1: \documentclass[12pt,preprint]{aastex}
2:
3: %% manuscript produces a one-column, double-spaced document:
4:
5: %\documentclass[manuscript]{aastex}
6:
7: %% preprint2 produces a double-column, single-spaced document:
8:
9: %\documentclass[preprint2]{aastex}
10: %\usepackage{emulateapj5}
11: %\usepackage{onecolfloat5}
12: \usepackage{epsf}
13:
14: \newcommand{\tx}[1]{\textrm{#1}}
15: \newcommand{\pc}[1]{\protect\citename{#1}}
16: \newcommand{\kms}{km~$\tx{s}^{-1}$}
17: \newcommand{\dv}{$r^{1/4}\,$}
18: \newcommand{\Io}{I$_{814}$}
19: \newcommand{\Vs}{V$_{606}$}
20: \newcommand{\sbe}{$SB_{\tx{e}}$}
21: \newcommand{\resec}{$r_{\tx{e}}$}
22: \newcommand{\Rekpc}{$R_{\tx{e}}$}
23: \newcommand{\mlsun}{M$_{\odot}$/L$_{\odot,V}$}
24: \newcommand{\mbh}{M$_{\rm BH}$}
25:
26: \newenvironment{inlinefigure}{
27: \def\@captype{figure}
28: \noindent\begin{minipage}{0.999\linewidth}\begin{center}}
29: {\end{center}\end{minipage}\smallskip}
30:
31: \newenvironment{inlinetable}{
32: \def\@captype{table}
33: \noindent\begin{minipage}{0.999\linewidth}\begin{center}}
34: {\end{center}\end{minipage}\smallskip}
35:
36: \slugcomment{ApJ Letters, submitted}
37:
38: \shorttitle{Black hole mass $\sigma$ at $z\sim0.37$ }
39: \shortauthors{Treu, Malkan \& Blandford}
40:
41: \begin{document}
42: %\twocolumn[
43: \title{The relation between black hole mass and velocity dispersion at $z\sim0.37$
44: }
45:
46: \footnotetext[1]{Based on data collected at Keck
47: Observatory, operated by Caltech and the University of California.}
48:
49: \author{Tommaso Treu$\!$\altaffilmark{2,3,4}, Matthew A. Malkan $\!$\altaffilmark{2} \& Roger D. Blandford $\!$\altaffilmark{4,5}}
50: \altaffiltext{2}{University of California at Los Angeles, CA 90095; ttreu@astro.ucla.edu, malkan@astro.ucla.edu}
51: \altaffiltext{3}{Hubble Fellow}
52: \altaffiltext{4}{California Institute of Technology. Pasadena, CA 91125}
53: \altaffiltext{5}{Kavli Institute for Particle Astrophysics and Cosmology
54: Stanford, CA; rdb@slac.stanford.edu}
55:
56: \begin{abstract}
57: The velocity dispersion of 7 Seyfert 1 galaxies at $z\sim0.37$ is
58: measured using high signal-to-noise Keck spectra. Black hole (BH)
59: mass estimates are obtained via an empirically calibrated
60: photoionization method. We derive the BH mass velocity dispersion
61: relationship at $z\sim0.37$. We find an offset with respect to the
62: local relationship, in the sense of somewhat lower velocity dispersion
63: at a fixed BH mass at $z\sim0.37$ than today, significant at the 97\%
64: level. The offset corresponds to $\Delta \log \sigma$ = $-0.16$ with
65: rms scatter of 0.13 dex. If confirmed by larger samples and
66: independent checks on systematic uncertainties and selection effects,
67: this result would be consistent with spheroids evolving faster than
68: BHs in the past 4 Gyrs and inconsistent with pure luminosity
69: evolution.
70: \end{abstract}
71:
72: \keywords{galaxies: elliptical and lenticular, cD --- galaxies: evolution --- galaxies: formation}
73:
74: \section{Introduction}
75:
76: The correlation of the mass of the central black hole (M$_{\rm BH}$)
77: with the spheroid velocity dispersion $\sigma$ (Ferrarese \& Merritt
78: 2000; Gebhardt et al.\ 2000; hereafter BHS) links phenomena at widely
79: different scales (from the pcs of the BH sphere of influence to the
80: kpcs of the bulge). This demonstrates that galaxy formation
81: and AGN activity are connected and several physical explanations have
82: been proposed (e.g. Kauffmann \& Haenhelt 2000; Monaco et al.\ 2000;
83: Volonteri et al.\ 2003; Haiman, Ciotti \& Ostriker 2004).
84:
85: Interesting clues can be obtained from the cosmic evolution of
86: empirical relations. Different scenarios -- all reproducing the local
87: BHS relation -- predict different evolution. For example, in a pure
88: luminosity evolution scenario of spheroids, $\sigma$ would not change
89: with time, while M$_{\rm BH}$ would increase as a result of
90: accretion. A typical $10^7 M_{\odot}$ BH accreting at an average rate
91: $0.01 M_{\odot} yr^{-1}$ (e.g., Sun \& Malkan 1989), would double its
92: mass in a Gyr. This would predict the \mbh\, at given $\sigma$ to
93: increase with time, possibly changing the slope of the BHS relation if
94: the accretion rate is a function of M$_{\rm BH}$ (e.g. Small \&
95: Blandford 1992). In contrast, if spheroids grew faster than BH,
96: $\sigma$ at fixed M$_{\rm BH}$ could increase with time.
97:
98: In this {\it Letter} we present the first results from an
99: observational program aimed at measuring the cosmic evolution of the
100: BHS relation. We measure the BHS relation for a sample of 7 Seyfert 1s
101: at $z\sim0.37$. This redshift is high enough to provide a time
102: baseline over which we might expect evolution, yet low enough to be
103: observationally practical.
104:
105:
106: \section{Observations and analysis}
107:
108: \subsection{Experiment Design and Sample Selection}
109:
110: The sphere of influence of supermassive BHs in galaxies at
111: cosmological distances cannot be resolved even with the Hubble Space
112: Telescope (HST). Therefore we target active galaxies, where \mbh\, can
113: be obtained from the integrated properties of the broad emission-line
114: region. In this paper, we will combine an empirically calibrated
115: photo-ionization estimate of the size of the broad line region
116: (hereafter ECPI; Wandel et al. 1999) with its kinematics measured from
117: the rms width of H$\beta$, to deduce the central BH mass. To measure
118: simultaneously the {\it stellar velocity dispersion} from absorption
119: lines, requires nuclei of relatively low luminosity, so that the
120: fraction of stellar light in the integrated spectrum is substantial.
121: Seyfert 1s provide the right balance between the two components:
122: absorption features typical of old stellar populations such as Mg5175
123: and Fe5270 are clearly visible in their high signal-to-noise
124: integrated spectra. In order to minimize the uncertainties from sky
125: subtraction and atmospheric absorption corrections, it is convenient
126: to select specific redshift windows where the relevant emission and
127: absorption lines fall in clean regions of the atmosphere. Accordingly,
128: we selected the ``clean window'' $z\sim0.37$, which corresponds to a
129: look-back time of $\sim 4$ Gyrs, for
130: H$_0=70$~km\,s$^{-1}$\,Mpc$^{-1}$, $\Omega_{\rm m}=0.3$, and
131: $\Omega_{\Lambda}=0.7$.
132:
133: A first object (MS1558+453; hereafter S99; Stocke et al.\ 1991) was
134: selected for a pilot study. When the Sloan Digital Sky Survey (SDSS)
135: became available, a larger sample of objects was selected according to
136: the following criteria: $0.35<z<0.37$, H$\beta$ equivalent width and
137: rms width greater than 5 \AA. These are sufficient to select only
138: broad H$\beta$ galaxies, but are loose enough that they should not
139: introduce significant bias in \mbh. The relevant properties of the
140: observed objects are listed in Table~\ref{tab:sample}.
141:
142: \subsection{Observations and data reduction}
143:
144: High signal-to-noise (Table~\ref{tab:sample}) spectra were obtained
145: using the Low Resolution Imaging Spectrograph (Oke et al. 1995) at the
146: Keck-I telescope on 2003 March 6 (S99) and 2003 Sep 3. Two exposures
147: were obtained for each object, with total exposure times ranging
148: between 1200s and 5280s. The 900/5500 grating with a $1\farcs5$ slit
149: provided a resolution of $\sigma_s= 55\pm5$ kms$^{-1}$ around Mg5175
150: and Fe5270, as measured from sky lines and arc lamps. Internal flat
151: fields were obtained after each object exposure, to correct the
152: fringing pattern of the red CCD. A set of A0V stars from the Hipparcos
153: catalog\footnote{URL
154: http://www.gemini.edu/sciops/instruments/niri/NIRISpecStdSearch.html}
155: to be $\lesssim$ 15 degrees from each target was observed during the
156: night as flux calibrators, and to measure the B-band atmospheric
157: absorption. Spectrophotometric standards were observed during
158: twilights. Internal tests, and comparison with SDSS spectra, show that
159: this procedure corrects the B-band absorption to a level of a few
160: parts in a thousand and relative flux calibration to a few
161: percent. The data reduction was similar to that in Treu et al. (2001).
162:
163: \subsection{Bulge Kinematics}
164:
165: The stellar velocity dispersion of the bulge was obtained by comparing
166: the spectral region corresponding to 5100-5300 \AA\ with spectra of
167: G-K giants as described in Treu et al.\ (2001). Briefly, the high
168: resolution template stars were redshifted and smoothed to match the
169: resolution of the instrumental setup, and convolved with gaussians in
170: $\log \lambda$ space to reproduce the kinematic broadening. Then a low
171: order polynomial representing the featureless continuum was added
172: (using the pixel fitting code by van der Marel 1994). Small AGN
173: emission features at vacuum wavelengths $\lambda$5160.33,
174: $\lambda$5200.53, and $\lambda$5310.34 \AA\, (van den Berk et
175: al. 2001) were masked out during the fit. This procedure yields a
176: velocity dispersion $\sigma_{\rm ap}$, a line strength $\gamma$, with
177: uncertainties (Tab.~1). For each object, we performed a number of
178: tests, varying the spectral range and the order of the polynomial used
179: for continuum fitting. The results were found to be sensitive to these
180: changes for 6/13 objects (the ones with shallower stellar absorption
181: features and typically with stronger FeII AGN emission). The other 7
182: objects (Fig.~2) yielded stable $\sigma$ -- changes much smaller than
183: estimated errors -- and were therefore considered reliable. We note
184: that excluding the Mg region changes the velocity dispersion by less
185: than the estimated errors (c.f. Barth, Ho \& Sargent 2003). For
186: simplicity and consistency with previous work we assume that the
187: central velocity dispersion $\sigma$ (i.e. within a circular aperture
188: of radius 1/8 of the effective radius; Ferrarese \& Merritt 2000) can
189: be obtained as $\sigma$ = ${\mathcal B} \sigma_{ap}$, with ${\mathcal
190: B}$=$1.1\pm0.05$. This assumes that the spheroids have an effective
191: radius of $0\farcs5$, are non rotating, their velocity dispersion
192: profile is similar to that of early-type galaxies, and the disk
193: contamination to the line profile is negligible.
194:
195:
196: \subsection{Black hole mass determination}
197:
198: Black hole masses were determined using the empirical correlation
199: between continuum luminosity and size of the broad line region (Wandel
200: et al. 1999; Kaspi et al. 2000, Vestergaard 2002) and the width of the
201: broad component of H$\beta$, summarized by: $M_{\rm BH}=4.9 \cdot 10^7
202: M_{\odot} L_{\rm 5100}^{0.5} W_{\rm 3000}^2$ (Shields et al.\ 2003),
203: where $L_{\rm 5100}$ is the luminosity of the continuum at 5100 \AA\,
204: in units of $10^{44}$ erg s$^{-1}$ and $W_{\rm 3000}$ is the FWHM of
205: the broad component H$\beta$ in units of 3000 \kms (Note that adopting
206: a different slope for $L_{\rm 5100}$, e.g. 0.66 in eq A5 in
207: Vestergaard 2002, does not change our result because of the luminosity
208: range spanned by our sample). This formula assumes that the regions
209: emitting broad H$\beta$ in all AGN are photoionized under the same
210: conditions, by a UV continuum of the same shape. The bulk velocities
211: of the emitting gas clouds are presumed to be dominated by gravity, an
212: assumption for which there is currently moderate observational support
213: (Peterson \& Wandel 1999).
214:
215: Absolute calibration of $L_{\rm 5100}$ was obtained by normalizing the
216: LRIS spectra to the model SDSS $r'$ magnitude. Since SDSS photometry
217: and LRIS spectra are obtained at different epochs, the flux
218: normalization is uncertain by $\sim 20$ \% due to AGN variability
219: \cite{WM00}. This introduces a random error component of $\sim 10$\%
220: on \mbh, which is negligible with respect to the intrinsic scatter of
221: the ECPI method (a factor of $\sim 2.5$, Vestergaard 2002). The
222: $L_{\rm 5100}$ flux was then corrected for Galactic extinction
223: (Schlegel et al.\ 1998) and subtracted the portion produced by
224: starlight, by comparing the measured line strength of the stellar
225: features with an assumed standard intrinsic value of
226: $\gamma=0.75\pm0.25$ (van der Marel 1994). The starlight falling
227: outside the slit, but included in the photometry, was estimated
228: from the measured seeing ($1\farcs0$) and a \dv\, profile for the host
229: galaxy with effective radius $0\farcs5$. The estimated fractions of
230: AGN-to-total light $f_{\rm AGN}$ are listed in Tab.~1.
231:
232: The high signal-to-noise and resolution of the spectra allowed us to
233: determine the width of the broad component of H$\beta$ using the
234: following procedure (Fig.~2): {\it i)} the continuum was subtracted by
235: fitting a straight line between the continua at 4700 \AA\, and 5100
236: \AA\, rest frame. {\it ii)} The [\ion{O}{3}] line at 5007\AA\, was
237: divided by 3 and blueshifted to remove the 4959\AA\ line. {\it iii)}
238: The [\ion{O}{3}] line at 5007\AA\, was blueshifted and rescaled to
239: remove the narrow component of H$\beta$. The line ratio H$\beta_{\rm
240: narrow}$ / [\ion{O}{3}]$\lambda$5007 was allowed to range between 1/20
241: and the maximum value consistent with the absence of ``dips'' in the
242: broad component (typically 1/10-1/7; e.g. Marziani et al.\ 2003). {\it
243: iv)} The second moment of the residual broad H$\beta$ component was
244: computed for the minimum and maximum narrow H$\beta$. The average and
245: semidifference of the two values were taken as best estimate and
246: uncertainty of the broad H$\beta$ rms. The rms was confirmed to be
247: much more insensitive than the FWHM to continuum and narrow component
248: subtraction (see also Peterson et al. 2004). {\it v)} The rms was
249: transformed into FWHM assuming FWHM = 2.35 rms, consistent with the
250: observed values (Peterson et al. 2004 report $2.03\pm0.59$. Their
251: value would lower \mbh\, by $\sim34$ \%).
252:
253: \section{The Black Hole mass velocity dispersion relation at $ z\sim0.37$ }
254:
255: The BHS relation is shown in Figure~\ref{fig:MBHs} and compared to the
256: local relations (Merritt \& Ferrarese 2001; Tremaine et al.\ 2002).
257: Local AGN with reliable $\sigma$ and \mbh\, from Ferrarese et
258: al. (2001) are also shown for comparison. We stress that the plot
259: should be interpreted with caution, since M$_{\rm BH}$ obtained with
260: the ECPI method is uncertain by a factor of $\sim 2.5$, as opposed to
261: the more precise estimates available for nearby objects. Furthermore,
262: the distribution of errors is non-gaussian and estimates for
263: individual objects may be significantly off, especially when based on
264: single epoch measurements (Vestergaard 2002). Given the relatively
265: large and non-gaussian uncertainties, selection effects could be
266: playing an important, and hard to quantify, role. Selection effects
267: enter in both quantities: $\sigma$ of bulges smaller or close to the
268: instrumental resolution ($\sigma_{s}$) cannot be reliably measured,
269: while very large $\sigma$ tend to dilute the absorption features and
270: therefore are harder to measure at a given S/N. Our observational
271: biases also limit the observable \mbh\, range. When we are
272: signal-to-noise limited, and therefore flux limited, we might exclude
273: objects with the smallest \mbh\, for a given $\sigma$. In contrast, by
274: requiring detectable stellar features in the integrated spectrum we
275: are excluding the brightest AGNs.
276:
277: At face value, this result indicates that 4 Gyrs ago BHs lived in
278: bulges with lower $\sigma$. The $\chi^2$ of the 7 points with respect
279: to the local relationship is 15.7, including errors on both
280: quantities, i.e. the probability that they are drawn from the local
281: relationship is 3\%. Assuming that BH growth was negligible and
282: adopting the local BHS slope, the average difference and scatter in
283: the intercepts corresponds to $-0.16\pm0.13$ dex in $\log \sigma$ at
284: fixed \mbh. Most of the offset and scatter is given by the galaxy
285: with the smallest $\sigma$ and largest fractional uncertainty
286: (S05). Discarding that point, the offset and scatter are
287: $-0.12\pm0.06$ dex in $\log\sigma$. In terms of \mbh, the scatter
288: around the relation is 0.6 dex, or 0.3 dex excluding S05. Since the
289: estimated uncertainty on M$_{\rm BH}$ alone is $\sim0.4$ dex, the
290: intrinsic component of the scatter did not increase dramatically. Our
291: result appears to be at variance with that by Shields et al.\ (2003),
292: who combined ECPI \mbh\, with [\ion{O}{3}] emission line width, to
293: measure the BHS relation out to $z\sim3$. Their result is consistent
294: with no evolution, albeit with large scatter. More data are needed to
295: understand the origin and significance of this discrepancy. Possible
296: explanations include the different velocity dispersions adopted
297: (stellar features vs [\ion{O}{3}]) and the different range in \mbh. At
298: this stage we refrain from quantifying the slope.
299:
300: In order to confirm this tantalizing and perhaps surprising result,
301: data for a larger and more complete sample will be needed, also to
302: address two important systematic effects that could conspire to
303: simulate the observed evolution. First, the local calibration of the
304: ECPI method might not be appropriate for the distant
305: universe. Reverberation mapping studies of distant galaxies could
306: verify this. Second, the assumed relationship between $\sigma$ and
307: $\sigma_{\rm ap}$ should be checked independently. For example the
308: contribution from a face-on cold disk could lower $\sigma_{\rm ap}$
309: requiring a higher value for ${\mathcal B}$, reducing the apparent
310: evolution: ${\mathcal B}$=1.59 would be required to bring the
311: $z\sim0.37$ and $z\sim0$ relationships in agreement. HST imaging is
312: needed to determine the inclination and relative luminosity of the
313: disk within the spectroscopic aperture, while higher signal-to-noise
314: ratio and spatially resolved spectroscopy are needed to study
315: departures from gaussianity of the line profile and rotational
316: support.
317:
318: \section{Conclusions}
319:
320: We have measured the velocity dispersion of the hosts of distant
321: active nuclei. In combination with \mbh\, estimates using the ECPI
322: method, we have obtained a first estimate of the BHS relation at
323: $z\sim0.37$. Since M$_{\rm BH}$ cannot decrease, this measurement,
324: taken at face value, suggests the $\sigma$ of bulges of Sy1s increased
325: by roughly 40\% in the past 4 Gyrs. Having provided a first
326: illustration of the method, it is now necessary to collect data for a
327: large sample of objects spanning a larger range of \mbh\, and
328: $\sigma$, and to test systematic uncertainties by obtaining
329: independent measures of \mbh\, and spatially resolved information on
330: the bulge and disk morphology and kinematics.
331:
332: {\acknowledgments We thank A. Barth, G. Bertin, L.Ciotti, B.Hansen,
333: L.V.E. Koopmans, S. Gallagher, and the referee for useful
334: suggestions. The use of software developed by R.~P.~van der Marel is
335: gratefully acknowledged. TT is supported by the NASA Hubble Fellowship
336: grant HF-01167.01. This project relied upon the SDSS Database. The
337: authors acknowledge the role of Mauna Kea within the Hawaiian
338: community.}
339:
340: \begin{thebibliography}{}
341:
342: \bibitem[Barth, Ho \& Sargent 2003]{BHS} Barth, A.J., Ho, L.~C., Sargent, W.L.W. 2003, ApJ, 583, 134
343: \bibitem[Ferrarese \& Merritt 2000]{FM00} Ferrarese, L. \& Merritt, D. 2000, ApJ, 539, L9
344: \bibitem[Ferrarese et al. 2001]{F01} Ferrarese, L., Pogge, R.W., Peterson, B.M., Merritt, D., Wandel, A. \& Joseph, C.L. 2001, ApJ, 555, L79
345: \bibitem[Gebhardt et al.\ 2000]{G00a} Gebhardt, K. et al. 2000 ApJ, 539, L13
346: \bibitem[Haiman, Ciotti \& Ostriker 2004]{HCO04} Haiman, Z., Ciotti, L., Ostriker, J.P. 2004, ApJ, 606, 763
347: \bibitem[Kaspi et al.\ 2000]{K00} Kaspi, S. et al. 2000 ApJ, 533, 631
348: \bibitem[Kauffmann \& Haenhelt 2000]{KH00} Kauffmann, G. \& Haenhelt, M. 2000, MNRAS, 311, 576
349: \bibitem[Marziani et al. 2003]{M03} Marziani, P., Sulentic, J. W., Zamanov, R., Calvani, M., Dultzin-Hacyan, D., Bachev, R., Zwitter, T. 2003, ApJS, 145, 199
350: \bibitem[Merritt \& Ferrarese 2001]{MF01} Merritt, D. \& Ferrarese L. 2001, ApJ, 547, 140
351: \bibitem[Monaco et al. 2000]{M00} Monaco, P. , Salucci, P. \& Danese, L. 2000, MNRAS, 311, 279
352: \bibitem[Oke et al. 1995]{O95} Oke, B. et al. 1995, PASP, 107, 375
353: \bibitem[Peterson \& Wandel]{PW99} Peterson, B.M. \& Wandel, A. 1999, ApJ, 521, L95
354: \bibitem[Peterson et al. 2004]{P04} Peterson, B.M. et al. 2004, preprint, astro-ph/0407299
355: \bibitem[Schlegel et al. 1998]{S98} Schlegel, D.J., Finkbeiner, D.P., Davis, M. 1998, ApJ, 500, 525
356: \bibitem[Shields et al. 2003]{S03} Shields, G.A., Gebhardt, K., Salviander, S., Wills, B.J., Xie, B., Brotherton, M.S., Yuan, J. \& Dietrich, M. 2003, ApJ, 583, 124
357: \bibitem[Small \& Blandford 1992]{SM92} Small, T. \& Blandford, R.D. 1992, MNRAS, 259, 725
358: \bibitem[Stocke et al. 1991]{S91} Stocke, J. T., Morris, S.L., Goia, I.M., Maccacaro, T., Schild, R., Wolter, A., Fleming, T.A., Henry, J.P. 2001, ApJS, 76, 813
359: \bibitem[Sun \& Malkan 1989]{SM89} Sun \& Malkan 1989, ApJ, 346, 58
360: \bibitem[Tremaine et al. 2002]{T02} Tremaine, S. et al.\ 2002, ApJ, 574, 740
361: \bibitem[Treu et al. 2001]{T01} Treu, T., Stiavelli, M., M{\o}ller, P., Casertano, S., Bertin, G., 2001, MNRAS, 326, 221
362: \bibitem[vanden Berk et al. 2001]{vdB01} vanden Berk, D.E. et al. 2001, AJ, 122, 549
363: \bibitem[van der Marel 1994]{vdm94} van der Marel, R.~P. 1994, MNRAS, 270, 271
364: \bibitem[Vestergaard 2002]{V02} Vestergaard, M. 2002, ApJ, 571, 733
365: \bibitem[Volonteri et al. 2003]{V03} Volonteri, M. Haardt F., \& Madau P. 2003, ApJ, 582, 559
366: \bibitem[Wandel Peterson \& Malkan 1999]{WPM99} Wandel, A., Peterson, B. \& Malkan, M. 1999, ApJ, 526, 579
367: \bibitem[Webb \& Malkan 2000]{WM00} Webb, W. \& Malkan, M.A. 2000, ApJ, 540, 652
368: \end{thebibliography}
369:
370: \clearpage
371: \begin{deluxetable}{lccllllllll}
372: %\rotate
373: %\tabletypesize{\small}
374: \tablecaption{Summary of relevant measurements \label{tab:sample}}
375: \tabletypesize{\tiny}
376: \tablehead{
377: \colhead{ID} & \colhead{$\alpha_{2000}$} & \colhead{$\delta_{2000}$} & \colhead{S/N} & \colhead{$z$} & \colhead{$\sigma_{\rm ap}$} & \colhead{$\gamma$} & \colhead{f$_{\rm 5100}$} & \colhead{rms(H$\beta$)}& \colhead{f$_{\rm AGN}$} & \colhead{$\log$ (M$_{\rm BH}$/M$_{\odot}$)}}
378: \startdata
379: S01 & 15:39:16.23 & $+$03:23:22.06 & 36 & 0.3593 & - & - & 6.00$\pm$0.01 & 50.0$\pm$0.4 & - & - \\
380: S02 & 16:11:11.67 & $+$51:31:31.12 & 51 & 0.3543 & - & - & 5.24$\pm$0.01 & 41.8$\pm$1.1 & - & - \\
381: S03 & 17:32:03.11 & $+$61:17:51.95 & 65 & 0.3583 & - & - & 10.72$\pm$0.02 & 35.7$\pm$0.2 & - & - \\
382: S04 & 21:02:11.51 & $-$06:46:45.03 & 53 & 0.3577 & 140$\pm$17 & 0.34$\pm$0.13 & 7.77$\pm$0.01 & 50.2$\pm$1.1 & 0.44 & 8.18 \\
383: S05 & 21:04:51.84 & $-$07:12:09.45 & 54 & 0.3533 & 81$\pm$32 & 0.18$\pm$0.05 & 9.44$\pm$0.02 & 62.2$\pm$1.6 & 0.68 & 8.52 \\
384: S06 & 21:20:34.18 & $-$06:41:22.24 & 36 & 0.3686 & 123$\pm$31 & 0.24$\pm$0.09 & 8.85$\pm$0.02 & 45.6$\pm$0.6 & 0.58 & 8.21 \\
385: S07 & 23:09:46.14 & $+$00:00:48.91 & 65 & 0.3517 & - & - & 11.69$\pm$0.02 & 53.5$\pm$0.3 & - & - \\
386: S08 & 23:59:53.44 & $-$09:36:55.53 & 62 & 0.3583 & 139$\pm$23 & 0.21$\pm$0.08 & 8.62$\pm$0.01 & 27.2$\pm$0.5 & 0.63 & 7.77\\
387: S09 & 00:59:16.11 & $+$15:38:16.08 & 46 & 0.3539 & 137$\pm$30 & 0.22$\pm$0.08 & 10.80$\pm$0.02 & 41.5$\pm$1.1 & 0.62 & 8.17\\
388: S10 & 01:01:12.07 & $-$09:45:00.76 & 61 & 0.3509 & - & - & 14.46$\pm$0.02 & 42.1$\pm$0.8 & - & - \\
389: S11 & 01:07:15.97 & $-$08:34:29.40 & 55 & 0.3555 & - & - & 9.07$\pm$0.02 & 32.8$\pm$0.1 & - & - \\
390: S12 & 02:13:40.60 & $+$13:47:56.06 & 46 & 0.3579 & 161$\pm$36 & 0.26$\pm$0.07 & 13.70$\pm$0.03 & 72.0$\pm$0.7 & 0.55 & 8.66 \\
391: S99 & 16:00:02.80 & $+$41:30:27.00 & 49 & 0.3676 & 182$\pm$18 & 0.43$\pm$0.10 & 5.83$\pm$0.01 & 72.2$\pm$2.6 & 0.31 & 8.33 \\
392: \enddata
393:
394: \tablecomments{For each object we list coordinates, average
395: signal-to-noise ratio per \AA\, of the region used for the kinematic
396: fit, redshift, velocity dispersion (in km s$^{-1}$), line strength,
397: total flux at 5100 \AA\, (rest frame; in 10$^{-17}$ erg
398: s$^{-1}$cm$^{-2}$\AA$^{-1}$; corrected as described in Section~3;
399: absolute flux calibration uncertainties are not included), rms width
400: of the broad component of H$\beta$ (in \AA, observed frame), estimated
401: fraction of AGN contribution to the flux at 5100\AA, estimated \mbh
402: (the related uncertainty is $\sim 0.4$ dex).}
403: \end{deluxetable}
404:
405: \clearpage
406: \begin{figure}
407: \plotone{f1.eps}
408: \caption{Portion of the spectra used to measure bulge kinematics. The
409: black line is the data, the red lines is the best fit, shaded areas
410: indicate regions masked out during the fit.}
411: \label{fig:veldisp}
412: \end{figure}
413:
414: \begin{figure}
415: \plotone{f2.eps}
416: \caption{Example of H$\beta$ width determination. The black line is
417: the original spectrum after continuum subtraction. The cyan and green
418: lines are the narrow components of [\ion{O}{3}]$\lambda$4959 and
419: H$\beta$ respectively, obtained by rescaling and blueshifting
420: [\ion{O}{3}]$\lambda$5007. The red spectrum is the residual broad line
421: used to compute the rms width. The yellow line underneath
422: [\ion{O}{3}]$\lambda$5007 is the reflection of the corresponding blue
423: part of H$\beta$ around its centroid.}
424: \label{fig:Hbeta}
425: \end{figure}
426:
427: \begin{figure}
428: \plotone{f3.eps}
429: \caption{Black hole mass velocity dispersion relation at
430: $z\sim0.37$ (solid squares with error bars). The local relations by
431: Merritt \& Ferrarese (2001) and Tremaine et al.\ (2002) are also shown
432: as solid and dashed lines. Since the latter adopts a slightly
433: different definition of velocity dispersion, it is overplotted
434: without corrections for comparison purposes only. Local AGN from
435: Ferrarese et al. (2001) are shown as open points.}
436: \label{fig:MBHs}
437: \end{figure}
438: \end{document}
439: