reddening} does not
408:   necessarily imply a small {\it extinction} since $R_V$ can be large.
409:   \label{ft:reddening}
410:   }
411: derived by fitting the afterglow photometry 
412: tends to be small since the SMC curve rises so rapidly 
413: with $\lambda^{-1}$ that a small $A_V$
414: would imply a large UV extinction.
415: This may explain the finding of ``a strong clustering 
416: toward low extinction ($A_V$\,$\simlt$\,0.2\,mag)''
417: in a detailed study of 19 GRBs by Kann et al.\ (2006),
418: and later by Kann et al.\ (2008) for 15 GRBs.
419: In contrast, for a flatter extinction law 
420: like that of Calzetti, Maiolino, MW with $R_V$\,$>$\,4,
421: or that derived by Chen et al.\ (2006), Li et al.\ (2008)
422: and Perley et al.\ (2008),
423: a relatively large $A_V$ is often obtained.
424: 
425: The visual extinction $A_V$ can also be 
426: inferred from the dust depletion method 
427: based on the gas-phase heavy-element abundances 
428: estimated from the afterglow optical absorption 
429: spectroscopy (Savaglio et al.\ 2003; 
430: Savaglio \& Fall 2004). %\footnote{%
431:   This analysis assumes both the dust depletion pattern
432:   and the visual extinction per unit dust column
433:   $A_V/N_{\rm dust}$ of GRB hosts to be the same as
434:   that of the MW. It is quite possible that GRB hosts may 
435:   have a different depletion pattern
436:   and/or a different $A_V/N_{\rm dust}$ conversion
437:   factor. The latter could result from a dust composition
438:   or size distribution differing from that of the MW.
439:   %\label{ft:depletion}
440:   %} 
441: 
442: One can also derive $A_V$ from
443: the neutral hydrogen column density $N_{\rm H}$
444: derived from Ly$\alpha$ absorption 
445: (Hjorth et al.\ 2003)
446: or the equivalent $N_{\rm H}$ obtained from 
447: soft X-ray absorption (mostly from 
448: oxygen K-shell absorption; Galama \& Wijers 2001,
449: Stratta et al.\ 2004, Watson et al.\ 2006).
450: %
451: There is a puzzling discrepancy between 
452: the optical reddening $E(B-V)$ derived from 
453: the afterglow SED fitting and the visual extinction
454: $A_V$ inferred from the dust depletion analysis
455: or from $N_{\rm H}$ measured from the Ly$\alpha$ 
456: or X-ray absorption spectra,
457: with the former considerably smaller than the latter.
458: 
459: This discrepancy problem could be alleviated
460: if one invokes a flat or gray extinction law.
461: %
462: This is because (1) gray dust ($\simgt 1\mum$) 
463: characterized with an extinction curve weakly 
464: dependent on $\lambda$ in the optical/UV  
465: could produce a high $A_V$ but little reddening 
466: (see Footnote-\ref{ft:reddening}),
467: and (2) per unit mass gray dust is not as effective 
468: as submicron-sized dust in absorbing and scattering 
469: optical light so that the $A_V/N_{\rm dust}$ conversion 
470: factor for gray dust is smaller than that of the MW dust 
471: (with a typical size of $\sim 0.1\mum$; see Li 2008). %\footnote{%
472: %  Alternatively, the mismatch between the X-ray-derived $A_V$
473: %  and that derived from the optical SED modeling
474: %  could be attributed to physically separate X-ray 
475: %  and optical emission regions 
476: %  (e.g. see Prochaska et al.\ 2006, Watson et al.\ 2007).
477: %  %Prochaska et al. (2006) placed a lower limit
478: %  %of $>$50\,pc on the locations of host galaxy absorption 
479: %  %system from GRBs! ... the X-ray opacity comes from 
480: %  %dust-free gas that is very local to the GRB
481: %  %(~1pc).
482: %  }
483: %%
484: The latter would imply that the methods based on 
485: %dust depletion (see Footnote-\ref{ft:depletion}) 
486: dust depletion (see above)
487: and Ly$\alpha$/X-ray absorption may overestimate $A_V$ 
488: if the dust size distribution of GRB host galaxies
489: is indeed biased toward large grains, as a result of
490: dust coagulational growth in the dense circumburst
491: environments or preferential destruction of small dust
492: by GRB emission (e.g. see Waxman \& Draine 2000,
493: Fruchter et al.\ 2001, Perna et al.\ 2003).
494: %
495: 
496: However, we should stress that the gray extinction hypothesis
497: should not be considered as the only solution to 
498: the discrepancy problem (after all, SMC-type or even
499: steeper extinction laws were derived for the hosts
500: of some GRBs; see Liang et al.\ 2008a).
501: Indeed, the mismatch between the X-ray-derived $A_V$
502: and that derived from the optical SED modeling
503: could be attributed to physically separate X-ray 
504: and optical emission regions (e.g. see Prochaska et al.\ 2006, 
505: Watson et al.\ 2007). Prochaska et al.\ (2006) argued that for 
506: GRB\,051111, the X-ray opacity comes from dust-free gas that is 
507: very local to the GRB ($\simali$1\,pc), while they placed 
508: a lower limit of $>$\,50\,pc on the host galaxy absorption 
509: systems from the GRB.
510: It has also been argued that $A_V$ is probably probing 
511: the dust outside of the dense molecular cloud around the GRB, 
512: since all dust within the cloud is likely to have been 
513: obliterated by the burst 
514: (e.g. see Perna \& Lazzati 2002, Prochaska et al.\ 2007).
515: Moreover, if the dust depletion pattern of GRB hosts is
516: different from that of the Milky Way, the discrepancy 
517: between the depletion-derived $A_V$ and that from 
518: the optical SED modeling could be alleviated.
519: 
520: 
521: \section{Our Approach\label{sec:approach}}
522: %
523: In view of the shortcomings of the prior assumption
524: of a template extinction law 
525: (see \S\ref{sec:status}) %\footnote{%
526: %  There is no reason for a prior
527: %  assumption of a known extinction law:
528: %  the composition and size distribution
529: %  (and therefore the extinction law) of 
530: %  the dust in the dense circumburst clouds
531: %  of GRB hosts with a wide range of metallicities
532: %  and evolutionary stages 
533: %  are not expected to resemble that of the MW, LMC, or SMC.
534: %  Indeed, it was shown that the afterglow SED
535: %  of GRB\,050904 at a redshift of $z\approx6.3$
536: %  cannot be explained by dust reddening with 
537: %  any of the conventional (MW, SMC, Calzetti) 
538: %  extinction curves; instead, it can be well 
539: %  reproduced by invoking the extinction curve 
540: %  inferred for a distant quasar at $z=6.2$ (Maiolino et al.\ 2004),
541: %  suggesting that the properties of dust may
542: %  evolve beyond $z=6$ (Stratta et al.\ 2007).
543: %  %
544: %  The fact that the SMC-type extinction 
545: %  is preferred in most of the GRB afterglow
546: %  SED modeling studies is probably just
547: %  because of the apparent lack of the 2175$\Angstrom$
548: %  extinction bump in the afterglow spectra
549: %  of most GRBs, while it is prominent in 
550: %  both the MW and LMC extinction laws.
551: %  }
552: and guided by Pei (1992), we propose a simple formula 
553: containing four dimensionless parameters 
554: ($c_1$, $c_2$, $c_3$, and $c_4$) for the wavelength-dependence 
555: of the extinction for
556: the dust in GRB host galaxies,
557: instead of adopting
558: any known extinction laws (see \S\ref{sec:status})
559: as a template,
560: %%%
561: \beqa
562: \nonumber
563: \label{eq:A2AV}
564: \Alambda/\AV & = & \frac{c_1}{\left(\lambda/0.08\right)^{c_2}
565: + \left(0.08/\lambda\right)^{c_2} + c_3} \\
566: \nonumber
567: & + & \frac{233\left[1 - c_1/\left(6.88^{c_2}+0.145^{c_2}+c_3\right)
568: - c_4/4.60\right]}{\left(\lambda/0.046\right)^2
569: + \left(0.046/\lambda\right)^2 + 90} \\
570: & + & \frac{c_4}{\left(\lambda/0.2175\right)^2
571: + \left(0.2175/\lambda\right)^2 -1.95} ~~~,
572: \eeqa
573: %%%
574: where $\lambda$ is in $\mu$m.\footnote{%
575:   Reichart (2001) proposed a seven-parameter formula
576:   for the dust extinction curve $A_\lambda/A_V$ of
577:   GRB hosts based on the expressions of 
578:   Cardelli et al.\ (1989; ``CCM''; for $\lambda >0.3\mum$)
579:   and of Fitzpatrick \& Massa (1990; ``FM''; 
580:   for $0.1\mum < \lambda < 0.3\mum$).
581:   The problem with the Reichart (2001) formula 
582:   (see his eqs.\,61,66) is that the CCM expression is
583:   only valid for the Galactic extinction curves,
584:   it is not suitable for the SMC or LMC extinction 
585:   (Gordon et al.\ 2003).
586:   Therefore, if a GRB host happens to have a SMC- or LMC-type
587:   extinction law, models based on the Reichart (2001) formula
588:   will not be able to restore the true extinction.  
589:   %Compared to the Pei (1992) formula
590:   %consisting of 12 free parameters and designed
591:   %for MW, LMC and SMC,
592:   %Eq.(\ref{eq:A2AV}), with four parameters,
593:   } 
594: %
595: While the first term in the right-hand side of Eq.(\ref{eq:A2AV})
596: represents the far-UV extinction rise, the second term 
597: and the third term respectively account for the near-IR/visible
598: extinction and the 2175$\Angstrom$ extinction bump.
599: We call this the ``{\it Drude}'' approach 
600: since Eq.(\ref{eq:A2AV}) looks like a sum of
601: Drude functions.
602: %
603: As shown in Figure \ref{fig:extcurv}, this formula,
604: with the free parameters $c_j$ ($j=1,...,4$) adjusted
605: using the Levenberg-Marquardt minimization algorithm
606: (Press et al.\ 1992; see Table \ref{tab:extcurv}),
607: can reproduce the extinction curves 
608: widely adopted as template extinction laws
609: in GRB afterglow SED modeling,
610: clearly demonstrating the advantages
611: of the proposed formula over any template
612: extinction laws with a fixed wavelength-dependence shape:
613: with the widely-adopted conventional extinction laws 
614: self-contained in Eq.(\ref{eq:A2AV}) 
615: and the capability of revealing extinction laws 
616: differing from the conventional ones, 
617: the proposed formula is more flexible 
618: and more powerful in modeling the afterglow SEDs.    
619: %
620: Indeed, as shown in Liang \& Li (2008a,b), 
621: dust reddening models based on this formula nicely 
622: reproduce the observed afterglow SEDs of distant
623: GRBs at $z>4$ (including GRB\,050904 at $z\approx 6.3$) 
624: and that of the ``troublesome'' 
625: GRB\,061126 (Perley et al.\ 2008) 
626: without resorting to an exotic extinction law. 
627: 
628: \section{GRB\,000301C and GRB\,021004: Test Cases\label{sec:test}}
629: %
630: We apply the above-described technique (\S\ref{sec:approach})
631: to the optical afterglows of GRB\,000301C 
632: at $z\approx 2.04$ (Jensen et al.\ 2001)
633: and GRB\,021004 at $z\approx 2.33$ (Fynbo et al.\ 2005).
634: They are selected mainly because they are among 
635: the best-observed in terms of sampling in the time
636: domain, and multiwavelength coverage. 
637: %(see Jensen et al.\ 2001, Fynbo et al.\ 2005).
638: We fit their broadband SEDs 
639: using Eqs.(\ref{eq:Fnu},\ref{eq:A2AV})
640: with $\beta$, $A_V$, $c_1$, $c_2$, and $c_3$ 
641: allowed to vary as free parameters [$\Fo$ is not really
642: a free parameter; for a given set of ($\beta$, $A_V$, 
643: $c_1$, $c_2$, $c_3$), $\Fo$ is uniquely determined by
644: the overall flux level. Therefore, in the SED modeling
645: we fit five free parameters to the six (seven) data points of 
646: GRB\,000301C (GRB\,021004)].\footnote{%
647:   We set $c_4$\,=\,0 based on a visual inspection
648:   of the observed SEDs which clearly suggest 
649:   the absence of a 2175$\Angstrom$ feature
650:   (see Figs.\,\ref{fig:GRB000301C},\ref{fig:GRB021004}).
651:   With $c_4$ treated as a free, positive parameter,
652:   even the best fits (given by $c_4\approx 0.0034,0.0018$
653:   for GRB\,000301C and GRB\,021004, respectively; 
654:   for comparison, $c_4\approx 0.051,0.039$ for
655:   MW and LMC, respectively) are not as good as 
656:   that provided by models with $c_4=0$.
657:   We place an upper limt of $c_4\approx 0.015$ (0.0073)
658:   for GRB\,000301C (GRB\,021004).
659:   }
660: We derive the best-fit parameters based on
661: the Levenberg-Marquardt minimization algorithm
662: (see Table \ref{tab:grbmod}).
663: As shown in Figure\,\ref{fig:GRB000301C}
664: for GRB\,000301C and in Figure\,\ref{fig:GRB021004}
665: for GRB\,021004, almost perfect fits to the observed
666: SEDs are achieved through this approach.
667: The inferred extinction curves differ substantially
668: from any of the template extinction laws.
669: 
670: \section{Discussion\label{sec:discussion}}
671: %
672: We have also fitted the afterglow
673: SEDs of GRB\,000301C and GRB\,021004
674: in terms of the MW, SMC, LMC, Calzetti,
675: and ``linear'' %($A_\lambda \propto \lambda^{-1}$)
676: template extinction curves (see Table \ref{tab:grbmod} 
677: and Figs.\,\ref{fig:GRB000301C},\ref{fig:GRB021004}).
678: Since for a given template extinction law 
679: the wavelength-dependence of the extinction
680: $A_\lambda/A_V$ is fixed, we are now left with 
681: only three parameters: $\Fo$, $\beta$, and $A_V$.
682: %As shown in Table \ref{tab:grbmod} 
683: %and Figures\,\ref{fig:GRB000301C},\ref{fig:GRB021004},
684: %none of these fits is as good as that obtained using
685: %the technique described in \S\ref{sec:approach}.
686: The models based on the MW and LMC extinction laws
687: could not fit the observed SEDs at all. 
688: This is because the 2175$\Angstrom$
689: extinction feature which is prominent in the MW and LMC   
690: curves is absent in the SEDs of GRB\,000301C and GRB\,021004.
691: In contrast, the SMC and ``linear'' models closely fit 
692: the afterglow SEDs of these two bursts, 
693: better than the Drude model proposed here 
694: as measured by $\chi^2/N_{\rm d.o.f.}$
695: (see Table \ref{tab:grbmod}).
696: 
697: While the Drude model has three more parameters 
698: than the SMC and ``linear'' models, the quality 
699: of the fitting of the Drude model is even not as
700: good as that of the SMC or ``linear'' model. 
701: Then, why do not we simply adopt the SMC 
702: or ``linear'' model? 
703: 
704: First of all, we should note that there are no physical
705: reasons for a prior assumption of a known extinction law,
706: either that of the SMC, LMC, ``linear'' or MW:
707: the composition and size distribution
708: (and therefore the extinction law) of 
709: the dust in the dense circumburst clouds
710: of GRB hosts with a wide range of metallicities
711: and evolutionary stages 
712: are not expected to resemble that of 
713: the MW, LMC, or SMC (e.g. see Dwek 2005).
714: In literature, a SMC-type extinction is often assumed
715: for low-metallicity environments. However, there is no
716: physical basis for this (except the lack of grain growth
717: in these regions because of the lack of raw dust materials
718: -- the SMC dust, on average, is substantially smaller than
719: that of the Milky Way [see Weingartner \& Draine 2001]).
720: Moreover, it is known that the GRB hosts have
721: a wide range of metallicities. 
722: %
723: Indeed, the reasons why the MW, LMC and SMC 
724: laws are often used for GRB afterglow SED
725: modeling are mainly (1) little is known about the extinction 
726: laws of other galaxies, and (2) the Pei (1992) formula for 
727: the MW, LMC and SMC extinction laws is numerically convenient
728: for computer implementation. 
729: 
730: Second, although the SMC-type extinction 
731: is preferred in most of the present afterglow 
732: SED modeling studies, only the Drude approach is 
733: capable of reproducing the SEDs of those reddened 
734: by gray extinction or by non-conventional extinction.
735: Indeed, it was shown that the afterglow SED
736: of GRB\,050904 at a redshift of $z$\,$\approx$\,6.3
737: cannot be explained by dust reddening with 
738: any of the conventional (MW, SMC, Calzetti) 
739: extinction curves; instead, it can be well 
740: reproduced by invoking the extinction curve 
741: inferred for a distant quasar at $z$\,=\,6.2 (Maiolino et al.\ 2004),
742: suggesting that the properties of dust may
743: evolve beyond $z$\,=\,6 (Stratta et al.\ 2007).
744: 
745: Third, the Drude model would at least complement 
746: the models using template extinction curves, 
747: particularly for those bursts for which the Drude 
748: model gives a larger $\chi^2/N_{\rm d.o.f.}$ 
749: (but still fits the observed SEDs well).  
750: Given that the derived extinction $A_V$ 
751: and the intrinsic spectral slope $\beta$ 
752: differ appreciably among different approaches
753: (see Table \ref{tab:grbmod}), the SMC model
754: (and other models) should be used along side
755: with the Drude model to gain insight into
756: the ``true'' extinction and the ``true'' spectral slope.
757: 
758: We finally demonstrate the uniqueness of the extinction
759: curve inferred from the Drude approach. 
760: To this end, we generate three sets of afterglow
761: ``photometry data'' by reddening the intrinsic afterglow 
762: spectrum $F_\nu\,(\mu {\rm Jy})$\,=\,$5.2\times 10^8 
763: \left(\nu/{\rm Hz}\right)^{-0.5}$ of a burst at $z\approx 2$ 
764: respectively with three template extinction laws: 
765: MW, SMC, and Calzetti, each with $A_V=0.5\magni$. 
766: We then apply the Drude approach to these three sets 
767: of artificially-created GRB afterglow data. 
768: As shown in Figure\,\ref{fig:test},
769: we uniquely restore the MW, SMC, and Calzetti extinction
770: laws: the inferred extinction curves are almost identical
771: to that used to redden the intrinsic spectrum
772: (the derived parameters [see Table\,\ref{tab:test}]
773: are essentially the same as those tabulated 
774: in Table\,\ref{tab:extcurv}).
775: %This suggests that although the SMC model is also able to
776: %reproduce the afterglow SEDs of GRB\,000301C and GRB\,021004, 
777: %it is probably just a coincidence: if their afterglow
778: %spectra are indeed reddened by SMC-type dust,
779: %the Drude approach should be able to restore 
780: %a SMC extinction curve.
781: %Moreover, if the intrinsic spectral slope $\beta$ is 
782: %known (say, from the fireball model), one should also 
783: %be able to break this degeneracy.
784: 
785: 
786: 
787: \acknowledgments
788: We thank the anonymous referee and J.X. Prochaska 
789: for very helpful comments.
790: A.L. and S.L. are supported in part by a NASA/Swift Theory Program, 
791: a NASA/Chandra Theory Program, and the NSFC Outstanding Oversea 
792: Young Scholarship. D.M.W. is supported by the NSFC grants
793: 10621303 and 10673034, and the National Basic Research Program of
794: China (973 Program 2007CB815404). D.A.K. and S.K. acknowledge 
795: financial support by DFG grant Kl 766/13-2.
796: 
797: \begin{thebibliography}{}
798: \bibitem[]{696} Berger, E., Cowie, L.~L., Kulkarni, S.~R., Frail, D.~A., 
799:              Aussel, H., \& Barger, A.~J.\ 2003, \apj, 588, 99
800: \bibitem[]{698} Calzetti, D., Kinney, A.~L., \& Storchi-Bergmann, T.\
801:              1994, \apj, 429, 582
802: \bibitem[]{783}Cardelli, J.~A., Clayton, G.~C., \& Mathis, J.~S.\ 
803:           1989, ApJ, 467, 334
804: \bibitem[]{700} Chen, S.~L., Li, A., \& Wei, D.~M.\ 2006, \apjl, 647, L13
805: \bibitem[]{701} Conselice, C.~J., et al.\ 2005, \apj, 633, 29
806: \bibitem[]{702} Djorgovski, S.~G., Kulkarni, S.~R., Bloom, J.~S.,
807:              Goodrich, R., Frail, D.~A., Piro, L.,
808:              \& Palazzi, E.\ 1998, \apjl, 508, L17
809: \bibitem[]{705} Djorgovski, S.~G., Frail, D.~A., Kulkarni, S.~R.,
810:              Bloom, J.~S., Odewahn, S.~C.,
811:              \& Diercks, A.\ 2001, \apj, 562, 654
812: \bibitem[]{}Dwek, E.\ 2005, in AIP Conf. Proc. 761, 
813:             The Spectral Energy Distributions of Gas-Rich Galaxies: 
814:             Confronting Models with Data, 
815:             ed. C. C. Popescu \& R. J. Tuffs (Melville: AIP), 103
816: \bibitem[]{708} Fiore, F., et al.\ 2005, \apj, 624, 853
817: \bibitem[]{709} Fiore, F., Guetta, D., Piranomonte, S.,
818:              D'Elia, V., \& Antonelli, L.~A.\ 2007, \aap, 470, 515
819: \bibitem[]{796}Fitzpatrick, E.~L., \& Massa, D.~L.\
820:              1990, ApJS, 72, 163
821: \bibitem[]{711} Fruchter, A., Krolik, J.~H., \& Rhoads, J.~E.\ 
822:              2001, ApJ, 563, 597
823: \bibitem[]{713} Fynbo, J.~P.~U., et al.\ 2005, \apj, 633, 317
824: \bibitem[]{714} Galama, T.~J., \& Wijers, R.~A.~M.~J.\ 2001,
825:              \apjl, 549, L209
826: \bibitem[]{716} Gordon, K.~D., Clayton, G.~C., Misselt, K.~A., 
827:              Landolt, A.~U., \& Wolff, M.~J.\ 2003, \apj, 594, 279 
828: \bibitem[]{718} Hjorth, J., et al.\ 2003, \apj, 597, 699
829: \bibitem[]{719} Holland, S.~T., et al.\ 2003, \aj, 125, 2291
830: \bibitem[]{720} Jakobsson, P., et al.\ 2004, A\&A, 427,785
831: \bibitem[]{} Jaunsen, A.~O., et al.\ 2008, ApJ, in press
832:              (astro-ph/0803.4017)  
833: \bibitem[]{721} Jensen, B.~L., et al.\ 2001, A\&A, 370, 909
834: \bibitem[]{722} Kann, D.~A., Klose, S., \& Zeh, A.\
835:              2006, \apj, 641, 993
836: \bibitem[]{724} Kann, D.~A., et al.\ 2008, \apj, submitted
837:              (astro-ph/0712.2186)
838: \bibitem[]{726} Klose, S., et al.\ 2000, \apj, 545, 271
839: \bibitem[]{814} Kr\"uhler, T., et al.\ 2008, ApJ, submitted
840: %\bibitem[]{727} Lazzati, D. \& Perna, R.\ 2002, MNRAS, 330, 383
841: \bibitem[]{728} Le Floc'h, E., Charmandaris, V.,
842:              Forrest, W.~J., Mirabel, I.~F., Armus, L.,
843:              \& Devost, D.\ 2006, \apj, 642, 636
844: \bibitem[]{731} Levan, A., et al.\ 2006, \apj, 647, 471
845: \bibitem[]{851} Li, A. 2004, in Penetrating Bars through Masks 
846:              of Cosmic Dust, ed. D.~L. Block, I. Puerari, 
847:              K.~C. Freeman, R. Groess, \& E.~K. Block 
848:              (Dordrecht: Kluwer), 535
849: \bibitem[]{732} Li, A.\ 2008, in: Small Bodies in Planetary Sciences 
850:              (Lecture Notes in Physics Series), I. Mann, A. Nakamura, 
851:              \& T. Mukai (eds.), Springer, in press
852: \bibitem[]{735} Li, Y., Li, A., \& Wei, D.~M.\ 2008, 
853:                 \apj, 678, 1136
854: \bibitem[]{737} Liang, S.~L., \& Li, A.\ 2008a, in preparation
855: \bibitem[]{738} Liang, S.~L., \& Li, A.\ 2008b, in preparation
856: \bibitem[]{739} Maiolino, R., Marconi, A., \& Oliva, E.\
857:              2001, \aap, 365, 37
858: \bibitem[]{741} Maiolino, R., Schneider, R., Oliva, E.,
859:              Bianchi, S., Ferrara, A., Mannucci, F., Pedani,
860:              M., \& Roca Sogorb, M.\ 2004, \nat, 431, 533
861: \bibitem[]{744} Osterbrock, D.~E., \& Ferland, G.~J.\ 2006,
862:              Astrophysics of Gaseous Nebulae and Active
863:              Galactic Nuclei
864:              (2nd ed; Sausalito: Univ. Sci. Books)
865: \bibitem[]{748} Pei, Y.~C.\ 1992, \apj, 395, 130
866: \bibitem[]{749} Perley, D.~A., et al.\ 2008, \apj, 672, 449
867: \bibitem[]{} Perna, R., \& Lazzati, D.\ 2002, \apj, 580, 261
868: \bibitem[]{750}Perna, R., Lazzati, D., \& Fiore, F.\
869:             2003, \apj, 585, 775
870: \bibitem[]{875} Press, W.~H., Teukolsky, S.~A., Vetterling, W.~T., 
871:              \& Flannery, B.~P. 1992, Numerical Recipes in 
872:              FORTRAN: The Art of Scientific Computing 
873:              (2nd ed.; Cambridge: Cambridge Univ. Press)
874: \bibitem[]{752} Prochaska, J.~X., Chen, H.-W., \& Bloom, J.~S.\ 
875:              2006, \apj, 648, 95 
876: \bibitem[]{} Prochaska, J.~X., Chen, H.-W., Dessauges-Zavadsky, M., 
877:              \& Bloom, J.~S.\ 2007, \apj, 666, 267 
878: \bibitem[]{754} Ramirez-Ruiz, E., Trentham, N., \& Blain, A.~W.\
879:              2002, \mnras, 329, 465
880: \bibitem[]{844}Reichart, D.~E.\ 2001, ApJ, 553, 235
881: \bibitem[]{756} Rhoads, J.~E., \& Fruchter, A.~S.\ 2001, \apj, 546, 117
882: \bibitem[]{757} Rol, E., et al.\ 2007, \apj, 669, 1098
883: %\bibitem[]{758} Sari, R., Piran, T., \& Narayan, R.\
884: %             1998, \apjl, 497, L17
885: \bibitem[]{760} Savaglio, S., \& Fall, S.~M.\
886:              2004, \apj, 614, 293
887: \bibitem[]{762} Savaglio, S., Fall, S.~M., \& Fiore, F.\
888:              2003, \apj, 585, 638
889: \bibitem[]{764} Schady, P., et al.\ 2007, \mnras, 377, 273
890: \bibitem[]{765} Stratta, G., Fiore, F., Antonelli, L.~A., Piro, L.,
891:              \& De Pasquale, M.\ 2004, \apj, 608, 846
892: \bibitem[]{767} Stratta, G., Perna, R., Lazzati, D., Fiore, F.,
893:              Antonelli, L.~A., \& Conciatore, M.~L.\ 2005, \aap, 441, 83
894: \bibitem[]{769} Stratta, G., Maiolino, R., Fiore, F.,
895:              \& D'Elia, V.\ 2007, \apjl, 661, L9
896: \bibitem[]{} Tanvir, N.~R., et al.\ 2008, MNRAS, in press
897:              (astro-ph/0803.4100)  
898: \bibitem[]{771} Vreeswijk, P.~M., et al.\ 2006, \aap, 447, 145
899: \bibitem[]{772} Watson, D., et al.\ 2006, \apj, 652, 1011
900: \bibitem[]{773} Waxman, E., \& Draine, B.T.\
901:              2000, \apj, 537, 796
902: \bibitem[]{908}Weingartner, J.C., \& Draine, B.T. 2001, ApJ, 548, 296
903: \bibitem[]{775} Woosley, S.~E., \& Bloom, J.~S.\ 2006, \araa, 44, 507 
904: %
905: \end{thebibliography}
906: 
907: %\clearpage
908: \begin{table}
909: {\footnotesize
910: \caption[]{\footnotesize
911:            ``Drude'' fits to known extinction curves
912:            for $\lambda$\,=\,0.1--1$\mum$
913:            widely adopted as ``templates'' in modeling 
914:            GRB afterglow SEDs to derive GRB host dust extinction.
915:            \label{tab:extcurv}}
916: \begin{center}
917: \begin{tabular}{ccccccc}
918: \tableline\tableline
919:  & $c_1$ & $c_2$ & $c_3$ & $c_4$ & $\chi^2/\dof$ \\
920: \cline{1-6}
921: MW       & 14.4 & 6.52  & 2.04  & 0.0519 & 1.66 \\
922: LMC      & 4.47 & 2.39  & -0.988 & 0.0221 & 1.19 \\
923: SMC      & 38.7 & 3.83  & 6.34 & 0. & 1.36 \\
924: Linear   & 66.2 & 4.97  & 22.1 & 0. & 1.42 \\
925: Calzetti & 44.9 & 7.56  & 61.2 & 0. & 1.68 \\
926: \tableline
927: \end{tabular}
928: \end{center}
929: }
930: \end{table}
931: 
932: \begin{table}
933: {\footnotesize
934: \caption[]{\footnotesize
935:            Results of fitting to the afterglow SEDs
936:            of GRB\,000301C and GRB\,021004 
937:            with the Drude approach 
938:            (see \S\ref{sec:approach}, \S\ref{sec:test})
939:            or various template extinction laws.
940:            Note that the Drude approach has more free
941:            parameters than the other approaches.
942:            \label{tab:grbmod}}
943: \begin{center}
944: \begin{tabular}{ccccccccccc}
945: \tableline\tableline
946: Extinction & $c_1$ & $c_2$ & $c_3$ & $c_4$  
947:            & $\AV$ & $\beta$ & $\Fo$     
948:            & $\chi^2/N_{\rm data}$ 
949:            & $\chi^2/N_{\rm d.o.f.}$ \\
950: Type       &  &  &   &   & (mag) &         & ($\mu$Jy) &    \\  
951: \cline{1-10}
952: &       &       &       &  &  GRB\,000301C & &  & &  &               \\ 
953: \cline{1-10} 
954: %Drude  & $c_1$ & $c_2$ & $c_3$  & $c_4$ & $R_V$ 
955: %        & $\AV$  & $\beta$ & $\Fo$  & $\chi^2/\dof$ \\
956: Drude &0.025 &0.048 &-2.00  & 0. &0.32 &0.61  &3.99E10 & 0.33 & 1.98\\
957: %
958: MW    & ...   & ...   & ...   & ... &0. & 0.85 &1.04E14 & 1.32 &2.64\\
959: SMC   & ...   & ...   & ...   & ... &0.11  & 0.62  &4.68E10  & 0.64 &1.28\\
960: LMC   & ...   & ...   & ...   & ... &0. & 0.85 &1.04E14  & 1.32 &2.64\\
961: Linear   & ... & ...  & ...   & ... &0.20  & 0.51  &1.23E9  & 0.47 &0.94\\
962: Calzetti & ... & ...  & ...   & ... &0. & 0.85 & 1.04E14 & 1.32 &2.64\\
963: \cline{1-10}
964: &       &       &       &   & GRB\,021004   & &  & &  &               \\ 
965: \cline{1-10}
966: %Type & $c_1$ & $c_2$ & $c_3$  & $c_4$ 
967: %      & $\AV$  & $\beta$ & $\Fo$ & \chi^2/\dof$ \\
968: Drude &0.015 &0.15 &-2.00  & 0. &0.13 &0.78  &6.13E12 &0.47 &1.64\\
969: %
970: MW    & ...   & ...   & ...   & ... &0.  & 1.06    &8.23E16  & 1.53 &2.68\\
971: SMC   & ...   & ...   & ...   & ... &0.15  & 0.67  &1.58E11   & 0.36 &0.53\\
972: LMC   & ...   & ...   & ...   & ... &0.  & 1.06  &8.23E16  & 1.53 &2.68\\
973: Linear   & ... & ...  & ...   & ... &0.26  & 0.54  &2.05E9   & 0.76 &1.33\\
974: Calzetti & ... & ...  & ...   & ... &0.95  & 0.    &46.7     & 0.80 &1.40\\
975: \tableline
976: \end{tabular}
977: \end{center}
978: }
979: \end{table}
980: 
981: 
982: \begin{table}
983: {\footnotesize
984: \caption[]{\footnotesize
985:            Results of Drude-fitting to the artificial SED
986:            generated by reddening the power-law afterglow           
987:            $F_\nu\,(\mu {\rm Jy})$\,=\,$5.2\times 10^8 
988:            \left(\nu/{\rm Hz}\right)^{-0.5}$ 
989:            with $A_V$\,=\,0.5\,mag extinction 
990:            of MW, SMC, and Calzetti-type
991:            (see Fig.\,\ref{fig:test}).
992:            \label{tab:test}}
993: \begin{center}
994: \begin{tabular}{ccccccccccc}
995: \tableline\tableline
996: Reddening & $c_1$ & $c_2$ & $c_3$ & $c_4$  
997:            & $\AV$ & $\beta$ & $\Fo$     
998:            & $\chi^2/N_{\rm data}$ 
999:            & $\chi^2/N_{\rm d.o.f.}$ \\
1000: Type       &  &  &   &   & (mag) &         & ($\mu$Jy) &    \\  
1001: \cline{1-10} 
1002: %Drude  & $c_1$ & $c_2$ & $c_3$  & $c_4$ & $R_V$ 
1003: %        & $\AV$  & $\beta$ & $\Fo$  & $\chi^2/\dof$ \\
1004: MW    & 14.3 & 6.49  & 2.02   & 0.0514
1005:       &0.501 & 0.499 &5.24E8 & 3.26E-4 &4.35E-4\\
1006: SMC   & 39.4 & 3.89  & 6.31  & 0.
1007:       &0.500 & 0.501 &5.26E8 & 1.32E-3 &1.76E-3\\
1008: Calzetti  & 45.2 & 7.51  & 61.7  & 0.
1009:       &0.497 & 0.502 &5.17E8 & 7.98E-4 &1.06E-3\\
1010: \cline{1-10}
1011: \tableline
1012: \end{tabular}
1013: \end{center}
1014: }
1015: \end{table}
1016: 
1017: 
1018: %\clearpage
1019: 
1020: \begin{figure}[ht]
1021: \begin{center}
1022: \includegraphics[width=\figwidth,angle=0]{f1.cps}
1023: \end{center}\vspace*{-1em}
1024: \caption{
1025:         \label{fig:extcurv}
1026:         Extinction laws widely adopted 
1027:         as {\it ``templates''} in GRB host extinction studies:
1028:         the SMC law (upper panel: dashed black line; Pei 1992),
1029:         the LMC law (upper panel: dashed blue line; Pei 1992), 
1030:         the linear $A_\lambda \propto \lambda^{-1}$ law
1031:         (upper panel: dashed cyan line),
1032:         the MW Galactic average extinction law 
1033:         ($R_V=3.1$; lower panel: dashed black line; Pei 1992),
1034:         the MW extinction law with $R_V=4.5$ for dense
1035:         clouds (lower panel: solid red line),
1036:         the Calzetti starburst attenuation law
1037:         (lower panel: dashed blue line),
1038:         and the Maiolino law for AGN dust tori
1039:         (lower panel: dashed cyan line; just like that of
1040:          the MW with $R_V=4.5$).
1041:         Also shown are the ``Drude'' fits to
1042:         these ``template'' extinction laws: 
1043:         SMC (upper panel: solid green line),
1044:         LMC (upper panel: solid red line),
1045:         Linear (upper panel: solid magenta line),
1046:         MW with $R_V=3.1$ (lower panel: solid green line),
1047:         and Calzetti (lower panel: solid magenta line). 
1048:         }
1049: \end{figure}
1050: 
1051: \begin{figure}[ht]
1052: \begin{center}
1053: \includegraphics[width=\figwidth,angle=0]{f2.cps}
1054: \end{center}\vspace*{-1em}
1055: \caption{
1056:         \label{fig:GRB000301C}
1057:         Upper panel: fitting the SED of the afterglow of 
1058:         GRB\,000301C (filled black circles) 
1059:         with the SMC (green) and LMC or MW (blue) template extinction
1060:         laws and the ``Drude'' approach (red; see Eq.\ref{eq:A2AV})
1061:         for the host extinction curve.
1062:         No extinction is allowed in the MW and LMC models 
1063:         (i.e. the best fit with a MW- or LMC-type extinction
1064:          is given by $A_V$\,$\approx$\,0): a small amount of
1065:         $A_V$ would lead to large deviations from the afterglow
1066:         SED since the 2175$\Angstrom$ bump prominent in
1067:         the MW and LMC laws is absent in the afterglow SED.
1068:         Lower panel: comparison of the SMC (green), LMC (blue), 
1069:         MW ($R_V=3.1$; black) extinction laws with
1070:         that derived from the Drude approach (red).
1071:         }
1072: \end{figure}
1073: 
1074: \begin{figure}[ht]
1075: \begin{center}
1076: \includegraphics[width=\figwidth,angle=0]{f3.cps}
1077: \end{center}\vspace*{-1em}
1078: \caption{
1079:         \label{fig:GRB021004}
1080:         Same as Figure \ref{fig:GRB000301C}
1081:         but for GRB\,021004.
1082:         }
1083: \end{figure}
1084: 
1085: \begin{figure}[ht]
1086: \begin{center}
1087: \includegraphics[width=\figwidth,angle=0]{f4.cps}
1088: \end{center}\vspace*{-1em}
1089: \caption{
1090:         \label{fig:test} 
1091:         Upper panel: Drude fits to the observer-frame
1092:         $UBVRIJHK$ ``photometry data''
1093:         artificially-generated by reddening the intrinsic afterglow
1094:         spectrum $F_\nu\propto \nu^{-\beta}$ of 
1095:         a burst at $z\approx 2$
1096:         (black line with red crosses superimposed for 
1097:          the observer-frame $UBVRIJHK$ bands) with 
1098:         the SMC (``data'': cyan squares; Drude fit: green line),
1099:         Calzetti (``data'': blue triangles; Drude fit: magenta line),
1100:         and MW (``data'': black circles; Drude fit: red line)
1101:         extinction laws (with $A_V=0.5\magni$ for each).
1102:         Lower panel: comparison of the SMC, Calzetti and MW
1103:         extinction curves (solid lines) with that inferred from 
1104:         the Drude approach (dashed lines).
1105:         }
1106: \end{figure}
1107: %%
1108: 
1109: \end{document}
1110: