1: %% Beginning of file 'sample.tex'
2: %%
3: %% Modified 03 Nov 99
4: %%
5: %% This is a sample manuscript marked up using the
6: %% AASTeX v5.0 LaTeX 2e macros.
7:
8: %% The first piece of markup in an AASTeX v5.0 document
9: %% is the \documentclass command. LaTeX will ignore
10: %% any data that comes before this command.
11:
12: %% The command below calls the default manuscript style,
13: %% which will produce a double-spaced document on one column.
14: %% Examples of commands for other substyles follow. Use
15: %% whichever is most appropriate for your purposes.
16:
17: %\documentclass{aastex}
18:
19: %% preprint produces a one-column, single-spaced document:
20:
21: \documentclass[preprint]{aastex}
22:
23: %% preprint2 produces a double-column, single-spaced document:
24:
25: % \documentclass[preprint2]{aastex}
26:
27: %% If you want to create your own macros, you can do so
28: %% using \newcommand. Your macros should appear before
29: %% the \begin{document} command.
30: %%
31: %% If you are submitting to a journal that translates manuscripts
32: %% into SGML, you need to follow certain guidelines when preparing
33: %% your macros. See the AASTeX v5.0 Author Guide
34: %% for information.
35:
36: %\newcommand{\vdag}{(v)^\dagger}
37: %\newcommand{\myemail}{skywalker@galaxy.far.far.away}
38:
39: %% You can insert a short comment on the title page using the command below.
40:
41: \slugcomment{Submitted to ApJ 2/26/01}
42:
43: %% If you wish, you may supply running head information, although
44: %% this information may be modified by the editorial offices.
45: %% The left head contains a list of authors,
46: %% usually a maximum of three (otherwise use et al.). The right
47: %% head is a modified title of up to roughly 44 characters. Running heads
48: %% will not print in the manuscript style.
49:
50: \shorttitle{Heckman et al.}
51: \shortauthors{Escaping Ionizing Radiation from Starbursts}
52:
53:
54: %% This is the end of the preamble. Indicate the beginning of the
55: %% paper itself with \begin{document}.
56:
57: \begin{document}
58:
59: %% LaTeX will automatically break titles if they run longer than
60: %% one line. However, you may use \\ to force a line break if
61: %% you desire.
62:
63: \title{On the Escape of Ionizing Radiation from Starbursts}
64:
65:
66: %% Use \author, \affil, and the \and command to format
67: %% author and affiliation information.
68: %% Note that \email has replaced the old \authoremail command
69: %% from AASTeX v4.0. You can use \email to mark an email address
70: %% anywhere in the paper, not just in the front matter.
71: %% As in the title, you can use \\ to force line breaks.
72:
73:
74: %% Notice that each of these authors has alternate affiliations, which
75: %% are identified by the \altaffilmark after each name. Specify alternate
76: %% affiliation information with \altaffiltext, with one command per each
77: %% affiliation.
78:
79: \author{T. M. Heckman\altaffilmark{1,2},
80: K. R. Sembach\altaffilmark{1}, and G. R. Meurer\altaffilmark{1}}
81: \affil{Department of Physics \& Astronomy, Johns Hopkins University,
82: Baltimore, MD 21218}
83:
84: \author{C. Leitherer\altaffilmark{1} and D. Calzetti\altaffilmark{1}}
85: \affil{Space Telescope Science Institute, Baltimore, MD 21218}
86:
87: \and
88:
89: \author{C. L. Martin\altaffilmark{1}}
90: \affil{Department of Astronomy, Caltech, Pasadena, CA 91125}
91:
92:
93: \altaffiltext{1}{Guest Investigators on the NASA-CNES-CSA Far
94: Ultraviolet
95: Spectroscopic Explorer. FUSE is operated for NASA by the Johns Hopkins
96: University under NASA contract NAS5-32985.}
97: \altaffiltext{2}{Adjunct Astronomer, Space Telescope Science Institute}
98:
99:
100: %% Mark off your abstract in the ``abstract'' environment. In the manuscript
101: %% style, abstract will output a Received/Accepted line after the
102: %% title and affiliation information. No date will appear since the author
103: %% does not have this information. The dates will be filled in by the
104: %% editorial office after submission.
105:
106: \begin{abstract}
107: Far-ultraviolet spectra obtained with $FUSE$ show that the
108: strong $CII\lambda$1036 interstellar absorption-line is essentially black in
109: five of the UV-brightest local starburst galaxies.
110: Since the opacity of the neutral ISM
111: below the Lyman-edge will be significantly larger than in the $CII$ line,
112: these data provide strong constraints on the escape of
113: ionizing radiation from these starbursts. Interpreted as a
114: a uniform absorbing slab, the implied optical depth at the Lyman edge is
115: huge ($\tau_0 \geq 10^2$). Alternatively, the areal covering factor of
116: opaque material is typically
117: $\geq$ 94\%.
118: Thus, the fraction of ionizing stellar photons that escape the
119: ISM of each galaxy
120: is small: our conservative estimates typically yield
121: $f_{esc} \leq 6\%$. Inclusion of extinction due
122: to dust will further decrease $f_{esc}$.
123: An analogous analysis of the rest-UV
124: spectrum of the star-forming galaxy $MS~1512-CB58$ at $z$ =2.7 leads
125: to similar constraints on $f_{esc}$.
126: These new results agree with
127: the constraints provided by
128: direct observations
129: below the Lyman edge in a few other local starbursts. However, they differ from
130: the recently reported properties of star-forming galaxies
131: at $z \geq$ 3.
132: We assess the idea that
133: the strong galactic winds seen in many powerful starbursts clear channels
134: through their neutral ISM. We show empirically that such
135: outflows may be a necessary - but not sufficient - part of the
136: process for creating
137: a relatively porous ISM. We note that
138: observations will soon
139: document the cosmic evolution in the contribution of star-forming galaxies
140: to the metagalactic ionizing background, with
141: important implications for the evolution of the IGM.
142: \end{abstract}
143:
144: %% Keywords should appear after the \end{abstract} command. The uncommented
145: %% example has been keyed in ApJ style. See the instructions to authors
146: %% for the journal to which you are submitting your paper to determine
147: %% what keyword punctuation is appropriate.
148:
149: \keywords{galaxies: starburst -- galaxies: intergalactic medium --
150: galaxies: formation}
151:
152:
153: \section{Introduction}
154:
155: The intergalactic medium (IGM) contains the bulk of the baryons in the
156: universe (e.g., Fukugita, Hogan, \& Peebles 1998).
157: Determining the source and strength of the metagalactic ionizing radiation field
158: and documenting its cosmic evolution is crucial to understanding
159: the fundamental properties of the IGM at both low- and high-redshift.
160: The two prime candidates for producing the background are QSOs
161: and star-forming galaxies. While the contribution to the ionizing background
162: from QSOs can be estimated with reasonable accuracy, considerably
163: less is known about the contribution from galaxies. QSOs
164: alone appear inadequate to produce the inferred background
165: (e.g, Madau, Haardt, \& Rees 1999), especially at $z \geq$ 3
166: where their co-moving space density declines
167: steeply with increasing redshift (Fan et al. 2000).
168: Moreover, there are only rather
169: indirect constraints on the contribution of galaxies to the ionizing
170: background in the low-redshift universe (e.g.,
171: Giallongo, Fontana, \& Madau 1997; Devriendt et al. 1998; Shull et al. 1999).
172:
173: The cosmic history of the formation of massive (ionizing) stars has
174: now been determined in broad terms (e.g., Madau, Pozzetti, \& Dickinson
175: 1998; Steidel et al. 1999;
176: Barger, Cowie, \& Richards
177: 2000). On-average
178: several thousand ionizing photons are produced for each nucleon
179: processed into stars, so the galactic contribution to the
180: ionizing background is potentially both substantial and calculable. However,
181: the greatest single uncertainty in this calculation is
182: the value for $f_{esc}$ - the fraction of the ionizing
183: photons that escape from star-forming galaxies.
184:
185: Uncertainties in the value of $f_{esc}$
186: also map directly into uncertainties in the use
187: of standard emission-line diagnostics to infer the basic properties
188: of star-forming galaxies. Most crucially, the luminosity of
189: the Hydrogen recombination-lines is a widely utilized
190: measure of the formation rate of massive stars
191: in both the local and high-redshift universe (e.g. Gallego et al.
192: 1995; Yan et al. 1999; Pettini et al. 2001). This approach
193: generally assumes that $f_{esc} << 1$.
194: However,
195: the smaller-than-predicted equivalent
196: widths of the Balmer emission lines in starbursts could imply
197: that this assumption is suspect
198: (e.g. Moy et al. 2000; Stasinska et al. 2001).
199:
200: An $HI$ column of only
201: 1.6 $\times$ 10$^{17}$ cm$^{-2}$ produces $\tau$ = 1 at the Lyman
202: edge. Mean galactic gas columns are much larger of course, ranging from
203: $\sim$ 10$^{21}$ cm$^{-2}$ in normal galactic disks to 10$^{24}$ cm$^{-2}$ in
204: nuclear starbursts (e.g., Kennicutt 1998). Starburst-like mean gas columns
205: are inevitable in the Lyman Break Galaxies (Heckman 2000), given their
206: high star-formation rates per unit area (Meurer et al. 1997). The leakage
207: of ionizing radiation out of galaxies must then be determined by the
208: topology of the ISM. As such, theory gives us scant guidance, and
209: so direct measurements of $f_{esc}$ are required.
210:
211: Leitherer et al. (1995) reported the first direct measurements of
212: $f_{esc}$ using the {\it Hopkins Ultraviolet Telescope} to observe
213: far-ultraviolet (far-UV) light
214: below the rest-frame Lyman edge in a sample of four local starbursts,
215: and these data were later reanalysed by
216: Hurwitz, Jelinsky, \& Dixon (1997). The
217: resulting upper limits on $f_{esc}$ were typically 3\% to 10\%.
218: In a very surprising development, Steidel, Pettini, \& Adelberger
219: (2001 - hereafter SPA) have reported
220: the detection of escaping Lyman continuum in
221: the combined spectrum of 29 Lyman Break Galaxies at a mean redshift
222: $<z>$ = 3.4. They estimate that the ratio of
223: $f_{esc}$ at 900 \AA\ and 1500 \AA\ ranges from 0.5 to 1 in this composite
224: spectrum (i.e., there is no definite detection of photoelectric opacity due to
225: $HI$).
226: Haehnelt et al.
227: (2001) reach similar conclusions, but point out that star-forming galaxies at
228: only slightly smaller redshifts can not be this porous
229: without violating constraints set by the observed HeII/HI opacity
230: ratio in the $Ly\alpha$ forest.
231:
232: It is clearly important to obtain more determinations
233: of $f_{esc}$, and to use these measurements to better understand the physical
234: processes that determine $f_{esc}$. This latter goal can be best accomplished
235: in the local universe, where detailed multiwaveband observations
236: of star-forming galaxies can be made. Unfortunately, direct observations
237: of the emerging Lyman continuum require observing
238: galaxies at redshifts greater than a few percent, so that the foreground
239: $HI$ opacity of the Milky Way is not significant at the relevant wavelengths
240: (Leitherer et al. 1995;
241: Hurwitz, Jelinsky, \& Dixon 1997). Given the modest sensitivity
242: of far-UV telescopes, this has limited such investigations to rather
243: small (and possibly unrepresentative) galaxy samples.
244:
245: Against this backdrop, we describe our analysis of new far-UV data obtained
246: with $FUSE$ for the UV-brightest local starbursts (section 2).
247: We point out that the strongest interstellar lines that trace the $HI$ phase
248: in these starbursts are black (or very nearly black)
249: at line-center. We show that
250: this implies low values for $f_{esc}$ (section 3). This not only more
251: than doubles the sample of local starbursts with good upper limits
252: on $f_{esc}$, it illustrates a technique that can potentially be used on a
253: much larger sample of galaxies. Indeed, similar
254: arguments can be applied to high-redshift galaxies (section 4), and
255: show that $f_{esc}$ is also low in the bright star-forming
256: galaxy $MS~1512-CB58$ at $z$ =2.7 (Pettini et al. 2000).
257: We discuss these results and
258: their implications in section 5.
259:
260: \section{Observations}
261:
262: \subsection{The Sample}
263:
264: We are carrying out a program with the
265: $Far~Ultraviolet~Spectroscopic~Explorer$
266: ($FUSE$ - Moos et al. 2000) to obtain spectra
267: of the six starburst galaxies from the Kinney et al. (1993)
268: `Ultraviolet Atlas of Star-Forming Galaxies' having the largest
269: UV flux at 1500\,\AA\ through the IUE
270: $10\arcsec\times20\arcsec$ aperture (F$_{\lambda}$ $>$
271: 10$^{-13}$ erg cm$^{-2}$ s$^{-1}$ \AA$^{-1}$). These targets span broad ranges
272: in: 1) metallicity, from 1/8 (NGC~1705) to 2.5 (M~83) times solar,
273: 2) starburst bolometric
274: luminosity, from $\sim$3$\times10^8$ (NGC~1705) to $\sim$2$\times10^{10}$
275: (NGC~3310) L$_{\odot}$,
276: 3) internal dust-reddening, from
277: A$_V \sim$ 0.0 (NGC~1705) to 0.6 (NGC~3310 and M~83),
278: and 4) host galaxy properties, from dwarfs (NGC~1705 and NGC~5253), to
279: irregulars
280: (NGC~4214 and NGC~4449), to spirals (NGC~3310 and M~83). The sample therefore
281: spans most of the multidimensional
282: parameter space of starbursts in the local universe (Heckman et al. 1998).
283: The selection on the basis of rest-frame $UV$ brightness makes this an
284: appropriate sample to compare to the UV-selected Lyman Break Galaxies at
285: high-redshift.
286:
287: \subsection{Observational Details}
288:
289: The observations undertaken to date are summarized in Table 1. Data
290: have been obtained for 5 of the 6 targets (only NGC~4449 has not yet
291: been observed).
292: Analyses of the dynamics of the ISM in individual galaxies
293: based on these data are being reported elsewhere:
294: Heckman et al. (2001a,b),
295: Martin et al. (2001).
296: The large (LWRS, $30\arcsec\times30\arcsec$) aperture on $FUSE$ was used
297: for the observations of 4 of the 5 targets, while NGC~5253 was observed
298: with the medium (MDRS, $4\arcsec\times20\arcsec$) aperture.
299: The corresponding physical sizes of the projected aperture are kpc-scale in
300: all cases but NGC~5253 (Table 1).
301:
302: The starburst was centered in the aperture of the
303: LiF1 (guiding) channel
304: for each observation by the standard guide-star acquisition procedure.
305: For four targets, flux was recorded through the LWRS
306: apertures in both long
307: wavelength (LiF, $\sim1000-1187$\,\AA) channels and both short wavelength
308: (SiC, $\sim900-1100$\,\AA) channels.
309: We obtained no useful data in the SiC channels during our MDRS
310: observations of NGC5253 (A0460404) due to a loss of mirror alignment.
311: We note that while the data do extend slightly below the wavelength
312: of the starburst's Lyman edge in the SiC1 channel, these data do not usefully
313: constrain
314: $f_{esc}$. This is because the opacity of foreground Galactic $HI$ is still
315: significant
316: at these wavelengths, largely due to the confluence of the high-order
317: Lyman series absorption lines (cf. Hurwitz et al. 1997).
318:
319: Most of the data for this program were obtained at night, which minimizes
320: \ion{O}{1} and \ion{N}{1} terrestrial airglow contamination in the spectra.
321: The percentage of night to total exposure time averaged $\sim65\%$, except for
322: NGC~1705 for which the ratio was $\sim92\%$.
323:
324: The velocity resolution of the data depends on how the far-UV light illuminates
325: the $FUSE$ apertures. The resolution for a filled $MDRS$ aperture (NGC~5253)
326: or for a point-like source in the $LWRS$ aperture (NGC~1705) is
327: $\sim30$ km~s$^{-1}$.
328: We estimate
329: a resolution of $\leq$50 to $\leq$70 km~s$^{-1}$ in
330: NGC~3310, M~83, and NGC~4214
331: based on the widths of the narrowest Galactic (foreground) absorption-lines.
332:
333: \subsection{Data Processing}
334:
335: The raw time-tagged photon event lists for each exposure were processed
336: with the standard $FUSE$ calibration software ($CALFUSE v1.8.5$) available at
337: the Johns Hopkins University as of January 2001. The lists were screened
338: for valid data, and corrections for geometric distortions, spectral motions,
339: and Doppler shifts were applied (see Sahnow et al. 2000). The individual
340: calibrated extracted spectra for each channel were cross-correlated, shifted
341: to remove residual velocity offsets due to image motion in the apertures,
342: and combined to produce a composite spectrum. These composite spectra in
343: the 1000--1070\,\AA\ region were then compared, and any remaining velocity
344: offsets were removed by referencing the Galactic absorption lines to the
345: appropriate Local Standard of Rest velocities.
346:
347: Since we are interested in measuring the fluxes in the cores of saturated
348: lines, accurate determinations of the residual fluxes are
349: essential. We applied the standard detector background corrections available
350: within $CALFUSE$ for each observation. These corrections account only for the
351: particle event backgrounds ($\sim0.75$ cnt cm$^{-2}$ s$^{-1}$) and do not
352: correct for scattered Ly$\alpha$ emission or other stray light. These
353: additional sources of background light are negligble in the LiF1 data at
354: the wavelengths of $\sim1040$\,\AA\
355: (typically $<5\times10^{-15}$ erg cm$^{-2}$ s$^{-1}$
356: \AA$^{-1}$, or less than $\lesssim2\%$ of the continuum flux). A description
357: of the FUSE backgrounds can be found in the $FUSE~Observers~Guide$
358: (Blair \& Andersson 2001)\footnotemark.
359: \footnotetext{This guide (v3.0) can be found on-line at
360: {\tt http://fuse.pha.jhu.edu/support/guide/guide.html}.}
361:
362: In the following discussion
363: we consider data from the LiF1 channel, which has the highest S/N at the
364: wavelengths of interest for this study ($\lambda\sim1020-1045$\,\AA).
365: The data for the other channels are consistent with our findings presented
366: below.
367:
368:
369: \section{Results from the FUSE Data}
370:
371: In Figure 1 we show the spectral region around the $Ly\beta$,
372: \ion{C}{2} $\lambda$1036.337
373: and \ion{O}{1} $\lambda$1039.230 absorption-lines in the five galaxies
374: in our sample.
375: It is immediately apparent that the cores of the \ion{C}{2} $\lambda$1036 and
376: Lyman lines are quite black. The residual normalized intensities
377: in the \ion{C}{2} line are typically
378: $\leq$ 6\% (Table 2). These \ion{C}{2} residual intensity estimates account
379: for overlying Galactic absorption due to the Milky Way \ion{O}{6}
380: $\lambda1037.617$ or H$_2$ R(1) $\lambda1037.149$ and P(1) $\lambda1038.157$
381: lines. There are no H$_2$ lines within the starbursts themselves that
382: contribute to the depth of the \ion{C}{2} line. We modeled the Galactic
383: absorption features using the shorter wavelength member of the \ion{O}{6}
384: doublet ($\lambda1031.926$) or other H$_2$ R(1) and P(1) lines
385: (typically P(1) $\lambda\lambda$1003.294, 1014.326 and
386: R(1) $\lambda$1013.435) to get both the velocity extents and expected
387: depths of the overlying lines.
388: We note that the calculation of the
389: residual intensity is also corrected for the effect of the maximum plausible
390: stellar contribution to the \ion{C}{2} line. Model starburst
391: spectra (Gonzalez Delgago, Leitherer, \& Heckman 1997) show that the stellar
392: photospheric line (which is entirely due to B stars) will depress the local
393: continuum by a factor of $\sim$ 1.2 to 2 (primarily depending on the age
394: of the starburst). In our sample, the stellar contribution
395: to \ion{C}{2} should be largest in the
396: B-star-dominated spectrum
397: of NGC~1705. Even in this case the \ion{C}{2} line is primarily
398: interstellar, since it (like the other interstellar lines) is blueshifted
399: by roughly 40 km s$^{-1}$ relative to the galaxy systemic velocity
400: (Heckman et al. 2001, and see also Heckman \& Leitherer 1997).
401:
402: \begin{figure}
403: \plotone{f1.eps}
404: \vskip -3cm
405: \caption{$FUSE$ spectra (LiF1A detector segment) of the
406: spectral region containing the interstellar $Ly\beta$,
407: \ion{C}{2} $\lambda$1036,
408: and \ion{O}{1} $\lambda$1039 absorption-lines in five UV-selected local
409: starbursts.
410: The lines due to the foreground (Galactic) and starburst ISM are indicated,
411: as is the fitted continuum (dotted line) to which the residual intensity
412: at the line core is referenced. Note that the $Ly\beta$ and
413: \ion{C}{2}\,$\lambda$1036
414: lines are effectively black in the line core, while the less
415: saturated \ion{O}{1} $\lambda$1039 line shows a range in residual intensity.
416: Tick marks indicate the positions (or expected positions) of $J=0-3$
417: H$_2$ lines in the Milky Way (upper, lighter ticks) and the starbursts
418: (lower, darker ticks).
419: All of the spectra shown are a combination of data taken during both
420: orbital day and night.}
421: \end{figure}
422:
423: The \ion{O}{1} $\lambda$1039 line shows behavior
424: ranging from a normalized residual intensity of about 15\% in NGC~1705
425: to $\leq$ 6\% in NGC~5253. Overlying Galactic absorption is not an issue
426: for the \ion{O}{1} at the resolution of the $FUSE$ data.
427: Both \ion{O}{1} and \ion{C}{2} arise in \ion{H}{1} regions
428: and contain direct information about the \ion{H}{1} column density
429: and areal covering factor of the neutral gas.
430:
431: We begin by writing the ratio of the optical depth at the Lyman edge
432: to the optical depth at line center. Following Spitzer (1977), this
433: is
434:
435: \begin{equation}
436: \tau_{Ly}/\tau_0 = 4\times10^{-16} (N_H/N_j) (b/f\lambda)
437: \end{equation}
438:
439: \noindent
440: where $(N_H/N_j)$ is the ratio of the \ion{H}{1} and
441: ground-state ionic columns,
442: $b$ is the normal Doppler parameter ($cm~s^{-1}$), $f$
443: is the oscillator strength, and the wavelength $\lambda$
444: is measured in $cm$.
445: Assuming that \ion{O}{1} and \ion{C}{2} are the dominant ionic species of
446: their respective element in the \ion{H}{1} phase, this can be written as
447:
448: \begin{equation}
449: \tau_{Ly}/\tau_0 = R Z_{gas}^{-1} (b/100~km/s)
450: \end{equation}
451:
452: \noindent
453: where $Z_{gas}$ is the gas-phase $C$ or $O$ abundance in solar units and
454: $R$ = 8.6 for \ion{C}{2} $\lambda$1036 and $R$ = 50 for \ion{O}{1}
455: $\lambda$1039
456: (Morton 1991). The important point is that {\it the optical depth at
457: the Lyman edge is much larger than the optical depth at line-center}
458: for either line for typical values of $Z_{gas}$ and $b$. Corrections for
459: grain-depletion will only increase the implied ratio of $\tau_{Ly}/\tau_0$.
460: Ionization corrections are unimportant for \ion{O}{1} since it is closely
461: coupled to \ion{H}{1} through charge exchange reactions, and will only
462: increase the implied ratio of $\tau_{Ly}/\tau_0$ for \ion{C}{2}.
463:
464: As an example, consider the case of NGC~1705 (Heckman et al. 2001).
465: Based on the nebular emission-lines, the gas-phase Oxygen abundance
466: is 1/8 solar, and this is consistent
467: with the abundances derived in the \ion{H}{1} gas for $Si$, $Ar$, and $Fe$
468: from the $FUSE$ spectra. The
469: measured $b$ value is 40$\pm$10 km s$^{-1}$. The residual intensity
470: at the core of the $OI\lambda$1039 line implies $\tau_0$ = 1.8
471: (for unit covering factor - see below). Equation 2 above then
472: implies that $\tau_{Ly}$ = 280. This is roughly consistent with the $HI$
473: column we derived from fitting the $Lyman$ series lines:
474: $N_{HI}$ = 1.5$\times$ $10^{20}$ cm$^{-2}$. Thus, in the
475: context of a homogeneous
476: absorbing slab, the escaping fraction of ionizing photons
477: will be almost identically zero in NGC~1705.
478: In principle, even more severe constraints on
479: $\tau_{Ly}$ can be placed using weaker absorption-lines
480: (transitions having smaller values
481: of $(N_j/N_H)f\lambda$). We do not take this approach, since even
482: the strong \ion{O}{1} $\lambda$1039 line already implies $f_{esc}\sim$0.
483:
484: Less stringent (more realistic) constraints on $f_{esc}$
485: can be placed
486: by adopting a ``picket fence'' model
487: in which the areal covering factor of optically-thick $HI$
488: is not unity (e.g., ionizing radiation can escape through ``holes''
489: in the $HI$). In this case, the upper limit on the residual intensity
490: at the center of the $CII\lambda$1036 line implies that the
491: covering factor must be $\geq$ 94\%, so that the upper bound
492: on the escape fraction for ionizing radiation is $f_{esc} \leq$ 6\%.
493:
494: We have undertaken this analysis for the other four galaxies in
495: our sample, and report the results in Table 2. In all cases, the
496: results are similar to those in NGC~1705, with implied values
497: for $f_{esc}$ that are almost identically zero (for the
498: homogeneous slab model) and $\leq$6\% for the picket-fence model.
499: The values listed in Table~2 should be considered stringent upper limits
500: for the following reasons: 1) We list only the
501: limits derived from the picket fence model.
502: 2) We measured $f_{esc}$ at velocities where the
503: residual flux was greatest after accounting for possible contamination
504: by other absorption features. 3) Scattered light may contribute at low
505: levels ($\lesssim2\%$) and only serves to strengthen the limit on $f_{esc}$.
506: 4) The FUSE line spread function (LSF) likely consists of two components: a
507: narrow component accounting for $\sim80\%$ of the LSF area and a broad
508: component accounting for $\sim20\%$ and having $\sim2-3$ times the width of the
509: narrow component. While the exact details of the LSF
510: are presently unknown, such a model more accurately describes the
511: shapes of the cores of the Galactic \ion{H}{1} lines toward nearby stars
512: than a single component LSF. As a result, it is possible that some of the
513: observed flux in the line cores is simply redistributed light from nearby
514: continuum regions within $\sim1$\,\AA\ of the line. 5) The continuum
515: estimates for the interstellar lines shown in Figure 1 are conservative
516: estimates and are unlikely to be lower than those shown.
517: 6) The
518: upper limits on $f_{esc}$ can be pushed to
519: even smaller values by including the opacity due
520: to dust (that is, our technique measures only the contribution
521: of the photoelectric opacity of \ion{H}{1}). Application of the
522: empirical starburst dust attenuation law (Calzetti 1997;
523: Meurer, Heckman, \& Calzetti 1999) to the $IUE$ spectra
524: of these galaxies implies an extinction due to dust at 1500\AA\ ranging
525: from negligible (NGC~1705) to $\sim$ 3 magnitudes in
526: M~83 and NGC~3310. The mean dust opacity in the Lyman continuum
527: (1 to 4 Ryd) should be larger still (Mathis 1990).
528:
529: We conclude that our UV-selected starbursts are quite opaque to
530: their own ionizing radiation. This confirms the results of
531: Leitherer et al. (1995) and Hurwitz, Jelinsky, \& Dixon (1997), but
532: generalizes and strengthens them.
533:
534: \section{Galaxies at High-Redshift}
535:
536: The technique described above can be applied to galaxies at high-redshift.
537: As an application, we consider the
538: star-forming galaxy $MS~1512-CB58$ at $z$ =2.7 (Yee et al. 1996).
539: Thanks to strong gravitational lensing, its flux has been
540: amplified by a factor of $\sim$ 30 (Seitz et al. 1998). Thus, the
541: rest-frame UV spectrum
542: discussed by Pettini et al. (2000) is by far the best of its kind.
543: We display the spectral region from $Ly\alpha$ to $CII\lambda$1335
544: in Figure 2. The quoted spectral resolution in these data is
545: $\sim$200 km s$^{-1}$ FWHM, while the interstellar lines are much broader
546: (500 to 700 km s$^{-1}$) and very deep.
547: It is instructive to note that the value
548: for $f\lambda$ (equation 1) for the $CII\lambda$1335 ($OI\lambda$1302)
549: line is 0.1 (0.8) dex
550: larger than for the $CII\lambda$1036 ($OI\lambda$1039) line in our $FUSE$
551: spectra. Thus, the spectrum of $MS~1512-CB58$ can
552: be straightforwardly compared to the local starbursts.
553:
554: \begin{figure}
555: \plotone{f2.eps}
556: \vskip -8cm
557: \caption{A portion of the $Keck$ LRIS spectrum of the
558: gravitationally-lensed
559: star-forming galaxy $MS~1512-CB58$ at $z =$2.7 published by
560: Pettini et al. (2000). The interstellar $Ly\alpha$, \ion{Si}{2} $\lambda$1260,
561: \ion{O}{1} $\lambda$1302, \ion{Si}{2} $\lambda$1304, and
562: \ion{C}{2} $\lambda$1335 absortion-lines
563: are indicated. The stellar continuum in the spectrum has been normalized
564: to unit intensity by fitting it with a low-order polynomial. Note that the
565: interstellar lines are nearly black, and much broader than the
566: instrumental resolution of 0.8 \AA\ (rest-frame).}
567: \end{figure}
568:
569: The measured residual intensites in the line cores in the
570: $MS~1512-CB58$ spectrum are: $Ly\alpha$ (1\%),
571: $SiII\lambda$1260 (7\%), $OI\lambda$1302 (6\%), and
572: $CII\lambda$1335 (2\%).
573: To apply equations 1) and 2) above, we adopt
574: a gas-phase metallicity of 1/4 solar
575: (Pettini et al. 2000;
576: Leitherer et al. 2001; Tremonti et al. 2001)
577: and take $b$ = 300 km s$^{-1}$ based
578: on the line-widths reported by Pettini et al. (2000).
579: The implied optical depth at the Lyman edge assuming a uniform
580: slab is $\tau_{Ly} \sim$ 300. Pettini et al. (2000) derive an $HI$
581: column of 7.5$\times$10$^{20}$ cm$^{-2}$ from the damped $Ly\alpha$
582: line, yielding an even larger value for $\tau_{Ly}$ for a uniform slab.
583: Alternatively, the residual intensity
584: at the core of the $CII\lambda$1335 line implies an areal
585: covering factor of optically-thick gas of 98\%. The implied
586: value for $f_{esc}$ ranges from $\sim$0 to 2\%. These values
587: are consistent with the results for the local starbursts in section 3 above.
588:
589: The situation is quite different in the faint Lyman Break Galaxies
590: at $z \geq$ 3. As SPA point out,
591: the interstellar
592: lines in their composite spectrum are far from black (the
593: apparent residual intensities are roughly 70\% at line center). The
594: composite spectrum of Lyman Break Galaxies at $<z>$ = 3.0 in
595: the Hubble Deep Field is similar (Lowenthal et al. 1997). It is possible
596: in these cases that spectra with higher spectral resolution and signal-to-noise
597: would show that the lines are in fact nearly black over narrow ranges
598: in velocity. However, the existing data are consistent with the
599: conclusion of SPA that the escape
600: fraction of ionizing radiation is large in these galaxies.
601:
602: \section{Discussion}
603:
604: The results we have reported for five of the six UV-brightest local starbursts
605: (NGC~1705, NGC3310, NGC~4214, NGC~5253, and M~83) implies
606: that UV-bright starburst galaxies today are highly opaque to their own ionizing
607: radiation. This agrees with results for a small sample of more distant
608: starbursts described by Leitherer
609: et al. (1995) and Hurwitz et al. (1997). These results in turn are consistent
610: with the estimate by Shull et al. (1999) that the strength of the
611: local metagalactic
612: ionizing background implies that the fraction of ionizing photons
613: that escape from
614: star-forming galaxies can not exceed a global average value of about 5\%.
615:
616: These results are in striking contrast to the detection of significant
617: escaping Lyman continuum radiation from Lyman Break Galaxies
618: at $z \sim$ 3 to 4 by SPA. Our analysis
619: shows that the star-forming galaxy $MS1512-CB58$ at $z=$2.7
620: is similar to the local starbursts: it too must be highly opaque below
621: the Lyman break.
622: Taking the current results at face value, we are left with a puzzle.
623: Why is the topology/structure of the neutral interstellar medium
624: apparently so different between the galaxies studied by
625: SPA and those in the present paper?
626:
627: In some ways this is a surprising result, because
628: local starbursts and
629: Lyman Break Galaxies are remarkably similar in most respects
630: (e.g., Heckman 2000).
631: Meurer et al. (1997)
632: showed that both have the same
633: bolometric surface-brightnesses, and therefore the same star formation rates
634: per unit area.
635: This implies that
636: the basic physical properties of the ISM should be similar.
637: The surface-mass-densities in the stars ($\Sigma_*$) and the
638: interstellar gas ($\Sigma_g$)
639: that fuels the star-formation must be similar.
640: The radiant energy density in the ISM must be similar
641: ($\sim$10$^3$ times higher than in the Milky Way), as must
642: the rate of mechanical energy deposition
643: (supernova heating) per unit area or volume.
644: Simple considerations of hydrostatic equilibrium imply correspondingly
645: high total pressures in the ISM in both:
646: $P \sim G \Sigma_g
647: \Sigma_{tot}$, or $P/k \sim$
648: few $\times$ 10$^7$ K cm$^{-3}$. This is several thousand times the value
649: in the local ISM in the Milky Way. Finally, the
650: characteristic dynamical time in the ISM
651: will be short in both cases: $t_{dyn}$
652: $\sim$ $(G\rho)^{-1/2}$ $\sim$ $(G\Sigma_{tot}H)^{-1/2}$
653: $\sim$ few Myr (where $H \sim$ 10$^2$ pc is the thickness of the disk).
654: Perhaps the most striking difference is the size-scale: starbursts
655: in the local universe are generally circum-nuclear events (sizes of $10^2$ to
656: $10^3$ pc) imbedded in a much larger galaxy, while the intense star-formation
657: in the Lyman Break
658: Galaxies appears to be a galaxy-wide event (sizes of $10^3$ to $10^4$ pc).
659: This might lead to fundamental differences in the global properties of the ISM.
660:
661: Such generic arguments aside, is there any empirical evidence that
662: the structure or dynamics in the ISM in our specific sample
663: of local starbursts differs from the Lyman Break Galaxies? We
664: consider two possibilities.
665:
666: First, our $FUSE$ sample galaxies
667: have considerably lower luminosities (star-formation rates)
668: than the typical Lyman Break Galaxies: $logL_{bol} \sim$ 8.5 to 10.3
669: $L_{\odot}$ {\it vs.} 10.7 to 11.7 $L_{\odot}$ (Adelberger \& Steidel 2000).
670: It seems physically plausible that the ISM might become more
671: porous in starbursts with higher star-formation (and energy deposition)
672: rates (cf. Lehnert \& Heckman 1996).
673: We have tested this idea by considering data on $f_{esc}$
674: for more powerful local starbursts. Two of the four starbursts observed
675: below the Lyman break by Leitherer et al. (1995) are powerful
676: systems with $L_{bol} \sim 10^{11} L_{\odot}$. Both are opaque in the
677: Lyman continuum:
678: $f_{esc} \leq$ 2\% and $\leq$ 4\% for Mrk~496 and IRAS~08339+6517 respectively
679: (Hurwitz et al. 1997).
680: We have also searched the HST
681: archives for UV spectra of powerful local starbursts taken with
682: a spectral resolution comparable to our $FUSE$ data and with a projected
683: aperture size ($\sim$kpc-scale) comparable to our $FUSE$ data. There are
684: three such starbursts in the sample investigated by Kunth et al. (1998), who
685: used the $GHRS$ echelle mode to observe the $Ly\alpha$, $OI\lambda$1302,
686: and $SiII\lambda$1304 lines with a resolution of $\sim$ 20 km s$^{-1}$.
687: These starbursts have luminosities ($L_{\odot}$) of
688: $logL_{bol}$ = 10.4 (ESO~400-G043), 11.0 (IRAS~08339+6517), and 11.2
689: (ESO~350-IG038). These interstellar lines are black in two cases
690: (ESO~400-G043 and ESO~350-IG038), and have
691: a residual intensity
692: of 60 to 70\% in the third case (IRAS~08339+6517). While a large value
693: for $f_{esc}$ is therefore {\it possible} in this galaxy, direct
694: observations below the Lyman break actually show that $f_{esc} \leq$ 4\%
695: in this case (Hurwitz et al. 1997).
696:
697: A second possibility is that large values for $f_{esc}$ are created
698: by supernova-driven galactic outflows
699: (superwinds) that clear out channels through which ionizing radiation
700: can escape (SPA). The tracers of superwinds in Lyman Break Galaxies are their
701: blueshifted interstellar absorption-lines (Franx et al. 1997;
702: Pettini et al. 1998; 2000; 2001).
703: Indeed, the composite
704: spectrum of the 29 Lyman Break Galaxies analyzed by SPA
705: shows the clear signature of these outflows: the interstellar
706: absorption-lines are blueshifted by 400 to 500 km s$^{-1}$ with
707: respect to the $Ly\alpha$ emission-line. This outflow signature is also seen in
708: the composite spectrum formed from the sum of the spectra
709: of 12 Lyman Break Galaxies in the HDF at $z \sim$ 3 (Lowenthal et al. 1997;
710: Franx et al. 1997).
711: As noted above, the relative
712: shallowness of the interstellar lines in both these composite spectra
713: is consistent with (but does not demand)
714: a large value for $f_{esc}$.
715:
716: Similar outflows are seen in the UV spectra of local starbursts
717: (Kunth et al. 1998; Gonzalez Delgado et al. 1998), and
718: Heckman et al. (2000) have shown that high-velocity outflows of
719: neutral gas are common in powerful ($L_{bol} \geq$ 10$^{11}$ $L_{\odot}$)
720: local starbursts. In contrast, our $FUSE$ data
721: reveal high-velocity ($\Delta$$v \geq$ 10$^2$ km s$^{-1}$) outflows in the
722: neutral phase of the ISM in only one of our five
723: galaxies: NGC~3310, in which the mean outflow
724: velocity is about -200 km s$^{-1}$ (see Figure 1).
725: In the other galaxies the neutral-phase absorption-lines
726: arise in relatively quiescent gas near the galaxy
727: systemic velocity (Figure 1; Martin et al. 2001; Heckman et al. 2001).
728: Based on the ISM dynamics, it is perhaps not surprising that
729: NGC~1705, NGC~4214,
730: NGC~5253, and M~83 are opaque to their Lyman continuum radiation.
731:
732: On the other hand, Gonzalez Delgado et al. (1998) have shown that
733: outflows at several hundred km s$^{-1}$ are present in the neutral
734: gas in at least three of
735: the four starbursts
736: observed below the Lyman break by Leitherer et al. (1995). In all
737: cases the galaxy is opaque below the Lyman break, with $f_{esc}$ typically
738: $\leq$ 3 to 10\%. Of the three powerful starbursts observed by
739: Kunth et al. (1998), two (ESO~400-G043 and ESO~350-IG038) show
740: black interstellar lines and outflows at $(v - v_{sys})$ =
741: -225 and -58 km s$^{-1}$
742: respectively. In the third (IRAS~08339+6517),
743: the lines have a residual intensity
744: of 60 to 70\%, and are strongly blueshifted by $\sim$-500
745: km s$^{-1}$ (Gonzalez Delgado et al. 1998).
746:
747: We conclude that the currently avalilable data do not demonstate
748: that galactic winds inevitably produce large values of $f_{esc}$
749: in local starbursts.
750: Such outflows appear to be a necessary - but not sufficient -
751: part of the process that creates an ISM that is porous to ionizing radiation.
752:
753: In the near future we will use
754: $FUSE$ to observe a sample of more powerful starbursts
755: ($L_{bol}$ = 4$\times$10$^{10}$ to 4$\times$10$^{11}$ $L_{\odot}$).
756: This will allow us to probe the Lyman continuum opacity in more starbursts
757: with luminosities (and star formation rates) that are similar to those
758: of typical Lyman Break Galaxies.
759: At the same
760: time, the extension of the investigation at high-redshift
761: to a wider range of Lyman Break
762: Galaxies can help clarify the situation. In particular,
763: it will be important to test the high values for
764: $f_{esc}$ by obtaining data on both the Balmer recombination-lines
765: and the Lyman continuum in the same set of galaxies (to provide a
766: consistency check).
767: Finally, the
768: $Galex$ mission (Martin et al. 1999) will directly measure the escaping
769: flux below the Lyman break for the population of star-forming galaxies in the
770: redshift range $z \sim$ 0.4 to 2 (the epoch that apparently dominates the
771: cosmic history
772: of star-formation).
773: Thus, over the next few years, we should be able to make the first
774: good direct measurement
775: of the contribution of star-forming galaxies to the metagalactic
776: ionizing background as a function of cosmic epoch. This will have
777: important implications for the history of the IGM.
778:
779:
780: \acknowledgments
781:
782: We thank the members of the $FUSE$ team for providing this superb
783: facility to the astronomical community. We thank Max Pettini
784: for enlightening discussions, and for kindly providing us
785: a copy of the composite Lyman Break Galaxy spectrum published
786: by SPA.
787: This work was supported in part
788: by NASA grants NAG5-6400 and NAG5-9012. KRS acknowledges support from
789: NASA Long Term Space Astrophysics grant NAG5-3485.
790:
791:
792: %% The reference list follows the main body and any appendices.
793: %% Use LaTeX's thebibliography environment to mark up your reference list.
794: %% Note \begin{thebibliography} is followed by an empty set of
795: %% curly braces. If you forget this, LaTeX will generate the error
796: %% "Perhaps a missing \item?".
797: %%
798: %% thebibliography produces citations in the text using \bibitem-\cite
799: %% cross-referencing. Each reference is preceded by a
800: %% \bibitem command that defines in curly braces the KEY that corresponds
801: %% to the KEY in the \cite commands (see the first section above).
802: %% Make sure that you provide a unique KEY for every \bibitem or else the
803: %% paper will not LaTeX. The square brackets should contain
804: %% the citation text that LaTeX will insert in
805: %% place of the \cite commands.
806:
807: %% We have used macros to produce journal name abbreviations.
808: %% AASTeX provides a number of these for the more frequently-cited journals.
809: %% See the Author Guide for a list of them.
810:
811: %% Note that the style of the \bibitem labels (in []) is slightly
812: %% different from previous examples. The natbib system solves a host
813: %% of citation expression problems, but it is necessary to clearly
814: %% delimit the year from the author name used in the citation.
815: %% See the natbib documentation for more details and options.
816:
817: \begin{thebibliography}{}
818:
819: \bibitem[]{}Adelberger, K., \& Steidel, C. 2000, ApJ, 544, 218
820: \bibitem[]{}Barger, A., Cowie, L., \& Richards, E. 2000, AJ, 119, 2092
821: \bibitem[]{}Calzetti, D. 1997, AJ, 113, 162
822: \bibitem[]{}Devriendt, J., Sethi, S., Guiderdoni, B., \& Nath, B. 1998, MNRAS,
823: 298, 708
824: \bibitem[]{}Fan, X., et al., AJ, 121, 54
825: \bibitem[]{}Franx, M., Illingworth, G., Kelson, D., van Dokkum, P., \& Tran,
826: K.-V. 1997, ApJL, 486, L75
827: \bibitem[]{}Fukugita, M., Hogan, C., \& Peebles, J. 1998, ApJ, 503, 518
828: \bibitem[]{}Gallego, J., Zamorano, J., Aragon-Salamanca, A., \& Rego, M.
829: 1995, ApJL, 459, L1
830: \bibitem[]{}Giallongo, E., Fontana, A., \& Madau, P. 1997, MNRAS, 289, 629
831: \bibitem[]{}Gonzalez Delgado, R., Leitherer, C., \& Heckman, T. 1997, ApJ,
832: 489, 610
833: \bibitem[]{}Gonzalez Delgado, R., Leitherer, C., Heckman, T., Lowenthal, J.,
834: Ferguson, H., \& Robert, C. 1998, ApJ, 495, 698
835: \bibitem[]{}Haehnelt, M., Madau, P., Kudritzki, R., \& Haardt, F. 2001,
836: ApJL, 549, L151
837: \bibitem[]{}Heckman, T., 2000, Phil. Trans. Roy. Soc., 358, No. 1772, 2077
838: \bibitem[]{}Heckman, T., \& Leitherer, C. 1997, AJ, 114, 69
839: \bibitem[]{}Heckman, T., Lehnert, M., Strickland, D., \& Armus, L. 2000,
840: ApJS, 129, 493
841: \bibitem[]{}Heckman, T., Robert, C., Leitherer, C., Garnett, D., \&
842: van der Rydt, F. 1998, ApJ, 503, 646
843: \bibitem[]{}Heckman, T., Sembach, K., Meurer, G., Strickland, D.,
844: Martin, C.L., Calzetti, D., \& Leitherer, C. 2001, ApJ, in press
845: (astro-ph/01022830
846: \bibitem[]{}Hurwitz, M., Jelinsky, P., \& Dixon, W. 1997, ApJ, 481, 31
847: \bibitem[]{}Kennicutt, R. 1998, ApJ, 498, 541
848: \bibitem[]{}Kinney, A., Bohlin, R., Calzetti, D.,
849: Panagia, N., \& Wyse, R. 1993, ApJS, 86, 5
850: \bibitem[]{}Kunth, D., Mas-Hesse, Terlevich, E., Terlevich, R., Lequeux, J.,
851: \& Fall, S.M. 1998, A\&A, 334, 11
852: \bibitem[]{}Lehnert, M., \& Heckman, T. 1996, ApJ, 462, 651
853: \bibitem[]{}Leitherer, C., Ferguson, H., Heckman, T., \& Lowenthal, J. 1995,
854: ApJL, 454, L19
855: \bibitem[]{}Leitherer, C., Leao, J., Heckman, T., Lennon, D., Pettini, M.,
856: \& Robert, C. 2001, ApJ, in press (astro-ph/0012358)
857: \bibitem[]{}Lowenthal, J., Koo, D., Guzman, R., Gallego, J., Phillips, A.,
858: Faber, S., Vogt, N., Illingworth, G., \& Gronwall, C. 1997, ApJ, 481, 673
859: \bibitem[]{}Madau, P., Pozzetti, L., \& Dickinson, M. 1998, ApJ, 498, 106
860: \bibitem[]{}Madau, P., Haardt, F., \& Rees, M. 1999, ApJ, 514, 648
861: \bibitem[]{}Martin, C.D., Bianchi, L., Donas, J., Heckman, T.,
862: Madore, B., Malina, R., Milliard, B., Friedman, P., Rich, M., Schiminovich, D.,
863: Siegmund, O., \& Szalay, A. 1999, in
864: Ultraviolet-Optical Space Astronomy Beyond HST, ASP Conference Series 164,
865: ed. J. Morse, J. M. Shull, \& A. Kinney, p. 182
866: \bibitem[]{}Martin, C.L., Heckman, T., Sembach, K., Meurer, G.,
867: Calzetti, D., \& Leitherer, C. 2001, in preparation
868: \bibitem[]{}Mathis, J. 1990, ARA\&A, 28, 37
869: \bibitem[]{}Meurer, G., Heckman, T., Lehnert, M., Leitherer, C., \& Lowenthal,
870: J. 1997, AJ, 114, 54
871: \bibitem[]{}Meurer, G., Heckman, T., \& Calzetti, D. 1999, 521, 64
872: \bibitem[]{}Moos, H.W. et al. 2000, ApJ, 538, L1
873: \bibitem[]{}Morton, D. 1991, ApJS, 77, 119
874: \bibitem[]{}Moy, E., Rocca-Volmerange, B., \& Fioc, M. 2000, A\&A, 365, 347
875: \bibitem[]{}Pettini, M., Kellogg, M., Steidel, C., Dickinson, M., Adelberger,
876: K., \& Giavalisco, M. 1998, ApJ, 508, 539
877: \bibitem[]{}Pettini, M., Steidel, C., Adelberger, K., Dickinson, M., \&
878: Giavalisco, M. 2000, ApJ, 528, 96
879: \bibitem[]{}Pettini, M., Shapley, A., Steidel, C., Cuby, J.-G., Dickinson, M.,
880: Moorwood, A., Adelberger, K., \& Giavalisco, M. ApJ, in press
881: (astro-ph/0102456)
882: \bibitem[]{}Sahnow, D.S., et al. 2000, ApJ, 538, L7
883: \bibitem[]{}Seitz, S., Saglia, R., Bender, R., Hopp, U., Belloni, P.,
884: \& Ziegler, B. 1998, MNRAS, 298, 945
885: \bibitem[]{}Shull, S.M., Roberts, D., Giroux, M., Penton, S., \& Fardal, M.
886: 1999, AJ, 118, 1450
887: \bibitem[]{}Spitzer, L. 1977, Physical Processes in the Interstellar Medium,
888: Wiley \& Sons
889: \bibitem[]{}Stasinska, G., Schaerer, D., \& Leitherer, C. 2001, A\&A, 370, 1
890: \bibitem[]{}Steidel, C., Adelberger, K., Giavalisco, M., Dickinson, M., \&
891: Pettini, M. 1999, ApJ, 519, 1
892: \bibitem[]{}Steidel, C., Pettini, M., \& Adelberger, K. 2001 (SPA), ApJ,
893: 546, 665
894: \bibitem[]{}Tremonti, C., Calzetti, D., Leitherer, C., \& Heckman, T. 2001,
895: ApJ, in press (astro-ph/0103432)
896: \bibitem[]{}Yan, L., McCarthy, P., Freudling, W.,
897: Teplitz, H., Malumuth, E.,
898: Weymann, R., \& Malkan, M. 1999, ApJL, 519, L47
899: \bibitem[]{}Yee, H., Ellingson, E., Bechtold, J., Carlberg, R., \& Cuillandre,
900: J.-C. 1996, AJ, 111, 1783
901:
902: \end{thebibliography}
903:
904:
905: %% Generally speaking, only the figure captions, and not the figures
906: %% themselves, are included in electronic manuscript submissions.
907: %% Use \figcaption to format your figure captions. They should begin on a
908: %% new page.
909:
910:
911: %% No more than seven \figcaption commands are allowed per page,
912: %% so if you have more than seven captions, insert a \clearpage
913: %% after every seventh one.
914:
915: %% There must be a \figcaption command for each legend. Key the text of the
916: %% legend and the optional \label in curly braces. If you wish, you may
917: %% include the name of the corresponding figure file in square brackets.
918: %% The label is for identification purposes only. It will not insert the
919: %% figures themselves into the document.
920: %% If you want to include your art in the paper, use \plotone.
921: %% Refer to the on-line documentation for details.
922:
923:
924: %% Tables should be submitted one per page, so put a \clearpage before
925: %% each one.
926:
927: %% Two options are available to the author for producing tables: the
928: %% deluxetable environment provided by the AASTeX package or the LaTeX
929: %% table environment. Use of deluxetable is preferred.
930: %%
931:
932: %% Three table samples follow, two marked up in the deluxetable environment,
933: %% one marked up as a LaTeX table.
934:
935:
936: %% In this first example, note that the \tabletypesize{}
937: %% command has been used to reduce the font size of the table.
938: %% Note also that the \label command needs to be placed
939: %% inside the \tablecaption.
940:
941: \clearpage
942:
943: \begin{deluxetable}{lcccccccc}
944: \tablecolumns{0}
945: \tablewidth{0pt}
946: \tablecaption{FUSE Starburst Observations}
947: \tablehead{Galaxy & $l$ & $b$ & $F_{1050\,\AA}$ & Dataset & Obs. Date
948: & T$_{exp}$\tablenotemark{a} & Aper.\tablenotemark{b} & $L$\tablenotemark{c}\\
949: &($\degr$)&($\degr$)&(erg\,cm$^{-2}$\,s$^{-1}$\,\AA$^{-1}$)&&(y/m/d)&(ksec)&&
950: (kpc)}
951: \startdata
952: NGC~1705 & 261.08 & $-$38.74 & $\sim4.2\times10^{-13}$ & A0460102 & 2000/02/04 &
953: \phn7.7 & LWRS & 0.9$\times$0.9 \\
954: & & & & A0460103 & 2000/02/05 &
955: 13.6 & LWRS\\
956: \\
957: NGC~3310 & 156.01 & +54.06 & $\sim2.6\times10^{-13}$ & A0460201 & 2000/05/05 &
958: 27.1 & LWRS & 2.56$\times$2.56\\
959: \\
960: NGC~4214 & 160.25 & +78.08 & $\sim1.6\times10^{-13}$ & A0460303 & 2000/05/12 &
961: 20.7 & LWRS & 0.50$\times$0.50\\
962: \\
963: M~83 & 314.58 & +31.97 & $\sim3.0\times10^{-13}$ & A0460505 & 2000/07/06 &
964: 26.5 & LWRS & 0.55$\times$0.55\\
965: \\
966: NGC~5253 & 314.86 & +30.11 & $\sim8.9\times10^{-14}$ & A0460404 & 2000/07/07 &
967: 27.4 & MDRS & 0.40$\times$0.08\tablenotemark{d}\\
968: \enddata
969: \tablenotetext{a}{On-target exposure time in the LiF1 channel.}
970: \tablenotetext{b}{Aperture sizes. MDRS:~4$\times$20 $arcsec$;
971: LWRS:~30$\times$30 $arcsec$. }
972: \tablenotetext{c}{Physical size of projected FUSE aperture.}
973: \tablenotetext{d}{Aperture position angle = 195.87$\degr$. No useful
974: SiC data obtained for this MDRS observation.}
975: \end{deluxetable}
976:
977: \clearpage
978:
979: \begin{deluxetable}{lccl}
980: \tablecolumns{4}
981: \tablewidth{0pt}
982: \tablecaption{Residual Intensities in the Core of the C II 1036.337\,\AA\ Line}
983: \tablehead{Galaxy & $v_{sys}$ & $f_{esc}$(C II) & \multicolumn{1}{c}{Notes on $f
984: _{esc}$ and nearby absorption}\\
985: & (km~s$^{-1}$) & (\%)\\}
986: \startdata
987: NGC~1705 & 569 & $\le 5.8$ & Weak Galactic high velocity \ion{O}{6} $\lambda1037
988: .617$ in red wing of profile.\\
989: & & & No H$_2$ absorption at nearby velocities. \\
990: \\
991: NGC~3310 & 997 & $\le 6-30$ & Value of $f_{esc}$ increases along redward wing of
992: absorption.\\
993: \\
994: NGC~4214 & 298 & $\le 4.0$ & Weak Galactic H$_2$ R(1) $\lambda1037.149$ on blue
995: side of profile.\\
996: \\
997: M~83 & 517 & $\le 6.0$ &
998: Weak Galactic H$_2$ P(1) $\lambda1038.157$ on red side of profile.\\
999: & & & Galactic \ion{O}{6} $\lambda1031.926$ on blue side of profile. \\
1000: \\
1001: NGC~5253 & 405 & $\le 6.0$ & Weak Galactic H$_2$ P(1) $\lambda1038.157$ on red s
1002: ide of profile.\\
1003: & & & Galactic \ion{O}{6} $\lambda1031.926$ on blue side of profile. \\
1004: \enddata
1005: \end{deluxetable}
1006:
1007:
1008:
1009:
1010: %% The following command ends your manuscript. LaTeX will ignore any text
1011: %% that appears after it.
1012:
1013: \end{document}
1014:
1015: