1: \documentclass{emulateapj}
2: \usepackage{graphicx,amsmath,amssymb}
3: \usepackage{natbib}
4: \usepackage{apjfonts}
5: \bibpunct[, ]{(}{)}{;}{a}{}{,}
6:
7:
8: \bibliographystyle{apj}
9:
10: \newcommand{\sect}{\S\,}
11:
12: \begin{document}
13:
14: \title{Very different X-ray to optical column density ratios in
15: $\gamma$-ray burst afterglows: ionisation in GRB
16: environments}
17:
18: \author{D.~Watson,\altaffilmark{1}
19: J.~Hjorth,\altaffilmark{1}
20: J.~P.~U.~Fynbo,\altaffilmark{1}
21: P.~Jakobsson,\altaffilmark{2}
22: S.~Foley,\altaffilmark{3}
23: J.~Sollerman,\altaffilmark{1,}\altaffilmark{4}
24: and R.~A.~M.~J.~Wijers\altaffilmark{5}
25: }
26: \altaffiltext{1}{Dark Cosmology Centre, Niels Bohr Institute, University of Copenhagen, Juliane Maries Vej 30, DK-2100 Copenhagen \O, Denmark; darach, jens, jfynbo, jesper @dark-cosmology.dk}
27: \altaffiltext{2}{Centre for Astrophysics Research, University of Hertfordshire, College Lane, Hatfield, Herts AL10 9AB, UK; P.Jakobsson@herts.ac.uk}
28: \altaffiltext{3}{School of Physics, University College Dublin, Dublin 4, Ireland; sfoley@bermuda.ucd.ie}
29: \altaffiltext{4}{Stockholm Observatory, Department of Astronomy, AlbaNova, S-106 91 Stockholm, Sweden}
30: \altaffiltext{5}{Astronomical Institute ``Anton Pannekoek'', University of Amsterdam, Kruislaan 403, 1098 SJ Amsterdam, The Netherlands; rwijers@science.uva.nl}
31: \begin{abstract}
32: We compare the \ion{H}{1} column densities from Ly$\alpha$ absorption
33: to the metal column densities from X-ray absorption in $\gamma$-ray
34: burst (GRB) afterglows. Eight bursts of seventeen entering the sample,
35: have significant extragalactic X-ray absorptions. Of these we find a
36: range in metallicities from sub-solar to a few hundred times the solar
37: value. There is a complete lack of correlation between the X-ray and
38: optical column densities. This, and the large range and extreme values
39: of these `metallicities', suggest that the column densities derived by
40: one or both methods are not a reliable indication of the true total
41: column densities towards GRBs. Ionisation of the GRB's gas cloud to
42: large distances along the line of sight seems the most likely cause.
43: From the lower limit on the total column density and the UV luminosity
44: of the GRBs, we derive a maximum distance to the majority of the gas
45: surrounding GRBs of $\sim3$\,pc, suggesting that the gas probed by
46: optical afterglow spectra is not the cloud in which
47: the burst occurs. This is an encouraging result for the use of GRB
48: optical afterglows as probes of the interstellar medium (ISM) in their
49: host galaxies, as the ISM observed is less likely to be strongly
50: affected by the GRB or its progenitor.
51: \end{abstract}
52: \keywords{ gamma rays: bursts -- X-rays: general -- X-rays: galaxies -- quasars: absorption lines
53: }
54:
55: \maketitle
56:
57: \section{Introduction\label{introduction}}
58:
59: It is known from studies of neutral hydrogen absorption in $\gamma$-ray burst
60: (GRB) afterglows \citep[a technique pioneered half a decade
61: ago,][]{2001A&A...370..909J,2001A&A...373..796F,2003ApJ...597..699H}, that
62: they probe much higher column densities than QSO sightlines
63: \citep{2004A&A...419..927V}, almost certainly because GRBs probe the
64: star-forming environments in which they are born
65: \citep{2006A&A...460L..13J}. The highest column density GRB damped
66: Ly$\alpha$ absorbers (DLAs) are higher than any known QSO-DLA
67: \citep{2006A&A...460L..13J,2006ApJ...652.1011W,2005A&A...442L..21S,2005ApJ...634L..25C},
68: in spite of the much larger QSO-DLA sample.
69:
70: Measurements of the \ion{H}{1} column density ($N($\ion{H}{1}$)$) are
71: needed to determine abundances in GRB environments where the metal column
72: densities are known either from absorption lines of non-refractory elements
73: such as Zn, or from the soft X-ray absorption which is dominated by
74: $\alpha$-chain elements in gas and solid phases. The optical
75: observations are complicated by the need to assume that certain elements are
76: not depleted by dust, and the need for high-resolution spectra. The X-ray
77: measurements do not suffer from these problems, but do not provide
78: a strong redshift constraint on the absorber
79: \citep[e.g.][]{2002A&A...395L..41W,2004ApJ...608..846S,2006A&A...455..803G}.
80: However using the X-ray metal column densities with Ly$\alpha$ columns has
81: been extremely difficult because of the conflicting redshift requirements
82: and has not yet been done in any object at $z>0.5$ other than a GRB. To
83: detect hydrogen Ly$\alpha$ 1216\,\AA\ with ground-based instruments we need
84: $z\gtrsim2$ to shift the line into the sensitivity range of optical/UV
85: spectrographs. But the soft X-ray absorption moves further and further out
86: of the bandpass as we go to higher redshift, such that once the neutral
87: oxygen edge at 0.52\,keV is no longer detectable, there is typically a
88: degeneracy between the redshift and the total absorption column that scales
89: roughly as $N_{\rm H}\propto (1+z)^{2.4}$. Thus we need a high
90: X-ray column density (greater than a few $\times10^{21}$) for it to
91: be detectable at $z\gtrsim2$.
92:
93: With the advent of the \emph{Swift} satellite and its highly successful
94: X-ray telescope (XRT), very early, largely complete, and high flux
95: observations of GRB X-ray afterglows have been obtained for about ninety
96: GRBs every year. This means that the sample of GRB afterglow X-ray
97: absorption column densities has increased enormously, such that we now have
98: limits on the extragalactic X-ray absorption for almost every new GRB
99: discovered and we finally have a few GRB damped Ly$\alpha$ systems
100: (GRB-DLAs) where we can compare with significant detections of extragalactic
101: soft X-ray absorption: see GRB\,050401 \citep{2006ApJ...652.1011W},
102: GRB\,050505 \citep{2006MNRAS.368.1101H}, and GRB\,050730
103: \citep{2005A&A...442L..21S}.
104:
105: In this \emph{Letter} we present the sample of GRB-DLAs with significant
106: extragalactic X-ray absorptions and compare the column densities obtained
107: with each method.
108: Uncertainties quoted are at the 68\% confidence level for one interesting
109: parameter unless otherwise stated. A cosmology where
110: $H_0=70$\,km\,s$^{-1}$\,Mpc$^{-1}$, $\Omega_\Lambda = 0.7$ and $\Omega_{\rm m}=0.3$ is assumed throughout.
111:
112:
113: \section{Sample and data reduction}\label{observations}
114:
115: \emph{Swift} GRBs at $z>2$ were examined by \citet{2006A&A...460L..13J}
116: to determine the properties of GRB-DLAs. From that sample bursts with
117: well-constrained \ion{H}{1} columns were selected. We analysed XRT data from
118: these bursts where they were believed to have X-ray absorption in excess of
119: Galactic from the literature or the \emph{Swift} data
120: table.\footnote{\url[http://swift.gsfc.nasa.gov/docs/swift/archive/grb_table/]{http://swift.gsfc.nasa.gov/docs/swift/archive/grb\_table/}}
121: Analysis was done in a standard way. Spectra were fit with a power-law with
122: absorption fixed at the Galactic value and a second absorber at the host
123: redshift. Where the count rate was high, regions with the PSF-core excised
124: were used for spectral analysis to account for pile-up.
125:
126: To ensure that only statistically significant excess absorption columns were
127: selected, two conservative criteria were applied. 1) GRBs at low Galactic
128: latitude ($|b|<20\degr$) were excluded. 2) The excess column density had to
129: be detected above the 99.7\% confidence level when the $z=0$ absorber was
130: fixed at $1\times10^{20}$cm$^{-2}$ or 20\% above the Galactic value,
131: whichever was larger
132: \citep[see][]{1986ApJ...310..291E,1989AJ.....97..777E,1990ARA&A..28..215D}.
133: Galactic column densities obtained from the newer \ion{H}{1} survey of
134: \citet{2005A&A...440..775K} were consistent with those obtained from
135: \citet{1990ARA&A..28..215D} within $\sim3\times10^{19}$\,cm$^{-2}$. At early
136: times an absorbed power-law model fit occasionally yields excess absorption
137: not present at later times
138: \citep[e.g.][]{2005A&A...442L..21S,2007A&A...462..565G,2007ApJ...654L..17C}.
139: \citet{2006astro.ph.12564B} have examined the reasons behind this and we
140: discuss it in \sect\ref{discussion}. Because of this effect we adopt the
141: lower value of absorption where inconsistent values were found at early and
142: later times. GRBs\,050730, 050820A, and 060714 are the bursts
143: affected by this, and we have adopted absorption values from the later PC
144: mode spectra for these bursts.
145:
146: The spectra of GRBs\,050401, 050505, 050730 and 050904 have been presented
147: elsewhere
148: \citep{2006ApJ...652.1011W,2006MNRAS.365.1031D,2006MNRAS.368.1101H,2005A&A...442L..21S,2006ApJ...637L..69W,2007A&A...462..565G,2007ApJ...654L..17C}.
149: For the bursts with persistent excess absorptions published
150: previously---050319 \citep{2006A&A...449...61C}, 050401, 050505, and
151: 050904---we obtain somewhat different values. This may be due to the more
152: recent calibration used in this analysis. However, in each case, the
153: authors claim a detection of excess absorption and adopting those values do
154: not change our conclusions.
155:
156:
157: \section{Results}\label{results}
158:
159: Seventeen GRBs enter our sample as set out in Table~\ref{tab:columns}. The comparison of
160: X-ray and and hydrogen Ly$\alpha$ absorption is shown in
161: Fig.~\ref{fig:xrayoptical}. If the metallicities of GRB formation sites are
162: low
163: \citep{2003A&A...400..499L,2003A&A...406L..63F,2006Natur.441..463F,2007astro.ph..1246M,2007AJ....133..882K},
164: we would expect to make no significant detections of excess absorption in
165: this sample. Instead, eight GRBs have significant excess
166: column densities: GRBs\,050319, 050401, 050505, 050904, 060210, 060607A,
167: 060714, and 060926 (see Table~\ref{tab:columns} and
168: Fig.~\ref{fig:xrayoptical}).
169:
170: \begin{table}
171: \caption{UV and X-ray estimates of the absorbing column densities in GRBs}
172: \label{tab:columns}
173: \setlength{\tabcolsep}{6pt}
174: \begin{center}
175: \begin{tabular}{@{}lccccc@{}}
176: \hline\hline
177: GRB & $z^a$ & $\log{N_{\rm H\,{\sc I}}}^a$ & Total $N_{\rm H}$ & Gal.\ $N_{\rm H}$ & $N_{\rm H}$ at $z$ \\
178: & & & at $z=0$ & & \\
179: & & & \multicolumn{3}{c}{($10^{20}$\,cm$^{-2}$)} \\
180: \hline
181: 050319 & 3.24 & $20.9$ & $5.1^{1.1}_{-1.0}$ & $1.1$ & $76^{+20}_{-18}$ \\ %1 %%Numbers from us. Campana et al. published slightly lower values
182: 050401 & 2.90 & $22.6$ & $16.9\pm0.8$ & $4.8$ & $196\pm13$ \\ %2 %%Total nh at z=0 estimated from the other values in our paper. Also estimated from simulations and gave same results.
183: 050505 & 4.27 & $22.1$ & $9.8\pm1.0$ & $2.0$ & $130\pm36$ \\ %3 %%Got data for 050505. !!REMEMBERED PILE-UP!! No WT!?
184: 050730 & 3.97 & $22.1$ & $3$ & $3.05$ & $<100$ \\ %5 %%Got from Rhaana's paper.
185: 050820A & 2.61 & $21.1$ & $4.0\pm1.0$ & $4.71$ & $<178$ \\ %6
186: 050904 & 6.30 & $21.3$ & $9.1\pm0.6$ & $4.9$ & $380\pm50$ \\ %7
187: 050908 & 3.34 & $19.2$ & $2.2\pm1.5$ & $2.1$ & $<122$ \\ %8 %%Very little data. Dominated by flares. No excess.
188: 050922C & 2.20 & $21.6$ & $6.5^b$ & $5.75$ & $<55^c$ \\ %9 %%*From Swift Table
189: 060206 & 4.05 & $20.9$ & $5.2^{+1.4}_{-1.3}$ & $0.9$ & $<280$ \\ %13 %%Did PC data for 060206 (no pileup correction done for PC). (Late starting observation?)
190: 060210 & 3.91 & $21.7$ & $15.2\pm0.5$ & $8.5$ & $169\pm18$ \\ %14 %%!!REMEMBERED PILE-UP!! Very close to b=20!
191: 060522 & 5.11 & $20.5$ & $4.6^b$ & $4.83$ & $170^c$ \\ %17 %%*From Swift Table
192: 060526 & 3.22 & $20.0$ & $6^b$ & $5.51$ & $<84^c$ \\ %18 %%*From Swift Table
193: 060607A & 3.08 & $<19.5$ & $5.6\pm0.4$ & $2.7$ & $55\pm7$ \\ %20
194: 060707 & 3.43 & $21.0$ & $1.8^b$ & $1.76$ & $<90^c$ \\ %21 %%*From Swift Table
195: 060714 & 2.71 & $21.8$ & $13.6^{+2.0}_{-1.9}$ & $6.7$ & $94^{+27}_{-25}$ \\ %22 %%NO pile-up problem... but variations in nH
196: 060906 & 3.69 & $21.9$ & $9.66^b$ & $9.66$ & $<100^c$ \\ %23 %%*From Swift Table
197: 060926 & 3.21 & $22.7$ & $20.5^{+4.8}_{-4.2}$ & $7.31$ & $250^{+110}_{-90}$ \\%25 %%
198: \hline
199: \end{tabular}
200: \end{center}
201: $^a$ See \citet{2006A&A...460L..13J} and references therein.
202:
203: $^b$ From \url[http://swift.gsfc.nasa.gov/docs/swift/archive/grb_table/]{http://swift.gsfc.nasa.gov/docs/swift/archive/grb\_table/}
204:
205: $^c$ Estimated using a redshift-corrected, $3\times10^{20}$ uncertainty in the
206: X-ray column and a $1\times10^{20}$ uncertainty in the Galactic column
207: density.
208: \end{table}
209:
210:
211:
212: \begin{figure}
213: \includegraphics[angle=-90,bb=63 54 559 386,clip=,width=\columnwidth]{f1.eps}
214: \caption{Neutral hydrogen column density (from the UV) as a function of
215: X-ray equivalent hydrogen column density. The ratio of these
216: values provides an estimate of the metal abundance of the
217: absorbing medium. The clear lack of correlation between these
218: column densities seems initially to indicate an extremely strong
219: variation in the metal enrichment of the absorbing clouds near
220: GRBs. These values are so extreme however, and many so much
221: greater than solar, that it seems likely that there are strong
222: systematic uncertainties related to either or both of the
223: absorbing column density estimates from X-rays and the UV,
224: possibly because of ionisation of most of the line-of-sight
225: hydrogen. Data from absorption toward a Galactic source (the Crab
226: pulsar) and a high column density DLA (foreground to the blazar
227: AO\,0235+164) are shown for comparison. Where the uncertainties on
228: the \ion{H}{1} column is not in the literature a value of 0.1 has
229: been plotted.}
230: \label{fig:xrayoptical}
231: \end{figure}
232:
233: The best limits that can be placed on the column density from
234: \emph{Swift}-XRT data above $z=2$ is roughly $\log{N_{\rm H}}\lesssim 21.5$.
235: Fig.~\ref{fig:xrayoptical} is therefore not populated below 21.5 in X-ray column.
236: Such a limit could only constrain the metal abundance to be less than solar
237: in the highest \ion{H}{1} column density cases.
238: Upper limits obtained in the optical/UV are more interesting: GRB\,060607A
239: has a tightly constrained, low \ion{H}{1} column density, but large
240: X-ray absorption. (The same is true for GRB\,060124,
241: but it does not enter our sample at $b=17\degr$.)
242:
243: There is evidence of a spread in the metallicities of GRB hosts, though
244: with most fairly close to $\sim10$\% of the solar value \citep[see][]{2006A&A...451L..47F}. If the metallicities of GRB
245: environments were all approximately similar we would expect some correlation
246: between the optical and X-ray column density determinations. It is clear
247: from Fig.~\ref{fig:xrayoptical} that there is no such correlation and that
248: the optical-to-X-ray ratios span a range of at least a few hundred.
249: Furthermore, the metal abundances derived by this comparison indicate
250: metallicities far above the solar values. The extreme abundances and the
251: lack of correlation between the optical and X-ray values are disturbing and
252: strongly imply that the X-ray absorption and the Ly$\alpha$ absorption are
253: sampling different environments. These data immediately show that for GRBs
254: the X-ray-to-optical absorption ratio is not a useful measure of the
255: metallicity.
256:
257: \subsection{Reliability of the measurements}
258: Metal abundances have been determined for intervening extragalactic
259: absorbers in a few radio-loud AGN using X-ray and Ly$\alpha$ absorption.
260: Those values are significantly below solar
261: \citep[see][]{2003ApJ...590..730T}. The metallicities are consistent with
262: extinction measures and expectations for these objects.
263: For the Crab
264: pulsar, this technique yields values consistent with solar
265: metallicity \citep{2000ApJ...537..861S,2001A&A...365L.212W}. Both these
266: results are plotted for comparison on Fig.~\ref{fig:xrayoptical} and
267: indicate that in general the comparison of column densities from X-ray
268: and Ly$\alpha$ is valid for both Galactic and extragalactic sources.
269:
270: GRB afterglows exhibit curvature close to the cooling frequency, which may
271: be at X-ray wavelengths at certain times and may mimic absorption.
272: \citet{2006astro.ph.12564B} have shown that in many cases where the soft
273: X-ray absorption apparently changes rapidly in the first few hundred second
274: after the burst, this may be better explained by curvature of the
275: continuum than ionisation of the absorber. This seems a reasonable
276: explanation in the early phases of some bursts. Such early curvature
277: however, should not affect the results presented here, as we have used
278: absorption values from later spectra where early (high) values of the
279: absorption are inconsistent with later data. Such an approach is
280: conservative, in the sense that it leads to lower column
281: densities. In general, spectral curvature is not responsible for the
282: detection of excess absorption. Most of the bursts in question have early WT
283: and PC mode data that agree on the absorption to within 1$\sigma$
284: despite large spectral changes.
285:
286: Given that the technique is reliable, could it be that the Galactic column
287: densities are underestimated? This explanation can be ruled out. First of
288: all we have excluded any excess column density that is not significant above
289: the 90\% variations of the Galactic columns
290: \citep{1986ApJ...310..291E,1989AJ.....97..777E,1990ARA&A..28..215D}.
291: Furthermore solar abundances were used to convert the Galactic 21\,cm
292: column densities \citep{1990ARA&A..28..215D}, resulting in a high
293: (conservative) estimate of the foreground Galactic column, higher
294: than is often argued for the Galactic gas by e.g.\
295: \citet{2000ApJ...542..914W}. A large underestimate of the Galactic
296: column density can also be excluded on the basis of observations of blazars with \emph{BeppoSAX}
297: \citep{2005A&A...433.1163D} that show no substantial absorption compared to
298: the values of \citet{1990ARA&A..28..215D}. Of nearly 90 blazars examined
299: with \emph{BeppoSAX} only 17 showed excess absorption in the
300: observed 0.1--50\,keV spectra and even these excesses are small,
301: $\sim1\times10^{20}$\,cm$^{-2}$.
302: Finally, if the excess columns were due to Galactic foreground
303: absorption, we would expect a systematic rise in observed column
304: densities with redshift, an effect we do not observe when we combine with
305: the lower-redshift values of \citet{2006A&A...449...61C}.
306:
307:
308:
309: \section{Discussion}\label{discussion}
310:
311: It is only with the success of \emph{Swift} that we have been
312: able to obtain Ly$\alpha$ and X-ray absorption measures in the same bursts.
313: Given the expectations for neutral hydrogen column densities surrounding
314: GRBs \citep{2002ApJ...565..174R}, few significant detections of excess
315: absorption in the X-ray afterglows of GRBs should have been expected above
316: $z\gtrsim2$.
317:
318:
319: \subsection{Intervening absorbers}
320:
321: It is expected that high redshift sources \citep[and GRBs have a very high
322: mean redshift,][]{2006A&A...447..897J} will often have intervening line of
323: sight absorbers. The X-ray absorption does not tell us about the redshift of
324: the absorber and some absorption could in principle be related to low-$z$
325: systems. The observed absorption in X-rays drops substantially as the
326: absorber is moved to higher redshift, so for this to be an effective
327: explanation any absorber would have to be at relatively low redshift.
328: However, very large column densities of metals would routinely be required
329: along most lines of sight at low redshifts. Recently,
330: \citet{2006ApJ...648L..93P} have shown that GRBs have intervening large
331: column density (W$_\lambda>1$\,\AA) \ion{Mg}{2} absorbers about four times
332: more frequently than QSOs. And it is a suggestive coincidence that GRBs also
333: have significant excess soft X-ray absorptions. However, to explain the
334: results observed here with intervening absorbers is not possible. This would
335: require an absorber with typically $\log(N_{\rm H})\simeq22.1$ at $z=0.3$
336: (assuming $Z/Z_\odot=0.1$) in half of the GRB sightlines---an extremely
337: large value. Assuming that an \ion{Mg}{2} absorber with
338: W$_\lambda\sim1-3$\,\AA\ corresponds to $\log(N_{\rm H})\lesssim21$ at this
339: metallicity \citep{2007MNRAS.tmp..112M}, the required absorption is roughly
340: two orders of magnitude above that provided by the \ion{Mg}{2} absorbers
341: seen in GRBs at $z\lesssim0.5$ \citep{2006ApJ...648L..93P}. To reinforce
342: this point, a similar excess of large equivalent width \ion{Mg}{2} absorbers
343: is also found in the foregrounds of blazars \citep{1997ApJ...489L..17S},
344: which, as noted above, do not have large X-ray absorptions
345: \citep{2005A&A...433.1163D}. This is an important difference in that it
346: seems likely that GRBs and blazars sample similar column densities of
347: intervening absorbers, but GRBs typically have large soft X-ray absorptions
348: and blazars do not,
349: showing that the large absorptions are indeed intrinsic
350: to the GRB hosts.
351:
352: It has been suggested \citep{2006A&A...460L..13J} that the lack of very high
353: column density and apparent overabundance of low column density
354: sources in comparison to model predictions may be due to ionisation of the
355: hydrogen near the GRB. This hypothesis explains the results found here; the
356: hydrogen may be ionised to large distances by the early afterglow, but
357: the metals, while also ionised, will not be stripped to such
358: an extent that they cease substantially to absorb the soft X-rays. This
359: hypothesis is also consistent with the lack of expected Wolf-Rayet features
360: in most GRB optical afterglows \citep{2006astro.ph.11079C}.
361:
362:
363: \subsection{Size of the gas cloud}
364:
365: It is possible to limit the size of the gas cloud where most of the hydrogen
366: has been ionised by the GRB. A maximum distance to the gas is found by
367: assuming the X-ray column density represents the total column density
368: (it cannot obviously be smaller than this, and if it is larger, then the
369: ionisation radius will be smaller) and that each ionising UV photon is
370: intercepted by a hydrogen atom (if this is not the case, the radius will
371: again, necessarily be smaller). Since we know the total UV fluence for
372: most of these bursts is a small fraction of the prompt energy, we can
373: derive a maximum distance to which the GRB could have ionised the gas. In
374: all cases, this is less than $\sim3$\,pc. Interestingly, if the gas cloud is
375: already ionised by the massive stars in the region, the typical densities of
376: \ion{H}{2} regions ($\lesssim10^4$\,cm$^{-3}$) yield sizes which are also at
377: most a few parsecs in radius. Smaller radii require higher densities. Radii
378: an order of magnitude greater than this make the total masses unfeasibly
379: large. This radius is similar to that obtained by
380: \citet{2007ApJ...654L..17C} based on the assumption that the apparent
381: decreasing absorption detected in GRB\,050904 is due to ionisation by the
382: GRB.
383:
384: Observations of UV absorption lines have allowed distances to low-ionisation
385: gas to be derived that are far larger than the maximum distance to the
386: majority of the gas column we derive above. \citet{2006ApJ...648...95P} find
387: $\gtrsim50$\,pc for \ion{Mg}{1} in GRB\,050111, and
388: \citet{2006astro.ph.11690V} find $1.7\pm0.2$\,kpc for \ion{Ni}{2} and
389: \ion{Fe}{2} in GRB\,060418. Gas with high ionisation states closer to the
390: GRB has been detected in the optical \citep{2006astro.ph..9825D}. Though
391: 1.7\,kpc seems large to be the typical distance scale between clouds in most
392: bursts, taken together, those results support our conclusion in this paper,
393: that the optical/UV spectra are probing gas that is outside the pc-scale
394: structure in which the GRB explodes.
395:
396:
397: \section{Conclusions\label{conclusions}}
398:
399: We have examined the X-ray afterglow spectra of the sample of \emph{Swift}
400: GRBs at $z>2$ as defined by \citet{2006A&A...460L..13J} and compared their
401: extragalactic soft X-ray absorbing column densities to the host \ion{H}{1}
402: column densities. We find no correlation between the column density values,
403: and the X-ray absorptions often far exceed the \ion{H}{1} column densities.
404: The most likely explanation for this discrepancy is that GRBs substantially
405: ionise the hydrogen in their immediate environments (at least along the line
406: of sight). This may be consistent with the sublimation of the dust that is
407: expected (by analogy with the local universe) to be associated with large
408: X-ray column densities, though it does not provide evidence in favour of
409: dust sublimation. It seems likely that significant information about the
410: immediate circumburst environment may be gleaned from high-resolution X-ray
411: spectroscopy of the very early afterglow, probably revealing the complex
412: absorption pattern associated with metal-rich warm absorbers, as observed in
413: some AGN. Finally, we find the maximum distance for the bulk of the
414: absorbing material surrounding the GRBs to be $\sim3$\,pc for these bursts,
415: similar to the sizes of compact \ion{H}{2} regions. Since the distances
416: inferred from optical observations are typically larger than this, it
417: implies that the optical/UV spectroscopy of GRB afterglows typically probes
418: environments that are little affected by the progenitor, the explosion or
419: the bright afterglow. This is particularly encouraging for the use of GRBs
420: as probes of star-forming regions at high redshift.
421:
422:
423: \begin{acknowledgements}
424: We thank Rhaana Starling, Tamara Davis, Nat Butler and Enrico Ramirez-Ruiz
425: for useful discussions. The Dark Cosmology Centre is funded by the DNRF.
426: \end{acknowledgements}
427:
428:
429:
430: \begin{thebibliography}{44}
431: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
432:
433: \bibitem[{{Butler} \& {Kocevski}(2006)}]{2006astro.ph.12564B}
434: {Butler}, N.~R., \& {Kocevski}, D. 2006, ArXiv:astro-ph/0612564
435:
436: \bibitem[{{Campana} {et~al.}(2007){Campana}, {Lazzati}, {Ripamonti}, {Perna},
437: {Covino}, {Tagliaferri}, {Moretti}, {Romano}, {Cusumano}, \&
438: {Chincarini}}]{2007ApJ...654L..17C}
439: {Campana}, S., et al. 2007, ApJ, 654, L17
440:
441: \bibitem[{{Campana} {et~al.}(2006){Campana}, {Romano}, {Covino}, {Lazzati}, {de
442: Luca}, {Chincarini}, {Moretti}, {Tagliaferri}, {Cusumano}, {Giommi},
443: {Mangano}, {Perri}, {La Parola}, {Capalbi}, {Mineo}, {Antonelli}, {Burrows},
444: {Hill}, {Racusin}, {Kennea}, {Morris}, {Pagani}, {Nousek}, {Osborne}, {Goad},
445: {Page}, {Beardmore}, {Godet}, {O'Brien}, {Wells}, {Angelini}, \&
446: {Gehrels}}]{2006A&A...449...61C}
447: {Campana}, S., et al. 2006,
448: A\&A, 449, 61
449:
450: \bibitem[{{Chen} {et~al.}(2006){Chen}, {Prochaska}, {Bloom}, {Ramirez-Ruiz},
451: {Dessauges-Zavadsky}, \& {Foley}}]{2006astro.ph.11079C}
452: {Chen}, H.-W., {Prochaska}, J.~X., {Bloom}, J.~S., {Ramirez-Ruiz}, E.,
453: {Dessauges-Zavadsky}, M., \& {Foley}, R.~J. 2006, ArXiv:astro-ph/0611079
454:
455: \bibitem[{{Chen} {et~al.}(2005){Chen}, {Prochaska}, {Bloom}, \&
456: {Thompson}}]{2005ApJ...634L..25C}
457: {Chen}, H.-W., {Prochaska}, J.~X., {Bloom}, J.~S., \& {Thompson}, I.~B. 2005,
458: ApJ, 634, L25
459:
460: \bibitem[{{De Pasquale} {et~al.}(2006){De Pasquale}, {Beardmore}, {Barthelmy},
461: {Boyd}, {Burrows}, {Fink}, {Gehrels}, {Kobayashi}, {Mason}, {McNought},
462: {Nousek}, {Page}, {Palmer}, {Peterson}, {Price}, {Rich}, {Roming}, {Rosen},
463: {Sakamoto}, {Schmidt}, {Tueller}, {Wells}, {Zane}, {Zhang}, \&
464: {Ziaeepour}}]{2006MNRAS.365.1031D}
465: {De Pasquale}, M., et al. 2006, MNRAS, 365, 1031
466:
467: \bibitem[{{D'Elia} {et~al.}(2006){D'Elia}, {Fiore}, {Meurs}, {Chincarini},
468: {Melandri}, {Norci}, {Pellizza}, {Perna}, {Piranomonte}, {Sbordone},
469: {Stella}, {Tagliaferri}, {Vergani}, {Ward}, {Angelini}, {Antonelli},
470: {Burrows}, {Campana}, {Capalbi}, {Caraveo}, {Cimatti}, {Costa}, {Cusumano},
471: {D'Avanzo}, {Della Valle}, {Filliatre}, {Frontera}, {Fugazza}, {Gehrels},
472: {Giannini}, {Giommi}, {Goldoni}, {Guetta}, {Israel}, {Lazzati}, {Malesani},
473: {Marconi}, {Mason}, {Mereghetti}, {Molinari}, {Moretti}, {Nousek}, {Perri},
474: {Piro}, {Stratta}, {Testa}, \& {Vietri}}]{2006astro.ph..9825D}
475: {D'Elia}, V., et al. 2006, ArXiv:
476: astro-ph/0609825
477:
478: \bibitem[{{Dickey} \& {Lockman}(1990)}]{1990ARA&A..28..215D}
479: {Dickey}, J.~M., \& {Lockman}, F.~J. 1990, ARA\&A, 28, 215
480:
481: \bibitem[{{Donato} {et~al.}(2005){Donato}, {Sambruna}, \&
482: {Gliozzi}}]{2005A&A...433.1163D}
483: {Donato}, D., {Sambruna}, R.~M., \& {Gliozzi}, M. 2005, A\&A, 433, 1163
484:
485: \bibitem[{{Elvis} {et~al.}(1986){Elvis}, {Green}, {Bechtold}, {Schmidt},
486: {Neugebauer}, {Soifer}, {Matthews}, \& {Fabbiano}}]{1986ApJ...310..291E}
487: {Elvis}, M., {Green}, R.~F., {Bechtold}, J., {Schmidt}, M., {Neugebauer}, G.,
488: {Soifer}, B.~T., {Matthews}, K., \& {Fabbiano}, G. 1986, ApJ, 310, 291
489:
490: \bibitem[{{Elvis} {et~al.}(1989){Elvis}, {Wilkes}, \&
491: {Lockman}}]{1989AJ.....97..777E}
492: {Elvis}, M., {Wilkes}, B.~J., \& {Lockman}, F.~J. 1989, AJ, 97, 777
493:
494: \bibitem[{{Fruchter} {et~al.}(2006){Fruchter}, {Levan}, {Strolger},
495: {Vreeswijk}, {Thorsett}, {Bersier}, {Burud}, {Castro Cer{\'o}n},
496: {Castro-Tirado}, {Conselice}, {Dahlen}, {Ferguson}, {Fynbo}, {Garnavich},
497: {Gibbons}, {Gorosabel}, {Gull}, {Hjorth}, {Holland}, {Kouveliotou}, {Levay},
498: {Livio}, {Metzger}, {Nugent}, {Petro}, {Pian}, {Rhoads}, {Riess}, {Sahu},
499: {Smette}, {Tanvir}, {Wijers}, \& {Woosley}}]{2006Natur.441..463F}
500: {Fruchter}, A.~S., et al. 2006, Nat, 441, 463
501:
502: \bibitem[{{Fynbo} {et~al.}(2001){Fynbo}, {Gorosabel}, {Dall}, {Hjorth},
503: {Pedersen}, {Andersen}, {M{\o}ller}, {Holland}, {Smail}, {Kobayashi}, {Rol},
504: {Vreeswijk}, {Burud}, {Jensen}, {Thomsen}, {Henden}, {Vrba}, {Canzian},
505: {Castro Cer{\'o}n}, {Castro-Tirado}, {Cline}, {Goto}, {Greiner}, {Hanski},
506: {Hurley}, {Lund}, {Pursimo}, {{\O}stensen}, {Solheim}, {Tanvir}, \&
507: {Terada}}]{2001A&A...373..796F}
508: {Fynbo}, J.~P.~U., et al. 2001, A\&A, 373, 796
509:
510: \bibitem[{{Fynbo} {et~al.}(2003){Fynbo}, {Jakobsson}, {M{\o}ller}, {Hjorth},
511: {Thomsen}, {Andersen}, {Fruchter}, {Gorosabel}, {Holland}, {Ledoux},
512: {Pedersen}, {Rhoads}, {Weidinger}, \& {Wijers}}]{2003A&A...406L..63F}
513: {Fynbo}, J.~P.~U., et al. 2003, A\&A, 406, L63
514:
515: \bibitem[{{Fynbo} {et~al.}(2006){Fynbo}, {Starling}, {Ledoux}, {Wiersema},
516: {Th{\"o}ne}, {Sollerman}, {Jakobsson}, {Hjorth}, {Watson}, {Vreeswijk},
517: {M{\o}ller}, {Rol}, {Gorosabel}, {N{\"a}r{\"a}nen}, {Wijers},
518: {Bj{\"o}rnsson}, {Castro Cer{\'o}n}, {Curran}, {Hartmann}, {Holland},
519: {Jensen}, {Levan}, {Limousin}, {Kouveliotou}, {Nelemans}, {Pedersen},
520: {Priddey}, \& {Tanvir}}]{2006A&A...451L..47F}
521: {Fynbo}, J.~P.~U., et al. 2006,
522: A\&A, 451, L47
523:
524: \bibitem[{{Galama} \& {Wijers}(2001)}]{2001ApJ...549L.209G}
525: {Galama}, T.~J., \& {Wijers}, R.~A.~M.~J. 2001, ApJ, 549, L209
526:
527: \bibitem[{{Gendre} {et~al.}(2006){Gendre}, {Corsi}, \&
528: {Piro}}]{2006A&A...455..803G}
529: {Gendre}, B., {Corsi}, A., \& {Piro}, L. 2006, A\&A, 455, 803
530:
531: \bibitem[{{Gendre} {et~al.}(2007){Gendre}, {Galli}, {Corsi}, {Klotz}, {Piro},
532: {Stratta}, {Bo{\"e}r}, \& {Damerdji}}]{2007A&A...462..565G}
533: {Gendre}, B., {Galli}, A., {Corsi}, A., {Klotz}, A., {Piro}, L., {Stratta}, G.,
534: {Bo{\"e}r}, M., \& {Damerdji}, Y. 2007, A\&A, 462, 565
535:
536: \bibitem[{{Hjorth} {et~al.}(2003){Hjorth}, {M{\o}ller}, {Gorosabel}, {Fynbo},
537: {Toft}, {Jaunsen}, {Kaas}, {Pursimo}, {Torii}, {Kato}, {Yamaoka}, {Yoshida},
538: {Thomsen}, {Andersen}, {Burud}, {Castro Cer{\' o}n}, {Castro-Tirado},
539: {Fruchter}, {Kaper}, {Kouveliotou}, {Masetti}, {Palazzi}, {Pedersen}, {Pian},
540: {Rhoads}, {Rol}, {Tanvir}, {Vreeswijk}, {Wijers}, \& {van den
541: Heuvel}}]{2003ApJ...597..699H}
542: {Hjorth}, J., et al. 2003, ApJ, 597, 699
543:
544: \bibitem[{{Hurkett} {et~al.}(2006){Hurkett}, {Osborne}, {Page}, {Rol}, {Goad},
545: {O'Brien}, {Beardmore}, {Godet}, {Burrows}, {Tanvir}, {Levan}, {Zhang},
546: {Malesani}, {Hill}, {Kennea}, {Chapman}, {Parola}, {Perri}, {Romano},
547: {Smith}, \& {Gehrels}}]{2006MNRAS.368.1101H}
548: {Hurkett}, C.~P., et al. 2006, MNRAS, 368, 1101
549:
550: \bibitem[{{Jakobsson} {et~al.}(2004){Jakobsson}, {Hjorth}, {Fynbo},
551: {Weidinger}, {Gorosabel}, {Ledoux}, {Watson}, {Bj{\" o}rnsson},
552: {Gudmundsson}, {Wijers}, {M{\" o}ller}, {Pedersen}, {Sollerman}, {Henden},
553: {Jensen}, {Gilmore}, {Kilmartin}, {Levan}, {Castro Cer{\' o}n},
554: {Castro-Tirado}, {Fruchter}, {Kouveliotou}, {Masetti}, \&
555: {Tanvir}}]{2004A&A...427..785J}
556: {Jakobsson}, P., et al. 2004, A\&A, 427, 785
557:
558: \bibitem[{{Jakobsson} {et~al.}(2006{\natexlab{a}}){Jakobsson}, {Fynbo},
559: {Ledoux}, {Vreeswijk}, {Kann}, {Hjorth}, {Priddey}, {Tanvir}, {Reichart},
560: {Gorosabel}, {Klose}, {Watson}, {Sollerman}, {Fruchter}, {de Ugarte Postigo},
561: {Wiersema}, {Bj{\"o}rnsson}, {Chapman}, {Th{\"o}ne}, {Pedersen}, \&
562: {Jensen}}]{2006A&A...460L..13J}
563: {Jakobsson}, P., et al. 2006{\natexlab{a}}, A\&A, 460, L13
564:
565: \bibitem[{{Jakobsson} {et~al.}(2006{\natexlab{b}}){Jakobsson}, {Levan},
566: {Fynbo}, {Priddey}, {Hjorth}, {Tanvir}, {Watson}, {Jensen}, {Sollerman},
567: {Natarajan}, {Gorosabel}, {Castro Cer{\'o}n}, {Pedersen}, {Pursimo},
568: {{\'A}rnad{\'o}ttir}, {Castro-Tirado}, {Davis}, {Deeg}, {Fiuza},
569: {Mykolaitis}, \& {Sousa}}]{2006A&A...447..897J}
570: {Jakobsson}, P., et al. 2006{\natexlab{b}}, A\&A, 447, 897
571:
572: \bibitem[{{Jensen} {et~al.}(2001){Jensen}, {Fynbo}, {Gorosabel}, {Hjorth},
573: {Holland}, {M{\" o}ller}, {Thomsen}, {Bj{\" o}rnsson}, {Pedersen}, {Burud},
574: {Henden}, {Tanvir}, {Davis}, {Vreeswijk}, {Rol}, {Hurley}, {Cline},
575: {Trombka}, {McClanahan}, {Starr}, {Goldsten}, {Castro-Tirado}, {Greiner},
576: {Bailer-Jones}, {K{\" u}mmel}, \& {Mundt}}]{2001A&A...370..909J}
577: {Jensen}, B.~L., et al. 2001, A\&A, 370, 909
578:
579: \bibitem[{{Kalberla} {et~al.}(2005){Kalberla}, {Burton}, {Hartmann}, {Arnal},
580: {Bajaja}, {Morras}, \& {P{\"o}ppel}}]{2005A&A...440..775K}
581: {Kalberla}, P.~M.~W., {Burton}, W.~B., {Hartmann}, D., {Arnal}, E.~M.,
582: {Bajaja}, E., {Morras}, R., \& {P{\"o}ppel}, W.~G.~L. 2005, A\&A, 440, 775
583:
584: \bibitem[{{Kewley} {et~al.}(2007){Kewley}, {Brown}, {Geller}, {Kenyon}, \&
585: {Kurtz}}]{2007AJ....133..882K}
586: {Kewley}, L.~J., {Brown}, W.~R., {Geller}, M.~J., {Kenyon}, S.~J., \& {Kurtz},
587: M.~J. 2007, AJ, 133, 882
588:
589: \bibitem[{{Le Floc'h} {et~al.}(2003){Le Floc'h}, {Duc}, {Mirabel}, {Sanders},
590: {Bosch}, {Diaz}, {Donzelli}, {Rodrigues}, {Courvoisier}, {Greiner},
591: {Mereghetti}, {Melnick}, {Maza}, \& {Minniti}}]{2003A&A...400..499L}
592: {Le Floc'h}, E., et al. 2003, A\&A, 400, 499
593:
594: \bibitem[{{Meiring} {et~al.}(2007){Meiring}, {Lauroesch}, {Kulkarni},
595: {P\'eroux}, {Khare}, {York}, \& {Crotts}}]{2007MNRAS.tmp..112M}
596: {Meiring}, J.~D., {Lauroesch}, J.~T., {Kulkarni}, V.~P., {P{\'e}roux}, C.,
597: {Khare}, P., {York}, D.~G.,
598: \& {Crotts}, A.~P.~S.\ 2007, MNRAS, 112
599:
600: \bibitem[{{Modjaz} {et~al.}(2007){Modjaz}, {Kewley}, {Kirshner}, {Stanek},
601: {Challis}, {Garnavich}, {Greene}, \& {Prieto}}]{2007astro.ph..1246M}
602: {Modjaz}, M., {Kewley}, L., {Kirshner}, R.~P., {Stanek}, K.~Z., {Challis}, P.,
603: {Garnavich}, P.~M., {Greene}, J.~E., \& {Prieto}, J.~L. 2007,
604: ArXiv:astro-ph/0701246
605:
606: \bibitem[{{Prochaska} {et~al.}(2006){Prochaska}, {Chen}, \&
607: {Bloom}}]{2006ApJ...648...95P}
608: {Prochaska}, J.~X., {Chen}, H.-W., \& {Bloom}, J.~S. 2006, ApJ, 648, 95
609:
610: \bibitem[{{Prochter} {et~al.}(2006){Prochter}, {Prochaska}, {Chen}, {Bloom},
611: {Dessauges-Zavadsky}, {Foley}, {Lopez}, {Pettini}, {Dupree}, \&
612: {Guhathakurta}}]{2006ApJ...648L..93P}
613: {Prochter}, G.~E., et al. 2006, ApJ, 648, L93
614:
615: \bibitem[{{Reichart} \& {Price}(2002)}]{2002ApJ...565..174R}
616: {Reichart}, D.~E., \& {Price}, P.~A. 2002, ApJ, 565, 174
617:
618: \bibitem[{{Sollerman} {et~al.}(2000){Sollerman}, {Lundqvist}, {Lindler},
619: {Chevalier}, {Fransson}, {Gull}, {Pun}, \& {Sonneborn}}]{2000ApJ...537..861S}
620: {Sollerman}, J., {Lundqvist}, P., {Lindler}, D., {Chevalier}, R.~A.,
621: {Fransson}, C., {Gull}, T.~R., {Pun}, C.~S.~J., \& {Sonneborn}, G. 2000, ApJ,
622: 537, 861
623:
624: \bibitem[{{Starling} {et~al.}(2005){Starling}, {Vreeswijk}, {Ellison}, {Rol},
625: {Wiersema}, {Levan}, {Tanvir}, {Wijers}, {Tadhunter}, {Zaurin}, {Gonzalez
626: Delgado}, \& {Kouveliotou}}]{2005A&A...442L..21S}
627: {Starling}, R.~L.~C., et al. 2005, A\&A, 442, L21
628:
629: \bibitem[{{Stocke} \& {Rector}(1997)}]{1997ApJ...489L..17S}
630: {Stocke}, J.~T., \& {Rector}, T.~A. 1997, ApJ, 489, L17
631:
632: \bibitem[{{Stratta} {et~al.}(2004){Stratta}, {Fiore}, {Antonelli}, {Piro}, \&
633: {De Pasquale}}]{2004ApJ...608..846S}
634: {Stratta}, G., {Fiore}, F., {Antonelli}, L.~A., {Piro}, L., \& {De Pasquale},
635: M. 2004, ApJ, 608, 846
636:
637: \bibitem[{{Turnshek} {et~al.}(2003){Turnshek}, {Rao}, {Ptak}, {Griffiths}, \&
638: {Monier}}]{2003ApJ...590..730T}
639: {Turnshek}, D.~A., {Rao}, S.~M., {Ptak}, A.~F., {Griffiths}, R.~E., \&
640: {Monier}, E.~M. 2003, ApJ, 590, 730
641:
642: \bibitem[{{Vreeswijk} {et~al.}(2004){Vreeswijk}, {Ellison}, {Ledoux}, {Wijers},
643: {Fynbo}, {M{\o}ller}, {Henden}, {Hjorth}, {Masi}, {Rol}, {Jensen}, {Tanvir},
644: {Levan}, {Castro Cer{\' o}n}, {Gorosabel}, {Castro-Tirado}, {Fruchter},
645: {Kouveliotou}, {Burud}, {Rhoads}, {Masetti}, {Palazzi}, {Pian}, {Pedersen},
646: {Kaper}, {Gilmore}, {Kilmartin}, {Buckle}, {Seigar}, {Hartmann}, {Lindsay},
647: \& {van den Heuvel}}]{2004A&A...419..927V}
648: {Vreeswijk}, P.~M., et al. 2004, A\&A, 419, 927
649:
650: \bibitem[{{Vreeswijk} {et~al.}(2006){Vreeswijk}, {Ledoux}, {Smette}, {Ellison},
651: {Jaunsen}, {Andersen}, {Fruchter}, {Fynbo}, {Hjorth}, {Kaufer}, {Moller},
652: {Petitjean}, {Savaglio}, \& {Wijers}}]{2006astro.ph.11690V}
653: {Vreeswijk}, P.~M., et al. 2006, ArXiv: astro-ph/0611690
654:
655: \bibitem[{{Watson} {et~al.}(2006{\natexlab{a}}){Watson}, {Fynbo}, {Ledoux},
656: {Vreeswijk}, {Hjorth}, {Smette}, {Andersen}, {Aoki}, {Augusteijn},
657: {Beardmore}, {Bersier}, {Castro Cer{\'o}n}, {D'Avanzo}, {Diaz-Fraile},
658: {Gorosabel}, {Hirst}, {Jakobsson}, {Jensen}, {Kawai}, {Kosugi}, {Laursen},
659: {Levan}, {Masegosa}, {N{\"a}r{\"a}nen}, {Page}, {Pedersen}, {Pozanenko},
660: {Reeves}, {Rumyantsev}, {Shahbaz}, {Sharapov}, {Sollerman}, {Starling},
661: {Tanvir}, {Torstensson}, \& {Wiersema}}]{2006ApJ...652.1011W}
662: {Watson}, D., et al. 2006{\natexlab{a}}, ApJ, 652, 1011
663:
664: \bibitem[{{Watson} {et~al.}(2006{\natexlab{b}}){Watson}, {Reeves}, {Hjorth},
665: {Fynbo}, {Jakobsson}, {Pedersen}, {Sollerman}, {Cer{\'o}n}, {McBreen}, \&
666: {Foley}}]{2006ApJ...637L..69W}
667: {Watson}, D., et al. 2006{\natexlab{b}}, ApJ, 637, L69
668:
669: \bibitem[{{Watson} {et~al.}(2002){Watson}, {Reeves}, {Osborne}, {Tedds},
670: {O'Brien}, {Tomas}, \& {Ehle}}]{2002A&A...395L..41W}
671: {Watson}, D., {Reeves}, J.~N., {Osborne}, J.~P., {Tedds}, J.~A., {O'Brien},
672: P.~T., {Tomas}, L., \& {Ehle}, M. 2002, A\&A, 395, L41
673:
674: \bibitem[{{Willingale} {et~al.}(2001){Willingale}, {Aschenbach}, {Griffiths},
675: {Sembay}, {Warwick}, {Becker}, {Abbey}, \&
676: {Bonnet-Bidaud}}]{2001A&A...365L.212W}
677: {Willingale}, R., {Aschenbach}, B., {Griffiths}, R.~G., {Sembay}, S.,
678: {Warwick}, R.~S., {Becker}, W., {Abbey}, A.~F., \& {Bonnet-Bidaud}, J.-M.
679: 2001, A\&A, 365, L212
680:
681: \bibitem[{{Wilms} {et~al.}(2000){Wilms}, {Allen}, \&
682: {McCray}}]{2000ApJ...542..914W}
683: {Wilms}, J., {Allen}, A., \& {McCray}, R. 2000, ApJ, 542, 914
684:
685: \end{thebibliography}
686:
687: \end{document}
688:
689:
690:
691: