1: \documentclass[apjl]{emulateapj}                     % for emulateapj

2: 

3: \newenvironment{inlinefigure}{

4: \def\@captype{figure}

5: \noindent\begin{minipage}{0.999\linewidth}\begin{center}}

6: {\end{center}\end{minipage}}

7: 

8: \newcommand{\ergscm}{ergs~s$^{-1}$~cm$^{-2}$}

9: \newcommand{\HST}{{\sl HST}}

10: \newcommand{\CXO}{{\sl CXO}}

11: \newcommand{\zband}{z_{850}}

12: \newcommand{\rp}{r_{\rm P}}

13: 

14: \slugcomment{{\it Astrophysical Journal Letters}, in press}

15: 

16: \shorttitle{AGN Morphology and Environment at $z\sim 0.4$--1.3}

17: \shortauthors{Grogin et al.}

18: 

19: \begin{document}

20: \setlength{\topmargin}{-0.1in}

21: \setlength{\textheight}{9.6in}

22: \title{AGN Host Galaxies at $z\sim$\,0.4--1.3\,: Bulge-Dominated and 

23: Lacking Merger--AGN Connection }

24: 

25: \author{N.~A.~Grogin\altaffilmark{1}, C.~J.~Conselice\altaffilmark{2},

26: E.~Chatzichristou\altaffilmark{3}, D.~M.~Alexander\altaffilmark{4},

27: F.~E.~Bauer\altaffilmark{5}, A.~E.~Hornschemeier\altaffilmark{1},

28: S.~Jogee\altaffilmark{6}, A.~M.~Koekemoer\altaffilmark{7},

29: V.~G.~Laidler\altaffilmark{7}, M.~Livio\altaffilmark{7},

30: R.~A.~Lucas\altaffilmark{7}, M.~Paolillo\altaffilmark{8},

31: S.~Ravindranath\altaffilmark{7}, E.~J.~Schreier\altaffilmark{9},

32: B.~D.~Simmons\altaffilmark{3}, and C.~M.~Urry\altaffilmark{3}}

33: 

34: \altaffiltext{1}{Department of Physics and Astronomy, Johns Hopkins

35: University, Charles and 34th Street, Baltimore, MD 21218.}

36: \altaffiltext{2}{Department of Astronomy, California Institute of

37: Technology, 1201 East California Boulevard, Pasadena, CA 91125.}

38: \altaffiltext{3}{Department of Astronomy, Yale University, P.O.~Box

39: 208101, New Haven, CT 06520-8101.}

40: \altaffiltext{4}{Institute of Astronomy, Madingley Road, Cambridge 

41: CB3 0HA, UK.}

42: \altaffiltext{5}{Department of Astronomy, Columbia University, 550 W. 120th

43: Street, New York, NY 10027.}

44: \altaffiltext{6}{Department of Astronomy, University of Texas,

45: 1 University Station, C1400, Austin, TX 78712-0259.}

46: \altaffiltext{7}{Space Telescope Science Institute, 3700 San Martin

47: Drive, Baltimore, MD 21218.}

48: \altaffiltext{8}{Dipartimento di Scienze Fisiche, Universit\'a degli Studi Federico II, 

49: C.~U.~Monte S.~Angelo, via Cintia, I-80126 Naples, Italy.}

50: \altaffiltext{9}{Associated Universities, Inc., 1400 16th Street, NW,

51: Washington, DC 20036.}

52: 

53: \clearpage

54: \begin{abstract}

55:   We investigate morphological structure parameters and local

56:   environments of distant moderate-luminosity active galactic nucleus

57:   (AGN) host galaxies in the overlap between the \HST/ACS observations

58:   of the Great Observatories Origins Deep Survey (GOODS) and the two

59:   Chandra Deep Fields.  We compute near-neighbor counts and $BViz$

60:   asymmetry ($A$) and concentration ($C$) indices for

61:   $\approx\!35,500$ GOODS/ACS galaxies complete to

62:   $\zband\approx26.6$, including the resolved hosts of $322$

63:   X-ray--selected AGNs.  Distributions of (1) $\zband$ asymmetry for

64:   130 $\zband<23$ AGN hosts and (2) near-neighbor counts for 173

65:   $\zband<24$ AGN hosts are both consistent with non-AGN control

66:   samples.  This implies no close connection between recent galaxy

67:   mergers and moderate-luminosity AGN activity out to appreciable

68:   look-back times ($z\lesssim 1.3$), approaching the epoch of peak AGN

69:   activity in the universe.  The distribution of $\zband$ $C$ for the

70:   AGN hosts is offset by $\Delta C\approx+0.5$ compared to the

71:   non-AGN, a 6.4 $\sigma$ discrepancy much larger than can be

72:   explained by the possible influence of unresolved emission from the

73:   AGN or a circumnuclear starburst.  The local universe association

74:   between AGN and bulge-dominated galaxies thus persists to

75:   substantial look-back time.  We discuss implications in the context

76:   of the low-redshift supermassive central black hole mass correlation

77:   with host galaxy properties, including concentration.

78: \end{abstract}

79: 

80: \keywords{galaxies: active---galaxies: structure---surveys---X-rays: galaxies}

81: 

82: \clearpage

83: \section{Introduction}

84: The connection between active galactic nuclei (AGNs) and their host

85: galaxies, and the evolution in that relationship over cosmic time,

86: have attracted great interest in recent years.  This includes the

87: discoveries that most nearby massive galaxies harbor central

88: supermassive black holes (SMBHs; Magorrian et al.~1998), that AGNs in

89: the local universe ($z\lesssim1$) reside predominantly in massive,

90: bulge-dominated host galaxies (Kauffmann et al.~2003), and that a

91: tight correlation exists locally between SMBH mass and host galaxy

92: properties such as bulge velocity dispersion and light-profile

93: concentration (Ferrarese \& Merritt~2000; Gebhardt et al.~2000; Graham

94: et al.~2001).  The 1--2 Ms Chandra Deep Fields (CDF-N and CDF-S;

95: Brandt et al.~2001; Giacconi et al.~2002), which have now resolved

96: much of the cosmic X-ray background into moderate-luminosity AGNs at

97: $z \sim 1$ (Alexander et al.~2003; Barger et al.~2003; Szokoly et

98: al.~2004), provide a unique AGN sample to probe these locally observed

99: SMBH-host relationships out to epochs nearing the peaks of star

100: formation and AGN activity in the universe.  This is one of the aims

101: of the Great Observatories Origins Deep Survey (GOODS; Giavalisco et

102: al.~2004), which has obtained deep multicolor \textsl{Hubble Space

103: Telescope} (\HST) Advanced Camera for Surveys (ACS) image mosaics

104: across the most sensitive regions of the CDF areas.  In this Letter we

105: report on the local environments and rest-frame optical morphologies

106: of AGN host galaxies in the GOODS fields as well as the implications for

107: SMBH-galaxy coevolution and the merger-AGN connection.

108: 

109: The largest pre-GOODS investigation of \HST-imaged CDF sources was

110: Koekemoer et al.~(2002), with 41 CDF-S 1~Ms sources in three

111: moderately deep Wide Field Planetary Camera 2 (WFPC2) pointings.  Grogin et al.~(2003, hereafter G03)

112: studied the \HST\ morphologies and local environments of these

113: faintest X-ray sources as compared with the X-ray undetected

114: population.  The AGNs were preferentially located in galaxies with

115: highly concentrated light profiles, but the AGN hosts could not be

116: differentiated from the non-AGN based on light-profile asymmetry or

117: frequency of near-neighbors.  The G03 conclusions were as follows: (1) distant

118: moderate-luminosity AGNs did not show a connection between recent

119: ($\lesssim\!1$~Gyr) galaxy merger/interaction and AGN activity; and (2)

120: the $z\sim1$ galaxy population already showed evidence of the SMBH-bulge

121: correlation.  Now that deeper and much larger area GOODS imaging is

122: available across both CDFs, we verify these results with much larger

123: samples of both CDF AGNs and quiescent galaxies.  We also place new

124: constraints on the \textit{evolution} in merger-AGN connection and

125: SMBH-bulge correlation with the extensive redshift information

126: now accumulated in these fields.  We adopt a cosmology with

127: $H_0=70$~km s${}^{-1}$ Mpc${}^{-1}$, $\Omega_m=0.3$, and

128: $\Omega_\Lambda \equiv 1 - \Omega_m = 0.7$.  Magnitudes are given in

129: the AB system.

130: 

131: \section{Observations and Sample Selection}\label{obsampsec}

132: Our analyses employ the \HST/ACS image mosaics in F606W ($V$), F775W ($i$), 

133: and F850LP ($z_{850}$) from the first three epochs of GOODS \HST\ observations

134: in both the northern (``GOODS-N'') and southern (``GOODS-S'') fields

135: (Giavalisco et al.~2004).  We use the $\zband$-detected source

136: catalog detailed in Ravindranath et al.~(2004), trimmed

137: by applying minimum thresholds in compactness ($>\!4$ pixels)

138: and signal-to-noise ratio ($>\!5$) optimized for removing spurious

139: detections.  The remaining 16,632 GOODS-S sources and 18,878

140: GOODS-N sources are complete to $\zband\approx26.6$ and form the basis of our

141: environmental and structural parameter analyses.

142: 

143: We identify candidate AGN hosts from source catalogs of the 1~Ms

144: CDF-S and 2~Ms CDF-N X-ray images reduced and source-extracted in a

145: consistent fashion (Alexander et al.~2003).  These two X-ray surveys

146: provide the deepest views of the universe in the 0.5--8.0~keV band.

147: Within the respective GOODS-N(S) areas, the CDF-N(S) contains

148: 324(223) X-ray sources down to comparable sensitivity limits of

149: $\approx\!1.0(1.3)\times10^{-16}$~\ergscm\ at

150: 0.5--2.0~keV and $\approx\!7.2(8.9)\times10^{-16}$~\ergscm\ at

151: 2--8~keV.  Coordinate-matching to the $\zband$ catalog

152: yields unambiguous counterparts for $>\!80$\% of the CDF sources

153: (F.~E.~Bauer et al.~2005, in preparation).  Many are comparatively nearby and optically

154: bright starbursts and ``quiescent'' galaxies contaminating our desired

155: X-ray--selected AGN sample.  The extensive redshift coverage of

156: $\zband\lesssim24$ CDF counterparts allows us to exclude these non-AGNs

157: with a luminosity threshold of $L(2$-$8~{\rm

158: keV})>10^{42}$~ergs s${}^{-1}$.  The resulting $L_X$-limited AGN sample

159: of 322 galaxies contains few CDF sources at $z<0.4$, so we choose this

160: as our lower limit for redshift-evolution analyses

161: (\S\ref{discsec}).

162: 

163: We investigate evolutionary trends in morphology and environment among

164: the GOODS AGN and non-AGN populations by constructing complete

165: volume-limited subsamples.  We estimate absolute magnitudes for the

166: GOODS-S galaxies by using the photometric redshift database of

167: Mobasher et al.~(2004), who claim an accuracy of $\Delta z/(1+z)

168: \lesssim 0.1$ at $\zband \lesssim 24.5$ for AGN and non-AGN alike.

169: Although we lack a comparable photometric redshift database for the

170: GOODS-N field, redshift measurements of the CDF-N sources

171: (Fernandez-Soto, Lanzetta, \& Yahil~1999; Cohen et al.~2000; Barger et

172: al.~2003) are largely complete to a similar depth ($\zband \lesssim

173: 24.5$).  Hence, we include another 135 CDF-N sources with measured

174: redshifts for our volume-limited AGN sample.

175: 

176: To probe the evolution of the {\em optical} morphology of AGN hosts

177: versus the field, we compute rest-frame $B$-band (hereafter $B_0$)

178: quantities out to the limit $z<1.3$ accessible to our reddest filter

179: ($\zband$).  To estimate $M_B$ and the $B_0$ structural parameters, we

180: linearly interpolate between $V$ and $i$ for galaxies at

181: $0.31<z\leq0.73$, between $i$ and $\zband$ for $0.73<z\leq1.09$, and

182: use $\zband$ quantities for $1.09<z<1.3$.  The redshift survey limit

183: of $\zband\approx24.5$ corresponds to $M_B\approx-19.5$ at $z=1.3$.

184: We adopt this as our limiting absolute magnitude, satisfied by 1090

185: GOODS-S galaxies in the non-AGN sample and another 37(42) AGN hosts

186: from CDF-N(S).  These volume-limited samples probe to

187: $\approx\!1$~mag fainter than $L_*$ and thus are not restricted to the

188: highest luminosity galaxies.  Coincidently, $z\sim1.3$ is the limit

189: for \textsl{Chandra} detection of $L(2$-$8~{\rm keV})>10^{42}$~ergs s${}^{-1}$

190: sources throughout the GOODS regions.  As a result, our AGN sample is

191: essentially complete.

192: 

193: \section{Concentration and Asymmetry Indices}\label{structsec}

194: We quantify the GOODS galaxy morphologies via non-parametric indices

195: of concentration $C$ and asymmetry $A$ (Conselice 2003 and references

196: therein).  Index $C$ scales with the ratio of radii containing 80\% and 20\%

197: of a source's total flux, $C\equiv 5 \log(r_{0.8}/r_{0.2})$, and

198: increases toward bulge-dominated morphologies.  Index $A$ is the

199: flux-normalized residual of the source pixels $S$ differenced with

200: their $180\arcdeg$-rotated counterpart $S_{180}$: $A \equiv

201: \min\left({{\sum_{\rm pix}{|S - S_{180}|\Big/\sum_{\rm

202: pix}{|S|}}}}\right) - A_0$, where $A_0$ is a background term and the

203: minimization is over a 0.2 pixel grid of possible centers of rotation.

204: While $A$ moderately increases toward disk-dominated morphologies, it is

205: driven to large values ($A\gtrsim0.35$) for galaxies with recent

206: or ongoing interaction.

207: 

208: Figure \ref{magconcfig} shows the $\zband$ indices $C$ (\textit{top

209: panel}) and $A$ (\textit{bottom panel}) for resolved sources in both GOODS

210: fields versus $\zband$ magnitude.  

211: \begin{inlinefigure}

212: \begin{center}

213: \resizebox{\textwidth}{!}

214: {\includegraphics{figure1.eps}} 

215: \figcaption{

216: Concentration index (\textit{top}) and asymmetry index

217: (\textit{bottom}) vs.~magnitude as measured in $\zband$ for resolved

218: GOODS sources.  The large symbols represent the AGN sample in the

219: north (\textit{red squares}) and south (\textit{blue triangles}); the

220: small dots represent the non-AGNs in the north (\textit{yellow}) and

221: south (\textit{magenta}).  The connected green crosses and their error

222: bars denote the median values and measurement errors for the non-AGNs

223: in successive 1 mag bins.

224: \label{magconcfig}}

225: \end{center}

226: \end{inlinefigure}

227: 

228: \noindent

229: The asymmetries of the

230: AGN hosts (\textit{large symbols}) and non-AGN (\textit{small dots}) are not clearly

231: separated.  

232: However the AGN host concentrations clearly populate the

233: upper end of the field distribution throughout the regime of good

234: signal-to-noise ratio ($\zband \lesssim 24$).  Such large $C$-values are

235: preferentially associated with massive early-type galaxies, although we

236: note that the AGN hosts span a broad range of morphology (e.g.,

237: Koekemoer et al.~2002).  

238: The declining $C$ with host magnitude is

239: interpreted as an evolutionary effect; galaxies at higher redshift

240: (and thus generally fainter) are intrinsically less concentrated

241: (Conselice et al.~2003).  When comparing the $C$ distributions for

242: $\zband$-limited samples of AGN and non-AGN (\S\ref{concdiscsec}), we

243: compensate for this magnitude-dependent bias in $C$ by resampling the

244: non-AGN population to match the AGN hosts' magnitude distribution.

245: 

246: \section{Near-neighbor Frequency}\label{nearfreqsec}

247: In assessing the role of environment in AGN activity, we complement

248: the $A$ analysis with a comparison of the near-neighbor counts around

249: $\zband<24$ AGN hosts versus the non-AGN.  Qualifying neighbors must

250: satisfy both proximity and relative brightness criteria.  We

251: investigate two different definitions of proximity threshold $d$: one

252: that scales with the Petrosian radius $\rp$ of the primary galaxy,

253: $d<3\rp$, and another that remains fixed for all galaxies,

254: $d<8\arcsec$.  The latter choice, consistent with the analysis of G03,

255: corresponds to $54$~kpc at $z=0.6$ and varies by only $\pm\!25$\% over

256: the range $0.4<z<1.3$.

257: 

258: To limit contamination of the neighbor statistics by chance

259: superpositions of background galaxies, a neighbor is rejected if more

260: than 2~mag fainter than the comparison galaxy.  This relative

261: magnitude cutoff, in the presence of steeply increasing galaxy number

262: counts with magnitude, introduces a bias towards more neighbors around

263: fainter galaxies.  When comparing AGN and non-AGN near-neighbor

264: counts, we therefore resample the non-AGN to match the magnitude

265: distribution of the AGN hosts, analogous to the procedure used in

266: comparing $C$ distributions (\S\ref{structsec}).  Because the resolved

267: CDF optical counterparts typically show only minor flux contribution

268: from the active nucleus (G03), the relative faintness of qualifying

269: AGN neighbors is not significantly biased with respect to the non-AGN.

270: We discuss the similarity of the AGN and non-AGN near-neighbor

271: frequency histograms in \S\ref{asymdiscsec}.

272: 

273: \section{Discussion} \label{discsec}

274: 

275: \subsection{Bulge-dominated AGN Hosts at $z\sim0.4$-$1.3$} \label{concdiscsec}

276: 

277: The top left panel of Figure \ref{ksevofig} notes the

278: Kolmogorov-Smirnov (K-S) test probabilities for the null hypothesis

279: that the $\zband<23$ AGN host and non-AGN $C$-values could

280: be drawn from the same underlying distribution.  The quoted

281: probabilities are the median values from 1000 resamplings of the

282: non-AGN sample matched to the magnitude distribution of $z<23$ AGN

283: hosts to remove magnitude-dependent $C$ bias (see \S\ref{structsec}).

284: The $C$ distributions are highly inconsistent at the

285: 6.4 $\sigma$ level ($P_{\rm K-S}=1.6\times10^{-10}$).  Moreover, both

286: the northern and the southern $\zband<23$ AGN host concentrations are

287: individually discrepant with their respective non-AGN counterparts at

288: $P_{\rm K-S} < 10^{-6}$.

289: 

290: \begin{inlinefigure}

291: \begin{center}

292: \resizebox{\textwidth}{!}{\includegraphics{figure2.eps}}

293: \figcaption{

294: Cumulative distribution functions of concentration index (\textit{left panels})

295: and asymmetry index (\textit{right panels}) for resolved GOODS AGN hosts from

296: the CDF-S (\textit{dotted blue line}), CDF-N (\textit{dashed red line}), and the combination of both

297: CDFs (\textit{thick solid black line}) for a flux-limited sample (\textit{top row}) and a

298: volume-limited sample divided into three redshift bins (\textit{bottom three

299: rows}).  The indices are compared in $\zband$ for the flux-limited

300: sample, and in (interpolated) rest-frame $B$ for the volume-limited

301: sample.  Null-hypothesis probabilities $P$ from K-S

302: comparisons with the control sample of non-AGN GOODS-South galaxies

303: (\textit{thin solid black line}) are also noted in each panel.

304: \label{ksevofig}}

305: \end{center}

306: \end{inlinefigure}

307: 

308: The median $C$ offset of $+0.5$ is consistent with the G03 measurement

309: based on $I<23$ sources in three \HST/WFPC2 pointings in the CDF-S

310: (including 25 X-ray--detected sources) and far exceeds the $\sim\!0.1$

311: enhancement in $C$ expected from nuclear point-source optical flux in

312: the AGN sample (see G03).  Moreover, $C$ enhancement by potential

313: circumnuclear starbursts is discounted because (1) \textit{unresolved}

314: starbursts reduce to the previous case for AGN point-source

315: contamination and (2) resolved ($\gtrsim\!500$ pc) central starbursts

316: with sufficient flux to bias $C$ would be inconsistent with the low

317: incidence of starburst-type spectral energy distributions (SEDs) among

318: the GOODS AGN hosts as compared to the field (Mobasher et al.~2004).

319: 

320: The traditional conception of ``local AGN $\equiv$ Seyfert $\equiv$

321: late-type host'' has been refuted by the Kauffmann et al.~(2003)

322: analysis of thousands of spectroscopically identified low-redshift

323: ($z\lesssim0.3$) AGN host galaxies from the Sloan Digital Sky Survey

324: (SDSS).  Local AGN of all luminosities reside almost exclusively in

325: massive hosts with sizes, stellar mass densities, and concentration

326: indices similar to ordinary early-type SDSS galaxies.  The enhanced

327: $C$ among the GOODS AGN hosts now indicates that nuclear activity

328: remains preferentially associated with bulge-dominated galaxies out to

329: substantial lookback times ($z<1.3$).  Our $z\sim0.4$--1.3 sample

330: largely bridges the span of cosmic time between the quasar epoch,

331: where accretion-driven luminosity is dominated by high-mass SMBHs, and

332: the recent-epoch AGN hosts probed by SDSS.  If the locally-observed

333: tight correlation between SMBH mass and host $C$ (Graham et al.~2001)

334: similarly extends to $z\sim0.4$--1.3, then our results newly suggest

335: that the accretion-driven luminosity of the universe is dominated by

336: the most massive SMBHs at virtually all times.

337: 

338: The discovery of an epoch beyond which SMBH mass and host-galaxy

339: properties (including $C$) lose their tight correlation could strongly

340: constrain theories of SMBH and host-galaxy co-evolution.  To ascertain

341: if the $C$ discrepancy between AGN and non-AGN shows any evolution

342: with redshift, we divide our volume-limited sample into three redshift

343: bins spanning $\approx\!550$~Mpc${}^3$ each: $0.4\leq z<0.85$,

344: $0.85\leq z < 1.1$, and $1.1\leq z <1.3$.  The AGN host $C$ is clearly

345: elevated in all three bins (Fig.~\ref{ksevofig}, \textit{three bottom left

346: panels}), reflected in the persistently low $\log P_{\rm K-S}\sim -6.8$

347: to $-4.7$.  Thus our GOODS AGN host-galaxy sample populates the high

348: end of both the $L_X$ distribution (by construction) and the $C$

349: distribution throughout the range $0.4 < z < 1.3$.  Pushing this

350: analysis beyond $z\approx1.3$ faces multiple obstacles, including (1)

351: identification of obscuration-unbiased moderate-luminosity AGNs ($L_X >

352: 10^{42}$~ergs s${}^{-1}$), requiring \textsl{Chandra} exposure depths of {\it

353: several} megaseconds; (2) construction of a large, complete $M_B\leq -19.5$

354: field sample in the so-called redshift desert; and (3)

355: well-resolved rest-frame optical light profiles out to meaningful

356: isophotes, requiring $\sim0\farcs1$ resolution $J$-band imaging at

357: extreme depths to overcome the sharply increasing surface brightness

358: dimming.

359:  

360: Although tantalizing to conclude that the Graham et al.~(2001)

361: SMBH-bulge correlation now persists to $z<1.3$, it is unclear whether

362: $L_X$ is a reasonable proxy for SMBH mass at these redshifts.  At low

363: redshift, AGNs with well-constrained SMBH mass are observed to have (1) $L_X

364: \lesssim L_{\rm Edd} \propto M_{\rm BH}$, but with a large scatter to

365: lower Eddington ratios (Woo \& Urry 2002) and (2) an AGN fundamental

366: plane in $L_X$, $L_{Radio}$, and $M_{\rm BH}$, but poor correlation

367: between $L_X$ and $M_{\rm BH}$ individually (Merloni et al.~2003).

368: However Barger et al. (2005) have recently claimed a tight

369: $L_X$-$M_{\rm BH}$ correlation at $z\sim1$.  Direct measurement of

370: the SMBH masses of GOODS galaxies would be ideal but extremely

371: challenging for moderate-luminosity (and often dust-obscured) AGNs at

372: these redshifts.

373: 

374: \subsection{No Merger-AGN Connection at $z\sim0.4$-$1.3$?} \label{asymdiscsec}

375: 

376: Unlike the disparate $C$ distributions, the asymmetry indices of the

377: 130 resolved $\zband<23$ GOODS AGN hosts are statistically

378: indistinguishable from the non-AGN ($P_{\rm K-S}=0.97$).  This result

379: reinforces the G03 findings based on a subset of 25 X-ray--detected

380: sources with $I<23$.  Figure~\ref{ksevofig}

381: (\textit{right panels}) shows that this similarity in $A$ exists in

382: $B_0$ throughout all three redshift bins and separately among

383: northern and southern AGN subsamples.  The lowest of the K-S test

384: probabilities, for the combined AGN sample at $0.4\leq z<0.85$, 

385: does not exceed a 2 $\sigma$ rejection of the null hypothesis.

386: 

387: Recent/ongoing galaxy mergers in the local universe

388: generally have large $A$ enhancement ($A>0.35$; Conselice et

389: al.~2003).  Furthermore, $N$-body simulations suggest that even minor mergers

390: have significant $A$ enhancement up to 1~Gyr from the onset of

391: the merger (Walker, Mihos \& Hernquist~1996).  The lack of

392: differentiation between AGN and non-AGN $A$ distributions therefore

393: implies that recent merging/interaction is no more prevalent among the

394: AGNs.  This in turn argues against the hypothesis that AGN fueling is

395: directly linked to recent ($\lesssim\!1$~Gyr) galaxy

396: merging/interaction, while favoring mechanisms such as low-level gas

397: accretion from the intergalactic medium and/or bar instability in

398: disks.

399: 

400: \HST\ imaging of 20 nearby ($z<0.3$) quasars by Bahcall et al.~(1997)

401: suggested that galaxy mergers/interactions are relatively common among the

402: highest luminosity AGNs ($\gtrsim\!10^{44.5}$ ergs s${}^{-1}$).

403: However, our finding of no $A$ enhancement among more distant

404: ($z\sim0.4$--1.3) AGNs at lower luminosities ($\lesssim\!10^{43.5}$ ergs

405: s${}^{-1}$) has precedent in the comparable low-redshift sample

406: analyzed by Corbin (2000).  The GOODS combination of deep \textsl{Chandra} and

407: \HST\ imaging now suggests a persistent merger-AGN disconnect

408: among moderate-luminosity AGNs out to

409: look-back times nearing the peak of AGN activity in the universe.

410: The limited GOODS solid angle provides too few

411: high-luminosity AGNs to test the Bahcall et al.~(1997)

412: conclusions over the same redshift range.  This may be remedied

413: by the wider area Galaxy Evolution from Morphology and SEDs project \HST\ imaging within the

414: extended CDF-S, where Sanchez et al.~(2004) have already noted a

415: higher incidence of merger/interaction among 15 optically selected,

416: $z\sim0.5$-$1.1$, type 1 AGNs with luminosities spanning the

417: Seyfert/quasar boundary.

418: 

419: Our comparison of the near-neighbor frequency histograms for AGNs and

420: non-AGNs yields a large $\chi^2$-test

421: probability of the null hypothesis both for the $d<3\rp$ threshold

422: [$P(\chi^2)=0.84$] and the $d<8\arcsec$ threshold [$P(\chi^2)=0.58$].

423: We note that the $\sim\!2\sigma$ discrepancy in near-neighbor

424: frequency previously observed in G03 is no longer apparent for the

425: substantially enlarged samples of the current work.  We conclude that

426: local environment, like host asymmetry, is not well correlated with

427: AGN activity.  This result, among moderate-redshift AGNs, now extends

428: similar findings of environment-AGN disconnect at low redshift from

429: recent analyses of the Southern Sky Redshift Survey (Maia, Machado, \&

430: Willmer~2003) and the SDSS (Miller et al.~2003).

431: 

432: \acknowledgements We acknowledge support for this work provided by

433: NASA through GO grants GO-09425 and GO-09583 from the Space Telescope

434: Science Institute, which is operated by AURA, Inc., under NASA

435: contract \hbox{NAS 5-26555}.  We thank the anonymous referee for 

436: helpful comments.

437: 

438: \clearpage

439: \begin{references}

440: 

441: \reference{} Alexander, D.~M., et al.~2003, AJ, 126, 539

442: 

443: \reference{} Bahcall, J.~N., et al.~1997, ApJ, 479, 642

444: 

445: \reference{} Barger, A.~J., et~al.~2003, AJ, 126, 632

446: 

447: \reference{} ---------.~2005, AJ, 129, 578

448: 

449: \reference{} Brandt, W.~N., et al.~2001, AJ, 122, 2810

450: 

451: \reference{} Cohen, J.~G., et al.~2000, ApJ, 538, 29

452: 

453: \reference{} Conselice, C.~J.~2003, ApJS, 147, 1

454: 

455: \reference{} Conselice, C.~J., et al.~2003, AJ, 126, 1183  

456: 

457: \reference{} Corbin, M. R.~2000, ApJ, 536, L73

458: 

459: \reference{} Fernandez-Soto, A., Lanzetta, K.~M., \& Yahil, A.~1999, 

460:         ApJ, 513, 34

461: 

462: \reference{} Ferrarese, L. \& Merritt, D.~2000, ApJ, 539, 9

463: 

464: \reference{} Gebhardt, K., et al.~2000, ApJ, 539, 13

465: 

466: \reference{} Giacconi, R., et al.~2002, ApJ, 139, 369

467: 

468: \reference{} Giavalisco, M., et al.~2004, ApJ, 600, L93

469: 

470: \reference{} Graham, A. W., et al.~2001, ApJ, 563, L11

471: 

472: \reference{} Grogin, N.~A., et al.~2003, ApJ, 595, 685 (G03)

473: 

474: \reference{} Kauffmann, et al.~2003, MNRAS, 346, 1055

475: 

476: \reference{} Koekemoer, et al.~2002, ApJ, 567, 657

477: 

478: \reference{} Magorrian, J., et al.~1998, AJ, 115, 2285

479: 

480: \reference{} Maia, M.~A.~G., Machado, R.~S., \& Willmer, C.~N.~A.~2003, AJ, 126, 1750

481: 

482: \reference{} Merloni, A., Heinz, S., \& Di Matteo, T.~2003, MNRAS, 345, 1057 

483: 

484: \reference{} Miller, C.~J., et al.~2003, ApJ, 597, 142

485: 

486: \reference{} Mobasher, B., et al.~2004, ApJ, 600, L167

487: 

488: \reference{} Ravindranath, S., et al.~2004, ApJ, 604, L9

489: 

490: \reference{} Sanchez, S.~F., et al.~2004, ApJ, 614, 586

491: 

492: \reference{} Szokoly, G.~P., et al.~2004, ApJS, 155, 271

493: 

494: \reference{} Walker, I.~R., Mihos, J.~C.,\& Hernquist~L.~1996, ApJ, 460, 121

495: 

496: \reference{} Woo, J-H. \& Urry, M.~C.~2002, ApJ, 579, 530

497: \end{references}

498: \clearpage

499: 

500: \end{document}

501: 

502: 

503: