1: \documentclass{emulateapj}
2:
3: \begin{document}
4:
5: \newcommand{\kms}{\,km\,s$^{-1}$}
6: \newcommand{\vej}{v_{\mathrm{ej}}}
7: \newcommand{\vmax}{v_{\mathrm{max}}}
8: \newcommand{\acd}{N_{\mathrm{a}}}
9: \newcommand{\zabs}{z_{\mathrm{abs}}}
10: \newcommand{\zem}{z_{\mathrm{em}}}
11:
12: \slugcomment{Accepted for Publication in {\it The Astrophysical
13: Journal}}
14:
15: \title{On the Fraction of Quasars with Outflows}
16: %
17: \author{Rajib Ganguly \& Michael S. Brotherton\altaffilmark{1}}
18: %
19: \altaffiltext{1}{Department of Physics \& Astronomy, The University
20: of Wyoming (Dept. 3905), 1000 East University Ave., Laramie, WY,
21: 82071}
22: %
23: \begin{abstract}
24: Outflows from active galactic nuclei (AGNs) seem to be common and
25: are thought to be important from a variety of perspectives: as an
26: agent of chemical enhancement of the interstellar and intergalactic
27: media, as an agent of angular momentum removal from the accreting
28: central engine, and as an agent limiting star formation in
29: starbursting systems by blowing out gas and dust from the host
30: galaxy. To understand these processes, we must determine what
31: fraction of AGNs feature outflows and understand what forms they
32: take. We examine recent surveys of quasar absorption lines,
33: reviewing the best means to determine if systems are intrinsic and
34: result from outflowing material, and the limitations of approaches
35: taken to date. The surveys reveal that, while the fraction of
36: specific forms of outflows depends on AGN properties, the overall
37: fraction displaying outflows is fairly constant, approximately
38: 60$\%$, over many orders of magnitude in luminosity. We emphasize
39: some issues concerning classification of outflows driven by data
40: type rather than necessarily the physical nature of outflows, and
41: illustrate how understanding outflows probably requires more a
42: comprehensive approach than has usually been taken in the past.
43: \end{abstract}
44:
45: \keywords{quasars: general --- quasars: absorption lines ---
46: galaxies: active --- accretion}
47:
48: \section{Introduction}
49:
50: The role of outflows from quasars and active galactic nuclei (AGN)
51: has recently become an important feature in the overall framework of
52: how galaxies and star formation processes evolve over cosmic time.
53: Mergers and other interactions triggering AGN seem to provide
54: feedback affecting the larger scale environment. Recent efforts to
55: include the effects of this so-called AGN feedback focus on two
56: modes: a ``radio'' mode whereby a relativistic jet heats the
57: surrounding interstellar and intercluster media
58: \citep[e.g.,][]{best07}, and a ``quasar'' mode whereby a lower
59: velocity but higher mass outflow also helps to clear out post-merger
60: shrouding gas and quenches star formation \citep*[e.g.,][]{tdm05}.
61: We focus on this second mode in this paper. For this mode, a number
62: of questions require addressing. How common are outflows? Do all AGN
63: have outflows? What drives outflows? Is there a single all-governing
64: structure of AGN? Answering these questions will help us to
65: understand the role AGN outflows with respect to issue of feedback,
66: and other important issues like chemical enrichment and accretion.
67:
68: In the ensuing sections we aim to achieve several goals: (1) to
69: review the ways in which outflows are detected in AGN over all
70: luminosity scales; (2) to comment on the merits of various catalogs
71: of outflows; and (3) to arrive at the true (possibly
72: property-dependent) observed frequency of outflows. In its most
73: basic interpretation, the observed frequency of outflows can be
74: equated with the fraction of solid angle (from the view point of the
75: central black hole) subtended by outflowing gas. This interpretation
76: assumes that all AGN feature outflows and that not all sight-lines
77: to the emitting regions are occulted by the outflow. Alternatively
78: (and equally simplistic), the frequency can be interpreted as the
79: fraction of the duty cycle over which AGN feature outflows (assuming
80: the outflow subtends 4$\pi$\ steradians). The actual conversion of
81: the fraction of AGN featuring spectroscopic evidence of outflow to
82: the solid angle subtend by such outflows has been treated by
83: \citet{cren99} and \citet*{cren03b}. This computation involves
84: further knowledge of the line-of-sight covering factor (that is, the
85: fraction of lines-of-sight that reach the observer that are occulted
86: by the outflow) as well as an understanding of range of solid angle
87: sampled by the AGN used (e.g., Type 1 versus Type 2 AGN). The true
88: situation is likely in between these two extremes, and may depend
89: also on properties we can not currently constrain, such as the time
90: since the AGN was triggered.
91:
92: Additionally, we strive here to build a case that more effort should
93: be made to consider outflows of all types together. Often data
94: limitations of one sort or another have led to the study of limited
95: parts of parameter space (e.g., outflow velocity or velocity
96: dispersion), creating artificial or at least biased divisions. There
97: appears to be a continuous range in properties of outflows and these
98: should only be regarded as fundamentally different when there is clear
99: evidence to reach such a conclusion. Below we discuss the
100: identification of outflows (\S 2) and the data-driven subcategories
101: (\S 3). We show an illustrative example of how combining the
102: different outflow subclasses may lead to a more unified physical
103: understanding of outflows (\S 4). Finally, we bring together the
104: different survey methodologies to determine an overall fraction of AGN
105: displaying the signatures of outflows (\S 5) and summarize the case
106: for more global studies of the outflow phenomenon. We adopt a
107: cosmology with $\Omega_M$ = 0.3, $\Omega_{\Lambda}$ = 0.7, and H$_0$ =
108: 70 km s$^{-1}$ Mpc$^{-1}$.
109:
110: \section{The Ideal Way To Select Quasars Exhibiting Outflows}
111:
112: Outflows from AGN are primarily detected in ultraviolet and X-ray
113: absorption against the compact continuum source (i.e, the inner
114: portions of the accretion disk) and/or the more extended broad
115: emission line region. In a few cases, outflows have been
116: demonstratively observed in emission both from the broad line region
117: in narrow-line Seyfert 1 galaxies
118: \citep[e.g.,][]{leighly04a,leighly04b,yuan07} and from the narrow
119: line region of Seyfert 1 galaxies \citep[e.g.,][]{das05,das06}.
120: [Arguably, the fact that broad emission lines in most AGN have only
121: a single peak is also a signature of outflowing gas
122: \citep[e.g.,][]{mur95}.] For emission-line gas, reverberation
123: mapping provides a direct means at establishing the location of the
124: gas. For absorption-line gas, placing a distance between the gas and
125: the ionizing continuum relies on using absorption-line diagnostics
126: to assess the photoionization parameter ($U$), and having other
127: information that constrains the density ($n$) of the gas. The
128: distance, $r$, is related to these quantities via
129: \begin{equation}
130: r = \sqrt{{{\int_{\nu_{\mathrm{LL}}}^{\infty} (L_\nu/\nu) d\nu}
131: \over {4 \pi h c n U}}},
132: \end{equation}
133: where $\nu_\mathrm{LL} = 3.3 \times 10^{15}$\,Hz is the frequency of
134: the Lyman limit. Constraints on the density can come from
135: time-variability (if ionization/recombination dominates the
136: variability timescale), or the presence of excited-state lines.
137: Density information is not typically available for intrinsic
138: absorbers, so secondary indicators must be employed to separate
139: intrinsic absorbers from absorption by interloping structures (e.g.,
140: IGM filaments, galaxy halos and disks).
141:
142: In order of decreasing utility and importance, the secondary
143: indicators of an intrinsic origin for an absorption-line system are:
144: (1) velocity width, (2) partial coverage, (3) time variability, (4)
145: high photoionization parameter, (5) high metallicity,
146: \citep[e.g.][]{bs97}. Not all intrinsic absorbers exhibit all of
147: these properties, but the probability of an intrinsic origin is
148: higher if an absorber exhibits more than one property. Likewise,
149: with the exception of the first two indicators (and the first only
150: in its most extreme, see \S\ref{sec:bal}), each of these indicators
151: have been observed in intervening material. Thus, by themselves, no
152: one indicator should be taken to imply an intrinsic origin.
153:
154: Historically, the first criterion has led to three divisions in the
155: classification of intrinsic absorbers. Outflows with the largest
156: velocity dispersions are termed ``broad absorption lines''
157: \citep[e.g.,][BALs, FWHM $\geq2000$\,\kms]{weymann91}. On the other
158: extreme, intrinsic absorbers where the velocity dispersion is
159: sufficiently small as to cleanly separate the \ion{C}{4} doublet are
160: called ``narrow absorption lines'' \citep[e.g.,][NALs, FWHM
161: $\leq500$\,\kms]{hf99}. Since there is a whole continuum of velocity
162: widths, this has led to an in-between class known as ``mini-BALs''
163: \citep[e.g.,][]{ham97c,cssg99}. Below we examine each of these
164: classes in terms of their observed frequency, and note various
165: issues in determining this number, including dependencies on quasar
166: physical properties.
167:
168: \section{Observations of the Incidence of Various Forms of Outflows}
169:
170: \subsection{Broad Absorption Lines (BALs)}
171: \label{sec:bal}
172:
173: Broad absorption lines in the spectra of quasars are the most easily
174: identifiable forms of outflows. The large velocity width is very
175: readily associated with accelerated, outflowing gas. As such, these
176: garnered more attention historically than their smaller
177: velocity-width kin
178: \citep[e.g.][]{weymann85,turn88,turn88b,weymann91,vwk93}.
179:
180: \citet{weymann91} established criteria, summarized in a number
181: called the BALnicity index (BI), for determining if an absorption
182: line constituted a BAL. The BI was a modified form of an equivalent
183: width whereby one counted absorption that fell below 90\% of the
184: true quasar continuum that was contiguously below this level for
185: more than 2000\,\kms. Moreover, no absorption within 3000\,\kms\ of
186: the quasar redshift was counted in order to remove possible
187: contamination by absorption-line gas not physically associated with
188: the quasar central engine (e.g., interstellar gas from the quasar
189: host galaxy, or intergalactic material from the host cluster).
190: [Note: With a minimum velocity of 3000\,\kms and a minimum
191: contiguous width of 2000\,\kms, this means that no absorption
192: falling entirely within 5000\,\kms\ of the quasar redshift is
193: counted.] This index was established using low-dispersion data of
194: high-redshift ($1.5 \leq z \leq 3.0$) objects from the Large Bright
195: Quasar Survey, or LBQS \citep{lbqs1,foltz89,lbqs3,lbqs6}, and was designed to
196: yield a pure sample of objects with bonafide outflows. We note here
197: that the utility of BI was driven purely by the data quality (signal-to-noise
198: ratio and resolution) of the LBQS spectra in conjunction with the desire
199: to remove false positives (at the expense of losing some true BAL quasars).
200: While the use of BI to define samples of BAL quasars has utility, especially
201: in comparing results between data sets of varying quality, it
202: excludes some fraction of real high-velocity dispersion outflows that
203: qualitatively appear to be BAL quasars but just fail to have positive BI.
204:
205: An improvement on the BI, termed the intrinsic absorption index (AI),
206: was developed by \citet{hallai} to alleviate the inadequacies of BI in
207: selecting objects where high velocity outflows were clearly observed
208: but were not included as BAL quasars by the BI criteria (e.g., UM 660,
209: PG 2302+029). The AI was designed to be more flexible and inclusive
210: and has been very useful in its application to newer and better
211: quality datasets like the Sloan Digital Sky Survey (SDSS). This
212: flexibility, while good at including objects not previously selected
213: by BI, has increased the contamination of samples of intrinsic
214: absorption while still not including other forms of intrinsic
215: absorption \citep[e.g.,][]{gan07b}.
216:
217: The incidence of BALs has primarily been determined using optical
218: spectra where, historically, large samples of high-redshift quasars
219: (to get rest-frame UV coverage) could efficiently be selected (e.g.,
220: with color-selection). In such surveys
221: \citep{hf03,reichard03,trump06,gan07b}, roughly 10-25\% of objects are
222: observed to host BALs. An issue with optical/UV surveys, however, is
223: potential biases in the selection of quasars against those hosting
224: BALs due to the fact that much of the continuum is absorbed
225: \citep[e.g.,][]{good95,goodrich97,kv98} and intrinsically reddened
226: \citep[e.g.,][]{reichard03}. Using the LBQS, where the observed
227: frequency of BAL quasars in the redshift range $1.5 \leq z \leq 3$\ is
228: 15\% using a BI criterion, \citet{hf03} estimated a true BAL frequency
229: of 22\% from comparisons in the $k$-corrections of BAL and non-BAL
230: quasars. The recent catalog of BAL quasars using an AI criterion from
231: \citet{trump06} found a BAL frequency of 26\% (in the redshift range
232: $1.7 \leq z \leq 4.38$). Both of these estimates are based on the
233: \ion{C}{4}\ $\lambda1548,1550$\ doublet, which is the most commonly
234: used species in selecting intrinsic absorption owing to the relatively
235: high abundance of carbon, the high ionization fraction of C$^{3+}$\ in
236: moderately-ionized gas, and the resonant absorption of the doublet.
237:
238: To combat possible selection biases in the optical, one can examine
239: quasar catalogs selected in other bands. \citet{becker00} examined
240: radio-selected quasars from the FIRST Bright Quasar Survey and found
241: a BAL quasar frequency of about 18\% (though it is only 14\% if only
242: BI$>0$\ objects are counted, comparable to other estimates based on
243: optical-selection). This again predominantly used the \ion{C}{4}
244: doublet and objects at $z > 1.7$.
245:
246: Incidentally, several studies \citep[e.g.,][]{bro98} have now
247: dismantled the myth that BALs are only observed in formally
248: radio-quiet (i.e.,
249: $f_\nu(\mathrm{5\,GHz})/f_\nu(\mathrm{3000\,\AA})<10$) objects, though
250: their frequency does significantly decrease among the most radio-loud
251: quasars \citep*{becker01,gregg06}. We note that a subset of
252: radio-selected BAL quasars can be identified as polar outflows
253: \citep*[e.g.,][]{zhou06,brotherton06,gp07}. At least one of these
254: objects, FIRST J155633.8$+$351758, appears to be an optically reddened
255: and beamed radio-quiet quasar \citep{rb07}. The presence of BAL
256: outflows in such objects as well in as edge-on FR II BAL quasars,
257: \citep[e.g., ][]{gregg06} indicates high-velocity outflows are present
258: in a variety of geometries. There is as yet no observational
259: signature in the absorption spectra that is correlated with
260: orientation indicators, so any geometrically restrictive model such as
261: those identifying BAL outflows solely with equatorial winds are either
262: wrong or incomplete. Any complete picture of outflows must reflect a
263: range of geometries. It has yet to be established observationally how
264: often polar outflows occur compared to equatorial, or if the location
265: or dynamics differ.
266:
267: In addition to radio-selection, one can examine the frequency of BAL
268: quasars from infrared selection. Recently, \citet{dai07} compared the
269: catalog of BAL quasars \citep{trump06} from the Third Data Release
270: (DR3) of SDSS and the parent sample of DR3 quasars \citep{sdssqso3}
271: with the Two Micron All-Sky Survey \citep[][2MASS]{2mass} Point-Source
272: Catalog (PSC). With some variation with redshift, they reported an
273: overall true BAL quasar fraction of $43\pm2$\%, markedly higher than
274: estimates based on UV/optical data alone. Presumably, this difference
275: accounts for the effects of dust and absorption that may bias
276: UV/optical selection techniques against find BAL quasars.
277:
278: We point out that this estimate relies heavily on the automated
279: techniques employed in finding BAL quasars in a large dataset such
280: as SDSS. From a critical look at 5088 $1.7 < z < 2$\ quasars from
281: SDSS DR2, \citet{gan07b} noted several instances of false (and
282: missed) classifications in the \citet{trump06} catalog. A comparison
283: of the \citet{gan07b} sample with the 2MASS PSC reveals a BAL
284: fraction of 66/287 (23\%), completely consistent with the analysis
285: of \citet{hf03}. Blindly using the \citet{trump06} catalog yields a
286: BAL fraction of 96/287 (33\%), consistent with the $z<2$\ points
287: from \citet[][see their Figure 4]{dai07}. At face value, this
288: implies that nearly 30\% of the \citet{trump06}-2MASS cross-matched
289: sample consists of false-positives. We return to the issue of
290: false-negatives below.
291:
292: %-- attempts in the hard x-rays??? soft X-rays??? \citet{green95} has
293: %observations of LBQS...
294:
295: %-- separate discussion of LoBALs/classification based on ionization?
296:
297:
298: \subsection{Narrow Absorption Lines (NALs) and mini-BALs}
299:
300: Intrinsic NALs and mini-BALs have, within the last decade, come to
301: light as a very powerful and complementary means of studying
302: outflows. Unlike their very broad kin, these absorbers are generally
303: not blended and, therefore, offer a means to determine ionization
304: levels and metalicities using absorption-line diagnostics. Thus,
305: NALs and mini-BALs are more useful as probes of the physical
306: conditions of outflows. The drawback, however, is that truly
307: intrinsic NALs and mini-BALs are more difficult to identify, since
308: interloping structures such as the cosmic web, galaxy clusters, and
309: galactic disks and halos also have comparable velocity spreads
310: ($\lesssim800$\,\kms). Historically, progress was made by
311: statistically identifying an excess of absorbers over what is
312: expected from randomly distributed intervening structures
313: \citep[e.g.,][]{wwpt}. With improved technologies (such as
314: high-resolution spectroscopy with large telescopes), we can now take
315: advantage of the other secondary indicators to separate intrinsic
316: from intervening absorption. In the following subsections, we
317: discuss the frequency of two subclasses based on both historical and
318: more recent studies. We distinguish between absorbers that appear
319: near the quasar redshift (associated absorbers), and those that
320: appear at large velocity separations.
321:
322: \subsubsection{Associated ($\zabs\sim\zem$) Absorbers (AALs)}
323: \label{sec:aal}
324:
325: The term ``associated'' refers to narrow velocity-dispersion
326: absorption-line systems that lie near the quasar redshift. It has
327: been shown that the frequency of such systems is much larger than
328: those at large velocity separations
329: \citep{wwpt,foltz87,and87,ald94,rich99,rich01b}. Typically,
330: associated absorbers are defined as those lying within 5000\,\kms\
331: of the quasar redshift \citep{foltz86}. As such, they were
332: historically very complementary to BAL quasars selected using BI.
333: Updating BAL classification to reflect the better data quality
334: usually available today does allow for some confusion among classes,
335: at least in some cases, and this should be kept in mind. The issue
336: of what types of quasars hosted AALs was the subject of much
337: scrutiny with some studies claiming to see an excess of AALs
338: \citep[e.g.,][]{foltz87}, while other studies claimed no excess
339: \citep*[e.g.,][]{sbs88}. It was surmised that strong AALs (i.e.,
340: those with a large \ion{C}{4} equivalent width) were preferentially
341: found in optically-faint, steep radio spectrum quasars
342: \citep{mj87,foltz88}. However, more recent studies have found that
343: AALs are found (with varying frequency) in all AGN subclasses from
344: Seyfert galaxies \citep[e.g.,][]{cren99,kriss06} to higher luminosity
345: quasars \citep[e.g.][]{gan01a,lb02,ves03,misawa07}, and from steep
346: to flat radio spectrum sources \citep{gan01a,ves03}.
347:
348: An important issue in the consideration of AALs as it relates to
349: outflows is where the absorbing gas originates. We note here a few
350: arguments for a direct association with outflows from the central
351: engine. While detailed studies of individual objects have shown
352: absorption-line {\it components} that must reside in the host galaxy
353: far from the central engine \citep[e.g.,][]{ham01,scott04,gan06a},
354: on the whole there have been no documented cases of AALs that are
355: truly redshifted with respect to the actual systemic velocity. If
356: AALs were to originate in the host galaxy, one would expect some
357: fraction of the absorbers to arise from infalling material. In fact,
358: the velocity distribution of \ion{C}{4} AALs is sharply peaked with
359: the \ion{C}{4} emission redshift \citep{gan01a}, implying a close
360: dynamical connection between AALs and the broad emission-line
361: region. In addition, blind studies of AALs using secondary
362: indicators find that $\geq20$\% are time-variable \citep{wise04},
363: and that $\sim33$\% show partial coverage \citep{misawa07}.
364:
365: From an analysis of 59 $z<1$\ quasars, \citet{gan01a} showed that
366: the overall frequency of AALs was 25$\pm$6\%, with some variation
367: with broad-band spectral properties. Similar frequencies have been
368: established at higher redshift by \citet[][$27\pm5$\%]{ves03} and
369: \citet[][23\%]{misawa07}, both of which made attempts to filter out
370: contamination by intervening absorbers. Oddly, these fractions are
371: lower than the recent study of \citet{gan07b}, who find an AAL
372: frequency of 1898/5088 (37\%), although the 5000\,\kms\ velocity
373: cutoff for traditional AALs was not strictly adhered to in that
374: survey. We note that 1478/1898 (78\%) AALs in that study were missed
375: by the AI selection used by \citet{trump06}. These certainly
376: constitute false-negatives from the standpoint of finding intrinsic
377: absorption, though not from the standpoint of finding only BAL
378: quasars. While \citet{ves03} did note that quasars with AALs are
379: redder on average, a comparison of the \citet{gan07b} sample with
380: the 2MASS PSC reveals that the frequency of AALs is similar to the
381: parent sample (107/287, 37\%). Thus, the selection of AALs quasars
382: is not affected by optical biases (e.g., reddening or large optical
383: absorption) like BAL quasars.
384:
385: \subsubsection{High Velocity NALs}
386:
387: The first observational evidence for intrinsic narrow
388: velocity-dispersion absorption appearing at high ejection velocity
389: (many tens of thousands of kilometers per second) came nearly a
390: decade ago and include: PG\,2302+029 \citep{kp96}, Q\,2343+125
391: \citep*{ham97b}, and PG\,0935+417 \citep{ham97c}. Models of quasar
392: winds generally are able to explain outflows with $\Delta v/v \sim
393: 1$, but are challenged by these $\Delta v/v << 1$\ systems. One idea
394: is that the sight-line cuts across the outflow that would produce a
395: BAL under a difference orientation \citep[e.g.,][]{elvis00,gan01a},
396: but this has yet to be demonstrated theoretically. These systems are
397: also interesting because they only absorb photons from the compact
398: continuum. Thus, partial coverage indicators provide severe
399: constraints on the geometry of the flow.
400:
401: In terms of demographics, the first assessment of the frequency of
402: these systems came from \citet{rich99} and \citet{rich01b}. From a
403: statistical analysis examining the variation in the velocity
404: distribution of \ion{C}{4} NALs with quasar radio-loudness and
405: spectral index, \citet{rich99} estimated that as many as 36\% of
406: \ion{C}{4} NALs may arise from outflowing gas. Recently,
407: \citet{misawa07} report that only 10-17\% of \ion{C}{4} NALs in the
408: velocity range 5000-70000\,\kms\ show evidence of partial coverage.
409: (Thus it is possible that \citet{rich99} overestimated the fraction
410: of high-velocity NALs, or that 50-70\% of intrinsic \ion{C}{4} NALs
411: do not show partial coverage.) This is not a statement, however, on
412: the fraction of quasars that host such outflows.
413:
414: \citet{ves03} reported that high velocity intrinsic NALs appeared in
415: 18$\pm$4\% of $1.5 < z < 3.6$\ quasars in the velocity range
416: 5000--21000\,\kms, with about a factor of two variation between
417: radio core-dominated (17$\pm$10\%) and radio lobe-dominated
418: (33$\pm$15\%) morphologies. In a recent survey of $1.8 < z < 3.5$\
419: SDSS sources, \citet{prh07} find about 12\% of quasars have
420: high-velocity NALs in the velocity range 5000-50000\,\kms, and
421: $\sim2.3$\% in the velocity range 25000-50000\,\kms. This latter
422: velocity range is often missed by surveys due purely to
423: observational cutoffs. Over this velocity range, absorption by
424: \ion{C}{4} can become confused with \ion{Si}{4} absorption.
425:
426: \section{An Example Illustrating the Merits of Comprehensive Outflow Studies}
427:
428: \begin{figure}
429: \epsscale{0.7} \rotatebox{-90}{\plotone{f1.eps}}
430: \protect\caption[vmax vs. luminosity]{We present a summary plot of
431: the maximum velocity of absorption versus the monochromatic
432: luminosity at 3000\,\AA. Data from several studies are included:
433: Seyfert 1 galaxies from \citet[][{\it filled blue circles}]{cren99};
434: $z<0.5$\ Palomar-Green quasars from \citet[][{\it open green
435: triangles}]{lb02}; SDSS DR2 BAL quasars from \citet[][{\it filled
436: yellow circles}]{gan07b}; LBQS BAL quasars from Gallagher et al.
437: (2006, {\it open red squares}); intrinsic NALs from \citet[][{\it
438: filled blue pentagons}]{ves03}; intrinsic NALs from \citet[][{\it
439: open cyan pentagons}]{misawa07}; polar BAL quasars from
440: \citet[][{\it pink stars}]{gp07}; UM 675, intrinsic NAL/mini-BALs in
441: Q\,2343+125, and PG\,0935+417 from \citet[][{\it black
442: stars}]{ham97c}; and the mini-BAL in PG\,2302+029 from \citet[][{\it
443: yellow star}]{kp96}.} \label{fig:vmaxlum}
444: \end{figure}
445:
446: When parameter space is truncated, either intentionally (e.g.,
447: through subclass segregation or the desire to avoid false
448: positives/negatives) or unintentionally (e.g., by data limitations),
449: real correlations that could lead to physical understanding may be
450: missed. The wide variety of observational techniques and the
451: improvements in sample size now make it possible to study the
452: outflow phenomenon in a more complete manner than ever before
453: possible.
454:
455: \input{tab1.tex}
456:
457: As discussed above, outflow velocity (both in terms of dispersion
458: and range) has often been limited in many studies. By putting
459: together different surveys, we can study the velocity properties of
460: outflows as a function of other potentially important parameters.
461: One recent example where this has been shown to be fruitful is in
462: plotting maximum velocity of absorption, $\vmax$, against
463: luminosity. This was originally done by \citet{lb02} for PG
464: quasars, who, finding an upper envelope formed by the soft X-ray
465: weak absorbers, argued for radiation-driven winds as quasar
466: outflows. More recently, others have added other samples of outflows
467: to this plot \citep[e.g.,][]{gallsc06,misawa07,gan07b} showing that
468: this envelope can be extrapolated to higher luminosities and
469: similarly constrains luminous BAL quasars.
470:
471: We have further added to this plot in Figure~\ref{fig:vmaxlum}
472: \citep[with the envelope fit from ][]{gan07b}, including all types
473: of intrinsic outflows and the most extreme in terms of both terminal
474: velocity and luminosity. Polar BAL quasars are included as well and
475: fall among the general BAL quasar population. Intrinsic NAL systems
476: are well represented throughout the full luminosity range. Some
477: data biases are present; for instance, the wavelength range of SDSS
478: spectra limited the observed $\vmax$\ to less than 30000 \kms\ for
479: most quasars studied by \citet{gan07b}, so the gap under the
480: envelope for luminosities above $10^{46}$\,erg\,s$^{-1}$\ is
481: potentially not real. Similarly, searching for intrinsic absorption
482: at velocities above 50000\,\kms\ becomes very difficult since the
483: \ion{C}{4} lines become blueshifted into the \ion{N}{5}/Lyman
484: $\alpha$ region where identification is especially challenging. Of
485: course, the empirical fit does not take into account relativistic
486: effects, so it is inappropriate to extrapolate it to arbitrarily
487: high luminosities. Taken at face value, the fit implies that a
488: quasar with luminosity $\lambda
489: L_\lambda$(3000\,\AA)$=10^{47.3}$\,erg~s$^{-1}$\ would be capable of
490: driving an outflow at the speed of light, but this is clearly
491: unphysical. More data, and insights establishing better criteria as
492: to which outflows sample the envelope are needed to improve our
493: understanding of the dependence on the terminal velocity on quasar
494: physical properties.
495:
496: The figure also illustrates another (though more subtle) issue.
497: While AALs are typically defined as those NALs appearing within
498: 5000\,\kms, some studies at lower redshift have opted for smaller
499: velocity differences, claiming that 5000\,\kms\ is an arbitrary
500: cut-off. The figure shows that there is a physical reason for this.
501: As the figure apparently shows, no outflow appears to be driven to a
502: velocity larger than is allowed by radiation pressure. At lower
503: redshifts, most objects studied are also lower luminosity (e.g.,
504: Seyfert galaxies). The figure clearly shows that objects with
505: $\lambda L_\lambda(3000\,\mathrm{\AA})\lesssim 3 \times
506: 10^{44}$\,erg\,s$^{-1}$ are not capable of radiatively driving
507: outflows with velocities larger than 5000\,\kms. An insight such as
508: this is not only interesting for understanding AGN outflow physics,
509: but is also of use to other fields that make use of intervening
510: absorption-line systems (as it presents an additional means at
511: separating intrinsic from intervening absorbers).
512:
513: \section{What is the True Fraction of Outflows?}
514:
515: Table~\ref{tab:demographics} summarizes the incidence of outflows in
516: AGN from several recent surveys (with the redshift and luminosity
517: ranges of the AGN listed in columns two and three, respectively, and
518: the ranges in outflow velocity and velocity width listed in columns
519: four and five, respectively). Inspection of the table shows that the
520: outflow fraction is dependent both on the characteristics of the
521: parent sample of AGN used, and on the forms of intrinsic absorption
522: included. If one only counts BALs observed in higher luminosity AGN
523: [$\lambda L_\lambda$(3000\,\AA)$\gtrsim 10^{45}$\,erg s$^{-1}$],
524: then the outflow fraction is 23\,\% \citep{hf03,gan07b}. However,
525: this is by no means a complete assessment of outflows.
526:
527: In order to compute a more complete outflow fraction, one must deal
528: with three issues: (1) cross-talk among classifications, (2)
529: dependence of frequencies on quasar properties, and (3) mode of the
530: outflow. Here, we ignore the third issue and focus on the first two.
531: \citet{cren99} surveyed outflows in lower luminosity [$\lambda
532: L_\lambda$(3000\,\AA)$\lesssim 10^{45}$\,erg s$^{-1}$] AGN and we
533: use their result, 59\%, as one benchmark \citep[see also][who finds
534: a similar percentage based on \ion{O}{6} absorption]{kriss06}. For
535: higher luminosity AGN, we start with the \citet{hf03} percentage of
536: 23\%, as it is the purest, and most well-defined sample of outflows.
537: To this, we must add in two things, the contribution from AALs, and
538: the contribution from high-velocity NALs/mini-BALs. There is general
539: agreement between \citet{gan01a} and \citet{ves03} that the AAL
540: fraction is 23--27\,\%. As noted by \citet{gan01a}, quasars that
541: host broad-absorption lines also tend to host associated absorption.
542: However, the sample of quasars employed in the \citet{ves03}
543: estimates explicitly does not include BAL quasars. Thus, while there
544: is likely some cross-talk between the class of BAL quasars and AAL
545: quasars, the above range should minimize this effect. Thus, the
546: outflow fraction counting BALs and AALs (integrated over ``all''
547: high luminosity AGN) is 46--50\%. For high velocity NALs/mini-BALs,
548: there is a more sizeable error margin (12--30\%) owing to cross-talk
549: and dependence on quasar property. Adding this uncertain number
550: gives our final tally: 57-80\%. This fraction is surprisingly
551: comparable to that of lower luminosity AGN.
552:
553: An alternative approach is to begin with the complete sample of
554: outflows from \citet{gan07b}. Correcting their overall outflow
555: fraction (2515/5088, 49\%) for quasar selection biases
556: \citep[following the strategy of][]{dai07}, we find an outflow
557: fraction of 60$\pm$5\% ($66/287 + 107/287$, see \S\ref{sec:bal}, and
558: \S\ref{sec:aal}). This falls in the above range, and the only
559: missing form of outflow from that sample is NALs/mini-BALs at
560: velocities larger than $\sim30000$\,\kms. We further note that this
561: explicitly eliminates cross-talk between quasars hosting BALs and
562: quasars hosting AALs.
563:
564: \section{Summary}
565:
566: We conclude that, largely independent of AGN luminosity, 60\% is a good
567: reference number for the percentage of AGN with intrinsic outflows. This
568: number may increase slightly with more thorough searches of parameter
569: space (e.g., very high velocity outflows). Until evidence suggests
570: otherwise, we recommend that quasar outflows be studied as a single phenomenon
571: whenever possible and that restrictions based on absorber subclass or data
572: limitations be clearly stated and considered in the interpretation of results.
573: Catalogs should clearly state their contamination issues and their limitations
574: for particular applications as appropriate. A more comprehensive understanding
575: of the outflow phenomenon awaits us.
576:
577: %\section{Mode of the Outflow}
578:
579: \acknowledgements
580:
581: We wish to thanks the anonymous referee for comments that improved
582: the quality of the paper. This publication makes use of data
583: products from the Two Micron All Sky Survey, which is a joint
584: project of the University of Massachusetts and the Infrared
585: Processing and Analysis Center/California Institute of Technology,
586: funded by the National Aeronautics and Space Administration and the
587: National Science Foundation. We acknowledge support from the US
588: National Science Foundation through grant AST 05-07781.
589:
590: %\bibliographystyle{apj}
591: %\bibliography{bibliography}
592:
593: \begin{thebibliography}{}
594:
595: \bibitem[\protect\citeauthoryear{{Aldcroft}, {Bechtold}, \& {Elvis}}{{Aldcroft}
596: et~al.}{1994}]{ald94}
597: {Aldcroft}, T.~L., {Bechtold}, J., \& {Elvis}, M. 1994, \apjs, 93, 1
598:
599: \bibitem[\protect\citeauthoryear{{Anderson} et~al.}{{Anderson}
600: et~al.}{1987}]{and87}
601: {Anderson}, S.~F., {Weymann}, R.~J., {Foltz}, C.~B., \& {Chaffee}, F.~H. 1987,
602: \aj, 94, 278
603:
604: \bibitem[\protect\citeauthoryear{{Barlow} \& {Sargent}}{{Barlow} \&
605: {Sargent}}{1997}]{bs97}
606: {Barlow}, T.~A., \& {Sargent}, W.~L.~W. 1997, \aj, 113, 136
607:
608: \bibitem[\protect\citeauthoryear{{Becker} et~al.}{{Becker}
609: et~al.}{2000}]{becker00}
610: {Becker}, R.~H., {White}, R.~L., {Gregg}, M.~D., {Brotherton}, M.~S.,
611: {Laurent-Muehleisen}, S.~A., \& {Arav}, N. 2000, \apj, 538, 72
612:
613: \bibitem[\protect\citeauthoryear{{Becker} et~al.}{{Becker}
614: et~al.}{2001}]{becker01}
615: {Becker}, R.~ H., {White}, R.~L., {Gregg}, M.~D., {Laurent-Muehleisen}, S.~A.,
616: {Brotherton}, M.~S., {Impey}, C.~D. {Chaffee}, F.~H., {Richards}, G.~T.,
617: {Helfand}, D.~J., {Lacy}, M., {Courbin}, F., \& {Proctor}, D. D. 2001, \apjs,
618: 135, 227
619:
620: \bibitem[\protect\citeauthoryear{{Berrington} et~al.}{{Berrington}
621: et~al.}{2007}]{rb07}
622: {Berrington}, R.~C., {Brotherton}, M.~S., {Gallagher}, S.~C.,
623: {Ganguly}, R. {Shang}, Zhaohui, {Lacy}, M., {Gregg}, M.~D., \& Hall,
624: P.~B. 2007, \apj, in preparation
625:
626: \bibitem[\protect\citeauthoryear{{Best}}{{Best}}{2007}]{best07}
627: {Best}, P.~N. 2007, New Astronomy Review, 51, 168
628:
629: \bibitem[\protect\citeauthoryear{{Brotherton}, {de Breuck}, \&
630: {Schaefer}}{{Brotherton} et~al.}{2006}]{brotherton06}
631: {Brotherton}, M.~S., {de Breuck}, C., \& {Schaefer}, J.~J. 2006, \mnras, 372,
632: L58
633:
634: \bibitem[\protect\citeauthoryear{{Brotherton} et~al.}{{Brotherton}
635: et~al.}{1998}]{bro98}
636: {Brotherton}, M.~S., {van Breugel}, W., {Smith}, R.~J., {Boyle}, B.~J.,
637: {Shanks}, T., {Croom}, S.~M., {Miller}, L., \& {Becker}, R.~H. 1998, \apjl,
638: 505, L7
639:
640: \bibitem[\protect\citeauthoryear{{Churchill} et~al.}{{Churchill}
641: et~al.}{1999}]{cssg99}
642: {Churchill}, C.~W., {Schneider}, D.~P., {Schmidt}, M., \& {Gunn}, J.~E. 1999,
643: \aj, 117, 2573
644:
645: \bibitem[\protect\citeauthoryear{{Crenshaw} et~al.}{{Crenshaw}
646: et~al.}{1999}]{cren99}
647: {Crenshaw}, D.~M., {Kraemer}, S.~B., {Boggess}, A., {Maran}, S.~P.,
648: {Mushotzky}, R.~F., \& {Wu}, C. 1999, \apj, 516, 750
649:
650: \bibitem[\protect\citeauthoryear{{Crenshaw}, {Kraemer}, \& {George}}{{Crenshaw}
651: et~al.}{2003}]{cren03b} Crenshaw, D.~M.,
652: Kraemer, S.~B., \& George, I.~M.\ 2003, \araa, 41, 117
653:
654: \bibitem[\protect\citeauthoryear{{Dai}, {Shankar}, \& {Sivakoff}}{{Dai}
655: et~al.}{2007}]{dai07}
656: {Dai}, X., {Shankar}, F., \& {Sivakoff}, G.~R. 2007, \apj, {submitted
657: (astro-ph/0704.2882)}
658:
659: \bibitem[\protect\citeauthoryear{{Das} et~al.}{{Das}
660: et~al.}{2005}]{das05}
661: {Das}, V., {Crenshaw}, D.~M., {Hutchings},
662: J.~B., {Deo}, R.~P., {Kraemer}, S.~B., {Gull}, T.~R., {Kaiser},
663: M.~E., {Nelson}, C.~H., {Weistrop}, D. 2005, AJ, 130, 945
664:
665: \bibitem[\protect\citeauthoryear{{Das} et~al.}{{Das}
666: et~al.}{2006}]{das06} {Das}, V., {Crenshaw}, D.~M., {Kraemer},
667: S.~B., {Deo}, R. P. 2006, AJ, 132, 620
668:
669: \bibitem[\protect\citeauthoryear{{Di Matteo}, {Springel}, \& {Hernquist}}{{Di
670: Matteo} et~al.}{2005}]{tdm05}
671: {Di Matteo}, T., {Springel}, V., \& {Hernquist}, L. 2005, \nat, 433, 604
672:
673: \bibitem[\protect\citeauthoryear{{Elvis}}{{Elvis}}{2000}]{elvis00}
674: {Elvis}, M. 2000, \apj, 545, 63
675:
676: \bibitem[\protect\citeauthoryear{{Foltz} et~al.}{{Foltz}
677: et~al.}{1989}]{foltz89}
678: {Foltz}, C.~B., {Chaffee}, F.~H., {Hewett}, P.~C., {Weymann}, R.~J.,
679: {Anderson}, S.~F., \& {MacAlpine}, G.~M. 1989, \aj, 98, 1959
680:
681: \bibitem[\protect\citeauthoryear{{Foltz} et~al.}{{Foltz}
682: et~al.}{1988}]{foltz88}
683: {Foltz}, C.~B., {Chaffee}, F.~H., {Weymann}, R.~J., \& {Anderson}, S.~F. 1988,
684: in QSO Absorption Lines; Probing the Universe; Proceedings of the QSO
685: Absorption Line Meeting, ed. J.~C. {Blades}, D.~A. {Turnshek}, \& C.~A.
686: {Norman} (Cambridge: Cambridge University Press), 53
687:
688: \bibitem[\protect\citeauthoryear{{Foltz} et~al.}{{Foltz} et~al.}{1987a}]{lbqs1}
689: {Foltz}, C.~B., {Chaffee}, F.~H., Jr., {Hewett}, P.~C., {MacAlpine}, G.~M.,
690: {Turnshek}, D.~A., {Weymann}, R.~J., \& {Anderson}, S.~F. 1987a, \aj, 94,
691: 1423
692:
693: \bibitem[\protect\citeauthoryear{{Foltz} et~al.}{{Foltz}
694: et~al.}{1987b}]{foltz87}
695: {Foltz}, C.~B., {Weymann}, R.~J., {Morris}, S.~L., \& {Turnshek}, D.~A. 1987b,
696: \apj, 317, 450
697:
698: \bibitem[\protect\citeauthoryear{{Foltz} et~al.}{{Foltz}
699: et~al.}{1986}]{foltz86}
700: {Foltz}, C.~B., {Weymann}, R.~J., {Peterson}, B.~M., {Sun}, L., {Malkan},
701: M.~A., \& {Chaffee}, F.~H. 1986, \apj, 307, 504
702:
703: \bibitem[\protect\citeauthoryear{{Gallagher} et~al.}{{Gallagher}
704: et~al.}{2006}]{gallsc06}
705: {Gallagher}, S.~C., {Brandt}, W.~N., {Chartas}, G., {Priddey}, R., {Garmire},
706: G.~P., \& {Sambruna}, R.~M. 2006, \apj, 644, 709
707:
708: \bibitem[\protect\citeauthoryear{{Ganguly} et~al.}{{Ganguly}
709: et~al.}{2001}]{gan01a}
710: {Ganguly}, R., {Bond}, N.~A., {Charlton}, J.~C., {Eracleous}, M., {Brandt},
711: W.~N., \& {Churchill}, C.~W. 2001, \apj, 549, 133
712:
713: \bibitem[\protect\citeauthoryear{{Ganguly} et~al.}{{Ganguly}
714: et~al.}{2007}]{gan07b}
715: {Ganguly}, R., {Scoggins}, B., {Cales}, S., {Brotherton}, M.~S., \&
716: {Vestergaard}, M. 2007, \apj, 665, 990
717:
718: \bibitem[\protect\citeauthoryear{{Ganguly} et~al.}{{Ganguly}
719: et~al.}{2006}]{gan06a}
720: {Ganguly}, R., {Sembach}, K.~R., {Tripp}, T.~M., {Savage}, B.~D., \& {Wakker},
721: B.~P. 2006, \apj, 645, 868
722:
723: \bibitem[\protect\citeauthoryear{{Ghosh} \& {Punsly}}{{Ghosh} \&
724: {Punsly}}{2007}]{gp07}
725: {Ghosh}, K.~K., \& {Punsly}, B. 2007, \apjl, 661, 139
726:
727: \bibitem[\protect\citeauthoryear{{Goodrich}}{{Goodrich}}{1997}]{goodrich97}
728: {Goodrich}, R.~W. 1997, \apj, 474, 606
729:
730: \bibitem[\protect\citeauthoryear{{Goodrich} \& {Miller}}{{Goodrich} \&
731: {Miller}}{1995}]{good95}
732: {Goodrich}, R.~W., \& {Miller}, J.~S. 1995, \apjl, 448, L73
733:
734: \bibitem[\protect\citeauthoryear{{Gregg}, {Becker}, \& {de Vries}}{{Gregg} et~al.}{2006}]{gregg06}
735: {Gregg}, M.~D., {Becker}, R.~H., \& {de Vries}, W. 2006, \apj, 641, 210
736:
737: \bibitem[\protect\citeauthoryear{{Hall} et~al.}{{Hall} et~al.}{2002}]{hallai}
738: {Hall}, P.~B., et~al. 2002, \apjs, 141, 267
739:
740: \bibitem[\protect\citeauthoryear{{Hamann} et~al.}{{Hamann}
741: et~al.}{1997}]{ham97c}
742: {Hamann}, F., {Barlow}, T., {Cohen}, R.~D., {Junkkarinen}, V., \& {Burbidge},
743: E.~M. 1997, in ASP Conf. Ser. 128: Mass Ejection from Active Galactic Nuclei,
744: ed. N.~{Arav}, I.~{Shlosman}, \& R.~J. {Weymann} (San Francisco: ASP), 19
745:
746: \bibitem[\protect\citeauthoryear{{Hamann}, {Barlow}, \& {Junkkarinen}}{{Hamann}
747: et~al.}{1997}]{ham97b}
748: {Hamann}, F., {Barlow}, T.~A., \& {Junkkarinen}, V. 1997, \apj, 478, 87
749:
750: \bibitem[\protect\citeauthoryear{{Hamann} \& {Ferland}}{{Hamann} \&
751: {Ferland}}{1999}]{hf99}
752: {Hamann}, F., \& {Ferland}, G. 1999, \araa, 37, 487
753:
754: \bibitem[\protect\citeauthoryear{{Hamann} et~al.}{{Hamann}
755: et~al.}{2001}]{ham01}
756: {Hamann}, F.~W., {Barlow}, T.~A., {Chaffee}, F.~C., {Foltz}, C.~B., \&
757: {Weymann}, R.~J. 2001, \apj, 550, 142
758:
759: \bibitem[\protect\citeauthoryear{{Hewett} \& {Foltz}}{{Hewett} \&
760: {Foltz}}{2003}]{hf03}
761: {Hewett}, P.~C., \& {Foltz}, C.~B. 2003, \aj, 125, 1784
762:
763: \bibitem[\protect\citeauthoryear{{Hewett}, {Foltz}, \& {Chaffee}}{{Hewett}
764: et~al.}{1995}]{lbqs6}
765: {Hewett}, P.~C., {Foltz}, C.~B., \& {Chaffee}, F.~H. 1995, \aj, 109, 1498
766:
767: \bibitem[\protect\citeauthoryear{{Hewett} et~al.}{{Hewett}
768: et~al.}{1991}]{lbqs3}
769: {Hewett}, P.~C., {Foltz}, C.~B., {Chaffee}, F.~H., {Francis}, P.~J., {Weymann},
770: R.~J., {Morris}, S.~L., {Anderson}, S.~F., \& {MacAlpine}, G.~M. 1991, \aj,
771: 101, 1121
772:
773: \bibitem[\protect\citeauthoryear{{Jannuzi} et~al.}{{Jannuzi}
774: et~al.}{1996}]{kp96}
775: {Jannuzi}, B.~T., et~al. 1996, \apjl, 470, L11
776:
777: \bibitem[\protect\citeauthoryear{{Kriss}}{{Kriss}}{2006}]{kriss06}
778: {Kriss}, G.~A. 2006, in Astronomical Society of the Pacific Conference Series,
779: Vol. 348, Astrophysics in the Far Ultraviolet: Five Years of Discovery with
780: FUSE, ed. G.~{Sonneborn}, H.~W. {Moos}, \& B.-G. {Andersson}, 499
781:
782: \bibitem[\protect\citeauthoryear{{Krolik} \& {Voit}}{{Krolik} \&
783: {Voit}}{1998}]{kv98}
784: {Krolik}, J.~H., \& {Voit}, G.~M. 1998, \apjl, 497, L5
785:
786: \bibitem[\protect\citeauthoryear{{Laor} \& {Brandt}}{{Laor} \&
787: {Brandt}}{2002}]{lb02}
788: {Laor}, A., \& {Brandt}, W.~N. 2002, \apj, 569, 641
789:
790: \bibitem[\protect\citeauthoryear{{Leighly}}{{Leighly}}{2004}]{leighly04b}
791: {Leighly}, K.~M. 2004, \apj, 611, 125
792:
793: \bibitem[\protect\citeauthoryear{{Leighly} \& {Moore}}{{Leighly} \&
794: {Moore}}{2004}]{leighly04a}
795: {Leighly}, K.~M., \& {Moore}, J.~R. 2004, \apj, 611, 107
796:
797: \bibitem[\protect\citeauthoryear{{Misawa} et~al.}{{Misawa}
798: et~al.}{2007}]{misawa07}
799: {Misawa}, T., {Charlton}, J.~C., {Eracleous}, M., {Ganguly}, R., {Tytler}, D.,
800: {Kirkman}, D., {Suzuki}, N., \& {Lubin}, D. 2007, \apjs, 171, 1
801:
802: \bibitem[\protect\citeauthoryear{{M{\o}ller} \& {Jakobsen}}{{M{\o}ller} \&
803: {Jakobsen}}{1987}]{mj87}
804: {M{\o}ller}, P., \& {Jakobsen}, P. 1987, \apjl, 320, L75
805:
806: \bibitem[\protect\citeauthoryear{{Murray} et~al.}{{Murray}
807: et~al.}{1995}]{mur95}
808: {Murray}, N., {Chiang}, J., {Grossman}, S.~A., \& {Voit}, G.~M. 1995, \apj,
809: 451, 498
810:
811: \bibitem[\protect\citeauthoryear{{Reichard} et~al.}{{Reichard}
812: et~al.}{2003}]{reichard03}
813: {Reichard}, T.~A., et~al. 2003, \aj, 125, 1711
814:
815: \bibitem[\protect\citeauthoryear{{Richards}}{{Richards}}{2001}]{rich01b}
816: {Richards}, G.~T. 2001, \apjs, 133, 53
817:
818: \bibitem[\protect\citeauthoryear{{Richards} et~al.}{{Richards}
819: et~al.}{1999}]{rich99}
820: {Richards}, G.~T., {York}, D.~G., {Yanny}, B., {Kollgaard}, R.~I.,
821: {Laurent--Muehleisen}, S.~A., \& {vanden Berk}, D.~E. 1999, \apj, 513, 576
822:
823: \bibitem[\protect\citeauthoryear{{Rodriguez Hidalgo} et~al.}{{Rodriguez
824: Hidalgo} et~al.}{2007}]{prh07}
825: {Rodriguez Hidalgo}, P., {Hamann}, F., {Nestor}, D., \& {Shields},
826: J. 2007, in Astronomical Society of the Pacific Conferene Series, The
827: Central Engine of Active Galactic Nuclei ed. L.~{Ho}, \& J.~M. {Wang}
828: (astro-ph/0701838)
829:
830: \bibitem[\protect\citeauthoryear{{Sargent}, {Steidel}, \&
831: {Boksenberg}}{{Sargent} et~al.}{1988}]{sbs88}
832: {Sargent}, W.~L.~W., {Steidel}, C.~C., \& {Boksenberg}, A. 1988, \apjs, 68,
833: 539
834:
835: \bibitem[\protect\citeauthoryear{{Schneider} et~al.}{{Schneider}
836: et~al.}{2005}]{sdssqso3}
837: {Schneider}, D.~P., et~al. 2005, \aj, 130, 367
838:
839: \bibitem[\protect\citeauthoryear{{Scott} et~al.}{{Scott}
840: et~al.}{2004}]{scott04}
841: {Scott}, J.~E., et~al. 2004, \apjs, 152, 1
842:
843: \bibitem[\protect\citeauthoryear{{Skrutskie} et~al.}{{Skrutskie}
844: et~al.}{2006}]{2mass}
845: {Skrutskie}, M.~F., et~al. 2006, \aj, 131, 1163
846:
847: \bibitem[\protect\citeauthoryear{{Trump} et~al.}{{Trump}
848: et~al.}{2006}]{trump06}
849: {Trump}, J.~R., et~al. 2006, \apjs, 165, 1
850:
851: \bibitem[\protect\citeauthoryear{{Turnshek}}{{Turnshek}}{1988}]{turn88b}
852: {Turnshek}, D.~A. 1988, in QSO Absorption Lines; Probing the Universe;
853: Proceedings of the QSO Absorption Line Meeting, ed. J.~C. {Blades}, D.~A.
854: {Turnshek}, \& C.~A. {Norman} (Cambridge: Cambridge University Press), 17
855:
856: \bibitem[\protect\citeauthoryear{{Turnshek} et~al.}{{Turnshek}
857: et~al.}{1988}]{turn88}
858: {Turnshek}, D.~A., {Grillmair}, C.~J., {Foltz}, C.~B., \& {Weymann}, R.~J.
859: 1988, \apj, 325, 651
860:
861: \bibitem[\protect\citeauthoryear{{Vestergaard}}{{Vestergaard}}{2003}]{ves03}
862: {Vestergaard}, M. 2003, \apj, 599, 116
863:
864: \bibitem[\protect\citeauthoryear{{Voit}, {Weymann}, \& {Korista}}{{Voit}
865: et~al.}{1993}]{vwk93}
866: {Voit}, G.~M., {Weymann}, R.~J., \& {Korista}, K.~T. 1993, \apj, 413, 95
867:
868: \bibitem[\protect\citeauthoryear{{Weymann} et~al.}{{Weymann}
869: et~al.}{1991}]{weymann91}
870: {Weymann}, R.~J., {Morris}, S.~L., {Foltz}, C.~B., \& {Hewett}, P.~C. 1991,
871: \apj, 373, 23
872:
873: \bibitem[\protect\citeauthoryear{{Weymann}, {Turnshek}, \&
874: {Christiansen}}{{Weymann} et~al.}{1985}]{weymann85}
875: {Weymann}, R.~J., {Turnshek}, D.~A., \& {Christiansen}, W.~A. 1985, in
876: Astrophysics of Active Galaxies and Quasi-Stellar Objects, ed. J.~S.
877: {Miller}, 333
878:
879: \bibitem[\protect\citeauthoryear{{Weymann} et~al.}{{Weymann}
880: et~al.}{1979}]{wwpt}
881: {Weymann}, R.~J., {Williams}, R.~E., {Peterson}, B.~M., \& {Turnshek}, D.~A.
882: 1979, \apj, 234, 33
883:
884: \bibitem[\protect\citeauthoryear{{Wise} et~al.}{{Wise} et~al.}{2004}]{wise04}
885: {Wise}, J.~H., {Eracleous}, M., {Charlton}, J.~C., \& {Ganguly}, R. 2004,
886: \apj, 613, 129
887:
888: \bibitem[\protect\citeauthoryear{{Yuan} et~al.}{{Yuan} et~al.}{2007}]{yuan07}
889: {Yuan}, Q., {Brotherton}, M., {Green}, R.~F., {Kriss}, G.~A., \& {Ganguly}, R.
890: 2007, \apj, {in preparation}
891:
892: \bibitem[\protect\citeauthoryear{{Zhou} et~al.}{{Zhou} et~al.}{2006}]{zhou06}
893: {Zhou}, H., {Wang}, T., {Wang}, H., {Wang}, J., {Yuan}, W., \& {Lu}, Y. 2006,
894: \apj, 639, 716
895:
896: \end{thebibliography}
897:
898: \end{document}
899: