1: \documentclass[12pt,preprint]{aastex}
2:
3: \shortauthors{Wang et al.}
4: \begin{document}
5:
6: \title{On the fraction of X-ray obscured quasars in the local universe}
7: \author{J. X. Wang \& P. Jiang}
8: \affil{Center for Astrophysics, University of Science and Technology of China, Hefei, Anhui 230026, P. R. China \\
9: Joint Institute of Galaxies and Cosmology, USTC and SHAO, CAS
10: }
11:
12: \email{jxw@ustc.edu.cn}
13:
14: \begin{abstract}
15: Recent wide area hard X-ray and soft Gamma ray surveys have shown that
16: the fraction of X-ray obscured AGNs in the local universe significantly
17: decreases with intrinsic luminosity.
18: In this letter we point out that two correction have to be made to the
19: samples:
20: 1) radio loud AGNs have to be excluded since their X-ray
21: emission might be dominated by the jet component;
22: 2) Compton thick sources have to be excluded too since their
23: hard X-ray and soft gamma ray emission are also strongly attenuated by Compton scattering.
24: The soft gamma-ray selected AGN samples obtained by $SWIFT$ and
25: $INTEGRAL$ provide the best opportunity to study the fraction of obscured
26: AGN in the local universe in the least biased way.
27: We choose these samples to check if the corrections
28: could alter the above result on the fraction of obscured AGNs.
29: We find that before the corrections both samples show significant
30: anti-correlation between $L_X$ and $N_H$, indicating obvious decrease in
31: the fraction of obscured AGNs with
32: luminosity. However, after the corrections, we find only marginal evidence
33: of anti-correlation (at 98\% confidence level) in the SWIFT sample, and no
34: evidence at all in the $INTEGRAL$ sample which consists of comparable number
35: of objects. We conclude that current samples only show a marginal decrease
36: in the fraction of obscured AGNs in the local universe,
37: and much larger samples are required to
38: reach a more robust conclusion.
39: \end{abstract}
40: \keywords{galaxies: active --- quasars: general --- X-rays: galaxies}
41:
42: \section{Introduction}
43: A large population of obscured powerful quasars called type-2 quasars has
44: long been predicted by the widely accepted unified model for active galactic
45: nuclei (AGNs; Antonucci 1993).
46: They are believed to be intrinsically the same as type-1
47: quasars but with strong obscuration in both optical and X-ray band due to
48: the presumable torus.
49: Most of such type-2 quasars, which might dominate the black hole growth
50: (e.g., Mart\'\i nez-Sansigre et al. 2005), have been missed by optical
51: surveys for quasars. Doubt has been expressed at times on the existence of
52: type-2 quasars (e.g. Halpern, Turner \& George 1999), mainly due to their
53: rareness and the fact that they are easily mimicked by other types of AGNs.
54: The hard X-ray emission is less affected by the photon-electric absorption,
55: making hard X-ray
56: surveys good approach to search for type-2 quasars.
57: Another advantage of X-day data is that X-ray spectral fitting can yield
58: intrinsic luminosity and absorption simultaneously, which are both
59: required to study the population of obscured quasars.
60: A number of obscured quasars
61: have been successfully revealed by recent deep and wide-area X-ray surveys
62: performed by $Chandra$ and XMM
63: (Norman et al. 2002; Stern et al. 2002;
64: Mainieri et al. 2002; Fiore et al. 2003; Caccianiga et al. 2004).
65:
66: However, many studies have claimed that the fraction of obscured quasars is
67: much smaller than expected from the simplified AGN unified model.
68: Ueda et al. (2003) computed the X-ray luminosity
69: function for 2 -- 10 keV selected AGN samples, and found that the fraction of
70: X-ray absorbed AGNs (with $N_H$ $>$ 10$^{22}$ cm$^{-2}$) drops from $\sim$
71: 0.6 at intrinsic $L_X$ around 10$^{42}$ ergs s$^{-1}$ to around 0.3
72: at $L_X$ above 10$^{44}$ ergs s$^{-1}$.
73: Several other studies based on hard X-ray surveys also
74: support the scheme that the fraction of type II AGNs (or X-ray absorbed
75: AGNs) decreases with intrinsic luminosity (Steffen et al. 2003; Barger et al.
76: 2005; La Franca et al. 2005; Shinozaki et al. 2006).
77: However, contrary results are also reported.
78: Perola et al. (2004) found that
79: the fraction of obscured AGNs in HELLAS2XMM does not change with X-ray
80: luminosity, although the fraction (40\%) appears smaller than expected.
81: Eckart et al. (2006) found that half of the AGNs identified by the
82: SEXSI program are X-ray obscured with $N_H$ $>$ 10$^{22}$ cm$^{-2}$,
83: and the fraction of obscured AGNs is independent of the unobscured luminosity.
84: The discrepancy can be attributed to the redshift evolution in the
85: fraction of obscured quasars (La Franca et al. 2005; Wang et al. 2006):
86: while shallow and wide surveys are probing quasars in the local universe
87: where probably only a small fraction of quasars are obscured,
88: deep surveys are detecting quasars at much higher redshift where
89: more than half of them are obscured (Wang et al. 2006).
90:
91: Soft gamma-ray selected AGN samples (Markwardt et al. 2005, hereafter M06; Bassani et al. 2006, hereafter B06; Beckmann, Gehrels \& Shrader 2006) obtained by $SWIFT$ and $INTEGRAL$
92: provide the best opportunity to study the fraction of obscured AGN in the local
93: universe in the least biased way. Note that for Compton-thick cases (N$_H$ $>$
94: 10$^{24}$ cm$^{-2}$), even soft gamma-ray emission could be significantly
95: attenuated, thus soft gamma-ray selected samples are not complete to Compton-thick sources.
96: Claims have been made based on the soft gamma-ray selected samples, that the
97: fraction of obscured AGNs in the local universe significantly decrease with
98: increasing luminosity.
99: In this letter we point out that two correction have to be made to the samples
100: to study of the fraction of obscured quasars. The
101: corrections are: a) radio loud AGNs have to be excluded since their X-ray
102: emission might be dominated by the jet component thus the measured luminosity
103: and the obscuration does not reflect the intrinsic values in the nuclei;
104: b) Compton thick sources have to be excluded too since their
105: soft gamma ray emission are also strongly attenuated by Compton scattering
106: and their intrinsic luminosities are hard to estimate.
107: After the corrections, we find only marginal decrease in the fraction of
108: obscured AGN with luminosity in the local universe. Larger samples
109: are required to reach a more robust conclusion.
110:
111: \section{The Samples}
112: \subsection{The $INTEGRAL$ Sample}
113: Bassani et al. (2006) provides an AGN sample selected in the 20 -- 100 keV
114: band with the IBIS on $INTEGRAL$. The sample
115: contains 62 active galactic nuclei (14 of them are unclassified) above a
116: flux limit of $\sim 1.5\times10^{-11}$ ergs cm$^{-2}$ s$^{-1}$.
117: B06 listed the available column densities (obtained from archival
118: X-ray spectra) for 35 sources with redshifts.
119: In this letter we provide an updated list of $N_H$
120: (and references) with more recent measurements
121: available in literature (mostly from $Chandra$ or $XMM$ observations, Table 1)
122: \footnote{We note that simply adopting the $N_H$ in B06 does
123: not significantly affect out main results in this paper.}.
124: Within this subsample, 10 are radio loud
125: (including five blazars), five are Compton-thick (N$_{H} \geqslant 10^{24}$
126: cm$^{-2}$), the rest 20 are radio quiet
127: and Compton-thin. The luminosities in
128: the 20 -- 100 keV band were
129: calculated by assuming a Crab-like spectrum.
130:
131: \subsection{The $SWIFT$ Sample}
132: The first 3 months of the $SWIFT$ Burst Alert Telescope (BAT) high Galactic
133: latitude survey provide a sample of 14 -- 195 keV band selected sources
134: (M05) with a flux limit of $\sim$ 10$^{-11}$ ergs
135: cm$^{-2}$ s$^{-1}$ and $\sim 2\arcmin.7$ (90\% confidence)
136: positional uncertainties for the faintest sources.
137: 86\% of the 66 high-latitude sources were identified.
138: Twelve are Galactic-type sources, and 44 are identified with known AGNs.
139: The luminosities in the 14 -- 195 keV band were
140: calculated by assuming a typical power-law spectrum ($\Gamma \sim 1.7$).
141: M05 listed the absorption column density derived from
142: literature or archive X-ray spectra for 39 AGNs.
143: We also provide an updated list of $N_H$ (and references) with more
144: recent measurements available in literature (Table 1). The major changes
145: are that two more Compton-thick sources (NGC 1365 and NGC 7582\footnote{Both
146: NGC 1365 and NGC 7582 show quick Compton-thick/Compton-thin transitions, see
147: Risaliti et al. 2005 and Turner et al. 2000.}) are classified.
148: Within the 39 sources, there are 7 radio loud sources (including three blazars),
149: 4 Compton-thick sources and 28 radio quiet Compton-thin Seyfert galaxies.
150: \subsection{The Combined Sample}
151: In order to increase the number of soft Gamma ray selected AGNs, we combine the
152: $INTEGRAL$ Sample and the $SWIFT$ Sample. Note that there are 10 objects which
153: were detected by both of them (see Table 1). Using these 10 objects, we obtained a linear scaling factor from the $SWIFT$ luminosity to the $INTEGRAL$ one (see Fig. 1).
154: With the scaling factor, we translate all the $SWIFT$ luminosity in 14 -- 195
155: keV to the the $INTEGRAL$ luminosity in 20 -- 100 keV. The combined
156: sample contains 42 radio quiet Compton-thin objects, 14 radio loud sources
157: (including 6 blazars) and eight Compton-thick objects.
158:
159:
160: \section {Results and Discussion}
161: In Fig. \ref{swift} we plot the X-ray absorption column density versus
162: the SWIFT luminosity for the 39 AGNs in the SWIFT sample. Consistent with
163: M05, we find a clear drop of $N_H$ with increasing luminosity.
164: Since soft gamma-ray selected samples are not biased to obscuration
165: (except for Compton thick sources), the sample selection is identical
166: for both obscured and unobscured sources. We can simply study the correlation
167: between $N_H$ and L$_X$ (or the luminosity distribution for obscured/unobscured
168: sources) in the samples without calculating the luminosity
169: function or folding luminosity function model with the
170: sample selection.
171: We performed the Spearman Rank (SR) statistic
172: to give the correlation between $N_H$ and L$_X$.
173: We found a significant
174: anti-correlation between $N_H$ and L$_X$ at 99.99\% level.
175: We note that while a flux-limited sample would introduce spurious
176: correlation between luminosity and redshift (or between luminosities
177: in different bands), such an effect will not affect our study since
178: the measurement of $N_H$ is independent of redshift for the local
179: samples.
180: With the radio loud and Compton thick sources excluded, the confidence
181: level of the anti-correlation significantly drops to 98\%.
182: 18 out of 28 radio quiet Compton-thin objects show
183: $N_H$ $\gtrsim$ 10$^{22}$ cm$^{-2}$, yielding a fraction of 64\%.
184:
185: In Fig. \ref{integral} we plot $N_H$ versus INTEGRAL luminosity for the
186: 35 AGNs in the INTEGRAL sample. As B06 has claimed,
187: the fraction of absorbed AGNs in the whole sample significantly decreases
188: with the 20 -- 100 keV luminosity. We found an obvious
189: anti-correlation between $N_H$ and L$_X$ with a confidence level of 99.7\%.
190: However, the anti-correlation is obviously dominated by the radio loud and
191: Compton-thick objects. Excluding radio loud and Compton-thick objects,
192: we found no more evidence of anti-correlation (the null hypothesis
193: gives a probability of 0.92).
194: 14 out of 20 radio quiet Compton-thin objects show
195: $N_H$ $\gtrsim$ 10$^{22}$ cm$^{-2}$, yielding a fraction of 70\%.
196:
197:
198: We perform the same statistical analysis to the combined sample described
199: in \S2 (see Fig. \ref{combine}). Similarly, we find strong anti-correlation
200: (with confidence level $>$ 99.99\%) between $N_H$ and L$_X$ in the whole sample, but the anti-correlation
201: is mainly due to the radio loud and Compton-thick objects. With them excluded,
202: the confidence level of the anti-correlation drops to 94\%.
203: A KS test also gives similar results that the luminosity distribution of obscured
204: sources (with $N_H$ $\gtrsim$ 10$^{22}$ cm$^{-2}$)
205: differs from that of unobscured ones at confidence level of
206: $>$99.99\% and 95\% respectively before and after excluding
207: the radio loud and Compton thick sources.
208: In the combined sample, 29 out of 42 radio quite Compton-thin objects
209: show $N_H$ $\gtrsim$ 10$^{22}$ cm$^{-2}$, yielding a fraction of 66.7\%.
210: For 8 luminous objects with L$_{20-100keV}$ $>$ 10$^{44}$ erg s$^{-1}$,
211: 3 show $N_H$ above 10$^{22}$ cm$^{-2}$, yielding a fraction of 38\%.
212: The drop of the fraction is marginal with a confidence level of 91\%, consistent
213: with those from the correlation test and KS test.
214: Note that assuming a powerlaw spectrum with photon index of $\sim$ 2.0,
215: L$_{20-100keV}$ is directly comparable to L$_{2-10keV}$.
216:
217:
218: A recent study by Shinozaki et al. (2006) built a complete flux-limited
219: sample of bright AGNs from HEAO-1 all-sky catalogs, with a maximum redshift
220: in the sample of 0.329. Such a sample is ideal to study the population of
221: obscured quasars in the local universe. Shinozaki et al. found no AGN
222: in the high luminosity high-intrinsic absorption regime (log L$_X[erg/s]$
223: $>$ 44.5, log $N_H[cm^{-2}]$ $>$ 21.5) in their sample, where $\sim$ 5 AGNs
224: are expected if assuming a constant fraction of obscured AGNs with luminosity.
225: Considering that there are 8 AGNs with log L$_X[erg/s]$
226: $>$ 44.5 and log $N_H[cm^{-2}]$ $<$ 21.5 in their sample, the
227: difference is at a confidence level of 98\%. We note that 3 out of 8
228: sources are obviously radio-loud (3C 273, III Zw2 and PG 1425+267).
229: Removing the radio-loud sources, the confidence level of the difference
230: drops to 90\%. This indicates that although the sample presents a trend
231: of decreasing fraction of obscured AGNs with luminosity, the decrease is
232: not statistically solid.
233:
234: We conclude that the significant decrease with luminosity in the fraction
235: of obscured AGNs
236: in the local universe might be mainly due to the radio loud and
237: Compton-thick objects in the samples. With these objects excluded, we find only
238: marginal evidence of decrease in the
239: SWIFT and Shinozaki sample, but no evidence of decrease in the INTEGRAL
240: sample which has a similar sample size.
241: Much larger samples are required to reach a more robust result on the fraction
242: of obscured quasars in the local universe.
243: \acknowledgments
244: The work was supported by Chinese NSF through NSFC10473009, NSFC10533050 and the CAS "Bai Ren" project at University of Science and Technology of China.
245: \clearpage
246:
247: \begin{references}
248: \reference{}Akylas, A. et al. 2002, MNRAS, 332, 23
249: \reference{}Antonucci, R. 1993, ARA\&A, 31, 473
250: \reference{}Ballantyne, D. R., Fabian, A. C., Iwasawa, K. 2004, MNRAS, 354, 839
251: \reference{}Barger, A. J., et al. 2005, ApJ, 129, 578
252: \reference{}Bassani, L. et al. 2006, ApJ, 636, L65, B06
253: \reference{}Beckmann, V. et al. 2004, ApJ, 614, 641
254: \reference{}Beckmann, V. et al. 2005, ApJ, 634, 939
255: \reference{}Beckmann, V., Gehrels, N., \& Shrader, C. R. 2006, ApJ, 638, 642
256: \reference{}Bianchi, S., Balestra, I., Matt, G., Guainazzi, M., Perola, G. C. 2003, A\&A, 402, 141
257: \reference{}Bianchi, S., Miniutti, G., Fabian, A. C., Iwasawa, K. 2005, MNRAS, 360, 380
258: \reference{}Caccianiga, A., et a. 2004, A\&A, 416, 901
259: \reference{}Colbert, E. J. M., Strickland, D. K., Veilleux, S., Weaver, K. A. 2005, ApJ, 628, 113
260: \reference{}Costantini, E., Kaastra, J. S., Steenbrugge, K. C., Arav, N., Gabel, J. R., Kriss, G. 2005, AIPC, 774, 321
261: \reference{}Dewangan, G. C., Griffiths, R. E., Schurch, N. J. 2003, ApJ, 592, 52
262: \reference{}Eckart, M. E. et al. 2006, ApJ in press, astro-ph0603556
263: \reference{}Fiore, F. et al. 2003, A\&A, 409, 79
264: \reference{}Foschini, L. et al. 2006, A\&A in press, astro-ph/0603268
265: \reference{}Gibson, R. R., Marshall, H. L., Canizares, C. R., Lee, J. C. 2005, ApJ, 627, 83
266: \reference{}Halpern, J. P., Turner, T. J., George, I. M. 1999, MNRAS, 307, L47
267: \reference{}Jiang, P., Wang, J. X., Wang, T. G. 2006, ApJ in press, astro-ph/0603339
268: \reference{}La Franca, F. et al. 2005, ApJ, 635, 864
269: \reference{}Lewis, K. T., Eracleous, M., Gliozzi, M., Sambruna, R. M., Mushotzky, R. F. 2005, ApJ, 622, 816
270: \reference{}Mainieri, V. et al. 2002, A\&A, 393, 425
271: \reference{}Markwardt, C. B., Tueller, J., Skinner, G. K., Gehrels, N., Barthelmy, S. D., Mushotzky, R. E. 2005, ApJ, 633, L77, M05
272: \reference{}Mart\'\i nez-Sansigre, A. et al. 2005, $Nature$, 436, 666
273: \reference{}Matsumoto, C. et al. 2004, ApJ, 617, 930
274: \reference{}Matt, G., Bianchi, S., D'Ammando, F., Martocchia, A. 2004, A\&A, 421, 473
275: \reference{}Matt, G., Bianchi, S., de Rosa, A., Grandi, P., Perola, G. C. 2006, A\&A, 445, 451
276: \reference{}Norman, C. A. et al. 2002, ApJ, 571, 218
277: \reference{}Perola, G. C. et al. 2004, A\&A, 421, 491
278: \reference{}Risaliti, G., Elvis, M., Fabbiano, G., Baldi, A., Zezas, A. 2005, ApJ, 623, L93
279: \reference{}Shinozaki, K., Miyaji, T., Ishisaki, Y., Ueda, Y., Ogasaka, Y. 2006, ApJ in press, astro-ph/0603699
280: \reference{}Steenbrugge, K. C. et al. 2003, A\&A, 408, 921
281: \reference{}Steenbrugge, K. C. et al. 2005, A\&A, 432, 453
282: \reference{}Steffen, A. T., Barger, A. J., Cowie, L. L., Mushotzky, R. F., Yang, Y. 2003, ApJ, 596, L23
283: \reference{}Stern, D. et al. 2002, ApJ, 568, 71
284: \reference{}Turner, T. J., Perola, G. C., Fiore, F., Matt, G., George, I. M., Piro, L., Bassani, L. 2000, ApJ, 532, 245
285: \reference{}Ueda, Y., Akiyama, M., Ohta, K., Miyaji, T. 2003, ApJ, 598, 886
286: \reference{}Vaughan, S., Fabian, A. C., Ballantyne, D. R., De Rosa, A., Piro, L., Matt, G. 2004, MNRAS, 351, 193
287: \reference{}Wang, J. X., et al. 2006, ApJ submitted
288: \reference{}Yaqoob, T., Reeves, J. N., Markowitz, A., Serlemitsos, P. J., Padmanabhan, U. 2005, ApJ, 627, 156
289: \end{references}
290:
291:
292: \begin{deluxetable}{lccccc}
293: \tabletypesize{\scriptsize}
294: \tablewidth{0pc}
295: \tablecaption{Soft Gamma-ray selected AGNs with an updated list of $N_H$}
296: \tablehead{
297: \colhead{Source} & \colhead{z} & \colhead{Type$^c$} & \colhead{N$_H$} & Ref. &
298: \colhead{$\log{L}$}
299: }
300: \startdata
301: Bassani et al. 06 & INTEGRAL &&&& log L$_{20-100keV}$ \\
302: \hline
303: QSO B0241+62 & 0.044 & RQ & 22.2 & 1 & 44.48 \\
304: ......\\
305: \hline
306: Markwardt et al. 2005 & SWIFT &&&& log L$_{14-195keV}$-0.018\\
307: \hline
308: Mrk 348 & 0.0150 & RQ & 23.1 & 14 & 43.68 \\
309: ......\\
310: \enddata\\
311: Note. -- The complete version of this table is in the electronic edition of
312: the Journal. The printed edition contains only a sample. $^a$ Sources included by both Bassani et al. 2006 and Markwardt et al. 2005. $^b$ QSO 0836+710. $^c$ RQ = radio quiet; RL = radio loud; Bl = blazar; Co = Compton thick. References.--(1) B06; (2) Matt et al. 2006; (3) Dewangan, Griffiths \& Schurch 2003; (4) Beckmann et al. 2005; (5) Beckmann et al. 2004; (6) M05; (7) Matt et al. 2004; (8) Steenbrugge et al. 2003; (9) Steenbrugge et al. 2005; (10) Matsumoto et al. 2004; (11) Lewis et al. 2005; (12) Bianchi et al. 2005; (13) Foschini et al. 2006; (14) Akylas et al. 2002; (15) Risaliti et al. 2005; (16) Ballantyne et al. 2004; (17) Vaughan et al. 2004; (18) Colber et al. 2005; (19) Jiang, Wang \& Wang 2006; (20) Yaqoob et al. 2005; (21) Costantini et al. 2005; (22) Gibson et al. 2005; (23) Turner et al. 2000; (24) Bianchi et al. 2003
313: \end{deluxetable}
314:
315: \begin{figure}
316: \plotone{f1.eps}
317: \caption{
318: The INTEGRAL luminosity in the 20 -- 100 keV band (from Bassani et al. 2006)
319: versus the $SWIFT$ luminosity in the 14 -- 195 keV band (from Markwardt et al.
320: 2005) for 10 AGNs which were detected by both instruments.
321: The solid line shows a linear correlation between two luminosities:
322: Log (L$_{20-100keV}$) = Log (L$_{14--195keV}$) - 0.018.
323: }
324: \label{scaling}
325: \end{figure}
326:
327: \begin{figure}
328: \plotone{f3.eps}
329: \caption{
330: The X-ray absorption column density $N_H$ versus the $SWIFT$ luminosity in the 14 -- 195 keV band for the 39 AGNs in Markwardt et al. (2005).
331: }
332: \label{swift}
333: \end{figure}
334:
335: \begin{figure}
336: \plotone{f2.eps}
337: \caption{
338: The X-ray absorption column density $N_H$ versus the INTEGRAL luminosity in the 20 -- 100 keV band for the 35 AGNs in Bassani et al. (2006).
339: }
340: \label{integral}
341: \end{figure}
342:
343: \begin{figure}
344: \plotone{f4.eps}
345: \caption{
346: The X-ray absorption column density $N_H$ versus the INTEGRAL luminosity
347: for the 65 AGNs in the combined sample.
348: }
349: \label{combine}
350: \end{figure}
351:
352: \end{document}
353: