A Search for the Smallest Supermassive Black Holes
Dissertation
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University
By
Himel Ghosh, M.A.
Graduate Program in Astronomy
The Ohio State University 2009
Dissertation Committee: Professor Smita Mathur, Advisor Professor Paul Martini Professor Andrew P. Gould Copyright by
Himel Ghosh
2009 ABSTRACT
Relations between supermassive black holes (SMBHs) and their host galaxies are now well known, but several questions remain: Do all galaxies harbor SMBHs?
Do correlations between BH mass and host galaxy properties extend to lower mass BHs and all galaxy types? Is the galactic bulge or the dark matter halo the defining component for the nuclear BH? Answering these questions requires a study of low mass SMBHs, in particular those that reside in the latest-type spiral galaxies. While the presence of an active galactic nucleus (AGN) provides certain proof of the existence of an SMBH, galaxies that are not previously known to host
AGNs may nevertheless have SMBHs at their centers. In other words, a galaxy may appear quiescent when in reality there is an accreting SMBH, if the accretion level is low enough. This thesis presents a search for such SMBHs by looking for the presence of low-level nuclear activity, as evidenced primarily by their x-ray emission, in a well-defined sample of nearby, optically quiescent spiral galaxies.
This work demonstrates that traditional methods of identifying AGNs, developed over the course of studying luminous (> 1042 erg s−1) AGNs, are inadequate for ∼ the population of low luminosity AGNs found in nearby galaxies, and develops the
ii techniques that must be used instead. These techniques are then applied to an x-ray survey of nearby, face-on spiral galaxies. The survey includes new snapshot observations of 37 galaxies, which are combined with archival data for a further
18 galaxies. Where available, multi-wavelength data are used to help distinguish
AGNs from other types of x-ray sources. These observations show the power of x-ray observations in detecting hidden AGNs, and also address the question of the prevalence of SMBHs in spiral galaxies that do not have bulges. This study has uncovered 14 previously unknown AGNs and strong AGN candidates, including two in galaxies of type Sd and Sdm. If the latter are confirmed as bona fide AGNs they will be only the fourth and fifth AGNs known to exist in bulgeless galaxies.
iii ACKNOWLEDGMENTS
I wish to thank my advisor, Prof. Smita Mathur, for her superb mentoring. She has been at all times supportive and encouraging, always optimistic but realistic, always solicitous but never condescending. It has been a pleasure to work with her, and I have learned much from her both as a scientist and as a person.
I am grateful to my family for their moral support during this endeavor, and especially to my wife Pampa for her support, very practical help, and unvarnished advice, all of which were invaluable.
Finally I would like to thank the faculty, staff, and other graduate students in the Department of Astronomy for making possible an exciting and enjoyable sojourn that now comes to an end.
iv VITA
1996 ...... A.B. cum laude Physics, Amherst College
1998 ...... M.A. Astronomy, Columbia University
1998 – 2001 ...... Computer Specialist, Smithsonian Astrophysical Observatory
2001 – 2003 ...... Research Astrophysicist, Smithsonian Astrophysical Observatory
2003 – present ...... Graduate Teaching and Research Associate, The Ohio State University
PUBLICATIONS
Research Publications
1. J. Kuraszkiewicz, B. J. Wilkes, G. Schmidt, H. Ghosh, P. Smith, R. Cutri, D. Hines, E. M. Huff, J. C. McDowell, and B. Nelson, “The Spectral Energy Distributions of Red 2MASS AGN”, ApJ, 629, 1143, (2009).
2. H. Ghosh, S. Mathur, F. Fiore, and L. Ferrarese, “Low-level Nuclear Ac- tivity in Nearby Spiral Galaxies”, ApJ, 687, 216, (2008).
3. H. Ghosh, R. W. Pogge, S. Mathur, P. Martini, and J. C. Shields, “Chandra Observations of Candidate ‘True’ Seyfert 2 Nuclei”, ApJ, 656, 105, (2007).
4. M. C. Bentz, K. D. Denney, E. M. Cackett, M. Dietrich, J. K. J. Fogel, H. Ghosh, K. D. Horne, C. Kuehn, T. Minezaki, C. A. Onken, B. M. Peterson, R. W. Pogge, V. I. Pronik, D. O. Richstone, S. G. Sergeev, M. Vestergaard, M. G. Walker, and Y. Yoshii, “NGC 5548 in a Low-Luminosity State: Implications for the
v Broad-Line Region”, ApJ, 662, 205, (2007).
5. K. D. Denney, M. C. Bentz, B. M. Peterson, R. W. Pogge, E. M. Cack- ett, M. Dietrich, J. K. J. Fogel, H. Ghosh, K. D. Horne, C. Kuehn, T. Minezaki, C. A. Onken, V. I. Pronik, D. O. Richstone, S. G. Sergeev, M. Vestergaard, M. G. Walker, and Y. Yoshii, “The Mass of the Black Hole in the Seyfert 1 Galaxy NGC 4593 from Reverberation Mapping”, ApJ, 653, 152, (2006).
6. M. C. Bentz, K. D. Denney, E. M. Cackett, M. Dietrich, J. K. J. Fogel, H. Ghosh, K. Horne, C. Kuehn, T. Minezaki, C. A. Onken, B. M. Peterson, R. W. Pogge, V. I. Pronik, D. O. Richstone, S. G. Sergeev, M. Vestergaard, M. G. Walker, and Y. Yoshii, “Reverberation-based Mass for the Central Black Hole in NGC 4151”, ApJ, 651, 775, (2006).
7. B. J. Wilkes, K. A. Pounds, G. D. Schmidt, P. S. Smith, R. M. Cutri, H. Ghosh, B. Nelson, and D. C. Hines, “XMM-Newton Observations of Red AGNs”, ApJ, 634, 183, (2005).
8. J. D. Silverman et al. (the ChaMP collaboration), “Hard X-Ray-emitting Active Galactic Nuclei Selected by the Chandra Multiwavelength Project”, ApJ, 618, 123, (2005).
9. H. Ghosh, D. L. DePoy, A. Gal-Yam, B. S. Gaudi, A. Gould, A., C. Han, Y. Lipkin, D. Maoz, E. O. Ofek, B.-G. Park, and 53 coauthors, “Potential Direct Single-Star Mass Measurement”, ApJ, 615, 450, (2004).
10. D.-W. Kim, et al. (the ChaMP collaboration), “Chandra Multiwave- length Project. I. First X-Ray Source Catalog”, ApJS, 150, 19, (2004).
11. D.-W. Kim, et al. (the ChaMP collaboration), “Chandra Multiwave- length Project II: First Results of X-ray Source Properties”, ApJ, 600, 59, (2004).
12. P. J. Green, et al. (the ChaMP collaboration), “The Chandra Multi- wavelength Project:Optical Followup of Serendipitous Chandra Sources”, ApJS, 150, 43, (2004).
13. S. Mathur, B. J. Wilkes, and H. Ghosh, “Chandra Detection of Highest Redshift (z 6) Quasars in X-rays”, ApJL, 570, L5, (2002). ∼ 14. B. J. Wilkes, G. D. Schmidt, R. M. Cutri, H. Ghosh, D. C. Hines, B. Nelson, and P. S. Smith, “The X-ray Properties of 2MASS Red Active Galactic Nuclei”, ApJL, 564, L65, (2002).
vi FIELDS OF STUDY
Major Field: Astronomy
vii Table of Contents
Abstract...... ii
Acknowledgments...... iv
Vita ...... v
ListofTables ...... xi
ListofFigures...... xiii
Chapter 1 Introduction ...... 1
1.1 ScopeoftheDissertation...... 6
Chapter 2 Proof of Principle: Six Galaxies from the Chandra Archive 7
2.1 Introduction...... 7
2.2 Sampleselection...... 8
2.3 Dataanalysis ...... 9
2.4 TheGalaxies ...... 11
2.4.1 NGC3169...... 11
2.4.2 NGC3184...... 13
2.4.3 NGC4102...... 18
2.4.4 NGC4647...... 26
2.4.5 NGC4713...... 30
2.4.6 NGC5457(M101) ...... 33
2.5 Discussion...... 43
viii Chapter 3 A strong candidate AGN in NGC 4713 ...... 58
3.1 Introduction...... 58
3.2 Observationsanddataanalysis ...... 59
3.3 Resultsanddiscussion ...... 61
3.4 The UV variability of NGC 4713 ...... 65
Chapter 4 The Chandra Survey: I. The Archival Data ...... 70
4.1 SampleSelection ...... 70
4.2 Dataanalysis ...... 71
4.3 TheGalaxies ...... 73
4.3.1 Group 1: The best AGN candidates ...... 74
4.3.2 Group 2: Candidates with large uncertainties ...... 84
4.3.3 Group 3: Nuclei with uncertain or no detection ...... 88
4.4 Discussion...... 90
Chapter 5 The Chandra Survey: II. The Detected Nuclei ...... 105
5.1 Introduction...... 105
5.2 The seven nuclear x-ray sources ...... 105
Chapter 6 The Chandra Survey: III. The Nondetections ...... 111
6.1 Introduction...... 111
6.2 Stackinganalysis ...... 112
6.3 Results...... 114
6.4 Discussion...... 120
Chapter 7 Conclusions and Future Directions ...... 135
ix 7.1 Futureresearch ...... 144
Bibliography ...... 149
x List of Tables
2.1 Targets and observation parameters ...... 46
2.1 Targets and observation parameters ...... 47
2.2 X-rayMeasurements ...... 48
2.3 NGC5457X-rayMeasurements ...... 49
2.3 NGC5457X-rayMeasurements ...... 50
2.3 NGC5457X-rayMeasurements ...... 51
2.4 X-RaySpectralFits...... 52
2.4 X-RaySpectralFits...... 53
2.4 X-RaySpectralFits...... 54
2.5 Inferred nuclear luminosities ...... 55
2.5 Inferred nuclear luminosities ...... 56
2.6 NGC 3184 luminosity in Spitzer bands...... 57
3.1 Spectral model parameters and goodness of fit...... 69
4.1 Targets and observation parameters ...... 96
4.1 Targets and observation parameters ...... 97
4.1 Targets and observation parameters ...... 98
4.2 X-rayMeasurements ...... 99
4.2 X-rayMeasurements ...... 100
4.3 X-RaySpectralFits...... 101
xi 4.3 X-RaySpectralFits...... 102
4.4 Luminositiesindifferentbands...... 103
4.5 EstimatedEddingtonratios...... 104
5.1 The seven galaxies with nuclear x-ray sources detected in the snapshot survey...... 109
5.2 Counts and hardness ratios of the six nuclear point sources...... 110
6.1 Targets and observation parameters ...... 128
6.1 Targets and observation parameters ...... 129
6.1 Targets and observation parameters ...... 130
6.1 Targets and observation parameters ...... 131
6.2 Stackedobservations ...... 132
6.3 Circumnuclearemission...... 133
6.4 Softandhardbandfluxesofthenuclei...... 134
7.1 The number of nuclear x-ray sources, and confirmed or candidate AGNs in the sample galaxies of types Sa–Sdm...... 148
xii List of Figures
2.1 SpectrumofNGC3169...... 14
2.2 NucleusofNGC3184...... 19
2.3 Spectrum of the southern component of the nucleus of NGC 3184 . . 20
2.4 NucleusofNGC4102...... 27
2.5 NGC4102:spectrumofthecore...... 28
2.6 NGC 4102: spectrum of the extended component ...... 29
2.7 X-ray and optical images of NGC 4647 ...... 31
2.8 NucleusofNGC5457(M101)...... 35
2.9 Variability and hardness ratio of the two NGC 5457 sources...... 36
2.10 CombinedspectrumofNGC5457 ...... 38
2.11 SpectrumofNGC5457inthelowstate...... 39
2.12 SpectrumofNGC5457inthehighstate...... 40
3.1 XMM-NewtonEPIC-PNspectrumofNGC4713 ...... 63
3.2 XMM-Newton OM U-band image of NGC 4713 ...... 66
3.3 Potential variability of NGC 4713 in U band...... 68
4.1 NGC3310hardnessratiomap...... 79
4.2 2MASS image of the nucleus of NGC 3310 ...... 80
4.3 FIRST image of the nucleus of NGC 3310 ...... 81
4.4 Multi-bandimagesofNGC4670...... 85
xiii 6.1 Stacked x-ray image of 6 galaxies ranging from types E to Sab . . . . 116
6.2 Stacked x-ray image of 6 galaxies ranging of types Sb and Sbc . . . . 117
6.3 Stacked x-ray image of 16 galaxies of types Sc through Sdm . . . . . 118
xiv Chapter 1
Introduction
The past decade has seen extraordinary improvement in our understanding of supermassive black holes (SMBHs) — their growth and evolution, and their links with their host galaxies. We now realize that galaxies hosting SMBHs at their centers are the rule rather than the exception. The observed correlations of the masses M• of the SMBHs with properties of their host galaxies, for example with the bulge stellar velocity dispersion (Ferrarese & Merritt 2000; Gebhardt et al. 2000), show that there is a close link between the formation and evolution of galaxies and of the
SMBHs they host. Furthermore, a comparison of the SMBH mass function required to explain the observed luminosity function of active galactic nuclei (AGNs) with estimates of the local SMBH mass function (e.g. Marconi et al. 2004; Shankar et al.
2004) shows that not only must SMBHs be very common in massive galaxies but that most, if not all, of these black holes are relics of AGNs active in previous epochs.
Thus knowledge of the local SMBH mass function enables us to put constraints on theories of galaxy and SMBH formation and growth, and AGN lifetimes.
1 Estimates of the local SMBH mass function are anchored by resolved stellar or gas dynamical measurements of the masses of 30 SMBHs (see, e.g., the review ∼ by Ferrarese & Ford 2005), and otherwise based on the distribution of host galaxy properties (luminosity of the bulge or bulge stellar velocity dispersion σ) and known scaling relationships between the mass of the SMBH and these properties
(most prominently M• σ and M• Mbulge ; but see Greene & Ho 2007a for a − − 8 more direct estimate ). Most of the measured SMBH masses are 10 M⊙ or ∼ greater, however, as the sphere of influence of a less massive SMBH is extremely hard to resolve even at moderate distances, even with the Hubble Space Telescope
6 (HST ). For example, the sphere of influence of a 10 M⊙ SMBH at 15 Mpc is 30 ∼ milliarcseconds (mas). As a result, while different estimates of the mass function
8 9 6 agree for 10 M⊙ < M• < 10 M⊙, the low-mass end (M• < 10 M⊙) often has ∼ ∼ ∼ discrepancies (see, e.g., Fig. 8 of Graham et al. 2007, for a comparison of different authors’ estimates of the mass function). A second source of uncertainty at the low-mass end is the fact that it is unknown how the scaling relationships extrapolate to very late-type spiral galaxies, which have little or no bulge component, and to very low mass galaxies (dE and dSph). Yet SMBHs do exist in very late-type spirals, e.g.
5 NGC 4395, a spiral galaxy of type Sdm, with M• 3 10 M⊙ (Peterson et al. 2005), ∼ × 5 and in very low mass galaxies, e.g. POX 52, a dwarf galaxy, with M• 3 10 M⊙ ∼ × (Barth et al. 2004). Questions that naturally arise at this point are: Do the scaling relationships break down at low masses? What determines the mass of an SMBH:
2 the mass of the bulge or the mass of the dark matter halo? Is there a lower bound to the local SMBH mass function? A well defined sample of low-mass SMBHs is needed to answer these questions.
Since we cannot detect low-mass SMBHs by their dynamical signature, looking for them by signs of their accretion activity may be the only viable way of detecting them. This of course limits detection to the subset of low-mass black holes that are active, but this fraction can be expected to be large, for the following reason.
We now understand that the “quasar era” is a function of luminosity, with the space density of the most luminous quasars peaking at high redshift and that of lower luminosity quasars peaking at progressively lower redshifts (Fiore et al. 2003;
Hasinger et al. 2005). This is often referred to as the “downsizing” of AGN activity with cosmic epoch. Moreover, at least some models of black hole growth (Marconi et al. 2004; Merloni 2004) require anti-hierarchical growth. That is, higher mass
SMBHs attain most of their mass at high redshift while lower mass SMBHs grow at progressively lower redshifts. The trends of AGN downsizing and anti-hierarchical growth, extended logically to the smallest mass SMBHs, imply that these objects were active in recent times and may still be accreting at the present epoch.
5 43 −1 The Eddington luminosity of a 10 M⊙ SMBH is only 10 erg s ; low-mass ∼ SMBHs, even if accreting at high rates, will be low-luminosity AGNs (LLAGNs).
The converse is not true; that is, not all LLAGNs have a low-mass SMBH. In addition to a low SMBH mass, the low luminosity of an LLAGN may be caused by a low rate
3 or radiatively inefficient mode of accretion to an SMBH of any mass (Soria et al.
2006a,b, present a study of massive SMBHs accreting at very low rates). Finally, obscuration may further lower the observed luminosity. Low-ionization nuclear emission-line region (LINER) nuclei have been studied at multiple wavelengths
(e.g. Eracleous et al. 2002; Satyapal et al. 2004; Dudik et al. 2005; Flohic et al.
2006) to identify LLAGNs among them, and these studies have demonstrated that these AGNs are not necessarily the same type of object with the same physical characteristics. Since we are interested specifically in the low-mass end of the SMBH mass function, it is necessary to identify among the AGNs those that can be expected to have the smallest black holes. One approach is to use mass estimators based on the luminosity of the AGN and the width of the broad component of emission lines in its spectrum. This is the technique that was used by Greene & Ho (2007b) in constructing their sample of low-mass SMBHs. A second approach, and the one we use, is to look for AGN activity in galaxies (late-type spirals, dwarf galaxies) where the known host galaxy-SMBH scaling relationships predict the lowest-mass SMBHs would reside. This approach has been used by Satyapal et al. (2007, 2008) to find candidate low-mass SMBHs. However, the scaling relationships are only statistical and cannot be used to estimate the mass of a particular SMBH; thus the second approach requires an independent estimate of the SMBH mass.
A search for active low-mass SMBHs requires confirmation of the presence of an AGN in each candidate nucleus, and measurement of the mass of the SMBHs in
4 the confirmed AGNs. The method used by Greene & Ho (2007b) has the virtue of effectively combining all of the above into a single step. However, as the luminosity of an AGN decreases, the optical spectrum of the galaxy nucleus becomes more and more dominated by host galaxy light, and the signature of the AGN becomes difficult to detect. Even when the optical spectrum shows no clear evidence of an
AGN, however, such evidence may still be present in other wavelengths, such as x-ray and radio (e.g. Filho et al. 2004), and infrared (e.g. Dale et al. 2006; Satyapal et al. 2007, 2008) (See Ho 2008, for a review of nuclear activity in nearby galaxies).
For the lowest-luminosity AGNs, therefore, it is possible that a system based on optical spectra would not classify the nuclei as AGNs at all. We choose to use x-ray selection to identify candidate AGNs for the following reasons: First, x-rays can penetrate obscuring material which may be hiding the line emitting regions.
Second, there are fewer sources of x-rays in a galaxy than there are of optical and
UV emission and so dilution of the AGN signature by host galaxy light is less of a problem. Where the luminosity of the AGN is low to begin with, even a modest amount of absorption may result in the signal being below the optical background imposed by the galaxy. Third, even if, as expected in some theories (Eracleous et al.
1995; Nicastro 2000; Nicastro et al. 2003; Laor 2003) AGNs that have luminosities or accretion rates below a cut-off value do not have broad-line regions, they should still be detectable in x-rays. X-ray observations have in fact detected AGNs in what were thought to be “normal” galaxies (e.g. Martini et al. 2002). The disadvantage, is
5 that x-ray observations by themselves cannot always distinguish between AGNs and other x-ray sources, such as x-ray binaries (XRBs) and ultraluminous x-ray sources
(ULXs). Multi-wavelength data are needed to determine the type of source.
1.1. Scope of the Dissertation
This dissertation aims to demonstrate the efficacy of x-ray observations in uncovering hidden low luminosity AGNs that are missed in optical surveys, and thereby take the first step in assembling a sample of the low-mass SMBHs that power these AGNs. This is done as follows: In Chapter 2 the methods used are demonstrated and tested by applying them to six galaxies for which data already existed in the archives of the Chandra X-ray Observatory at the time this work was begun. Chapter 3 presents a deeper follow-up observation of a particularly interesting candidate AGN from those presented in Chapter 2. Chapters 4, 6, and 5 then define the characteristics of a larger, quasi-volume-limited sample and examine the galaxies in the sample. Chapter 7 draws conclusions from the results of this work and points out possible directions for future study.
6 Chapter 2
Proof of Principle: Six Galaxies from the Chandra Archive
2.1. Introduction
The motivation for this project was to evaluate the feasibility of detecting low-mass SMBHs in late-type spiral galaxies, which may still be accreting at the current epoch and if so should be detectable in x-rays. The six galaxies studied in this chapter (listed in Table 2.1) are not all late-type, but span the range Sa–Sd. The observations studied here examine both aspects of a search for accreting low-mass
SMBH: First, are these objects really detectable, given that the accretion rate is expected to be low? Second, are any sources detected in the very latest type spiral galaxies that have small or no bulges?
7 2.2. Sample selection
Starting with Third Reference Catalogue of Bright Galaxies1 (RC3; de
Vaucouleurs et al. 1991), we imposed the following main filters. (1) Morphological type: we selected spiral galaxies of type Sa through Sd (1.0 T 7.0); (2) Distance: ≤ ≤ galaxies that had recession velocity cz 3000 km s−1; (3) Galactic latitude: ≤ b 30◦ to avoid x-ray absorption by gas in our own Galaxy; (4) Inclination: we | | ≥ required log(a/b) 0.4, where a and b are the projected lengths of the semi-major ≤ and semi-minor axes, respectively. This was to avoid obscuration by the disk of the host galaxy; (5) Nuclear inactivity: the galaxy must not be known to be an
AGN or starburst. Starbursts were excluded since the vigorous star formation also increases the likelihood of the presence of XRBs and ULXs. LINERs were included as there is still debate about whether the fundamental source of energy in these nuclei is an AGN or star formation. We also required that the object have the value of the “goodposition” flag set to 1 in the RC3 table in the Sloan Digital Sky Survey
(SDSS) database, which indicated that the coordinates of the galaxy were accurate to 0.1 s in RA and 1′′ in Dec. All objects passing these filters were cross-correlated with SDSS Data Release 3, with a match radius of 30′′. This resulted in a list of 76 galaxies. This list was then checked against publicly available data in the Chandra archive. The final sample consists of six galaxies: NGC 3169 (Sa), NGC 3184 (Scd),
NGC 4102 (Sb), NGC 4647 (Sc), NGC 4713 (Sd), and NGC 5457 (Scd).
1In practice, we accessed the RC3 table in the SDSS database.
8 2.3. Data analysis
Details of the observations that were analyzed are given in Table 2.1. X-ray data were downloaded from the Chandra archive and analyzed using version 3.4 of
Chandra Interactive Analysis of Observations (CIAO; Fruscione et al. 2006) software.
In each case the Level 2 event list was processed using the observation-specific bad-pixel file and the latest calibration files, and filtered to exclude times when the instrument experienced a background flare. The event lists were then converted to
FITS images and the CIAO wavelet source detection tool wavdetect was run on them to determine source positions. For each observation, source counts were extracted from a circle centered on the wavdetect source position, and with a radius equal to the greater of 4.67 pixels (2.3′′) and the 95% encircled-energy radius at 1.5 keV on
ACIS-S at the position of the source. Background counts were taken from an annulus with an inner radius of twice, and an outer radius of five times, the source circle radius after excising any sources that happened to fall in the background region.
Counts were extracted in the 0.3 – 8.0 keV (Broad), 2.5 – 8.0 keV (Hard) and 0.3 –
2.5 keV (Soft) bands. A hardness ratio (HR) was defined as HR =(H S)/(H + S), − where H and S are the counts in the hard and soft bands, respectively. Spectral
fits were performed with the CIAO tool Sherpa. Uncertainties reported in fit values represent the 90% confidence level for one parameter. In all estimations of the AGN
Eddington ratio we have assumed that the bolometric luminosity is ten times the
2–10 keV luminosity of the AGN.
9 Where available, optical and UV data from HST (WFPC2) were downloaded from the archive and analyzed using IRAF to extract nuclear fluxes. We note here that we use “nuclear flux” to refer to the total background-subtracted flux at the location of the nucleus. We use this flux, and colors if observations in more than one band exist, only as a consistency check, not to obtain photometry of any
38 40 −1 putative AGN. At these extremely low luminosities (Lbol 10 –10 erg s ), the ∼ contribution of the AGN, even if present, will be a small fraction of the nuclear
flux in the optical and UV bands. In most cases, therefore, we expect the fluxes and colors to be consistent with the absence of an AGN. We check for anomalous colors or fluxes, since these may indicate the presence of an additional source of
flux, possibly an AGN. As such, we perform simple aperture photometry for resolved nuclear sources, with the aperture size matched to the source size. For point sources background-subtracted source counts were determined using imexam, using a 2-pixel radius aperture for the source, and an annulus with an inner radius of 7 pixels and an outer radius of 12 pixels for the background. These were corrected for CTE effects
(Dolphin 2002) then extrapolated to a 0.5′′ radius aperture using the encircled-energy tables of Holtzman et al. (1995). A correction of 0.10 mag was added to represent − the magnitude in an infinite aperture. Proprietary ACS data for NGC 4713 were made available by Martini et al. (2008, in preparation). The flux of the resolved nuclear source was estimated using aperture photometry on the drizzled image. We
10 use infrared fluxes from the Two-Micron All Sky Survey (2MASS; Skrutskie et al.
2006) Point Source Catalog (PSC) for similar consistency checks.
Individual targets are discussed below. For each target, we first describe the observed properties in the x-ray, and then in other wavebands when such observations exist. We then consider whether the evidence supports the hypothesis that the x-ray source is an AGN.
2.4. The Galaxies
2.4.1. NGC 3169
This is a galaxy of type Sa, at a distance of 19.7 Mpc (Tully 1988), and its nucleus is classified as a LINER (Ho et al. 1997a). The Chandra observation of the nucleus of NGC 3169 was analyzed by Terashima & Wilson (2003) and the nucleus identified as a low luminosity AGN, but the x-ray data are analyzed and presented here again for completeness. NGC 3169 was observed by Chandra for
2 ks and detected with 159 counts. Its hardness ratio HR = +0.86 makes it the hardest of the six sources discussed here and the only one to have a positive HR.
Its spectrum, shown in Fig. 2.1, can be reasonably fit by an absorbed power-law,
23 −2 with NH 10 cm and Γ 2. Spectral fit parameters are shown in Table 2.4, ≃ ≃ with the first row showing a fit using the Cash statistic and data binned to five
11 counts per bin, and the second row a fit using the Gehrels χ2 statistic and data binned by 20 PHA channels. Best-fit parameter values from both fits are consistent within errors with each other and with the results obtained by Terashima & Wilson
(2003). Fluxes used in the analysis are from the Cash fit. The unabsorbed broad band flux is f(0.3 8 keV) = 1 10−11 erg cm−2 s−1, implying a luminosity − × L(0.3 8 keV) = 5 1041 erg s−1 for the assumed distance. − ×
Observations with the Very Large Baseline Array (VLBA) show the nucleus is a mas-scale radio source at 5 GHz (Nagar et al. 2005). The nuclear flux density
−26 −2 −1 −1 37 −1 is f(5 GHz) = 6.6 10 erg cm s Hz , or νLν = 1.5 10 erg s . The × × measured brightness temperature exceeds 107.7 K and rules out starbursts and supernova remnants (Nagar et al. 2005).
The nucleus was detected by 2MASS. The observed magnitudes J = 11.2,
42 −1 H = 10.5, Ks = 10.0 imply luminosities νLν (4 6) 10 erg s in these bands. ≈ − ×
The nature of the nuclear emission
The combination of being a sub-parsec scale radio source as well as a luminous hard x-ray source points to the source being an AGN. Further evidence comes from the high luminosity. The source is obscured, as made clear from the x-ray spectrum.
The infrared luminosity therefore is a better indication of the true bolometric
12 luminosity. This luminosity, a few 1042 erg s−1, is higher than is expected from × XRBs or nuclear star formation regions.
Dong & De Robertis (2006) perform a bulge-disk decomposition of NGC 3169 and use the 2MASS K-band luminosity of the galaxy to estimate a central black hole mass log (M•/M⊙) = 8.2. H´eraudeau & Simien (1998) report the central stellar
−1 velocity dispersion σ∗ = 163 km s , which implies log (M•/M⊙) 7.8. Assuming ≈ 46 −1 log (M•/M⊙) 8 for simplicity, the corresponding LEdd 10 erg s . The inferred ∼ ∼ 42 −1 −4 Lbol 10 erg s then implies Lbol/LEdd 10 . ∼ ≈
2.4.2. NGC 3184
This is of type Scd, classified as having an H II nucleus by Ho et al. (1997a), and is at a distance of 8.7 Mpc (Tully 1988). It was observed with Chandra twice, one month apart, for 40 ks and 25 ks. Both observations used the ACIS-S3 chip. The nuclear x-ray emission appears to consist of two components, a fainter, point-like source in the north and a brighter, more extended source in the south. The total extent of the nuclear emission is about 4′′ ( 170 pc). Both components are soft and ∼ have HR 0.8. ∼ −
Since the northern component has only 30 counts, detailed spectral fitting ∼ is not possible. We check for consistency of the data with different models using unbinned data. An absorbed power-law can be fitted to the data, and despite the
13 Fig. 2.1.— Spectrum of NGC 3169. Squares show the data binned by 20 PHA channels. The solid line shows the best-fit model, an absorbed power-law with 23 −2 NH 10 cm ,andΓ=2.6. The lower panel shows the residuals from the fit. ∼
14 21 −2 degeneracy in the parameters, an intrinsic NH > 6 10 cm is inconsistent at the ∼ ×
3-σ level. Fixing NH at the Galactic value gives a range of acceptable power-laws with 1.2 Γ 2.8 (3-σ limits). An unabsorbed blackbody with temperature ≤ ≤ 0.28 keV kT 0.56 keV (3-σ limits) is also consistent with the data. The ≤ ≤ observed broadband luminosity of this component, corrected for Galactic absorption,
37 −1 corresponds to L0.3−8 keV 2 10 erg s . The difference in count rates between ∼ × the two observations is not statistically significant, so this source does not vary on month timescales.
There are archival HST observations in the UV (WFPC2, F300W) and optical
(WFPC2, F606W) of this galaxy. The F606W image has the nucleus on the WF4 chip, and clearly shows a central point source. The nucleus is on the PC chip in the
F300W image. There is no central point source detected. Both images also show diffuse nuclear emission and a prominent spiral arm in the central 1′′. Other than the nuclear source in the F606W image, there are no obvious counterparts of the x-ray sources in either the UV or the optical. We note here that the coordinates in the
HST images are misaligned with each other and with the x-ray image. The images were therefore aligned by cross-correlating point sources in each individual image with SDSS sources that are classified as stars. The aligned HST images match each other to within 0.5′′, and match the Chandra image to within 0.7′′. These offsets are comparable to the absolute astrometric accuracy of both Chandra and HST .
15 The nature of the nuclear emission
The fact that we see x-ray emission in the soft band but almost none in the hard band argues against the emission being the continuum from the AGN, whether direct or scattered. The source is not seen in the UV image. Both of these points suggest the presence of heavy obscuration. The observed x-rays could instead be soft emission from circumnuclear ionized gas, if the AGN is completely obscured and what is visible is mostly re-processed radiation. This is in fact the case in some heavily obscured Seyfert 2s (e.g. Bianchi et al. 2006; Levenson et al. 2006; Ghosh et al. 2007), where the dominant emission is the soft emission from the Narrow-Line
Region. The Chandra and HST observations, therefore, are inconclusive regarding whether the source is an AGN, but do not rule out that possibility either.
A stronger argument that the source is an AGN derives from infra-red data.
This galaxy is part of the Spitzer Infrared Nearby Galaxies Survey (SINGS;
Kennicutt et al. 2003). Dale et al. (2006) have used the equivalent width of the PAH feature at 6.2µm and the fluxes in a mix of high- and low-ionization lines ([S IV]
10.51µm, [Ne II] 12.81µm, [Ne III] 15.56µm, [S III] 18.71µm, [O IV] 25.89µm, [S III]
33.48µm, [S II] 34.82µm) to create diagnostic diagrams that distinguish between
AGN and star-forming galaxies. This nucleus falls into the “transition” region between AGNs and H II regions. This suggests an AGN component to the emission exists that may have been diluted because of the large aperture used ( 20′′) to ∼
16 extract the fluxes. In IRAC images the nuclear source is resolved ( 3.5′′ FWHM ∼ compared to the 1.7′′ FWHM of the PSF). Nuclear fluxes were extracted using ∼ 3′′ apertures in the IRAC channels (D. A. Dale, private communication). Host galaxy emission appears to dominate the MIPS (24, 70, and 160 µm) fluxes, but the observed IRAC colors, [3.6] [4.5] = 0.26 0.16 and [5.8] [8.0] = +0.59 0.16 − − ± − ±
(magnitudes in ABν system), are redder than more than 80% of normal late-type galaxies (Assef et al. 2008). This is expected if there is an AGN, as AGN power-law emission falls off more slowly than galactic emission in the NIR. Thus, the IR line ratios and IRAC colors strongly argue for the source to be an AGN.
The nucleus was also detected by 2MASS. The J, H, and Ks magnitudes, from the 2MASS Point Source Catalog (2MASS PSC), are J = 13.3, H = 12.7,
41 −1 and Ks = 12.5, which correspond to νLν 10 erg s . Such a luminosity can be ∼ easily produced by an AGN. While all the observations are consistent with there being an AGN in NGC 3184, we cannot rule out a nuclear super-starcluster as the source. The x-ray emission could be from one or more XRBs in such a cluster. The x-ray luminosity ( 1037 erg s−1) is in the range seen from XRBs. The presence of a ∼ nuclear star formation region is indicated by the optical line ratios observed by Ho et al. (1997a). In addition, Larsen (2004) reports a candidate nuclear star cluster in
NGC 3184.
In conclusion, there are two scenarios that are consistent with the observations.
In both there is a nuclear star cluster that dominates the optical emission. In
17 the first, there is no AGN and the x-rays are produced by one or more XRBs.
In the second, there is a low-luminosity AGN in the center of the star cluster, so heavily obscured that we do not observe any direct or scattered emission. The x-ray emission arises from photoionized gas immediately surrounding the AGN.
Using the luminosity derived using the Bremsstrahlung model above, and assuming that approximately 1% of the AGN luminosity is reprocessed into the plasma x-ray emission, the AGN has a luminosity 1041 erg s−1. The infrared line ratios and ∼ IRAC colors from Spitzer strongly argue for this scenario.
2.4.3. NGC 4102
This is an Sb galaxy, classified as H II in Ho et al. (1997a) and as H II/LINER in the NASA/IPAC Extragalactic Database (NED), at a distance of 17 Mpc (Tully
1988). The Chandra observation of this galaxy was presented by Dudik et al. (2005) and also by Tzanavaris & Georgantopoulos (2007) . There are about 350 counts in the x-ray image of the nucleus. As a hardness ratio map (Fig. 2.4) shows, the hard and soft emission are well segregated into a fairly hard (HR = 0.20 0.08), − ± +0.03 point-like source with extended, very soft (HR = 0.88− ) emission to the west − 0.06 of it. Wavdetect, run on hard and soft band images, detects a small hard source and an encompassing soft source whose centers are 1′′ apart. We therefore fit the spectra of the harder “core” and the softer extended emission separately, using counts from the regions shown in Fig. 2.4. The spectrum of the core has 171
18 Fig. 2.2.— Nucleus of NGC 3184, which appears to consist of two components, labeled N and S in the image. North is up and east to the left in the image. About 117 and 36 counts were detected in the southern and northern components, respectively. The image is 15′′ on a side. The black bar in the lower left represents a projected distance of 4′′, corresponding to 170 pc. ∼
19 Fig. 2.3.— Spectrum of the southern component of the nucleus of NGC 3184, together with an absorbed power-law model. Data points were binned so that there were at 21 least 5 counts in each bin. The model shown above (red line) has NH = 3.0 10 cm−2 and Γ = 2.6. The bottom panel shows the residuals. ×
20 counts, and shows a broad line between 6 and 7 keV, suggestive of reflection. The spectrum was fit after binning to 5 counts per bin, using the Cash statistic (cstat in Sherpa). A simple absorbed power-law model, fit to the energy range 0.3–3 keV, shows increasingly positive residuals at energies above 3 keV, again suggestive of a reflection component. The fit quality is poor, with a statistic value of 200 for 27 d-o-f. Using a reflection model (xspexrav) reduces the statistic to 60 for 26 d-o-f.
The possible line at 6.4 keV is not well constrained because of the lack of counts on the high energy side of the line (there are 8 counts with E > 7 keV). For the line, we added a Gaussian at the fixed position of 6.4 keV and of fixed FWHM 0.3 keV, allowing only the amplitude to vary. This reduces the statistic further, to 40 for 25
+0.6 +283 d-o-f. The final parameters are Γ = 2.2−0.5 and reflection scaling factor R = 129−85 .
The formal equivalent width (EW) of the line is 2.5 keV, but the uncertainty in the amplitude of the line is of the order of 50% and in the normalization of the ∼ reflected component, 30%. The flux, excluding the line and corrected for Galactic ∼ absorption, is F (0.3 8 keV) = 4.2 10−13 erg cm−2 s−1. A fit using the χ2 statistic − × (shown in Table 2.4) gives best-fit parameter values consistent with those from the
Cash fit, with the exception of the equivalent width of the line, which is 3.3 keV in this fit. Given the poor determination of the continuum near the line the difference is not significant. The equivalent width of the Fe Kα in type 1 AGNs is typically of the order of 300 eV, and a large EW implies that the direct continuum is suppressed and the source is reflection dominated. In NGC 4102 the formal EW is an order of
21 magnitude larger; therefore even after accounting for the large uncertainty we can conclude that the true EW is greater than the typical type 1 value.
The extended emission is very soft, with just 5 counts (out of 115) above 2.5 keV. The spectrum was fit after binning to 5 counts per bin and using both the χ2 and Cash statistics. A bremsstrahlung model provides an acceptable fit (χ2 = 9.6 and Cash statistic = 30 for 18 d-o-f), and the MEKAL model is less favored
(χ2 = 15.6 and Cash statistic = 41 for 18 d-o-f). Best-fit parameter values obtained using the two statistics are consistent within the errors. The best-fit temperature kT 1 keV is higher than what may be expected of 100 pc-scale circumnuclear ∼ gas. The plasma surrounding an AGN may be photoionized rather than collisionally ionized, and there may be complexity in the spectrum that is hidden because of the poor quality. The flux in the extended emission is F (0.3 8 keV) 2 10−13 erg − ≈ × cm−2 s−1. Spectral models and best-fit parameter values for both the core and the extended emission are shown in Table 2.4. The χ2 fits for the core and extended components are shown in Figs. 2.5 and 2.6.
HST imaging observations in visual and near-infrared (NIR) bands show that the nucleus is clumpy and has large amounts of dust, and that there is circumnuclear star formation (Carollo et al. 1997). High absorption and reddening towards the nucleus is also indicated by the Balmer decrement which gives E(B V ) = 1.00 − (Ho et al. 1997a). The nucleus was detected by 2MASS. The 2MASS PSC gives
J = 10.9, H = 9.8, and Ks = 9.2. For a distance of 17 Mpc, these correspond to
22 42 −1 42 −1 42 νLν(J) 6 10 erg s , νLν(H) 8 10 erg s , and νLν(Ks) 7 10 erg ≈ × ≈ × ≈ × s−1. The nucleus is highly luminous in the far-infrared (FIR). Infrared Astronomical
44 −1 Satellite (IRAS) observations indicate LFIR 10 erg s (Mouri & Taniguchi 1992), ∼ where LFIR is the luminosity between 40µm and 120µm, albeit in the arcminute-scale
IRAS apertures. The nucleus was also detected in radio by the FIRST survey
(1.4 GHz). The stated radio position is within 0.5′′ of the Chandra position, and therefore within Chandra astrometric accuracy. The radio beam size, however, was
3.75′′ 2.84′′. The integrated radio flux density was 223 mJy, with peak flux density × of 167 mJy. This is the same integrated 1.4 GHz flux density reported by Condon et al. (1982), fifteen years prior to the 1997 FIRST observation. Condon et al. (1982) also report a higher resolution 4.9 GHz observation. The 4.9 GHz map shows radio emission extended in the northeast-southwest direction with a total extent of about
5.2′′ ( 420 pc), and with two peaks separated by 0.9′′ ( 70 pc) and also aligned ∼ ∼ ∼ along a NE-SW axis. The position of the southwest peak coincides with the hard x-ray “core” seen in the Chandra observation. The radio spectral index α = 0.7 − α (Sν ν ) suggests a synchrotron rather than thermal origin for the radio emission. ∝
The nature of the nuclear emission
We first consider the case where there is no AGN and all of the observed emission is due to star formation, and in particular consider the radio emitting region, which, as mentioned above, has a size of the order of 400 pc. If this
23 region obeys the radio-FIR correlation for non-AGN galaxies (Condon 1992), then it has an FIR luminosity of 1044 erg s−1, exactly what was measured by ∼ IRAS. Therefore the radio and FIR observations are consistent with each other and with the star-formation-only hypothesis, requiring only the assumption that this region dominated the FIR emission within the IRAS aperture. Following
Condon (1992), we use the radio emission to estimate a supernova rate νSN and consequently a star formation rate (SFR). The luminosity density measured by
28 −2 −1 −1 −1 FIRST, Lν(1.4GHz) 8 10 erg cm s Hz , implies νSN 0.08 yr , and ≈ × ≈ −1 SFR(M 5M⊙) 2M⊙ yr . In the samples of Grimm et al. (2003) and Ranalli ≥ ≈ et al. (2003), the galaxies that have this star formation rate have x-ray luminosities ranging from a few 1039 erg s−1 to a few 1040 erg s−1. The observed nuclear x-ray × × 40 −1 luminosity of NGC 4102, LX 1.5 10 erg s is really a lower limit because the ∼ × amount of absorption is undetermined, but for the purpose of the argument here may be considered to be consistent with the LX –SFR relationship of Ranalli et al. and
Grimm et al. On the face of it, therefore, all of the observations are consistent with a starburst origin to the emission, but this explanation requires an extraordinarily intense starburst in a non-interacting galaxy. The radio, FIR, and x-ray luminosities
−1 require 2 M⊙ yr of massive (M 5M⊙) star formation within a 400 pc region. ≥ For comparison, the massive star formation rate in the interacting starburst galaxy
−1 M 82 integrated over the whole galaxy is 2.2 M⊙ yr (Neff & Ulvestad 2000), and
24 the total SFR (M 5M⊙) in the merging pair NGC 4038/4039 is estimated to be ≥ −1 between 5 and 10 M⊙ yr (Neff & Ulvestad 2000; Grimm et al. 2003).
In favor of the AGN hypothesis, NGC 4102 shows an approximately conical region of outflowing gas that has a higher [O III]/Hβ ratio than the surrounding star forming regions (Ganda et al. 2006), suggesting exposure to a harder ionizing radiation than stellar continuum emission, and reminiscent of the ionization cones sometimes seen in Seyfert 2s (e.g. Pogge 1988). The [O III] line profile in this region is broader than in the surrounding regions and cannot be fit with a single Gaussian
(Ganda et al. 2006). Gon¸calves et al. (1999) reported a broad component (FWHM
560 km s−1) to the [O III] line as well. The broad component is weak, comprising ≈ just 5–7% of the flux in the line in the case of the Hα and Hβ lines. Using this component gives [N II]λ 6583/Hα = 1.57, similar to LINERs and Seyferts, and thus
Gon¸calves et al. argue for the presence of a very weak Seyfert 2 in NGC 4102. In this respect NGC 4102 is similar to NGC 1042, where Shields et al. (2008) recently demonstrated the existence of a broad component in [N II], and that considering only the broad component moves the nucleus into the Seyfert/LINER regions in line-ratio diagnostic diagrams. Finally, the combination of a point-like hard source, soft extended circumnuclear emission, and an Fe Kα line with a large EW, is one often seen in type 2 AGNs.
There is undeniably strong star formation occurring in the nucleus of NGC
4102, and we cannot rule out the extremely large SFR implied if all of the observed
25 emission is imputed to star formation alone. Nevertheless, the evidence is strong that there is an AGN in NGC 4102. We conclude that NGC 4102 is another example where an AGN and strong star formation co-exist at the nucleus.
2.4.4. NGC 4647
NGC 4647 is a galaxy of type Sc, its nucleus classified as H II in Ho et al.
(1997a), at a distance of 16.8 Mpc (Tully 1988). The Chandra observation is of the elliptical galaxy NGC 4649, and NGC 4647 is on the chip, at an off-axis angle of
2.5′. Two factors complicate the detection of the nucleus of NGC 4647. First, the ∼ nucleus lies within the extended emission from NGC 4649. Second, the nucleus falls on a node boundary of the CCD. Wavdetect at its default filter for source significance
( 1 false source per 106 pixels) does not detect the nucleus. However, once the ∼ elliptical galaxy is modeled and subtracted out of the image, there is a positive residual at the location of the nucleus of NGC 4647 (see Fig. 2.7). To extract source counts we used a circular region centered on the centroid of the residual and with radius 2.3′′, which is approximately the 95% encircled-energy radius at 1.5 keV at that position. To extract background counts we located another region on the same node boundary that was at the same distance from the center of NGC 4649 as was the source circle. The source region has 11 counts after background subtraction, ∼ but this number necessarily has a large uncertainty. We take the radius of the source circle, 2.3′′, as the Chandra positional uncertainty of this source.
26 Fig. 2.4.— Nucleus of NGC 4102 with the regions used to extract counts overlaid on a hardness ratio map. North is up and east to the left in the image. The lightest colored pixels have HR = 1; the darkest pixels have HR = +1. The circle, 2.9 pixels in radius, was used for the− core. The partial annulus is the region used to extract counts from the extended emission. The compact core has harder emission than the extended component. The bar on the lower left represents 2′′, or a projected distance of 160 pc. The image is 10′′ on a side. ∼
27 Fig. 2.5.— Spectrum of the core of the NGC 4102 nucleus. Data have been grouped to 5 counts per bin. The model is a reflected power-law and a Gaussian line, with no absorption in excess of Galactic. The best-fit parameters are Γ 2.3 0.6 and reflection factor R 130. The line was fixed at 6.4 keV with FWHM≈ = 0.3± keV. The lower panel shows the≈ residuals from the fit.
28 Fig. 2.6.— Spectrum of the extended component of NGC 4102. Data have been grouped to 5 counts per bin. The model is absorbed bremsstrahlung emission, with 2.8 21 −2 +8.6 NH = (1.2−1.2) 10 cm and kT = 1.2−0.8 keV. The lower panel shows the residuals from the fit. ×
29 The nucleus may also have been detected in x-rays by XMM-Newton (Randall et al. 2006). While the source was detected with S/N = 11, the positional uncertainty was 3′′. The XMM-Newton source is soft, similar to the Chandra residual. The XMM-Newton and Chandra source positions differ by 5.2′′. There is
′′ a nuclear 2MASS point source (Ks 12.3) at a distance of 1.1 from the Chandra ≈ position, but it is embedded in diffuse emission and the flux from the point source is poorly constrained. In the radio, there is a 5 σ upper limit to the nuclear emission of 0.5 mJy at 5 GHz (Ulvestad & Ho 2002). There is an older report of a radio detection of the nucleus with a flux density of 16 mJy at 1.4 GHz (Willis et al. 1976;
Kotanyi 1980), but with positional uncertainty (∆α, ∆δ)=(0.16s, 11.2′′).
The nature of the nuclear emission
The data are inconclusive at this time as to whether there is an AGN in NGC
4647. The faintness of the source and the positional uncertainties in the existing observations prevent a firm identification and characterization of the source.
2.4.5. NGC 4713
This galaxy is of type Sd, at a distance of 17.9 Mpc (Tully 1988). Its nucleus was classified as T2 (transition object with type 2 spectrum) by Ho et al. (1997a).
The nucleus is clearly detected by Chandra, but with only ten counts. It is a very
30 Fig. 2.7.— On the left are shown the residuals in the Chandra image of NGC 4647 after the elliptical galaxy NGC 4649 is modeled and subtracted. The residual image has been smoothed by convolving it with a Gaussian with σ = 5 pixels. The black circle, whose radius is 2.3′′ or roughly the 95% encircled-energy radius at 1.5 keV at that position, shows the probable nucleus of NGC 4647. On the right, a circle (white) of the same radius, whose center has the same celestial coordinates as the one on the left, is superposed on a DSS image of NGC 4647. Both images are shown on the same scale and have north up and east to the left. The bar in the lower left corresponds to 10′′, or a projected distance of 800 pc. ∼
31 soft source, with nine of the ten counts below 2.5 keV. This object was also analyzed by Dudik et al. (2005), but since they were looking for hard-band sources, this object was not counted as a detection. The observed count rate corresponds to a flux of
(5–10) 10−14 erg cm−2 s−1 in the 0.3–8 keV band, or luminosity (3–5) 1038 erg × × s−1 for our assumed distance. The 2–10 keV flux depends strongly on the assumed
−17 −2 −1 model, from fX 1 10 erg cm s for a thermal bremsstrahlung model with ∼ × −15 −2 −1 kT = 0.3 keV, to fX 5 10 erg cm s for a power-law with Γ = 2 and ∼ × Galactic absorption.
NGC 4713 was observed by HST in December 2006 (Martini et al. 2008, in preparation). The nucleus is resolved; thus only a nuclear star cluster or star forming region, and no AGN, is detected. The cluster is 0.40′′ in diameter in the image, ∼ corresponding to a physical size of 35 pc. Within this aperture the flux density is ∼ −17 −2 −1 −1 40 −1 fλ = 5.34 10 erg cm s A˚ , which implies νLν = 1.2 10 erg s . The × × 41 −1 nucleus is also in the 2MASS PSC, with J, H, Ks luminosities νLν 10 erg s . ∼ However, the nucleus is embedded in diffuse extended emission and therefore the reported magnitudes may not be accurate estimates of the nuclear emission. There is an upper limit of 1.1 mJy to the nuclear radio emission at 15 GHz (Nagar et al.
2005), and a similar limit, 1.0 mJy, to the emission at 1.4 GHz from the FIRST survey.
32 The nature of the nuclear emission
As in the case of NGC 4647, the data are inconclusive regarding the presence of an AGN. The data are consistent with the AGN hypothesis: the “transition object” classification by Ho et al. (1997a) implies there may be an AGN component in the optical spectrum; the nucleus is without doubt an x-ray source, though it is not possible to distinguish between emission from circumnuclear gas and the nucleus proper in the current Chandra imaging. The hardness ratio (with large uncertainty) is consistent with a Γ = 2 power-law. NGC 4713 may be similar to NGC 3184 in having a low-luminosity AGN inside a nuclear star cluster.
2.4.6. NGC 5457 (M 101)
NGC 5457 is a galaxy of type Scd at a distance of approximately 7 Mpc
(Freedman et al. 2001; Stetson et al. 1998). The nucleus was classified as H II by Ho et al. (1997a). It has been observed multiple times by Chandra for a total observation time of about 1.1 Ms. Our analysis omits the shorter, and hence low signal-to-noise, exposures. The observations included here are listed in Table 2.3.
The total usable exposure time is 695 ks. Chandra clearly resolves two sources in the nuclear region (Fig. 2.8), which we label N and S. The northern source, N, is the nucleus, while the southern, S, is a star cluster (Pence et al. 2001). Table 2.3 shows the counts and hardness ratios of the two sources in each of the observations. Source
33 N (the nucleus) varies in brightness by about a factor of 9 over the course of about 8 months (see Table 2.3 and Fig. 2.9). There is no significant change in hardness ratio.
In our analysis and discussion below we consider only the nucleus (source N).
Three nuclear x-ray spectra were extracted: one (“merged”) from the event list obtained by merging all observations done in 2004, and fit using the instrument response from ObsID 5339; one (“high state”) from merging two observations where the source had a high count rate (ObsIDs 4736 and 6152) and fit using the response from ObsID 4736; and one (“low state”) from merging five observations where the count rate was low (ObsIDs 5300, 5309, 4732, 5322, 5323), fit using the response from ObsID 4732. The spectra are shown in Figs. 2.10, 2.11 and 2.12; fit models and parameters are shown in Table 2.4. The low state spectrum can be fit with an absorbed power-law (χ2/dof = 18.7/30) but the fit is improved slightly with the addition of a plasma component (xsmekal) at temperature kT = 0.3 keV
(∆χ2 = 4.5 for two fewer d-o-f). The best-fit power-law slope Γ = 1.7 0.5 is − ± typical of unabsorbed AGN, and best-fit intrinsic absorption is consistent with zero.
The unabsorbed 0.3–8 keV luminosity is 3 1037 erg s−1. The high state spectrum ∼ × +0.4 can also be fit by an absorbed power-law. The best-fit slope is Γ = 2.2−0.3, steeper but consistent with the low state slope within the uncertainties. The differences from
21 the low state spectrum are, first, that the fit requires intrinsic absorption (NH 10 ∼ cm−2), and second, that there is no evidence for the MEKAL component. If a plasma component exists in the high state its flux falls below that of the power-law
34 Fig. 2.8.— Nucleus of NGC 5457 (M 101). North is up and East to the left in the image. The nucleus is resolved into two sources, marked N and S in the image. The northern source N is the candidate active nucleus. The southern source S is a known star cluster (Pence et al. 2001). The black bar in the lower left represents a projected distance of 2′′, or 70 pc. The image is 8′′on a side. ∼
35 5.0 N S )
-1 4.0
3.0
2.0 Count rate (ks 1.0
0.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Date (2004) -0.3 N -0.4 S -0.5 -0.6
HR -0.7 -0.8 -0.9 -1.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Date (2004) Fig. 2.9.— Variability and hardness ratio of the two NGC 5457 sources. Only observations performed in 2004 are shown here. The nucleus (source N) varies by a factor of 9 between March and November. Uncertainty in count rate was derived assuming √∼n uncertainty in the counts.
36 component. The unabsorbed 0.3–8 keV luminosity in the high state is 3 1038 ∼ × erg s−1. A fit where the power-law slope is forced to be identical in the low and high states is of similar statistical significance and produces best-fit values similar to the separate fits. This fit is shown in Table 2.4 in the rows labeled “sim. low” and “sim. high”. The merged spectrum is similar to the high state one in that an absorbed power-law provides a good fit without the need for additional components.
We analyzed archival HST WFPC2 images of NGC 5457, a 2400 s exposure using the F336W filter (referred to as U band below; nucleus on WF3) and a 1600 s exposure using the F547M filter (referred to as V band below; nucleus on PC1). The nucleus and the star cluster are detected in both images. In 2-pixel radius apertures, the U band nuclear flux is 6.3 10−17 erg cm−2 s−1 A˚−1 and the V band flux is × −17 −2 −1 −1 8.3 10 erg cm s A˚ , corresponding to mF336W = 19.3 and mF547M = 19.0 × in the STMAG system. The 2MASS PSC contains a source at the position of the nucleus (x-ray source N) but none at the location of source S. The given magnitudes
41 −1 J = 13.1, H = 12.5, and Ks = 11.8 all correspond to νLν 10 erg s . The ≈ nucleus was not detected by FIRST. Heckman (1980) gives an upper limit to the
35 −1 nuclear luminosity at 6 cm corresponding to νLν < 6 10 erg s . ×
−1 6 McElroy (1995) reports σ∗ 78 km s which implies log (M•/M⊙) 2 10 . ≈ ≈ × 44 −1 The corresponding Eddington luminosity is LEdd 3 10 erg s . Assuming the ≈ × source is an AGN, and assuming bolometric correction factors of 10 and 1 in ∼ ∼
37 Fig. 2.10.— Spectrum of NGC 5457 obtained from merging all 2004 observations. Data have been binned to 15 counts per bin. The line shows the best fit model, an absorbed power-law (Γ = 1.9) and thermal plasma (kT = 0.44 keV). The line-like feature at 4 keV may be due to statistical fluctuation. The lower panel shows the residuals.
38 Fig. 2.11.— Spectrum of NGC 5457 in the low state. Data have been binned to five counts per bin. The best-fit model is a power-law plus MEKAL plasma with no absorption in excess of Galactic. The lower panel shows the residuals from the fit.
39 Fig. 2.12.— Spectrum of NGC 5457 in the high state. Data have been binned to five counts per bin. The best-fit model is a power-law with both intrinsic and Galactic absorption but no plasma. The lower panel shows the residuals from the fit.
40 the x-ray and IR, respectively, implies that the bolometric luminosity is in the range
39 41 −1 −5 −3 10 –10 erg s , or that L/LEdd 10 –10 . ∼
The nature of the nuclear emission
The nuclear source shows several properties typical of an AGN. First, variability:
The source varies by a factor of nine in eight months in 2004, and also varied between
2000 and 2004. Second, the x-ray spectrum: The spectrum is an absorbed power-law with slope 2. In particular, the spectrum cannot be fit by a thermal component ∼ alone — a power-law is necessary. The low state spectrum is reminiscent of highly obscured AGN where an unabsorbed but diminished spectrum is seen via scattering.
Third, x-ray colors: The ratio of 0.3–2 keV, 2–5 keV, and 5–8 keV counts puts this object into the Compton-thick AGN part of the Levenson color-color diagram (Fig.
9 in Levenson et al. 2006). In the context of the latter two points it is worth noting that if the source really is a highly absorbed AGN then energy emitted by the AGN may show up as re-processed radiation in the infrared. The existence of an infrared source with luminosity 1041 erg s−1 is consistent with this picture, though it does ∼ not argue for the presence of an AGN to exclusion of other types of sources.
The x-ray properties of the nucleus are consistent with both the XRB and AGN hypotheses, but an AGN is not ruled out. The inferred x-ray luminosity (1037−38 erg s−1) is in the range seen from XRBs, and the observed x-ray colors put this source in the LMXB region of the Prestwich et al. (2003) color-color diagram. In favor of
41 the XRB scenario is that the variability of the source is similar to the spectral state changes in XRBs. The low state may be interpreted as being the XRB state known as “hard” or “low/hard”, and the best-fit power law slope, Γ 1.7 is the value seen ∼ in XRBs in this state. The high state would then be the XRB state known as the
“very high” or “steep power law” state. The best-fit power-law slope Γ 2.2 in ∼ this case is less steep than is usually seen in XRBs in this state (Γ > 2.4) but the ∼ steeper value is included in the 90% confidence range. However, it must be kept in mind first, that at the quality of the spectra analyzed here the power law slopes are consistent with being identical in the two states, and second, that AGNs can show the same state behavior (e.g. 1H 0419-577, Pounds et al. 2004). In XRBs, the steep power-law state is associated with the presence of quasi-periodic oscillations in the x-ray emission, while the hard state is associated with the presence of a radio jet. In principle the presence of these features could provide additional evidence supporting the XRB hypothesis, but it is not currently feasible to detect them in XRBs at the distance of NGC 5457.
Thus, there are again two possible scenarios, as in the case of NGC 3184. The x-ray source could be an HMXB in a super-star cluster. The star cluster would dominate the optical and IR emission. The large variation observed in the x-ray flux rules out the source’s being more than one HMXB, as otherwise they would have to be varying in concert. The actual amount of obscuration is important, however, for the plausibility of the HMXB hypothesis since the inferred x-ray luminosity ( 1038 ∼
42 erg s−1 in the high state) is already at the high end of the range of XRB luminosities and there is not much room for a significantly higher absorption-corrected intrinsic luminosity. The alternative scenario is an AGN together with a nuclear star cluster.
In AGNs both the intrinsic luminosity and the amount of obscuration are known to vary. The low state x-ray spectrum may be explained as a truly under-luminous
(1037 erg s−1) unabsorbed AGN, or as an AGN where the obscuration is so high that only scattered light, and no direct emission, is seen. The fact that the x-ray colors are similar to those of Compton-thick AGN (Levenson et al. 2006) supports the latter view, but we also note the absence of the 6.4 keV Fe Kα line in the x-ray spectrum that is often present in the reflected component. Though both the HMXB and AGN hypotheses are possible, on balance the AGN appears to be the more plausible one.
2.5. Discussion
Of the six galaxies presented here, all six show nuclear x-ray sources. This implies that it is a very common occurrence. NGC 3169 and NGC 4102 were regarded as low luminosity AGNs. None of the remaining four nuclei, however, was known to have an accreting SMBH, of any mass. NGC 3169 and NGC 4102 are of type Sa and Sb, which have massive bulges and are expected to have massive
SMBHs. For galaxies of types Scd and Sd, on the other hand, the lack of a luminous
43 AGN could mean either that there is no SMBH or that the mass of the SMBH is low. Given that the sample presented in this chapter consists of only six galaxies, we do not draw statistical conclusions here of the prevalence of very low-luminosity
AGNs in nearby galaxies. But we note that, as shown in 2.4.1–2.4.6, of the six §§ nuclear x-ray sources, NGC 3169 is almost certainly an AGN, and NGC 4102, NGC
3184, and NGC 5457 have very strong, though not conclusive, arguments in favor of their being AGNs. The two remaining galaxies, NGC 4713 and NGC 4647, are ambiguous but AGNs are not ruled out.
A detailed discussion of the issues involved in identifying low luminosity AGNs using x-ray observations is deferred to Chapter 7. Here we briefly note that, first, the method is feasible for uncovering previously unknown AGNs or candidates, and second, multi-wavelength data are crucial when the luminosity of the putative
AGN is very low. For the six galaxies in this chapter multi-wavelength photometry is summarized in Tables 2.5 and 2.6. NGC 3184 provides a good example where different modes of observing, imaging and spectroscopy, in two wavebands, x-ray and
IR, together make a compelling argument for the presence of an AGN where each observation individually is ambiguous. Of the six galaxies here, NGC 3169, NGC
4102, and NGC 5457 have measurements of either the stellar velocity dispersion or the luminosity of the bulge, thus allowing an estimate of their SMBH masses
(assuming here that the source in M 101 is an AGN). The scatter in the M•–σ and M•–Lbulge relations, together with the uncertainty in the observed flux and the
44 bolometric correction, result in uncertainties of about an order of magnitude in the
−4 inferred Eddington ratio, but all three objects have L/LEdd 10 . This is in the ∼ −5 range seen in low-luminosity AGNs (e.g. L 10 LEdd for M 81; Young et al. 2007), ∼ −9 and much higher than the L 10 LEdd seen in truly quiescent SMBHs. ∼
45 Morph. Nucleus Coordinates (J2000) Dist.b Scale Exp.
Target Type Typea RA Dec (Mpc) (pc/′′) Obs. Date ObsID (ks)
NGC 3169 Sa L2 10 14 15.0 +03 27 58 19.7 96 2001 May 2 1614 2.0
46 NGC 3184 Scd H II 10 18 17.0 +41 25 28 8.7 42 2000 Jan 8 804 39.8
2000 Feb 3 1520 21.3
NGC 4102 Sb H II 12 06 23.1 +52 42 39 17.0 82 2003 Apr 30 4014 4.9
NGC 4647 Sc H II 12 43 32.3 +11 34 55 16.8 81 2000 Apr 20 785 36.9
(cont’d) Table 2.1. Targets and observation parameters Table 2.1—Continued
Morph. Nucleus Coordinates (J2000) Dist.b Scale Exp.
Target Type Typea RA Dec (Mpc) (pc/′′) Obs. Date ObsID (ks) 47 NGC 4713 Sd T2 12 49 57.9 +05 18 41 17.9 87 2003 Jan 28 4019 4.9
NGC 5457c Scd H II 14 03 12.6 +54 20 57 7.2 35 c c 750 ··· ···
aFrom Ho et al. (1997a). L2: Type 2 LINER, T2: Type 2 Transition object. bFrom Tully (1988) except from Stetson et al. (1998) for NGC 5457. cNGC 5457 (M 101) was observed multiple times. The total observation time analyzed here is given in this table. The individual observations are listed in Table 2.3 Counts
Broad Hard Soft Bkg/Src
Target Src Bkg Src Bkg Src Bkg Area Ratio HRa
+0.05 NGC 3169 159 23 148 5 11 18 21.4 +0.86−0.03
+0.08 NGC 3184 N 36 95 4 33 32 62 92 0.75− − 0.13 +0.05 S 117 95 13 33 104 62 60 0.77− − 0.07 +0.04 NGC 4102 all 354 48 78 8 276 40 6.8 0.55− − 0.05 +0.08 core 171 48 68 8 103 40 52.5 0.20− − 0.07 +0.03 ext 115 48 6 8 109 40 26.1 0.88− − 0.06 +0.04 NGC 4647 15 38 1 15 14 23 10.1 0.80− − 0.20 +0.09 NGC 4713 10 4 1 0 9 4 21.4 0.69− − 0.25 NGC 5457 314 256 23 45 291 211 20.8 0.86 0.03 − ±
aHardness ratio, HR = (H S)/(H+S), where H and S are the counts in the hard and soft bands respectively, calculated− using the method described in Park et al. (2006). The tool used is available at http://hea-www.harvard.edu/AstroStat/BEHR/.
Table 2.2. X-ray Measurements
48 North source South source
Exp Time Counts Rate HR Counts Rate HR
ObsID Obs Date (ks) (ks−1) (ks−1)
+0.03 934 2000 Mar 26 97.8 262 2.68 0.81− 264 2.70 0.70 0.04 − 0.04 − ± a +0.07 +0.06 4731 2004 Jan 19 56.2 92 1.64 0.64− 136 2.41 0.58− − 0.09 − 0.07 49 +0.08 +0.06 5300 2004 Mar 7 52.1 36 0.69 0.75− 117 2.24 0.60− − 0.13 − 0.08 +0.10 +0.05 5309 2004 Mar 14 70.8 38 0.54 0.62− 158 2.24 0.64− − 0.14 − 0.07 +0.07 +0.06 4732 2004 Mar 19 69.8 56 0.80 0.77− 151 2.17 0.58− − 0.09 − 0.07 +0.06 5322 2004 May 3 64.7 54 0.83 0.79− 144 2.22 0.69 0.06 − 0.10 − ± +0.09 +0.06 5323 2004 May 9 42.4 42 0.99 0.64− 120 2.83 0.54− − 0.14 − 0.08
(cont’d) Table 2.3. NGC 5457 X-ray Measurements Table 2.3—Continued
North source South source
Exp Time Counts Rate HR Counts Rate HR
ObsID Obs Date (ks) (ks−1) (ks−1)
5338a 2004 Jul 6 28.6 61 2.14 0.56+0.09 41 1.45 0.50+0.12
50 − − − 0.12 − 0.14 b +0.13 +0.06 5339 2004 Jul 7 14.0 18 1.26 0.60− 27 1.90 0.79− − 0.22 − 0.15 +0.06 +0.04 5340 2004 Jul 9 54.4 81 1.48 0.72− 110 2.02 0.74− − 0.08 − 0.07 +0.13 +0.09 4734 2004 Jul 11 35.5 31 0.87 0.53− 63 1.77 0.61− − 0.17 − 0.11 +0.08 +0.08 6114 2004 Sep 5 38.2 91 2.38 0.60− 64 1.68 0.64− − 0.09 − 0.11
(cont’d) Table 2.3—Continued
North source South source
Exp Time Counts Rate HR Counts Rate HR
ObsID Obs Date (ks) (ks−1) (ks−1) 51 +0.05 +0.07 6115 2004 Sep 8 35.4 96 2.72 0.78− 63 1.79 0.70− − 0.07 − 0.10 +0.07 +0.06 4735 2004 Sep 12 28.8 70 2.43 0.70− 49 1.71 0.77− − 0.10 − 0.11 ab +0.03 +0.06 4736 2004 Nov 1 77.4 290 3.75 0.78− 122 1.58 0.67− − 0.04 − 0.07 a +0.05 +0.09 6152 2004 Nov 7 26.7 133 4.99 0.78− 51 1.92 0.63− − 0.06 − 0.12
aSource extended, or possibly image smeared. bPossible residual contamination from background flare. Galactic Spectral Fit Parameters
a b c d e c c 2 Target NH Model kT NH Γ Refl. Line EW χν(dof)
+46 +1.2 f NGC 3169 2.86 ab(pl) 99− 2.0− 1.1(26) ··· 36 1.1 ········· +90 +2.1 ab(pl) 116− 2.6− 0.3(22) ··· 52 1.5 ········· +1.4 f
52 NGC 4102 core 1.79 ab(pl) 0 2.0 0.5 7.4(27) ··· ± ········· +0.4 +149 f ab(rn) 0 1.8 0.4 108− 2.3(26) ··· ± 61 ······ +1.9 +0.6 g ab(rn+g) 0 2.3− 129 6.4 3.3 0.6(25) ··· 0.5 +0.6 +0.6 +283 f ab(rn+g) 0 2.2− 129− 6.4 2.5 1.6(25) ··· 0.5 85 +8.6 +2.8 ext ab(br) 1.2− 1.2− 0.5(18) 0.8 1.2 ············
(cont’d) Table 2.4. X-Ray Spectral Fits Table 2.4—Continued
Galactic Spectral Fit Parameters
a b c d e c c 2 Target NH Model kT NH Γ Refl. Line EW χν(dof)
NGC 5457 sep. low 1.15 ab(me+pl) 0.3+0.2 0+1.8 1.7 0.5 0.5(28)
53 − 0.1 ± ········· +1.1 +0.4 sep. high ab(pl) 1.5− 2.2− 0.5(40) ··· 1.5 0.3 ········· +0.3 +1.9 +0.4 sim. low ab(me+pl) 0.3− 0.42− 2.0− 0.5(28) 0.1 0.42 0.2 ········· +0.4 sim. high ab(pl) 1.1 0.6 2.0− 0.5(41) ··· ± 0.2 ·········
(cont’d) Table 2.4—Continued
Galactic Spectral Fit Parameters
a b c d e c c 2 Target NH Model kT NH Γ Refl. Line EW χν(dof)
NGC 5457 merged ab(pl) 0.6 0.4 1.9 0.2 0.6(70) ··· ± ± ········· 54 aIn units of 1020 cm−2. bModel labels: ab=xswabs, photoelectric absorption; ga=xswabs with value frozen at Galactic column density towards this target; br=xsbremss, thermal bremsstrahlung; g=gauss1d, one-dimensional Gaussian; me=xsmekal, thermal plasma; pl=powlaw1d, one-dimensional power law; rn=xspexrav, power-law reflected by neutral material. cIn units of keV. dIn units of 1021 cm−2. eReflection scaling factor. f Cash statistic (cstat in Sherpa). gUnconstrained by fit. Filter NGC 3169 NGC 4102 NGC 4647 NGC 3184 NGC 5457 NGC 4713
Wavelength or Band (Sa) (Sb) (Sc) (Scd) (Scd) (Sd)
X-ray 0.3–8 keV 41.7 40.2 39.0ab 37.3c 37.5–38.5d 38.6c
UV & Opt. F300W < 39.4 ········· ······ 55 F336W 39.0 ············ ··· F547M 39.5 ············ ··· F606W 39.5 40.1 ········· ···
IR Ks 42.6 42.8 38.8 40.9 41.0 41.0
(cont’d) Table 2.5. Inferred nuclear luminosities Table 2.5—Continued
Filter NGC 3169 NGC 4102 NGC 4647 NGC 3184 NGC 5457 NGC 4713
Wavelength or Band (Sa) (Sb) (Sc) (Scd) (Scd) (Sd)
Radio 15 GHz < 36.8 ··············· 5 GHz 37.2 < 35.8 ········· ··· 56 1.4 GHz 38.0 36.9b < 35.1 < 34.9 < 35.7 ···
−1 −1 Note. — Values are log(L/erg s ) in the specified bandpass in the x-ray and log(νLν/erg s ) where a single filter, wavelength, or frequency is given. Note that the values are derived from observations with varied instruments, PSFs, apertures, and signal-to-noise ratios, and the uncertainties can be as large as a factor of two. a0.3–12 keV. bLarge positional uncertainty makes it unclear whether the detected source is really the nucleus. cBased on a small number of counts. A power-law spectrum was assumed. Intrinsic absorption is unknown.
dVariable source. a Band Aperture log(νLν)
(µm) Radius (erg s−1)
3.6 3′′ 40.6
4.5 3′′ 40.4
5.8 3′′ 40.7
8.0 3′′ 40.8
24 6′′ 41.1
70 10′′ 41.5
160 10′′ 41.7
aUncertainties are 10% in the 3.6µm, 4.5µm, 5.8µm,
8.0µm, and 160µm bands; 4% at 24µm; 12% at 70µm.
Table 2.6. NGC 3184 luminosity in Spitzer bands.
57 Chapter 3
A strong candidate AGN in NGC 4713
3.1. Introduction
The previous chapter (see 2.4.5) presented a Chandra observation of the § nucleus of the bulgeless galaxy NGC 4713 (type Sd), which showed that the nucleus is indeed an x-ray source. However, only 10 counts were obtained, making it impossible to characterize the nature of the source. At the time only two AGNs were known in Sd galaxies: NGC 4395 (Ho et al. 1997a) and NGC 3621 (Satyapal et al.
2007). The nature of the nuclear source in NGC 4713 was therefore worth examining in more detail to see whether it was another example of an accreting SMBH in a bulgeless galaxy. In order to analyze the x-ray spectrum of this source we proposed for and obtained a 70 ks observation with XMM-Newton. The reasons for using
XMM-Newton rather than Chandra were, first, that XMM-Newton has an effective area 10 times larger than that of Chandra, making it possible to obtain sufficient ∼ counts for spectral analysis without needing prohibitively long exposure times; and second, that XMM-Newton by virtue of its Optical Monitor (OM) telescope can take simultaneous x-ray and optical/UV exposures. As emphasized in Chapter 7,
58 multiwavelength data are often crucial in identifying low luminosity AGNs. The observations and results are presented below. As can be seen, though the data at present fall short of proving conclusively that the source is a bona fide AGN they nevertheless provide very strong arguments for believing that is the case. The UV data from the OM have the potential for providing the necessary conclusive evidence
(see 3.4) and the present uncertainty is due not to the data but inaccuracies in the § data analysis software.
3.2. Observations and data analysis
NGC 4713 was observed by XMM-Newton on 2007 December 22 for 69.2 ks.
The primary instrument was EPIC, with the thin filter. Data processing was done using the Science Analysis System v. 9.0.0. Current calibration files were obtained by using an online version of the tool cifbuild 1. EPIC PN and MOS data were processed using the tools epproc and emproc respectively. Standard filters were applied to the event lists: PATTERN 12, 200 PI 15000, and FLAG = #XMMEA EP for ≤ ≤ ≤ PN, and PATTERN 12, 200 PI 12000, and FLAG = #XMMEA EM for ≤ ≤ ≤ MOS. To check for background flares, lightcurves were made for each camera (PN,
MOS1, MOS2) using events from the whole instrument with nominal energy > 10 keV. For obtaining good time intervals (GTIs) an iterative filtering scheme was applied to the lightcurves whereby all time bins having a count rate greater than 1.2
1http://xmm2.esac.esa.int/external/xmm_sw_cal/calib/cifbuild.shtml
59 times the mean count rate were discarded. The resultant GTIs were then used to
filter the event lists. The final usable exposure times were 50.6 ks for PN, 61.4 ks for
MOS 1, and 62.8 ks for MOS2.
The nucleus is clearly detected in all three EPIC cameras. There are two sources near the nucleus, at distances of 24′′ and 30′′ respectively. To reduce ∼ ∼ contamination from these sources, we extracted source counts from a circle centered on the nuclear position and with a radius of 15′′ (approximately 60% encircled energy radius). In all cases background counts were extracted from a rectangular source-free region on the same chip and near the nucleus. For the PN detector the background region was chosen such that the range of RAWY coordinate values was the same as for the source region, to ensure that the CCD response function in the background region was similar to that at the source location. The net broad band (0.3–12 keV) source counts thus obtained were 508 in PN, 144 in MOS 1, and 140 in MOS 2.
Before extracting spectra the event lists were filtered to match the criterion
FLAG=0 to ensure that only clean events were selected. Spectra were fit using
Sherpa, the spectral fitting tool part of the Chandra Interactive Analysis of
Observations (CIAO; Fruscione et al. 2006) package. Background spectra were subtracted from source spectra, and all fitting was done between 0.3 and 10 keV.
Spectra were binned by counts in two ways, first with 5 counts per bin and then with 20 counts per bin. Both types of binning produced consistent fitting results.
Spectra from the PN, MOS1, and MOS2 instruments were fit simultaneously. The
60 best-fit values and uncertainties quoted here come from the fit with 20 counts per bin. The quoted uncertainty represents the 90% confidence level for one parameter.
3.3. Results and discussion
The spectrum shows significant excess over a power law at energies below 1 keV. There is also a marginal excess between 6 and 7 keV. The excess at low energies is peaked at 0.6 keV rather than having the broad distribution of blackbody or ∼ bremsstrahlung radiation. It can be fit, on a purely descriptive basis, by a Gaussian.
Physically it may be a blend of emission lines from circumnuclear ionized gas. The
final model used to fit the whole spectrum (0.3–10 keV) consists of an absorbed power law and two Gaussians (Fig. 3.1). The best fit parameter values are given in Table 3.1. The x-ray spectrum is essentially a power law with slope Γ = 2.
The flux in the power law is F (0.3 10 keV) = (3.0 0.9) 10−14 erg cm−2 s−1, − ± × which corresponds to a luminosity L(0.3 10 keV) 1 1039 erg s−1. Since the − ≈ × spectrum was extracted from within the 60% encircled energy radius the true ∼ luminosity is about a factor of two higher. The best-fit absorbing column density is only a few 1020 cm−2, and is in fact consistent with zero. However, the presence × of soft emission in addition to the power law component introduces a caveat to the determination of the amount of absorption present in the system. This is because a part of the power law emission lost to absorption may be compensated for by the
61 soft component. The number of counts in the current spectrum is inadequate to separate the contributions of the power law and the additional soft component.
The primary question is whether the nuclear x-ray source is an AGN or not.
The two pieces of evidence to consider here are the shape of the x-ray spectrum and the x-ray luminosity. The luminosity, 2 1039 erg s−1, by itself is consistent ∼ × with XRBs, ULXs, and AGNs, but values higher than 1039 erg s−1 are usually ∼ associated with ULXs and AGNs rather than XRBs. Of these two possibilities, ULX and AGN, we argue that an AGN provides a more plausible explanation when x-ray and optical data are considered together.
ULX spectra typically consist of a soft component that is interpreted to be blackbody emission from the inner accretion disk, and a hard power law component.
In addition, it has recently been suggested that the hard emission is not a single power law, but that there is a break at 4–5 keV (Stobbart et al. 2006). However, ∼ attempts to fit the spectrum of NGC 4713 with disk blackbody plus power law or broken power law models failed to find any fits with meaningful values of the disk temperature. Formally a broken power law is not preferred over a single power law; however, the number of counts at energies > 4 keV is too low to distinguish between ∼ the two models (Stobbart et al. 2006).
The shape of the spectrum is suggestive of a very heavily obscured AGN where little or none of the direct emission reaches the observer. In such cases the observed
62 Fig. 3.1.— EPIC PN spectrum of the nucleus of NGC 4713. The squares in the upper panel show the data, binned such that there are at least 5 counts in each bin. The solid line shows the best-fit model, which is a power law with slope Γ = 2. The lower panel shows the residuals from the fit.
63 emission is the small fraction of the total emission that has been scattered into the line of sight. The observed spectrum is an apparently unabsorbed power law, and sometimes a 6.4 keV Fe Kα line is present. The x-ray spectrum of NGC 4713 clearly shows the first two characteristics, namely a power law spectrum and low absorption, and marginal evidence for the presence of a 6.4 keV line. If the hypothesis of a heavily obscured AGN is correct, it implies that the true luminosity of NGC 4713 is much higher than the apparent luminosity. It is noteworthy that Ho et al. (1997a) classified the nucleus as a “transition” object based on its optical spectrum. That is, the emission line ratios were intermediate between the values expected for pure star forming nuclei and those for pure AGNs. The optical data, therefore, are consistent with the presence of a heavily absorbed AGN in the nucleus. In this interpretation of the data, the bolometric luminosity of the AGN is high, 1041 erg s−1, given that ∼ scattered x-ray emission is itself of the order of 1039 erg s−1. The luminosity of NGC
4713 would be similar to the only three known AGNs in Sd galaxies (NGC 4395,
NGC 3621, NGC 4178) all of which also have bolometric luminosities of a few 1041 × erg s−1.
NGC 4713 provides a second example (the other is NGC 3310; Ghosh et al.
2009b) where a candidate AGN is detectable in x-rays but not in infrared. The presence of high ionization emission lines, for example, [NeV] 14µm and 24µm, are taken as indicators of the presence of AGNs. However, Spitzer Space Telescope spectroscopy of the nucleus of NGC 4713 failed to detect [NeV] emission lines
64 (Satyapal et al. 2009). This demonstrates the efficacy of using x-ray observations to detect hidden AGNs in nearby galaxies. Longer x-ray observations would enable a
firmer detection of the 6.4 keV Fe K4α line and a test for x-ray variability, both of which are characteristic features of AGNs. However, the current evidence (see also
3.4) while not conclusive, strongly points to NGC 4713 being the fourth known § example of an Sd galaxy hosting an accreting SMBH.
3.4. The UV variability of NGC 4713
The XMM-Newton telescope can conduct optical and UV observations concurrently with x-ray observations using its 30 cm diameter Optical Monitor telescope. In this section we present 14 consecutive exposures in the U filter (effective wavelength 3440 A),˚ each of 2500 s duration except for the final exposure which was
1340 s. Fig. 3.2 shows one of images, which is representative of all of them.
The angular resolution of the OM is about 0.5′′, and NGC 4713 is known to have a nuclear star cluster (see 2.4.5). For both these reasons the flux of the § AGN cannot be directly measured from the OM images. However, the multiple exposures present an opportunity to look for variability in the nuclear emission.
Emission from star forming regions and star clusters is not expected to vary on the timescales probed by these images, whereas emission from AGNs is known to vary
65 Fig. 3.2.— One of the U-band images of NGC 4713 obtained with the Optical Monitor. The green circle at the center shows the location of the nucleus. The image is 2′ on a side. The bar in the lower right represents 10′′, or a projected distance of 870 pc. North is up and east to the left in the image. ∼
66 at all timescales. Therefore, variability in the total emission from the nucleus would provide conclusive evidence that there is an AGN in NGC 4713.
Unfortunately, in the course of processing the OM images a bug was discovered in the SAS OM photometry tool omsource that called into question the reliability of the calculated fluxes. As of this writing this bug has not been fixed and there is no known workaround. With this caveat, and for the sake of completeness, a plot of flux vs. time is shown in Fig. 3.3 for the nucleus and an off-nuclear star cluster. There is reason to suspect that the individual measurements may be incorrect by 0.1–0.2 mag. As measured, the nucleus shows greater variability than the star cluster. As it is uncertain whether this result will hold true once the issue with photometry is resolved, we do not use this result in any of the arguments presented here.
67 17.5 nucleus cluster 17.6
17.7
mag 17.8 λ AB
17.9
18
18.1 18 20 22 24 26 28 30 32 Image number Fig. 3.3.— Measured fluxes from the OM U-band images of NGC 4713. The measurements are shown here only for completeness, as there are questions about the accuracy of the calculated flux (see text). The true flux from a star cluster is expected not to vary on the timescales shown here, the images being consecutive 2500 s exposures. The nucleus appears to vary in comparison to the star cluster. ∼
68 Model
Component Parameter Best-fit value
+9.2 20 −2 Absorption NH 2.6− 10 cm 2.6 × Power law Γ 2.0 0.4 ± F (0.3 10 keV) 3.0 0.9 erg cm−2 s−1 − ± +0.07 Gaussian 1 Position 0.65−0.14 keV
+0.22 FWHM 0.25−0.12 keV
+0.9 Eq. width 0.3−0.2 keV
Gaussian 2 Position 6.4 0.1 keV ± +1.4 FWHM 0.9−0.9 keV
Eq. width 3.2 2.7 keV ± χ2/dof PN 16.4/21
MOS1 7.2/5
MOS2 2.6/5
Table 3.1. Spectral model parameters and goodness of fit.
69 Chapter 4
The Chandra Survey: I. The Archival Data
4.1. Sample Selection
A quasi-volume-limited sample is drawn from the Nearby Galaxies Catalog
(Tully 1988) by selecting all galaxies that have (1) distance R 20 Mpc, (2) ≤ morphological types Sdm and earlier, as denoted by HubCode values < 9 in the catalog, (3) inclination angle i< 35◦, and (4) Galactic latitude b 30◦. The third | | ≥ and fourth criteria minimize the probability of obscuration by the host galaxy and the Milky Way, respectively. There are 98 galaxies in the catalog that meet these criteria. However, three galaxies have more recent distance estimates that put them farther than 20 Mpc, so these galaxies were excluded1. Of the 95 remaining galaxies, the following were excluded: (1) E and S0 galaxies brighter than MB = 18.50, − as massive bulge-dominated galaxies are unlikely to host low-mass SMBHs, (2) starburst galaxies, as they have large populations of x-ray binaries (XRBs) and often have ultra-luminous x-ray sources (ULXs), making it difficult to identify and study
1A fourth galaxy with a distance error was found after the observations were already made. This galaxy, NGC 1400, is discussed in Chapter 5.
70 any putative AGN, and (3) galaxies that were already known to host AGNs (all of the AGNs had been identified based on their optical spectra). These criteria and
filters resulted in a sample of 56 galaxies, consisting of 54 spiral galaxies ranging in type from S0 to Sdm, 1 of type E, and 1 classified as BCD/E. Of the 56 galaxies,
18 already had observations in the Chandra archive (the “archival sample”). We obtained new Chandra observations for 37 of the unobserved galaxies. One galaxy was awarded to a different Chandra program. The new observations are presented elsewhere (Chapters 5 and 6, and Ghosh et al. 2009a,c). In this chapter we focus on the archival sample. Two galaxies in the archival sample, NGC 3184 and NGC
5457, have already been analyzed in Chapter 2 as part of a different sample and the data are not presented here again, but they are included when the archival sample is discussed as a whole.
4.2. Data analysis
Data were downloaded from the Chandra archive and analyzed using version
3.4 of the Chandra Interactive Analysis of Observations (CIAO; Fruscione et al.
2006). Parameters of the observations are given in Table 4.1. Level 1 event lists were re-processed with observation-specific bad-pixel files and the latest calibration
files to create new Level 2 event lists. The Level 2 event lists were filtered to exclude times when the detector experienced background flares. The event lists were then
71 converted to FITS images, and the CIAO source detection tool wavdetect was run on them to determine source positions. Source counts were extracted from a circle centered on the wavdetect source position and with a radius equal to the greater of
4.67 pixels (2.3′′) and the 95% encircled-energy radius at 1.5 keV on ACIS-S at the position of the source. Background counts were taken from an annulus with an inner radius of two times, and an outer radius of five times, the source circle radius after excising any sources that happened to fall in the background region. Counts were extracted in the 0.3 – 8.0 keV (Broad), 2.5 – 8.0 keV (Hard) and 0.3 – 2.5 keV (Soft) bands. A hardness ratio (HR) was defined as HR = (H S)/(H + S), where H − and S are the counts in the hard and soft bands, respectively. Counts and hardness ratios are given in Table 4.2.
Spectral fits were performed with the CIAO tool Sherpa. Uncertainties reported in fit values represent the 90% confidence level for one parameter. If the nuclear point source was embedded in diffuse soft emission we obtained the power law index and absorbing column density in three steps. First, we fit the spectrum of the diffuse emission with a thermal plasma (MEKAL) model. Next, assuming an absorbed power law model for the nuclear spectrum, the slope of the power law was obtained by fitting the spectrum between 3 and 8 keV, where both the plasma component and the effects of absorption are minimal. Finally, the amount of intrinsic absorption was then obtained by fitting the complete spectrum (0.3–8 keV) with plasma
72 temperature and power law slope fixed at the best-fit values already obtained. All spectral fits are listed in Table 4.3.
Where spectral fitting was not possible because of insufficient counts estimates of fluxes were made using PIMMS2 using the broad band count rate, assuming
20 −2 Galactic NH = 10 cm and a power law source spectrum with index Γ = 2. No correction was made for intrinsic absorption, which cannot be determined for these sources. In the case of a nondetection, we take 3√Ct as the upper limit to the number of counts produced by the source, where Ct is the total number of counts in the source circle at the position of the nucleus as determined from observations at other wavelengths.
4.3. The Galaxies
The galaxies in the archival sample are presented in this section, broken into three groups. The first group consists of those galaxies which have strong arguments in favor of their hosting AGNs. The second group is made up of the galaxies where
AGNs are a plausible explanation of the data but are not favored over XRBs. The galaxies in the third group have inadequate data to make arguments either for or
2Portable, Interactive Multi Mission Simulator v3.9i. http://cxc.harvard.edu/toolkit/pimms.jsp
73 against the presence of an AGN. Within each group the galaxies are ordered by their catalog number.
4.3.1. Group 1: The best AGN candidates
NGC 628 NGC 628 was observed twice with Chandra (Table 4.1). The nuclear source is isolated and is fairly luminous, with L(0.3 8) 1038 erg s−1 in both − ∼ 38 Chandra observations. An XMM observation in 2002 also found LX 2 10 erg ∼ × s−1 (Soria & Kong 2002). The x-ray spectra obtained from the Chandra data are well
fit by a power law with Γ 1.7 and do not show signs of intrinsic absorption. This ≈ galaxy is part of the Spitzer Infrared Nearby Galaxies Survey (SINGS; Kennicutt et al. 2003). Dale et al. (2006) have used SINGS data to create diagnostic diagrams that distinguish between AGN and star-forming galaxies, using the equivalent width of the PAH feature at 6.2µm and the fluxes in a mix of high- and low-ionization lines.
The nucleus of NGC 628 falls into the “transition” region between AGNs and H II nuclei when using the [Si II]34.9µm/[Ne II]12.8µm line ratio, and in the H II nucleus region when using the [Si II]34.9µm/[S III]33.5µm ratio. The nucleus was unresolved
41 in HST/NICMOS observations at 1.6µm and had luminosity νLν 1.5 10 erg ∼ × s−1 (Quillen et al. 2000). The x-ray luminosity and the power law x-ray spectrum are suggestive of an AGN; the IR line ratios suggest that there is an AGN component to the emission; the fact that the nucleus was unresolved by NICMOS and the high
IR luminosity all support the hypothesis of the presence of an AGN. In addition,
74 the data are consistent with the AGN being highly obscured. The high absorption is suggested by the following features. First, the x-ray colors, that is, the ratio of counts in the 0.3–2, 2–5, and 5–8 keV bands, is consistent with that expected for
Compton thick AGNs (Levenson et al. 2006). Second, the power law spectrum is apparently unabsorbed, as is often the case when the absorption is so heavy that no direct emission is seen, and what is observed is emission that is scattered from other directions into our line of sight. Third, the IR luminosity is 3 orders of magnitude ∼ higher than the apparent x-ray luminosity. And fourth, there were no strong optical emission lines in the spectrum of the nucleus obtained by Ho et al. (1995). We note that an XRB in a nuclear star cluster cannot be ruled out, but the x-ray emission became harder when the source flux was higher, which is opposite to the expected behavior of XRBs.
NGC 1073 The nucleus of NGC 1073 was detected in both soft and hard bands, but the total number of counts was 28, too few for spectral fitting. According to
PIMMS, for a Γ = 2 power law and only Galactic absorption, the observed count rate corresponds to a flux of F (0.3 8) 5 10−14 erg cm−2 s−1, or L(0.3 8) 1.4 1039 − ∼ × − ∼ × erg s−1. Since the amount of intrinsic absorption is unknown, this luminosity must be taken as a lower limit. The high x-ray luminosity by itself is sufficient for NGC
1073 to be considered a good candidate AGN.
75 NGC 1291 The nucleus of NGC 1291 was identified as a possible LLAGN by
Irwin et al. (2002) based on its x-ray luminosity and variability. In addition, Irwin et al. (2002) point to unresolved 60µm and 100µm emission (Rice et al. 1988) without the expected CO emission (Bregman et al. 1995) as another reason to think there is an AGN. The nucleus was also part of a recent survey by Zhang et al.
(2009). We summarize the x-ray properties here. NGC 1291 was observed twice by
Chandra. The nucleus is a hard source, surrounded by other point sources. The point sources are embedded in diffuse soft emission. The nuclear spectrum is well
22 −2 fit by an absorbed power law (Γ = 1.3–1.6, NH 10 cm ) plus additional soft ≈ emission from the diffuse background. In ObsID 795 the unabsorbed flux in the power law is F (0.3 8keV) = 5.1 10−13 erg cm−2 s−1, which corresponds to a − × luminosity of L(0.3 8 keV) = 4.5 1039 erg s−1 for our adopted distance of 8.6 − × Mpc. In ObsID 2059 the flux in the power law is a factor of two lower. The nucleus is also a radio source, with integrated flux density at 1.49 GHz of 3.9 mJy (Condon
35 −1 1987), implying νLν 5 10 erg s . ∼ ×
NGC 2681 Chandra observations of NGC 2681 were analyzed in Kilgard et al.
(2005), and the nucleus has been identified as a candidate AGN multiple times
(Satyapal et al. 2005; Flohic et al. 2006; Gonz´alez-Mart´ın et al. 2006; Zhang et al.
2009). Our re-analysis of the data gives results consistent with the previous authors’.
A brief description of its x-ray properties is given here for completeness. The nucleus
76 is a hard point source embedded in diffuse soft emission. The nuclear spectrum can be fit with a power law with 1.2 Γ 1.9. The best-fit column density is ≤ ≤ 21 −2 NH (3 1) 10 cm (Flohic et al. 2006). The 0.3–8 keV flux in the power law ≈ ± × component is approximately 2 10−14 erg cm−2 s−1 with an uncertainty of 35%. × This flux correponds to a luminosity of L(0.3 8 keV) 3 1038 erg s−1. In addition − ≈ × to the high x-ray luminosity there are also a few arguments in favor of an AGN from optical data. The nucleus was classified as a LINER 1.9 by Ho et al. (1997a).
Cappellari et al. (2001) report that nuclear Hα+[N II] emission cannot be coming from a size substantially larger than their HST/FOS aperture, 0.21′′ 0.21′′, or 13 × ∼ pc 13 pc. The nucleus was not detected in the radio at 5 GHz (Filho et al. 2006) × or at 15 GHz (Nagar et al. 2005).
NGC 3310 The nucleus is very prominent and is the brightest x-ray source in the galaxy. Even though nucleus is embedded in diffuse emission, it is a hard source
(HR = 0.29) and quite distinct, as can be seen in the hardness ratio map (Fig. 4.1). − We extracted spectra of both the diffuse emission and the nucleus. Spectral models and fit parameters are shown in Table 4.3. We assume the spectrum extracted from the nuclear source region (marked with a circle in Fig. 4.1) consists of an absorbed power law with the diffuse emission superimposed. The best-fit column
21 −2 density is NH = (8.7 1.7) 10 cm . The flux in the power-law component is ± × F (0.3 8 keV) = 5.5 10−13 erg cm−2 s−1, with an uncertainty of about 11% in − × 77 the power law normalization. For our adopted distance this translates to luminosity
L(0.3 8 keV) = 2.3 1040 erg s−1, making it almost certain the source is an AGN. − × This conclusion is in agreement with Tzanavaris & Georgantopoulos (2007) who also analyzed the Chandra observation of NGC 3310, but the results of their spectral
fitting were different, in that they obtained a flatter power law (Γ 1.3) and lower ∼ 21 −2 absorption (NH 2.3 10 cm ). This difference may arise from the fact that we ∼ × fit the spectrum in multiple steps, as described in 4.2, and obtained the power law § slope and absorption separately whereas they obtained both values from a single fit.
They also report a 2σ detection of a 6.4 keV Fe Kα line, which we do not detect. We cannot account for this difference with certainty, but it may be an effect of differing temporal filters and binning that are applied to the data. Additional support for the
AGN hypothesis comes from the fact that the nucleus, but not the circumnuclear x-ray sources, is clearly seen in a 2MASS Ks band image (Fig. 4.2). The nucleus is also a radio source, visible in a FIRST image (Fig. 4.3). The luminosity of the
36 −1 36 compact nuclear source is νLν = 2.1 10 erg s at 1.49 GHz and νLν = 3.3 10 × × erg s−1 at 4.86 GHz (Vila et al. 1990).
+0.14 NGC 4561 The nucleus is a hard source, with HR = 0.53− . The number − 0.13 of counts detected is 103, which makes spectral fitting possible but with large uncertainties in the best-fit values. The spectrum can be fit with a power law with
Γ = 1.5 0.3 and no intrinsic absorption. The flux F (0.3 8 keV) = 2.5 10−13 ± − × 78 Fig. 4.1.— X-ray hardness ratio map of the nuclear region of NGC 3310. The color of each pixel represents the hardness of the emission detected in that pixel, with harder emission denoted by darker colors. There is an extended region of diffuse soft emission, and several harder sources are visible embedded in it. The nucleus is marked with a circle. The image is 35′′ on a side and the bar in the lower left corresponds to a projected distance of 900 pc. North is up and east to the left. ∼
79 Fig. 4.2.— 2MASS Ks band image of the nucleus of NGC 3310 with the Chandra source circle overplotted. The nucleus is prominent in the 2MASS image. The image is on the same scale as Fig. 4.1.
80 Fig. 4.3.— FIRST images of the nucleus of NGC 3310 with the Chandra source circle overplotted. The nucleus is clearly a radio source. The two extended off-nuclear radio sources coincide with off-nuclear x-ray sources in Fig. 4.1 and are likely to be star forming clumps. The image is on the same scale as Fig. 4.1.
81 erg cm−2 s−1 corresponds to L(0.3 8 keV) 5 1039 erg s−1. This luminosity is − ≃ × consistent with the luminosity reported in Zhang et al. (2009) and Desroches & Ho
(2009). The spectral hardness of the source suggests that the emission is absorbed and the low absorption indicated by the spectral fit is a result of the poor quality of the spectrum because of the low number of counts. The inferred luminosity should be considered a lower limit to the true luminosity. The hardness of the emission and the high luminosity makes the nucleus a candidate AGN.
NGC 4670 The nuclear region of NGC 4670 has a clumpy morphology and is known to be a site of intense star formation (Hunter et al. 1994). The source detection tool wavdetect detects two sources lying on an east-west axis and separated by 2.9′′ (Fig. 4.4). The eastern source (source E hereafter) is the nucleus. The 95% encircled-energy radius at 1.5 keV at the location of the sources is 2′′. Therefore ∼ the overlap of the PSFs of the two sources is small at low energies, but because the PSF becomes larger at higher energies the amount of overlap also increases.
The nominal (that is, using the standard source circle described in 4.2) number § of counts in source E and the western source (source W hereafter) are 44 and 18, respectively. A proper correction for the overlapping PSFs would require knowledge of the spectrum of each source. Since there are not enough counts to determine spectra, we use the nominal counts in our analysis, with the caveat that there is residual uncertainty from the overlapping PSFs. However, because (1) source E is
82 the brighter of the two sources, and (2) it is a harder source (nominal HR = 0.7) − than source W (nominal HR = 0.9) and the PSF is larger at higher energies, its − count extraction region loses more photons to the region belonging to source W than it gains. Therefore the flux of source E is more likely to be underestimated than overestimated, and a complete correction for the overlapping PSFs can be expected to make the argument for an AGN stronger by increasing the inferred luminosity of the source.
The observed count rate of source E corresponds to L(0.3 8 keV) 1039 erg − ∼ s−1 according to PIMMS, correcting only for Galactic absorption, and assuming a power law spectrum with Γ = 2 and a distance of 11 Mpc (Tully 1988). About 70% of the luminosity is in the hard energy band, 2.5–8 keV. Source E, but not source
W, is detected in the hard band image. In SDSS and HST/WFPC2/F336W images of the nucleus the brightest knot coincides with the position of source E to within
Chandra’s angular resolution of 0.5′′. Source E is also the brightest x-ray source in the galaxy. This source is therefore distinctive in having the highest luminosity in x-rays as well as in the optical and UV among the multiplicity of star forming knots in the nuclear region. This makes an AGN explanation plausible. It is possible that the source is instead a super star cluster and the x-rays are from a ULX, but the nucleus is underluminous in radio if the UV is attributed to star formation.
In an analysis of HST/FOC data, Meurer et al. (1995) noted the presence of a
−1 central compact UV source, and estimated a star formation rate of 0.2 M⊙ yr .
83 −1 Following Condon (1992), a star formation rate of 0.2 M⊙ yr corresponds to a 1.4
GHz luminosity density of 8 1027 erg s−1 Hz−1 from the compact source alone. ∼ × However, a FIRST observation detected a luminosity density lower by a factor of six, even though the beam size of 12.5′′ encompassed multiple star forming knots ∼ in addition to the central compact UV source.
4.3.2. Group 2: Candidates with large uncertainties
NGC 1493 The nucleus of NGC 1493 has 49 background-subtracted counts, which corresponds to F (0.3 8 keV) = 3 10−14 erg cm−2 s−1 according to PIMMS, − × assuming a power law spectrum with Γ = 2 and correcting only for Galactic absorption. For a distance of 11.3 Mpc this implies L(0.3 8 keV) 5 1038 erg − ≈ × s−1. Seth et al. (2008) report L(2 10) = 3.5 1038 erg s−1 using XMM data. The − × nuclear x-ray source is the second brightest in galaxy (assuming the brightest source is not a background AGN). All of these facts make the presence of an AGN plausible.
However, NGC 1493 is known to have a nuclear cluster (B¨oker et al. 2002) with a
6 mass of (2 – 3) 10 M⊙ (Walcher et al. 2005). There may be one or more LMXBs × in such a cluster and therefore the detected x-ray source may be the combined x-ray emission from LMXBs in the cluster. No significant Hα emission is associated with nuclear source (Phillips et al. 1996), ruling out a young cluster and HMXBs.
84 Fig. 4.4.— NGC 4670. Clockwise from top left: X-ray broad-band image; X-ray hard- band image; FIRST image; X-ray soft-band image. The green circles and ellipses show sources detected by wavdetect. The cross-hair shows the position of the hard-band source in each image.
85 NGC 2500 The nucleus is detected with only 6 counts. Based on the count rate and assuming a power law spectrum with Γ = 2, PIMMS estimates a flux of
F (0.3 8 keV) 2 10−14 erg cm−2 s−1, implying L(0.3 8 keV) 2 1038 erg s−1. − ∼ × − ∼ × Given the small number of counts these values necessarily have large uncertainties.
Our results are consistent with those of Zhang et al. (2009) and Desroches & Ho
(2009).
NGC 3344 The Chandra detection of the nucleus consists of only 10 counts (but the nucleus is still the brightest x-ray source on the chip). The flux was therefore estimated using the count rate with PIMMS. The estimated flux for a power law spectrum (Γ = 2) is F (0.3 8 keV) 5 10−14 erg cm−2 s−1, which implies − ∼ × L(0.3 8 keV) 2 1038 erg s−1 for our adopted distance of 6.1 Mpc. Though − ∼ × the inferred x-ray luminosity is consistent with an LLAGNs, it is well within the range of luminosities of XRBs. The x-ray data are inconclusive. In UV the nucleus is unresolved in HST/FOC imaging (Maoz et al. 1995). The upper limit put on the size of the source by Maoz et al. (1995) is FWHM < 0.05′′, corresponding to < 1.5 pc for our assumed distance of 6.1 Mpc. An unresolved UV source is consistent with an AGN, but the upper limit on the size is also consistent with a nuclear star cluster.
NGC 4136 NGC 4136 was observed twice by Chandra, three months apart, and detected in both. One of the observations is also presented in Zhang et al.
86 (2009). The first observation had a count rate of = 1.0 ct ks−1 and hardness ratio 0.90 HR 0.35 while the second had a count rate of = 0.62 ct ks−1 − ≤ ≤ − 0.50 HR +0.37 (90% CL). Even the higher count rate corresponds to only − ≤ ≤ 37 −1 5 10 erg s for Galactic NH and a Γ = 2 power law spectrum. The low ∼ × luminosity and the hardening of the spectrum when the source becomes fainter are both consistent with XRB behavior.
NGC 4245 In the very short (1.8 ks) Chandra observation of NGC 4245 only one x-ray source is detected, with 4 counts, in the nuclear region. However, there is a 2′′ ∼ offset between the Chandra position and stated NED position and also the peak of the emission in a 2MASS Ks band image. According to PIMMS the observed count rate corresponds to a 0.3–8 keV flux F 3 10−14 erg cm−2 s−1, assuming a Γ = 2 ∼ × power law and correcting for Galactic absorption. This flux implies a luminosity
(0.3–8 keV) L 3 1038 erg s−1. An HST/STIS observation of the nucleus (Shields ∼ × et al. 2007) detected no optical emission lines. This galaxy has a nuclear star forming ring, at a distance of about 4′′ (190 pc) from the nucleus. The classification of the nucleus as an H II nucleus is based on ground-based spectroscopy (Ho et al.
1995) which used a 2′′ 4′′ aperture. The fact that no emission lines are detected × in the much narrower STIS aperture indicates that the emission lines are probably from the circumnuclear star forming ring, not the nucleus itself. Unfortunately the
Chandra source position is almost exactly in the middle, between the nucleus and
87 the ring. An HST/WFPC2/F606W observation exists and shows the nucleus and ring but coordinates are misaligned by 8′′ with respect to Chandra, 2MASS, and ∼ SDSS images, so cannot be used to independently check for the optical counterpart to the Chandra source. In the WFPC2-PC image the nucleus is not an isolated point source. Instead there is a point source in the center of a structure showing axial symmetry. The luminosity (albeit based on the count rate of the 4 counts detected) is high enough to suggest a ULX or an AGN rather than an ordinary XRB. However, the presence of the circumnuclear star forming ring means that a ULX cannot be ruled out. This, combined with the positional offset of the x-ray source, makes it impossible to conclude that the nucleus has been detected in x-rays.
4.3.3. Group 3: Nuclei with uncertain or no detection
NGC 4204 The upper limit to the 0.3–8 keV luminosity is log(LX ) < 37.7 for our adopted distance of 7.9 Mpc. This galaxy was also part of the samples of Desroches
& Ho (2009) and Zhang et al. (2009). The galaxy is known to have a nuclear star cluster (B¨oker et al. 2002).
NGC 4314 NGC 4314 was observed twice with Chandra (ObsIDs 2062 and 2063).
ObsID 2062 was strongly affected by background flares but after filtering it was possible to extract 10.9 ks of usable exposure time. A nuclear source embedded in extended clumpy emission is detected in ObsID 2062, and multiple sources are
88 detected in ObsID 2063. However, this galaxy is known to have a circumnuclear star-forming ring. The ring and the clumpy emission surrounding the detected point sources make it difficult to determine whether the true nucleus has in fact been detected. Flohic et al. (2006) also analyzed ObsID 2062 and claim there are no nuclear point sources. Dudik et al. (2005) looked at both ObsIDs and say there are multiple point sources of comparable brightness near the nucleus. Zhang et al.
(2009) describe the nuclear morphology as a point source surrounded by diffuse emission. Optically the nucleus is classified as a LINER type 2 in Ho et al. (1997a).
NGC 5068 The nucleus of NGC 5068 was not detected. For a distance of 6.7 Mpc the upper limit to the 0.3–8 keV luminosity is log(LX ) < 37.2. The galaxy is known to have a nuclear star cluster (B¨oker et al. 2002). NGC 5068 was also part of the samples of Desroches & Ho (2009) and Zhang et al. (2009).
IC 5332 No nuclear x-ray source was detected. Among the galaxies in the sample this galaxy has the most stringent upper limit to any nuclear x-ray (0.3–8 keV) luminosity, with log(LX ) < 36.5 (consistent with Desroches & Ho 2009; Zhang et al.
2009). An HST/WFPC2/F606W observation with the PC chip suggests there is a nuclear star cluster. The FWHM of the nuclear source is 0.077′′, or 3 pc using ∼ ∼ our adopted distance of 8.4 Mpc to IC 5332.
89 4.4. Discussion
For the purposes of this discussion we include all 18 galaxies that had pre-existing data in the Chandra archive: the 16 galaxies in this chapter, and
NGC 3184 and NGC 5457, which were presented (as part of a different sample) in
Chapter 2 and in Ghosh et al. (2008). A nuclear x-ray source was detected in 12 of the 16 galaxies presented in this paper, as well as in NGC 3184 and NGC 5457, or in a total of 14 of the 18 galaxies. Of the 14 nuclear x-ray sources, 9 are good candidates to be AGNs, as explained in 4.3.1 and in Ghosh et al. (2008). Proving § that a particular candidate is a bona fide AGN is difficult when the luminosity of the AGN is low, especially when considering just the x-ray emission, because at
39 −1 LX < 10 erg s the x-ray luminosity is in the range produced by x-ray binaries. ∼ 39 −1 Ultra-luminous x-ray sources can even exceed LX 10 erg s . Nevertheless, when ∼ the sample as a whole is considered, it becomes statistically unlikely that all of the detected nuclear x-ray sources are highly luminous XRBs and ULXs which happen to be at the center of the galaxy. In the archival sample presented here the x-ray luminosity of the nuclear sources ranges from 1038 to 1040 erg s−1, and many ∼ ∼ of them are the brightest x-ray source in the galaxy (Tables 4.2 and 4.4). We can conclude that the nuclear x-ray sources are accreting SMBH in many, if not most, of the galaxies presented here.
90 The bolometric correction Lbol/LX applicable to the x-ray band in very low luminosity AGNs has not been firmly established. A recent estimate (Ho 2009) of the bolometric correction was 16. The intrinsic x-ray luminosity itself is uncertain ∼ for most of the sources here because the amount of absorption cannot be estimated owing to the lack of sufficient data for spectral fitting. We assume a conservative value of Lbol/LX = 10. Where an estimate of the absorption is available from a spectral fit we use the inferred luminosity after correcting for intrinsic and Galactic absorption. Where spectral fitting was not possible we use the luminosity inferred after correcting the flux for Galactic absorption alone; the flux itself is estimated from the observed count rate. In this case the estimated LX is a lower limit to the true x-ray luminosity. The estimated x-ray luminosity of these sources is in the
38 40 −1 39 41 −1 range 10 –10 erg s corresponding to Lbol of 10 –10 erg s . Stellar velocity dispersions for 12 of our galaxies are available from Ho et al. (2009) and HyperLeda
(Paturel et al. 2003). We have calculated the mass of the SMBH in these objects using the M-σ relationship (Ferrarese & Ford 2005) (using the Tremaine et al. 2002 relationship results in black hole masses that are different by 0.3 dex, but our ∼ results are not sensitive to the difference) and thus obtained an estimate of the
Eddington ratio (Table 4.5). Objects that have large ( 100%) uncertainties in ∼ either LX or M• have a question mark appended to the Eddington ratio in the last column of Table 4.5. The inferred Eddington ratios span approximately three orders
91 of magnitude, from 10−6 to 10−3 (if the objects with large uncertainty are excluded) and possibly going up to 10−2 (if all objects are considered).
In two objects in our sample (the S0/a galaxy NGC 1291 and the Sbc galaxy
NGC 3310) spectral fitting indicates absorbing column densities of approximately
1022 cm−2. The faint nuclear x-ray source in the nearby Sd galaxy NGC 3621 (not
23 in our sample) is estimated to have an absorbing column density of NH 10 ∼ cm−2 (Gliozzi et al. 2009). In two other objects in our sample, NGC 628 (Sc) and
NGC 4561 (Sdm), the x-ray spectra are consistent with no intrinsic absorption.
Such apparently unobscured spectra are also produced when the obscuration is so high that no direct emission is observable, and the only observed emission is that scattered into the line of sight by reflecting material. For NGC 628 and NGC 4561 the two cases, that of zero absorption and that of total absorption of direct emission, cannot be distinguished with current data. Since the galaxies in the archive do not constitute a well-defined sample we do not draw statistical conclusions about the distribution of absorbing column densities and intrinsic luminosities in these very low-luminosity AGNs. Zhang et al. (2009) find a correlation between the absorption corrected x-ray luminosity and the absorbing column density, such that objects
37 38 −1 with LX 10 –10 erg s have little or no intrinsic absorption and those with ∼ 42 −1 25 −2 LX 10 erg s can have NH as high as 10 cm . Qualitatively this is the ∼ behavior expected if all very low-luminosity AGNs are very heavily obscured, but have a range of luminosities. Then at low LX , no direct emission from the AGN
92 is seen. X-rays from “outside” the AGN, for example from the narrow-line region
(NLR), can still be seen, and perhaps reflected radiation scattered into the line of sight as in NGC 1068. In this case the observed emission will look unabsorbed and the absorption corrected luminosity will also be low. As the value of LX increases, at some point the direct AGN emission will start leaking through, hard emission
first. The observed emission will now contain some hard emission from the AGN and some soft emission from the circumnuclear region. The total spectrum looks like an unabsorbed AGN spectrum but with a higher luminosity. At even higher values of
LX , more and more of the direct emission is seen, but this emission bears the effects of absorption. As the proportion of the direct component rises (in the observed emission), the amount of the missing (absorbed) soft emission that is filled in by the soft NLR (or other circumnuclear) emission falls in proportion, so the total observed emission begins to look more and more absorbed and consequently the absorption corrected luminosity is closer to the true high value of LX . Above some LX (which means above a certain Lbol) the known anti-correlation between L and absorption will once again hold, as the AGN blows away the gas obscuring it. Therefore, though the amount of absorption may be an apparent function of the luminosity it does not imply that there is a real relationship between the quantities.
The issue of absorption is related to the detectability and identification of targets as candidate AGNs, and ultimately to the physical conditions in LLAGNs
— that is, where in the AGN does the emission in any particular waveband arise,
93 and what is the feature that is primarily responsible for the absorption in that waveband? In the simplest picture an unabsorbed LLAGN should be detectable in the optical wavelengths (if host galaxy light can be subtracted). A more absorbed
LLAGN may be undetectable in the optical but still detectable in x-rays, whence surveys like this one and that of Desroches & Ho (2009) and Zhang et al. (2009).
A highly absorbed LLAGN may be undetectable even in x-rays but still detectable in the infrared; a feature exploited by surveys looking for high-ionization IR lines
(Satyapal et al. 2007, 2008; Goulding & Alexander 2009). In reality optical, x-ray, and IR surveys do not result in successively larger, inclusive sets of LLAGNs, and no one wavelength appears sufficient to detect all candidate LLAGNs. An illustrative example is provided by the galaxies NGC 3310 and NGC 3621, as IR and x-ray data are available for both. NGC 3621 was identified as an AGN by the presence of the
IR lines [Ne V] 14µm, 24µm (Satyapal et al. 2008; Goulding & Alexander 2009). The ionization potential of Ne V is 97.1 eV. Since even the hottest stars produce very few photons with energies that high, the presence of Ne V is taken to indicate the presence of an AGN, which can easily produce the required hard ionizing radiation.
NGC 3621 is faint in x-rays, with F (0.5 8keV) 6.5 10−15 erg cm−2 s−1. Based − ∼ × on its IR luminosity Gliozzi et al. (2009) estimate the bolometric luminosity of NGC
41 −1 3621 to be Lbol 10 erg s , and conclude that the faintness in x-rays is due to ∼ high absorption. The x-ray data by themselves are inadequate to determine the absorbing column density; corrected only for Galactic absorption, the inferred x-ray
94 luminosity is only L(0.5 8keV) few 1037 erg s−1. Based on the x-ray data − ∼ × alone it would be hard to argue that NGC 3621 was a bona fide AGN. In contrast to NGC 3621, the [Ne V] lines were not detected in NGC 3310 (Satyapal et al.
2007). Yet, as argued in 4.3.1 it too is almost certain to be an AGN: the nucleus § 40 −1 of NGC 3310 is a luminous (LX 2 10 erg s ), hard x-ray source. In addition ∼ × the nucleus is also a radio source. The best-fit absorbing column density obtained from fitting the x-ray spectrum is an order of magnitude lower than that of NGC
3621. The x-ray luminosity of NGC 3310 suggests that its bolometric luminosity is comparable to that of NGC 3621. The point that is underscored here is that even if the nearby LLAGNs are all intrinsically similar, the observable characteristics can differ so greatly that multi-wavelength data are essential to have a true census of nuclear activity in nearby spiral galaxies.
95 Nucleara Coordinates (J2000) Dist.b Scale Exp.c
Target Type Activity R.A. Decl. (Mpc) (pc arcsec−1) Obs. Date ObsID (ks)
NGC 628 Sc 01 36 41.77 +15 47 00.5 9.7 47 2001 Jun 19 2057 43.2 ··· 2001 Oct 19 2058 46.2
NGC 1073 Sc 02 43 40.51 +01 22 34.0 15.2 74 2004 Feb 9 4686 5.7 ···
96 NGC 1291 S0/a 03 17 18.60 41 06 29.1 8.6 42 2000 Jun 27 795 37.2 ··· − 2000 Nov 7 2059 22.0
NGC 1493 Scd 03 57 27.39 46 12 38.6 11.3 55 2006 Jun 17 7145 9.7 ··· − NGC 2500 Sd Hii 08 01 53.21 +50 44 13.6 10.1 49 2005 Dec 18 7112 2.5
NGC 2681 S0/a L1.9 08 53 32.73 +51 18 49.3 13.3 64 2001 Jan 30 2060 77.1
2001 May 2 2061 76.8
(cont’d) Table 4.1. Targets and observation parameters Table 4.1—Continued
Nucleara Coordinates (J2000) Dist.b Scale Exp.c
Target Type Activity R.A. Decl. (Mpc) (pc arcsec−1) Obs. Date ObsID (ks)
NGC 3310 Sbc Hii 10 38 45.89 +53 30 12.3 18.7 91 2003 Jan 25 2939 39.7
NGC 3344 Sbc Hii 10 43 31.15 +24 55 20.0 6.1 30 2006 Jan 25 7087 1.7
NGC 4136 Sc Hii 12 09 17.71 +29 55 39.4 9.7 47 2002 Mar 7 2920 18.2 97
2002 Jun 8 2921 19.7
NGC 4204d Sdm 12 15 14.36 +20 39 32.4 7.9 38 2006 Apr 2 7092 2.0 ··· NGC 4245 S0/a Hii 12 17 36.77 +29 36 28.8 9.7 47 2006 Apr 5 7107 1.9
NGC 4314d Sa L2 12 22 31.99 +29 53 43.3 9.7 47 2001 Apr 2 2062 10.9
2001 Jun 22 2063 14.5
(cont’d) Table 4.1—Continued
Nucleara Coordinates (J2000) Dist.b Scale Exp.c
Target Type Activity R.A. Decl. (Mpc) (pc arcsec−1) Obs. Date ObsID (ks)
NGC 4561 Sdm 12 36 08.14 +19 19 21.4 12.3 60 2006 Mar 15 7125 3.5 ···
98 NGC 4670 BCD 12 45 17.14 +27 07 31.8 11.0 53 2006 Mar 5 7117 2.6 ··· NGC 5068d Sd 13 18 54.81 21 02 20.8 6.7 32 2006 Mar 17 7149 6.7 ··· − IC 5332d Sd 23 34 27.49 36 06 03.9 8.4 41 2001 May 2 2066 51.8 ··· − 2001 Jul 10 2067 50.2
aFrom Ho et al. (1997a) and NED. bFrom Tully (1988) cUsable exposure time after filtering for background flares.
dNot detected in Chandra observation. Count rate
Targeta Net Counts (ks−1) HR
NGC 628 (2057) 94.9 2.2 0.2 0.52 0.14 ± − ± (2058) 71.7 1.6 0.2 0.70 0.14 ± − ± +0.22 NGC 1073 27.9 4.9 0.9 0.60− ± − 0.24 NGC 1291 (795) 1087.6 29 1 0.22 0.05 ± − ± (2059) 352.9 16 1 0.38 0.08 ± − ± NGC 1493 49.2 5.1 0.7 0.70 0.16 ± − ± +0.43 NGC 2500 5.9 2.4 1.0 0.51− ± − 0.46 NGC 2681b(2060) 953.7 12 0.4 0.84 0.03 ± − ± +0.03 (2061) 868.9 11 0.4 0.78− ± − 0.04
+0.05 NGC 3310 1344.9 34 1 0.29− ± − 0.04
+0.21 NGC 3344 9.8 5.8 1.8 0.83− ± − 0.17
+0.27 NGC 4136 (2920) 18.4 1.0 0.2 0.62− ± − 0.28
+0.43 (2921) 12.3 0.62 0.18 0.06− ± − 0.44
(cont’d) Table 4.2. X-ray Measurements
99 Table 4.2—Continued
Count rate
Targeta Net Counts (ks−1) HR
NGC 4245c 3.9 2.1 1.0 ± ··· +0.14 NGC 4561 102.7 29 3 0.53− ± − 0.13
b +0.16 NGC 4670 44.4 17 3 0.70− ± − 0.17
aObsIDs are given in parentheses when a target has more than one observation. bCounts and HR may be affected by a nearby source. cAmbiguity in source position. Too few counts to obtain meaningful HR.
100 Galactic Spectral Fit Parameters
NH kT NH Γ
Target (1020 cm−2) Modela (keV) (1021 cm−2) χ2(dof)
NGC 628 (Jun) 4.81 ga(pl) 1.6 0.6 7.5(15) ······ ± +0.7 (Oct) ga(pl) 1.7− 3.3(11) ······ 0.6
101 +0.18 NGC 1291 bkg 2.24 ga(me+pl) 0.33− 1.9 0.5 18(28) 0.05 ··· ± +0.9 bkg ga(bb+pl) 0.18 0.02 0.6− 25(28) ± ··· 1.0 high E ga(pl) 1.6 0.6 12(27) ······ ± nucleus ga(me+ab(pl)) 0.33 12 2 1.6 76(91) ± +0.04 +5 nucleus ga(bb+ab(pl)) 0.22−0.03 18−4 1.6 52(90)
+0.3 NGC 2681 2.48 ga(me+pl) 0.75 0.08 1.6− 24(37) ± ··· 0.4
(cont’d) Table 4.3. X-Ray Spectral Fits Table 4.3—Continued
Galactic Spectral Fit Parameters
NH kT NH Γ
Target (1020 cm−2) Modela (keV) (1021 cm−2) χ2(dof) 102 +0.9 +0.6 NGC 3310 bkg 1.12 ga(me+ab(pl)) 0.62 0.04 0.6− 2.4− 70(89) ± 0.6 0.5 +0.69 high E ga(pl) 1.89− 34(48) ······ 0.64 nucleus ga(me+ab(pl)) 0.62 8.7 1.7 1.89 108(147) ± NGC 4561 2.11 ab(pl) 0+1.7 1.5 0.3 4.8(16) ··· ±
aModel labels: ab=xswabs, photoelectric absorption; ga=xswabs with value frozen at Galactic column density towards this target; bb=bbody, blackbody; me=xsmekal, thermal plasma; pl=powlaw1d, one-dimensional power law; log(L) erg s−1
Target X-ray Opt. & UV IR Radio
NGC 628 38.1 41.2 ··· ··· NGC 1073 39.1 <35.1 ······ NGC 1291 39.7 35.7 ······ NGC 1493 38.7 40.5 ··· ··· NGC 2500 38.3? 39.7 40.1? <35.3
NGC 2681 38.5 <36.6 ······ NGC 3310 40.4 36.5 ······ NGC 3344 38.3 40.6 35.1 ··· NGC 4136 37.7 ········· NGC 4245 38.5? <35.6 ······ NGC 4314 <36.2 ········· NGC 4561 39.7 <35.4 ······ NGC 4670 39.0 40.8 36.3 ··· NGC 4204 < 37.7 ········· NGC 5068 < 37.2 ········· IC 5332 < 36.5 ·········
Table 4.4. Luminosities in different bands.
103 σ M•
−1 a 6 b Galaxy (km s ) Ref. (10 M⊙) log (10LX /LEdd)
NGC 628 72 7.8 1 1.2 0.8 5.1 ± ± − +0.012 NGC 1073 24.8 8.7 1 0.007− 1.9? ± 0.007 − NGC 1291 172 20 2 80 48 5.3 ± ± − +0.3 NGC 2500 47.1 22.1 1 0.1− 3.8? ± 0.1 − NGC 2681 109.1 6.7 1 8.7 3.7 5.6 ± ± − NGC 3310 84 1 1 2.4 1.0 3.1 ± ± − NGC 3344 73.6 9.1 1 1.3 1.0 4.9 ± ± − +0.07 NGC 4136 38.4 8.7 1 0.05− 4.1? ± 0.05 − NGC 4245 82.7 8.6 1 2.3 1.5 5.0? ± ± − NGC 3184 43.4 8.9 1 0.1 0.1 4.8? ± ± − +0.01 c NGC 5457 23.6 8.7 1 0.005− 3.3, 2.3? ± 0.005 − −
aStellar velocity dispersion σ from (1) Ho et al. (2009), (2) HyperLeda. bHighly uncertain values are marked with a question mark. cVariable source.
Table 4.5. Estimated Eddington ratios.
104 Chapter 5
The Chandra Survey: II. The Detected Nuclei
5.1. Introduction
This chapter discusses the seven nuclear x-ray sources detected in the Chandra snapshot observations. Data analysis was performed as described in 4.2. The § results of these observations, in conjunction with the results presented in the other chapters, are discussed in Chapter 7.
5.2. The seven nuclear x-ray sources
The seven galaxies in which nuclear x-ray sources were detected in the short
Chandra observations are individually presented below. NGC 7552 is optically classified as a LINER, but the remaining six nuclei show no signs of activity in the optical. In six cases the nuclear source is point-like, while in one (NGC 7552) the source is extended. The point sources are all faint and the number of counts is too small to allow spectral fitting. Fluxes were therefore estimated using the count rates using PIMMS, assuming a power law spectrum with slope Γ = 2 and correcting
105 only for Galactic absorption. Intrinsic absorption is unknown and not corrected for; therefore the inferred luminosities are lower limits. As shown below, despite the low number of counts, five of the seven nuclei are strong candidate AGNs by virtue of their high inferred luminosities.
NGC 1302 NGC 1302 is an Sa galaxy at a distance of 20 Mpc. The broad band count rate of 0.0029 ct s−1 corresponds to a flux F (0.3 8 keV) = 1.9 10−14 erg − × cm−2 s−1, which implies L(0.3 8 keV) = 9.1 1038. This luminosity is sufficient, − × independent of other evidence, to consider the source a good candidate AGN.
NGC 1400 The distance to the S0 galaxy NGC 1400 is given as 5 Mpc
(approximately the Hubble flow distance) in Tully (1988), but subsequent measurements have put the distance at between 20 and 30 Mpc (e.g. Jensen et al.
2003). This galaxy was included in the sample in error and will not be considered when looking at sample properties. Nevertheless, the nucleus is indeed an x-ray point source, with flux F (0.3 8 keV) = 1.9 10−14 erg cm−2 s−1. At a distance of − × 5 Mpc the luminosity inferred would be L(0.3 8 keV) = 6.9 1037 erg s−1, which − × is in the range expected of XRBs. If the galaxy is > 20 Mpc instead, the inferred luminosity is L(0.3 8 keV) > 1039 erg s−1, making an AGN plausible. In addition, − the nucleus is a hard x-ray source, with HR = 0.25. We can conclude that the − nucleus is a good AGN candidate but should not be included in this sample.
106 NGC 1640 This Sb galaxy is at a distance of 19.1 Mpc. The nuclear x-ray point source is a very hard source, with HR = +0.30. As is evident from the hardness ratio and from the observed counts (Table 5.2), the soft emission is substantially absorbed. Therefore the broad band flux was calculated using the hard band count rate. The inferred flux is F (0.3 8 keV) = 1.3 10−13 erg cm−2 s−1. The − × corresponding luminosity L(0.3 8 keV) = 5.7 1039 erg s−1 is high enough that we − × can conclude that this source is almost certainly an AGN.
NGC 3887 NGC 3887 is an Sbc galaxy at a distance of 19.3 Mpc. The flux estimated from the broad band count rate is F (0.3 8 keV) = 9.0 10−15 erg cm−2 − × s−1, which implies a luminosity L(0.3 8 keV) = 4.1 1038 erg s−1. The high value − × of the luminosity makes this source a good candidate to be an AGN.
NGC 4492 NGC 4492 is an Sa galaxy at a distance of 16.8 Mpc. The broad band count rate of 0.0014 ct s−1 implies a flux F (0.3 8 keV) = 9.1 10−15 erg cm−2 − × s−1, which in turn implies a luminosity L(0.3 8 keV) = 3.1 1038 erg s−1. The − × luminosity is high enough for this object to be considered a candidate AGN.
NGC 4900 NGC 4900 is at a distance of 17.3 Mpc. The galaxy is of morphological type Sc, making this the latest type spiral galaxy to have a detected nuclear x-ray source among the targets of the snapshot observations. The observed broad band count rate of 0.002 ct s−1 corresponds to a flux F (0.3 8 keV) = 1.3 10−14 erg − × 107 cm−2 s−1, and therefore a luminosity L(0.3 8 keV) = 4.7 1038 erg s−1. The high − × luminosity makes it a good AGN candidate and the fact that it is in an Sc galaxy makes it a particularly interesting target to follow up.
NGC 7552 This Sab galaxy is at distance of 19.5 Mpc. Optically classified as a LINER, the nucleus shows bright extended soft emission covering the central
7′′. Though there may be point sources, including the nucleus, embedded in the ∼ diffuse emission, none were detected by the source detection tool. NGC 7552 is known to have a high nuclear star formation rate, so it is possible that the LINER classification derives solely from the effects of stellar sources.
108 Morph. Nucleus Coordinates (J2000) Dist.a Scale Exp.b
Target Type Type RA Decl (Mpc) (pc arcsec−1) Obs. Date (ks)
NGC 1302 Sa 03 19 51.21 26 03 38.1 20.0 97 2007 May 27 4.9 ··· − NGC 1400 S0 03 39 30.84 18 41 17.4 5.0c 24c 2007 Jul 11 4.9 ··· − NGC 1640 Sb 04 42 14.5 20 26 05.2 19.1 93 2006 Dec 22 4.8 ··· −
109 NGC 3887 Sbc 11 47 04.57 16 51 16.6 19.3 94 2007 Mar 14 5.1 ··· − NGC 4492 Sa 12 30 59.74 +08 04 40.6 16.8 81 2007 Feb 22 4.9 ··· NGC 4900 Sc 13 00 39.13 +02 30 05.3 17.3 84 2007 Apr 15 5.1 ··· NGC 7552 Sab LINER 23 16 10.77 42 35 05.4 19.5 95 2007 Mar 31 5.1 −
aFrom Tully (1988). bUsable exposure time after filtering for background flares. cMore recent measurements indicate it may be farther. See text.
Table 5.1. The seven galaxies with nuclear x-ray sources detected in the snapshot survey. Net counts
Target Broad Hard Soft HR Fluxa Lum.b
NGC 1302 14.4 1.8 12.6 0.76 1.9 9.1 − NGC 1400 15.7 5.9 9.8 0.25 2.3 0.69c − NGC 1640 18.4 12.0 6.4 +0.30 13.0 57
NGC 3887 6.7 0.8 5.9 0.76 0.9 4.1 − NGC 4492 6.7 0.9 5.8 0.73 0.9 3.1 − NGC 4900 10.0 0.9 9.1 0.83 1.3 4.7 −
a0.3–8 keV flux in units of 10−14 erg cm−2 s−1. b0.3–8 keV luminosity in units of 1038 erg s−1. cLuminosity based on a distance of 5 Mpc.
Table 5.2. Counts and hardness ratios of the six nuclear point sources.
110 Chapter 6
The Chandra Survey: III. The Nondetections
6.1. Introduction
Of the 37 galaxies targeted by the Chandra snapshot observations, 30 did not have a detected nuclear x-ray source. However, because the nuclear positions are known a priori from observations at other wavelengths, we may still examine the x-ray properties at the location of the nucleus. In particular, it is possible to stack the x-ray data such that nuclear positions of the galaxies in the stack coincide. Even if the signal-to-noise ratio (SNR) of the individual sources is too low to confidently distinguish the sources from the background, it is possible that by stacking the data the SNR at the source position is enhanced to the point where a source can be unambiguously detected. The caveat to the use of stacking analysis is that the properties of the resulting stacked “source” are an amalgam of those of the individual sources that constitute the stack. An analysis of the stacked source is only meaningful when there is some confidence that the individual sources are similar to each other, so that it is one set of properties that is being amplified. If the sources are disparate then the properties of the stacked source is unlikely to be
111 representative of any real source. However, the very fact that the sources need to be stacked to be detected means that there is not enough data to be certain that the sources are similar. The 30 nuclei presented here are therefore grouped by the
Hubble type of the host galaxy. An assumption is made that galaxies of the same morphological type provide similar environments to the SMBHs they host, and in turn the emission from the SMBHs, if they are active, is also similar.
6.2. Stacking analysis
Data were analyzed using CIAO 3.4 (Fruscione et al. 2006) and CALDB 3.4.0.
Level 1 event lists were re-processed to create new level 2 event lists to ensure that the latest calibration and observation-specific bad pixel files were used. The event lists were filtered to include only photons with energies between 0.3 and 8 keV. FITS images were produced from the filtered event lists and the CIAO tool wavdetect was run on the images at the default significance threshold of 10−6, implying that wavdetect should produce 1 false detection per 1024 1024 chip. In none of the ∼ × observations presented here did wavdetect detect a source at the location of the galactic nucleus.
The location of the galactic nucleus in each Chandra image was determined by examining Digitized Sky Survey (DSS) and 2MASS images of the same field, and SDSS and HST images in addition if they were available. In most cases the
112 nucleus was clearly visible in at least one of the optical or infrared images, and the location of the nucleus was marked on the x-ray image. If there were counterparts to detected x-ray sources on one or more of the comparison images then these were used to verify that the coordinate systems of the x-ray images matched those of the comparison images. It did not prove necessary to transform the coordinates of any of the x-ray images. In UGC 5460 and NGC 4299 the nuclear region is clumpy
(in the optical or IR bands) with multiple sources of similar brightness and it was not possible to determine with certainty which source was the nucleus. These two objects were not included in the image stacks.
A square region centered on the nucleus and 100 pixels ( 50′′) on a side was ∼ extracted from each x-ray image. These extracted images were combined by Hubble type into groups for the stacking analysis, as described in the sections below and in Table 6.2. Source counts were extracted from a circle centered on the center of the image and with a radius equal to 4.67 pixels (2.3′′). This radius was chosen to include the larger Chandra PSF at high energy. Background counts were taken from an annulus with an inner radius of 2 times and an outer radius of 5 times the source circle radius. Estimates of fluxes were made using PIMMS using the
20 −2 broad band count rate, assuming Galactic NH = 10 cm and a power law source spectrum with index Γ = 2. No correction was made for intrinsic absorption, which cannot be determined for these sources. A total luminosity for each stack was calculated from the total flux by using the exposure time weighted mean distance
113 of the galaxies in the stack. Counts, fluxes, and luminosities are in the 0.3–8 keV band. All measurements of counts were assumed to have a 1σ error of √n where n is the number of counts. This uncertainty was propagated to calculations of fluxes and luminosities and is the uncertainty quoted in these quantities. A source was taken to be real if the Poisson probability of obtaining the total number of counts
−3 Ct in the source circle solely from background fluctuations was less than 10 . The
CIAO tool wavdetect was run on each 100 100 pixel stacked image, with the × sigthresh parameter now set to 3 10−5, and confirmed the detections. In the case × of a nondetection (after stacking), if Ct > 0 we take 3√Ct as the upper limit to the number of counts produced by the source. If Ct = 0, the upper limit to the source count rate is taken to be 1/texp where texp is the observation exposure time.
6.3. Results
The stacked image of the six early type (E–Sab) galaxies is shown in Fig. 6.1.
The combined image has an effective exposure time of 27.1 ks. A nuclear source is clearly detected. The source circle contains 12 counts, while the expected background is 2.3 counts. The Poisson probability of obtaining 12 counts when the mean is 2.3 is 4.4 10−6. Thus the source has 10 counts, and the signal-to-noise ∼ × ∼ ratio (SNR) of the detection is 2.8. The count rate of (3.7 1.2) 10−4 ct ∼ ± × s−1 corresponds to an observed flux, not corrected for Galactic absorption, of
114 F (0.3 8 keV) = (2.2 0.7) 10−15 erg cm−2 s−1, assuming a power law spectrum − ± × with photon index Γ = 2. For an absorbed source with intrinsic column density
21 −2 37 −1 NH = 10 cm , this flux corresponds to L(0.3 8 keV) = (8.7 2.8) 10 erg s . − ± × 22 −2 For intrinsic column density of NH = 10 cm the required luminosity increases to
38 −1 Lavg(0.3 8 keV) = (2.3 0.7) 10 erg s . − ± ×
The stack of the six Sb and Sbc galaxies has a combined exposure time of 29.6 ks. The source circle contains 18 counts whereas the expected background is 2.7 counts. The Poisson probability of obtaining 18 background counts is 5 10−10. ∼ × The nucleus is detected with 15 net counts, at an SNR of 3.6. The mean observed
−15 −2 −1 21 flux is Favg = (3.2 0.8) 10 erg cm s . For intrinsic absorption NH = 10 ± × −2 22 −2 38 cm and NH = 10 cm the implied luminosities are Lavg = (1.6 0.4) 10 ± × erg s−1 and (4.4 1.1) 1038 erg s−1, respectively. The stacked image is shown in ± × Fig. 6.2.
The late-type galaxies are not detected even after stacking. This is true both of individual stacks of the 9 Sc/Scd galaxies and 7 Sd/Sdm galaxies separately, as well as of a stack combining all 16 galaxies (Fig. 6.3). The latter stack has a combined exposure time of 75.3 ks. Analogous to the previous cases, a source circle centered on the center of the stacked image, and a background annulus are defined. The expected number of background counts in the source circle is 6.1 whereas 4 counts are detected. The upper limits to the fluxes and luminosities are given in Table 6.2.
115 Fig. 6.1.— Stacked x-ray image of 6 galaxies ranging from types E to Sab. The image represents a combined exposure time of 27 ks. The image has been smoothed with a Gaussian of radius 3 pixels. A nuclear source is clearly present. The image is 100 pixels (49.2′′) on a side. The apparent source near the left edge is a background fluctuation.
116 Fig. 6.2.— Stacked x-ray image of 6 galaxies ranging of types Sb and Sbc. The image represents a combined exposure time of 29.6 ks. The image has been smoothed with a Gaussian of radius 3 pixels. A nuclear source is clearly present. The source to the north of the nucleus is a real off-nuclear source in one of the galaxies, NGC 4037. The image is 100 pixels (49.2′′) on a side.
117 Fig. 6.3.— Stacked x-ray image of 16 galaxies of types Sc through Sdm. The nucleus is not detected in a combined exposure time of 75 ks. The image is 100 pixels (49.2′′) on a side. North is up and east to the left.
118 To look for differences in circumnuclear x-ray emission in the different stacks, we used the annular regions already defined for nuclear background subtraction.
The annuli have an inner radius of 4.6′′ and an outer radius of 11.5′′. For each stack, count rates in the soft and hard bands were normalized by the projected area at the mean distance of the galaxies in the stack. These count rates are presented in
Table 6.3. For comparison, we also show the approximate background count rate in the ACIS-S3 chip as given in Table 6.9 of the Chandra Proposers’ Observatory
Guide, normalized the same way as the observed count rates. For comparison with our soft band (0.3–2.5 keV), we used the listed background for 0.5–2 keV, and for our hard band (2.5–8 keV) we used the 2–7 keV background (calculated as the difference between the given backgrounds for 0.5–2 keV and 0.5–7 keV). As seen from
Table 6.3, in all stacks the hard band count rate is consistent with the background count rate, and the soft band shows a significant excess over background. The soft band count rates in the E-Sab, Sb-Sbc, and Sc-Scd stacks are consistent with each other within 3σ. However, the Sd-Sdm galaxies have a lower count rate and the difference between the lowest (Sd-Sdm) and highest (E-Sab) count rates is 5σ. ∼ Given the small number of galaxies in each stack it is possible that the difference in count rates is a statistical fluctuation instead of representing a real property of
Sd–Sdm galaxies.
119 6.4. Discussion
With fewer than 20 counts in the stacked images, it is impossible to identify the nature of the sources unambiguously. The estimated average luminosities of the sources range from 1038 to 4 1038 erg s−1 (assuming the column density ∼ ∼ × 22 −2 responsible for intrinsic absorption is 0 NH 10 cm ). The luminosities ≤ ≤ are thus consistent with the sources being x-ray binaries (XRBs), AGNs, or circumnuclear ionized gas, though if they are XRBs then all of them are drawn from the bright end of the XRB luminosity function. The soft band fluxes and the upper limit to the hard band fluxes show that the spectra are consistent with an unabsorbed power law with photon index Γ = 2 (see Table 6.4), and also with a thermal bremsstrahlung model, which has lower hard band emission than a power law. Finally, the angular resolution of Chandra confines the position and extent of the source to the central 100 pc of the galaxy, which is consistent with all ∼ three types of sources mentioned above. At this point, therefore, the data cannot distinguish between AGNs, XRBs, and circumnuclear ionized gas. However, excess emission from ionized gas very close to the nucleus is itself suggestive of an AGN, so the only viable non-AGN-related source is an XRB. For the source detected after stacking to be composed entirely of XRBs it would require the galaxies to have an excess of XRBs at the nucleus compared to the circumnuclear region used to estimate the background. Such an excess is possible if there happens to be a compact star cluster at the nucleus but less likely if the XRBs simply follow the
120 underlying stellar distribution, as low-mass XRBs (LMXBs) can be expected to do. We check for consistency of the data with two extreme cases: first, that the only sources of nuclear x-rays are XRBs in nuclear clusters, and galaxies without nuclear clusters do not contribute any x-ray emission to the stacks; second, that the only source of x-rays is accretion onto an SMBH, with zero contribution from
XRBs. Surveys have shown that nuclear star clusters are present in approximately
60% of Sa–Sbc galaxies (Carollo et al. 1997, 1998, 2002) and in 80% of Scd–Sm ∼ galaxies (B¨oker et al. 2002, 2004). The four galaxies in this chapter that were part of the above mentioned surveys (NGC 1325A, NGC 2082, NGC 4411B, NGC 4299) are known to have nuclear clusters, but none has a nuclear x-ray source detected individually. Rossa et al. (2006) find that the average mass of a nuclear cluster in an early type spiral is a factor of about 20 higher than the average mass of a nuclear cluster in a late type spiral. If the x-ray emission is from LMXBs then the x-ray luminosities of the clusters in early type spirals should be higher by a similar factor than that of clusters in late type spirals, as the number of LMXBs should trace the total stellar mass. If we compare a stack of the 10 S0–Sbc galaxies with the stack
38 of 16 Sc–Sdm galaxies (Table 6.2), we see that the former have LX 1.5 10 erg ≈ × −1 37 −1 s while the latter have LX < 1.9 10 erg s . Thus the early types are at least a × factor of 8 brighter (factor of 11 if the respective occurrences of clusters, 60% and
80%, are taken into account), which is consistent with the nuclear cluster LMXB hypothesis. However, two points should be noted that complicate a comparison of
121 the expected x-ray emission from clusters in different types of galaxies. First, even though the average masses of clusters, binned by the Hubble type of the host galaxy, differ by a factor of 20, the individual cluster masses as calculated in Rossa et al.
(2006) show a wide range, from log(MNC /M⊙) = 5.43 to log(MNC /M⊙) = 8.63, or a factor of 1600. Second, this comparison is necessarily affected by small number ∼ statistics. For example, Gilfanov (2004) finds that the number of LMXBs with
37 −1 11 luminosity greater than 10 erg s is of the order of 140 per 10 M⊙ in stellar
6 9 mass. Accordingly, for cluster masses in the range of 10 –10 M⊙ we may expect
0.01–1 LMXBs. The actual number will be higher because the high stellar density in clusters is conducive to the formation of binaries. Nevertheless, it is likely that the number of LMXBs in any cluster that contains LMXBs is of order unity. Thus, no matter how small the probability, the occurrence of an LMXB of the required luminosity cannot be ruled out in any particular nucleus. On the other hand, looking at the Sb–Sbc stack in detail, we find that of the six galaxies two (NGC 1325A and NGC 4689) have zero counts within the source circle, and the remaining four if considered by themselves have an average luminosity of L(0.3 8 keV) 3 1038 − ≈ × −1 38 −1 erg s . LMXBs with LX 3 10 erg s are not common. For example, in the ∼ × survey of globular clusters by Sivakoff et al. (2007) 1%–2% of the clusters had ∼ 38 −1 LMXBs brighter than LX = 3.2 10 erg s . Thus, LMXBs by themselves may × not be a sufficient explanation for the nuclear x-ray emission in all cases, unless a nuclear cluster preferentially forms high luminosity XRBs.
122 For the Sc–Sdm galaxies, Table 6.3 shows that the cause of the nondetection was not a higher background level. It is interesting that the circumnuclear emission
( 0.8–1.9 kpc) appears to be the same, within the uncertainties, in all morphological ∼ types in this sample, with the possible exception of the Sd-Sdm galaxies. The large scale star formation rate (SFR) increases towards later type galaxies, but this trend is not seen in the circumnuclear x-ray emission. This may be a statistical effect because of the small sample size. If the data are taken at face value, there are several possible interpretations of the observed circumnuclear emission: (1) The x-ray emission is predominantly from unresolved HMXBs. This would imply that the SFR close to the nucleus may not be linked to the large scale SFR in the galaxy. (2) The emission is instead dominated by unresolved LMXBs. This would indicate that the underlying old stellar population in the nuclei of the S0–Scd galaxies in this sample is similar despite differences in Hubble type. The Sd galaxies may have a lower mass in the old population, as is to be expected. (3) The emission is dominated by circumnuclear gas that has been ionized by a central AGN. This of course requires that there be an AGN, even in the late type galaxies where no nuclear point source was detected after stacking. However, the relevant physical parameters, for example the AGN luminosity, the amount of obscuration, the shape of the ionizing spectrum, and the gas density are unknown or not well constrained.
We consider next the possibility that the nuclear x-ray sources are all accreting
SMBHs. In the case of the Sc-Scd galaxies a useful comparison can be made with
123 NGC 3184 and NGC 5457. These galaxies fulfill the same selection criteria that were used to select the sample presented in this chapter, but both have detected nuclear x-ray sources. Both galaxies are of type Scd, and both are cases where strong arguments can be made that they host AGNs (Ghosh et al. 2008). Both of these galaxies are 8 Mpc away. In the Chandra observations that were taken of ∼ these nuclei, NGC 3184 had a 0.3–8 keV count rate of 0.7 ct ks−1, while NGC 5457 ∼ had a count rate that varied between 0.8 and 5 ct ks−1. At a distance of 16 Mpc, representative of the distance to most of the galaxies in this chapter, the count rate would be 0.2 ct ks−1, and so a 5 ks observation would obtain 1 count. NGC 3184 ∼ ∼ and NGC 5457 (in the low state) would therefore be not detected individually at the distance of the galaxies in this chapter. A stack of 9 galaxies identical to NGC 3184, with a total exposure time of 45 ks (the total exposure time of the Sc–Scd stack), can be expected to contain 9 3 counts. The actual number of counts observed in ± the Sc-Scd stack source circle was 3, which would be a 2σ deviation. Thus, the data at present do not rule out the presence of AGNs, and therefore of SMBHs, in the late-type galaxies. The presence of SMBHs in the galaxies of type earlier than Sc is expected given their large bulge components.
The fact that soft band photons are detected but hard band photons are not puts an upper limit to the amount of intrinsic absorption that the observed emission has suffered if the input spectrum is assumed to be a power law. For a Γ = 2 power
20 −2 law with fixed Galactic NH = 10 cm , PIMMS predictions show that intrinsic
124 22 −2 NH cannot be greater than 10 cm for both the E-Sab and Sb-Sbc stacks. ∼ Thus, the x-ray emission can be (1) direct, unabsorbed emission from very low luminosity AGNs, or (2) scattered emission from completely absorbed AGNs whose true luminosities are much higher, or (3) entirely from circumnuclear ionized gas, where neither direct nor scattered emission from the AGN reaches us. If the entire observed emission is due to accretion, then the inferred rates of accretion are very
−8 −1 low, with M˙ 2 10 L38 (0.1/η) M⊙ yr , where L38 is the bolometric luminosity ≃ × 2 38 −1 Lbol = ηMc˙ in units of 10 erg s . Assuming a bolometric correction of 10 for
−5 x-ray emission, the bolometric luminosity of one of these nuclei is 10 LEdd for ∼ 6 a 10 M⊙ SMBH. The luminosity upper limits in Table 6.2 for the Sc-Sdm galaxies
−6 6 correspond to an upper limit of 10 LEdd for a 10 M⊙ SMBH. From Table 6.2, it ∼ can be seen that in our sample the average luminosity of a nucleus in the Sc–Sdm stack is a factor of 10 times lower than the average luminosity of a nucleus in the ∼ stacks consisting of galaxies of type Sbc and earlier. In the Palomar spectroscopic survey of the nuclei of nearby galaxies (Ho et al. 1995), the fraction of nuclei in which
AGNs were detected dropped from 50%–70% for galaxies Sbc and earlier to < 20% ∼ for types Sc and later (Ho 2008). These observations suggest the possibility of a fundamental difference between the nuclei of spiral galaxies earlier and later than type Sc. Whether the difference can be attributed entirely to higher obscuration (the nuclei are gas rich), or whether there are systematic differences in black hole masses,
125 accretion modes, or other aspects of the nuclear environment remains an interesting problem to be solved.
If the current accretion rate is as low as calculated above, then it raises the question of identifying the epoch in which these black holes gained most of their
5 mass. A 10 M⊙ SMBH radiating at even 1% of the Eddington limit has a bolometric luminosity of 1041 erg s−1. If the downsizing hypothesis holds to the lowest ∼ AGN luminosities, therefore, there should exist at very low redshift a population of late-type spiral galaxies with nuclei whose luminosities approach Seyfert levels at the high end and ULX levels at the low end. If, as proposed for their more luminous counterparts, these AGNs are heavily enshrouded for part of their life
(Hopkins et al. 2005), then they may not be found in an x-ray or optical survey, but should still be detectable via their re-processed radiation in the infrared. The ratio of direct to indirect detections would indicate the fractions of the lifetimes spent in the unobscured versus the enshrouded states. On the other hand, if there is no such population, it may mean that the mode of accretion in quasars (and Seyferts) and that in very low luminosity AGNs are different; in the latter the accretion rate instead of peaking at any particular epoch may be low but steady for the entire
Hubble time.
As mentioned above, four of the galaxies in this chapter are already known to have nuclear star clusters, though it is not known whether these particular clusters have XRBs. One galaxy, NGC 4411B, has been confirmed as an AGN (Seth et al.
126 2008) since its inclusion in our sample. Thus, the true picture undoubtedly involves both XRBs and AGNs. In fact, Seth et al. (2008), who explicitly studied the coincidence of nuclear clusters and AGNs, found that galaxies with more massive nuclear clusters are also more likely to have AGNs. As is always the case with extremely low luminosity AGNs, confirming that a particular source is an AGN requires multiwavelength data. Radio observations are the most helpful, as the detection of a compact source with high brightness temperature unambiguously identifies a source as an AGN.
127 Morph. Nucleus Coordinates (J2000) Dist.b Scale Exp.c
Target Type Typea RA Decl (Mpc) (pc arcsec−1) Obs. Date ObsID (ks)
NGC 4308 E 12 21 56.9 +30 04 27 9.7 47 2008 Feb 27 7853 4.7 ··· NGC 7077 BCD/E H II 21 29 59.6 +02 24 51 13.3 64 2007 Aug 23 7854 5.1
NGC 3073 S0 H II 10 00 52.1 +55 37 08 19.3 94 2006 Dec 27 7851 4.7
128 NGC 1315 S0 03 23 06.6 21 22 31 17.1 83 2007 Mar 13 7850 3.0 ··· − NGC 4665 S0/a 12 45 06.0 +03 03 21 17.9 87 2007 Apr 15 7852 4.7 ··· NGC 1341 Sab H II 03 27 58.4 37 09 00 16.9 82 2007 Mar 14 7846 4.9 − NGC 4037 Sb 12 01 23.7 +13 24 04 17.2 84 2007 Apr 13 7842 4.7 ··· NGC 4394 Sb LINER 12 25 55.6 +18 12 50 16.8 81 2007 Nov 11 7864 5.1
NGC 4689 Sbc H II 12 47 45.6 +13 45 46 16.8 81 2007 May 07 7865 4.9
(cont’d) Table 6.1. Targets and observation parameters Table 6.1—Continued
Morph. Nucleus Coordinates (J2000) Dist.b Scale Exp.c
Target Type Typea RA Decl (Mpc) (pc arcsec−1) Obs. Date ObsID (ks)
NGC 255 Sbc 00 47 47.3 11 28 07 20.0 97 2007 May 27 7844 4.7 ··· − NGC 1325A Sbc 03 24 48.5 21 20 10 17.1 83 2007 Jun 16 7841 5.1 ··· −
129 IC 5325 Sbc 23 28 43.4 41 20 01 18.1 88 2007 Jul 25 7843 5.1 ··· − NGC 991 Sc 02 35 32.7 07 09 16 18.8 91 2007 Jun 18 7861 5.0 ··· − NGC 1703 Sc H II 04 52 52.2 59 44 32 17.4 84 2007 Sep 16 7839 4.9 − NGC 2082 Sc 05 41 51.1 64 18 04 12.1 59 2007 Jun 13 7838 5.1 ··· − NGC 3938 Sc H II 11 52 49.5 +44 07 15 17.0 82 2007 Oct 15 7862 4.9
NGC 4254 Sc H II 12 18 49.6 +14 24 59 16.8 81 2007 Nov 21 7863 5.1
(cont’d) Table 6.1—Continued
Morph. Nucleus Coordinates (J2000) Dist.b Scale Exp.c
Target Type Typea RA Decl (Mpc) (pc arcsec−1) Obs. Date ObsID (ks)
NGC 4411 Sc 12 26 30.1 +08 52 20 16.8 81 2007 Nov 13 7837 5.1 ··· NGC 1536 Sc pec H II 04 10 59.9 56 28 50 13.4 65 2007 May 23 7836 5.0 − d
130 NGC 4411B Scd AGN 12 26 47.2 +08 53 05 16.8 81 2007 Feb 13 7840 4.9
NGC 4688 Scd H II 12 47 46.5 +04 20 10 17.1 83 2007 Feb 19 7859 4.7
NGC 685 Sd 01 47 42.8 52 45 43 15.2 74 2007 Mar 12 7857 2.0 ··· − UGC 5460e Sd 10 08 09.3 +51 50 38 19.9 96 2007 Mar 10 7835 5.1 ··· NGC 3913 Sd H II 11 50 38.9 +55 21 14 17.0 82 2007 Jul 12 7856 4.6
UGC 6930 Sd 11 57 17.4 +49 16 59 17.0 82 2007 Sep 24 7855 4.5 ···
(cont’d) Table 6.1—Continued
Morph. Nucleus Coordinates (J2000) Dist.b Scale Exp.c
Target Type Typea RA Decl (Mpc) (pc arcsec−1) Obs. Date ObsID (ks)
NGC 4393 Sd 12 25 51.2 +27 33 42 9.7 47 2008 Apr 01 7833 4.7 ··· NGC 4571 Sd 12 36 56.4 +14 13 03 16.8 81 2008 Feb 14 7858 4.7 ··· 131 NGC 3299 Sdm 10 36 23.8 +12 42 27 5.4 26 2007 Feb 10 7831 5.1 ··· NGC 4299e Sdm 12 21 40.9 +11 30 12 16.8 81 2007 Nov 18 7834 5.1 ··· UGC 9912 Sdm 15 35 10.5 +16 32 58 19.7 96 2007 Apr 26 7832 5.1 ···
aClassification from Ho et al. (1997a) for NGC 3073, NGC 4665, NGC 4394, NGC 4689, NGC 3938, NGC 4254, and NGC 4688, from Seth et al. (2008) for NGC 4411B and NGC 3913. All others from NED. bFrom Tully (1988). cUsable exposure time after filtering for background flares. dIdentification occurred after sample selection. See 2.2. § eNot included in stacking analysis. Position of nucleus uncertain. a c Included No. of Net Exposure Mean Dist. Obs. LX
Morph. Types Galaxies Detection Counts SNR (ks) (Mpc) Fluxb (1038 erg s−1)
E–Sab 6 Y 10 2.8 27.1 15.6 2.2 0.7 2.3 0.7 ± ± Sb–Sbc 6 Y 15 3.6 29.6 17.6 3.2 0.8 4.4 1.1 ± ± Sc–Scd 9 N 44.6 16.2 < 0.74 < 0.86 132 ······ S0–Sbc 10 Y 22 4.3 46.9 17.7 2.9 0.6 4.2 0.9 ± ± Sc–Sdm 16 N 75.3 15.4 < 0.49 < 0.50 ······
aExposure time weighted mean. bObserved flux in the 0.3–8 keV band in units of 10−15 erg cm−2 s−1 assuming a power law spectrum with photon index Γ = 2. cThe 0.3–8 keV luminosity required to produce the observed flux, for a source at the mean distance and assuming 20 −2 22 −2 Galactic NH = 10 cm and intrinsic NH = 10 cm .
Table 6.2. Stacked observations Total Mean Norm. obs. count rate Norm. ACIS-S3 bkg. rate
No. of Exp. Scale (ct ks−1kpc−2) (ct ks−1kpc−2)
Stack gal. (ks) (pc arcsec−1) Soft Hard Soft Hard 133 E-Sab 6 27.1 76 0.70 0.11 0.19 0.06 0.08 0.28 ± ± Sb-Sbc 6 29.6 85 0.47 0.08 0.28 0.06 0.06 0.22 ± ± Sc-Scd 9 44.6 79 0.68 0.08 0.38 0.06 0.07 0.26 ± ± Sd-Sdm 7 30.6 69 0.29 0.08 0.25 0.07 0.09 0.33 ± ±
Table 6.3. Circumnuclear emission. Soft band Observed Γ = 2 predicted Hard band Observed
Stack count rate (ks−1) softflux hardflux countrate(ks−1) hardflux
E-Sab 0.34 2 10−15 9 10−16 < 0.11 < 2 10−15 × × ×
134 Sb-Sbc 0.45 2 10−15 1 10−15 < 0.18 < 3 10−15 × × × Sc-Sdm < 7.9 10−2 < 4 10−16 < 2 10−16 < 1.3 10−2 < 3 10−16 × × × × ×
Note. — All fluxes are in units of erg cm−2 s−1 and refer to the flux observed from the stacked 20 −2 nuclei assuming a power law spectrum with Γ = 2, Galactic NH = 10 cm , and no intrinsic absorption. Soft band is 0.3–2.5 keV, hard band 2.5–8 keV.
Table 6.4. Soft and hard band fluxes of the nuclei. Chapter 7
Conclusions and Future Directions
We first note that the galaxies presented here show nuclear x-ray sources are a very common occurrence. In a survey of late-type spiral galaxies such as our
Chandra survey, therefore, the predominant concern is not detection efficiency, but rather identification of the AGNs among the detected sources. We discuss below the issues such surveys will face when attempting to identify the nature of the detected sources.
The diagnostic tools traditionally used to distinguish AGNs from non-AGNs
(e.g. optical line ratios, Baldwin et al. 1981; Veilleux & Osterbrock 1987) were developed in the course of studying luminous AGN. Dilution of the AGN emission by host galaxy light was not a serious problem and observations with low spatial resolution (several arcseconds) sufficed. In the study of AGN that are either intrinsically less luminous or are heavily obscured, however, host galaxy light becomes increasingly problematic, and surveys relying on optical spectra (e.g. Ho et al. 1995; Greene & Ho 2004) require careful subtraction of the starlight using galactic spectral templates. In the weakest AGNs, however, signs of AGN emission
135 may not be detected by the usual diagnostics. This problem will persist until optical observations with angular resolution of 1–10 mas become possible so that host galaxy light can be effectively excluded, though it can be mitigated by using regions of the spectrum where host galaxy emission is negligible, for example very high energy x-rays (tens to hundreds of keV). The detection of a compact radio source unresolved at milliarcsecond-scales, especially if the source is accompanied by jets, would also unambiguously identify the nucleus as an AGN. This has been the motivation for radio surveys like that of Nagar et al. (2002). Some AGNs obscured in the optical and UV may be detectable using infrared emission line strengths and ratios (e.g.
Dale et al. 2006; Satyapal et al. 2007, 2008).
While the methods listed above allow the unambiguous identification of AGNs, observations often do not have the angular resolution or sensitivity to distinguish
AGN and non-AGN flux. Other sources of radiation in the vicinity of the AGN are, for example, plasma photoionized by the AGN itself, or a nuclear star cluster.
The targeted AGNs have very low luminosity and thus even moderate amounts of obscuration may cause a significant decrement in the observed flux. In most cases, therefore, the AGN contribution should not be expected to dominate the total observed flux. Consequently, identifying these AGNs requires a different approach than what can be used in the case of the more luminous AGNs (Seyferts and QSOs). AGNs can be identified using x-ray observations (e.g. with Chandra and XMM-Newton) solely, but only if they are point sources whose inferred
136 luminosities are greater than 1041 erg s−1. Below that value, AGNs can become ∼ indistinguishable from ULXs and XRBs in x-rays. ULXs can have luminosities of a few times 1040 erg s−1 (e.g. Soria et al. 2007) and have x-ray spectra that look similar to AGN spectra. It has been suggested (Stobbart et al. 2006) that ULX spectra show a break at 5 keV. AGNs are not known to show this break. In high quality ∼ spectra with a large number of counts it may be possible to exploit this difference to separate ULXs and AGNs. XRBs have power law spectra with Γ 2, can show ∼ an Fe Kα emission line, and emit hard x-rays, all characteristics of AGNs as well.
Additionally, even with Chandra’s angular resolution, the physical space probed ranges from the central 10 to 100 pc of the galaxy. The existence of one or more ∼ XRBs within that region would be unsurprising. However, the fact that the inferred luminosities are as high as 1038 erg s−1 severely constrains the expected number of
XRBs. For example, for NGC 5457, Pence et al. (2001) provide a log N-log S relation as well as the surface density of x-ray point sources as a function of radius (their
Figs. 3 and 4). Approximately 12.5% of the sources have luminosities exceeding 1037 erg s−1. The surface density of sources in the innermost 0.5′ is 4.75 arcmin−2. ∼ Therefore we may expect 0.6 sources arcmin−2 above the luminosity cut-off in the ∼ central 0.5′, or 4 10−3 such sources within the Chandra source circle of radius ∼ × 2.3′′ that has been used here. Therefore, invoking XRBs and ULXs alone to explain the x-ray observations leads to physically implausible conditions, such as requiring that most, or all, quiescent, non-starburst spiral galaxies have a ULX or 100 XRBs ∼
137 in the central 0.5′′ 2′′. Nevertheless, x-ray observations by themselves will usually − be inadequate to distinguish AGNs from non-AGNs in any particular instance.
Information from other wavebands is thus crucial for the identification process.
But here too, traditional methods of identifying AGNs using flux ratios such as αOX ,
αKX , and fX /fR, must be used with caution. The low luminosities of the AGNs mean that the emission in the two bands being compared may not be from the same object. For instance, for an obscured AGN surrounded by a nuclear star cluster, the observed x-ray flux may be from the AGN but the optical flux may be dominated by the cluster. The existence of an AGN must instead be inferred by consistency and plausibility checks considering as much of the spectral energy distribution as possible, and the goal is the rejection of the hypothesis that all of the observed properties can be explained without requiring the presence of an AGN. As noted before, NGC 3184 is a good example where data in two wavebands, x-ray and IR, together provide strong arguments for the presence of an AGN even though each observation individually is inconclusive.
Nuclear star clusters are fairly common in spiral galaxies (e.g. B¨oker et al. 2002;
Walcher et al. 2005; Seth et al. 2008). A cluster poses two main problems. First, it makes the presence of XRBs more likely. Second, if the cluster is young and contains many O and B stars, it may overwhelm AGN emission in the UV in addition to the optical (e.g. NGC 4303, Colina et al. 2002). It may be possible in some cases, as for
NGC 1042 (Shields et al. 2008) and NGC 4102 (Gon¸calves et al. 1999), to attempt a
138 separation of the cluster and AGN components in the optical spectrum. In addition, stellar spectra, even for late-type stars, fall faster towards the infrared than the power-laws typical of AGNs. The mid-infrared colors of AGNs, therefore, tend to be redder than those of stellar populations and this color difference can be used to infer the presence of an AGN (Stern et al. 2005).
In addition to the spectral energy distribution, source variability can be a discriminant, as AGNs are known to vary in all wavelengths, whereas a star cluster, say, would not. Conversely, if the variation is periodic it would rule out an AGN and argue for an XRB instead. In addition, XRBs often display quasi-periodical oscillations. We note here that if the claimed QPO in RE J1034+396 (Gierli´nski et al. 2008) is real, then the mere presence of a QPO will no longer be a discriminator, but the characteristic frequencies will still be different for XRBs and AGNs.
Despite the uncertainties described above, this survey has resulted in a number of strong AGN candidates. The archival sample had nonuniform exposure times, so it is necessary to consider how many of the archival targets would have been detected in the snapshot survey. The limiting count rate for detection in the 5 ks snapshot survey was 1 ct ks−1 in the 0.3–8 keV band (this corresponds to a flux 6 10−15 × erg cm−2 s−1, or to a luminosity of 3 1038 erg s−1 for a galaxy at the survey limit × of 20 Mpc). Of the 14 archival galaxies that were detected, all except 3 have count rates that exceed this limit. The three galaxies that would not have been detected in the snapshot survey are (1) NGC 5457 (Scd) in its low state, when the count rate
139 was 0.5 ct ks−1, (2) NGC 3184 (Scd), which had count rates of 0.61 and 0.73 ct ∼ ks−1 in the two Chandra observations, (3) NGC 4136 (Sc), which had a count rate of 0.62 ct ks−1 in one observation (the count rate was the threshold value 1 ct ks−1 in the second observation). A fourth galaxy, NGC 4204 (Sd), was observed for only
2 ks and was not detected. It is unknown, therefore, whether NGC 4204 would be detected in 5 ks. It is unsurprising that the faintest nuclei belong to very late type spirals.
We next focus particularly on the galaxies of type Sa–Sdm, as those of type
S0 and S0/a had an additional selection criterion that did not apply to the later galaxies, namely, that MB > 18.50. For the purpose of studying the frequency of − nuclear x-ray sources and AGNs we consider not only the Chandra targets presented here but also include the galaxies rejected because they were either already known to be AGNs (5 galaxies) or because they were starburst galaxies (3 galaxies). The
final numbers of galaxies in each morphological class are given in Table 7.1. The fractions of AGNs (in this part of the discussion we will use AGN to refer to both confirmed AGNs and to strong AGN candidates) are comparable in Sa–Sab, Sb–Sbc, and Sc–Scd galaxies. It is highest in Sb–Sbc galaxies ( 46%), and markedly ∼ lower in Sd–Sdm galaxies ( 7%). If it is assumed that all detected x-ray nuclei ∼ are in fact AGNs and conversely that all AGNs are x-ray sources, then we obtain the numbers shown in parentheses in the third column of Table 7.1. It is striking that the number of x-ray nuclei in Sc–Scd galaxies, which are nearly bulgeless, is
140 comparable to the number in earlier types. The fraction of nuclei for which there are strong arguments in favor of the AGN hypothesis is lower, but as the nuclei are fainter deeper observations are needed to characterize the source. As late type spiral galaxies often have nuclear star clusters, multiwavelength data may be needed to distinguish between an x-ray binary and an AGN.
In brief, this study has showed the following: (1) Nuclear x-ray sources are common in nearby late type galaxies; (2) Many of the nuclei are almost certain to be low luminosity AGNs. Of the 53 galaxies of type Sa–Sdm that passed the distance, inclination and Galactic latitude cuts, 5 were known to be AGNs. This survey has added a further 12 strong AGN candidates. (3) Sc–Scd galaxies are as frequently likely to host an x-ray source and/or AGN as galaxies of earlier types.
These AGNs were missed in earlier surveys because of their faintness. (4) X-ray sources and AGNs are truly rarer in Sd galaxies than they are in earlier types. This is consistent with the fact that until very recently NGC 4395 was the only known example of an AGN in an Sd galaxy. Since then only two more AGNs have been found in Sd galaxies, in NGC 3621 (Satyapal et al. 2007) and in NGC 4178 (Satyapal et al. 2009). This study provides strong candidates for what would be the fourth and fifth known AGNs in Sd galaxies, NGC 4713 and NGC 4561. (5) The data are consistent with heavy absorption being a common phenomenon in these AGNs.
There are multiple instances of x-ray spectra showing the characteristics of very heavy absorption where no direct emission reaches the observer. These x-ray spectra
141 show apparently unabsorbed but very faint power law spectra, often with a 6.4 keV
Fe Kα emission line (e.g. NGC 3169, NGC 4102, NGC 4713). In addition, several nuclei are faint in x-rays but show AGN-like IR luminosities (e.g. NGC 628, NGC
3184). Finally, it should be noted that NGC 3310 and NGC 4713 show AGN-like x-ray emission but no [Nev] IR lines (Satyapal et al. 2009; Goulding & Alexander
2009). The ionization potential required to produce Nev is 97 eV, in the soft x-ray band. This raises the possibility that absorption is so high that almost all of the soft x-ray emission is getting absorbed. (6) The accretion rate of the AGNs in the sample galaxies is low. The estimated Eddington rates range from 10−6 to 10−3.
A survey of the type discussed here finds AGNs and strong AGN candidates, but does not measure the masses of the SMBHs in those AGNs. Measurement of the mass of one of these SMBHs will be a difficult endeavor, since the sphere of influence of the black hole cannot be resolved with current technology and resources.
None of the six objects studied in this paper shows broad optical emission lines whose widths could be used to estimate the BH mass, and this is likely to be typical behavior. Spectropolarimetry may uncover broad lines in polarized light in some of the AGNs. Estimates of the SMBH mass may be made by using known scaling relationships, with the caveat that the correlations are all based on SMBHs two or more orders of magnitude more massive than the ones expected to be found by the survey. The least indirect method is an application of the observed correlation between the x-ray power law slope and Eddington ratio (Williams et al. 2004). The
142 Eddington ratio and the luminosity in turn provide an estimate of the mass of the
SMBH. Other relationships that can be used are: (a) the M•–σ relation (Ferrarese
& Merritt 2000; Gebhardt et al. 2000), but this method becomes more and more uncertain as the bulge itself becomes ill-defined in the latest-type spirals; (b) the
M•–Lbulge relationship (Kormendy & Richstone 1995; McLure & Dunlop 2001;
Marconi & Hunt 2003), which has more scatter and also faces the problem of the definition of the bulge; (c) the M•–vcirc relation (Ferrarese 2002; Baes et al. 2003), which has the advantage that it does not require the presence of a well-defined bulge; (d) the M•–C relation (Graham et al. 2001), where C is the concentration of light. There is also a reported relationship between black hole mass and core radio power (Snellen et al. 2003; McLure et al. 2004), but this relationship is not as well established as the others, and is based on observations of elliptical galaxies, so its applicability to the spiral galaxies here is uncertain. Overall, there is unlikely to be one standard method of measurement that can be applied to these galaxies, but one or more of the above methods may provide useful estimates of or limits to the masses of the SMBHs in these AGNs. Mass measurements independent of the scaling relationships are possible if an object turns out to have broad emission lines, like NGC 4395, in which case line widths or reverberation mapping may be used, or if it has maser emission, like NGC 4258, in which case gas dynamics can be used.
Mass measurement in other cases will have to await mas-scale angular resolution in bands other than radio to resolve the sphere of influence of these black holes.
143 These observations indicate that there is indeed a population of accreting
SMBHs in nearby spiral galaxies that do not show optical signs of activity but can be uncovered by looking for their x-ray emission. Such a population will answer the question of whether the bulge is the dominant component that determines the existence, and mass, of a nuclear SMBH. The discoveries of AGNs in the Sd galaxies
NGC 4395 (Ho et al. 1997b) and NGC 3621 (Satyapal et al. 2007), and the strong evidence for AGNs in the Scd galaxies NGC 3184 and NGC 5457 suggest it is not, at least as far as existence is concerned. Among the SMBHs discovered in the latest-type spirals (with small or no bulges) and in the lowest-mass galaxies should
6 be a population of SMBHs with masses less than 10 M⊙, enabling a systematic study of the low-mass end of the local supermassive black hole mass function.
7.1. Future research
Identifying candidate AGNs is only the first step towards the goal of estimating the local low-mass SMBH mass function. The logical next step is to identify the bona fide AGNs among the candidates. Subsequently, efforts should be directed at measuring the masses of the SMBHs in the AGNs.
A major step forward would be the determination of the spectral energy distributions (SEDs) of these objects, especially including radio emission. As mentioned above, the way to identify the very low luminosity AGNs present in nearby galaxies is to use as much of the SED as possible to rule out the hypothesis
144 that the SED can be explained without requiring an AGN component. Knowing the
SED ensures that it will be possible to estimate the true luminosity, and therefore the mass accretion rate, of the AGNs. Of particular interest are the constraints an
SED can put on the mode of accretion when the accretion rate becomes very low. If both x-ray and radio data are available, the relationship between LX /LR and mass accretion rate (Yuan & Cui 2005) can be tested.
The candidate AGNs exhibit a large dynamic range in the fluxes in different bands. Therefore the aim should be to obtain spectra in the wavebands where a source is bright, and single-band fluxes where a spectrum would be prohibitively expensive. X-ray spectra need to be obtained for detected sources that are bright enough. Here XMM-Newton is the preferred telescope because of its much larger effective area compared to Chandra. The usefulness of XMM-Newton observations was demonstrated in Chapter 3 for the case of NGC 4713. Long-duration x-ray observations will also allow variability analysis, which is a powerful tool for SMBH mass determination (McHardy et al. 2004).
Radio follow-up is important because of the high angular resolution achievable in radio observations, and because radio detections are one of the few ways to unambiguously distinguish an AGN from other x-ray sources. This is because the probability of a single XRB or supernova remnant (SNR) being luminous enough to be detected is low. A collection of XRBs or SNRs may account for the emission, but positing a certain number of XRBs or SNRs requires a corresponding star formation
145 rate. It is then necessary that the entire spectral energy distribution be consistent with the inferred star formation rate. If star formation can be ruled out, then a luminous nuclear radio source is probably an AGN. The argument for an AGN is of course strengthened if radio jets are present.
Despite the fact that these nuclei are obviously not bright in the optical, high spatial resolution optical spectroscopy is still desirable. If dilution by host galaxy light can be minimized enough to isolate broad components in the emission lines, then it may be possible to prove the existence of an AGN, as was done recently by
Shields et al. (2008) in the case of NGC 1042. Measuring the line width of a broad component in the Hα or Hβ line would simultaneously provide an estimate of the
SMBH mass. Such spectroscopy requires an instrument like the Space Telescope
Imaging Spectrograph (STIS), on Hubble, which has a narrow 0.2′′ slit width.
Infrared spectroscopy may also prove to be a powerful technique to detect hidden AGNs. Near infrared coronal lines like [S viii] 0.99 µm and [Si x] 1.43 µm provide support for an AGN as stellar emission cannot produce these highly ionized species. Infrared lines also have the advantage that they are much less affected by extinction than are lines in the UV and optical.
Finally, there are related questions regarding the accretion modes and physical structure of AGNs whose answers may also emerge as the luminosity of the AGNs studied and the masses of the SMBHs in them are pushed lower. Do stellar mass
146 black holes in x-ray binaries and SMBHs in AGNs have the same accretion physics?
Is an ultra-low luminosity AGN a scaled-up version of an XRB in the low/hard state? Does accretion mode change suddenly in host galaxies later than type Sbc? In the (optical) Palomar survey the detection rate is 50%–70% in galaxies of types Sbc and earlier and < 20% in types Sc and later (Ho 2008), and similarly in our snapshot ∼ Chandra observations of 37 galaxies, 6 of 16 galaxies of types S0–Sbc are detected, but only 1 of 19 galaxies of types Sc–Sdm is. The Sc–Sdm galaxies are undetected even after stacking, whereas the earlier types are detected in stacked images.
In conclusion, research into the activity of the SMBHs in late-type spiral galaxies will not only help us understand the specific AGNs studied, but will almost certainly further our understanding of the AGN phenomenon in general.
147 Morphology Total Number X-ray Sourcesa AGNsb
Sa–Sab 8 4(5) 3
Sb–Sbc 13 7(7) 6
Sc–Scd 18 8(10) 7
Sd–Sdm 14 2(2) 1
aAll nuclei detected in x-rays are counted. The number in parenthesis is the number of nuclei that are x-ray sources if it is assumed that all confirmed AGNs in the sample are x-ray sources, whether or not the x-ray source was detected. The two numbers are different when a nucleus is a confirmed AGN based on observations in other wavelengths but was either not detected in x-rays or does not have an x-ray observation. bIncludes confirmed AGNs and those identified as strong AGN candidates in the previous chapters.
Table 7.1. The number of nuclear x-ray sources, and confirmed or candidate AGNs in the sample galaxies of types Sa–Sdm.
148 BIBLIOGRAPHY
Assef, R. J. et al. 2008, ApJ, 676, 286
Baes, M., Buyle, P., Hau, G. K. T., & Dejonghe, H. 2003, MNRAS, 341, L44
Baldwin, J. A., Phillips, M. M., & Terlevich, R. 1981, PASP, 93, 5
Barth, A. J., Ho, L. C., Rutledge, R. E., & Sargent, W. L. W. 2004, ApJ, 607, 90
Bianchi, S., Guainazzi, M., & Chiaberge, M. 2006, A&A, 448, 499
B¨oker, T., Laine, S., van der Marel, R. P., Sarzi, M., Rix, H.-W., Ho, L. C., & Shields, J. C. 2002, AJ, 123, 1389
B¨oker, T., Sarzi, M., McLaughlin, D. E., van der Marel, R. P., Rix, H.-W., Ho, L. C., & Shields, J. C. 2004, AJ, 127, 105
Bregman, J. N., Hogg, D. E., & Roberts, M. S. 1995, ApJ, 441, 561
Cappellari, M., Bertola, F., Burstein, D., Buson, L. M., Greggio, L., & Renzini, A. 2001, ApJ, 551, 197
Carollo, C. M., Stiavelli, M., de Zeeuw, P. T., & Mack, J. 1997, AJ, 114, 2366
Carollo, C. M., Stiavelli, M., & Mack, J. 1998, AJ, 116, 68
Carollo, C. M., Stiavelli, M., Seigar, M., de Zeeuw, P. T., & Dejonghe, H. 2002, AJ, 123, 159
Colina, L., Gonzalez Delgado, R., Mas-Hesse, J. M., & Leitherer, C. 2002, ApJ, 579, 545
Condon, J. J. 1987, ApJS, 65, 485
——. 1992, ARA&A, 30, 575
Condon, J. J., Condon, M. A., Gisler, G., & Puschell, J. J. 1982, ApJ, 252, 102
149 Dale, D. A. et al. 2006, ApJ, 646, 161 de Vaucouleurs, G., de Vaucouleurs, A., Corwin, Jr., H. G., Buta, R. J., Paturel, G., & Fouque, P. 1991, Third Reference Catalogue of Bright Galaxies (Volume 1-3, XII, 2069 pp. 7 figs.. Springer-Verlag Berlin Heidelberg New York)
Desroches, L.-B., & Ho, L. C. 2009, ApJ, 690, 267
Dolphin, A. E. 2002, in The 2002 HST Calibration Workshop : Hubble after the Installation of the ACS and the NICMOS Cooling System, Proceedings of a Workshop held at the Space Telescope Science Institute, Baltimore, Maryland, October 17 and 18, 2002. Edited by Santiago Arribas, Anton Koekemoer, and Brad Whitmore. Baltimore, MD: Space Telescope Science Institute, 2002., p.301, ed. S. Arribas, A. Koekemoer, & B. Whitmore, 301
Dong, X. Y., & De Robertis, M. M. 2006, AJ, 131, 1236
Dudik, R. P., Satyapal, S., Gliozzi, M., & Sambruna, R. M. 2005, ApJ, 620, 113
Eracleous, M., Livio, M., & Binette, L. 1995, ApJ, 445, L1
Eracleous, M., Shields, J. C., Chartas, G., & Moran, E. C. 2002, ApJ, 565, 108
Ferrarese, L. 2002, ApJ, 578, 90
Ferrarese, L., & Ford, H. 2005, Space Science Reviews, 116, 523
Ferrarese, L., & Merritt, D. 2000, ApJ, 539, L9
Filho, M. E., Barthel, P. D., & Ho, L. C. 2006, A&A, 451, 71
Filho, M. E., Fraternali, F., Markoff, S., Nagar, N. M., Barthel, P. D., Ho, L. C., & Yuan, F. 2004, A&A, 418, 429
Fiore, F. et al. 2003, A&A, 409, 79
Flohic, H. M. L. G., Eracleous, M., Chartas, G., Shields, J. C., & Moran, E. C. 2006, ApJ, 647, 140
Freedman, W. L. et al. 2001, ApJ, 553, 47
Fruscione, A. et al. 2006, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 6270, Observatory Operations: Strategies, Processes, and Systems, ed. D. R. Silva & R. E. Doxsey
Ganda, K., Falc´on-Barroso, J., Peletier, R. F., Cappellari, M., Emsellem, E., McDermid, R. M., de Zeeuw, P. T., & Carollo, C. M. 2006, MNRAS, 367, 46
150 Gebhardt, K. et al. 2000, ApJ, 539, L13
Ghosh, H., Mathur, S., Ferrarese, L., & Fiore, F. 2009a, ApJ, submitted
——. 2009b, ApJ, submitted
——. 2009c, ApJ, in prep.
Ghosh, H., Mathur, S., Fiore, F., & Ferrarese, L. 2008, ApJ, 687, 216
Ghosh, H., Pogge, R. W., Mathur, S., Martini, P., & Shields, J. C. 2007, ApJ, 656, 105
Gierli´nski, M., Middleton, M., Ward, M., & Done, C. 2008, Nature, 455, 369
Gilfanov, M. 2004, MNRAS, 349, 146
Gliozzi, M., Satyapal, S., Eracleous, M., Titarchuk, L., & Cheung, C. C. 2009, ArXiv e-prints, 0906.0019
Gon¸calves, A. C., V´eron-Cetty, M.-P., & V´eron, P. 1999, A&AS, 135, 437
Gonz´alez-Mart´ın, O., Masegosa, J., M´arquez, I., Guerrero, M. A., & Dultzin-Hacyan, D. 2006, A&A, 460, 45
Goulding, A., & Alexander, D. 2009, ArXiv e-prints, 0906.0772
Graham, A. W., Driver, S. P., Allen, P. D., & Liske, J. 2007, MNRAS, 378, 198
Graham, A. W., Erwin, P., Caon, N., & Trujillo, I. 2001, ApJ, 563, L11
Greene, J. E., & Ho, L. C. 2004, ApJ, 610, 722
——. 2007a, ApJ, 667, 131
——. 2007b, ApJ, 656, 84
Grimm, H.-J., Gilfanov, M., & Sunyaev, R. 2003, MNRAS, 339, 793
Hasinger, G., Miyaji, T., & Schmidt, M. 2005, A&A, 441, 417
Heckman, T. M. 1980, A&A, 87, 152
H´eraudeau, P., & Simien, F. 1998, A&AS, 133, 317
Ho, L. C. 2008, ARA&A, 46, 475
——. 2009, ApJ, 699, 626
Ho, L. C., Filippenko, A. V., & Sargent, W. L. 1995, ApJS, 98, 477
151 Ho, L. C., Filippenko, A. V., & Sargent, W. L. W. 1997a, ApJS, 112, 315
Ho, L. C., Filippenko, A. V., Sargent, W. L. W., & Peng, C. Y. 1997b, ApJS, 112, 391
Ho, L. C., Greene, J. E., Filippenko, A. V., & Sargent, W. L. W. 2009, ApJS, 183, 1
Holtzman, J. A. et al. 1995, PASP, 107, 156
Hopkins, P. F., Hernquist, L., Martini, P., Cox, T. J., Robertson, B., Di Matteo, T., & Springel, V. 2005, ApJ, 625, L71
Hunter, D. A., van Woerden, H., & Gallagher, III, J. S. 1994, ApJS, 91, 79
Irwin, J. A., Sarazin, C. L., & Bregman, J. N. 2002, ApJ, 570, 152
Jensen, J. B., Tonry, J. L., Barris, B. J., Thompson, R. I., Liu, M. C., Rieke, M. J., Ajhar, E. A., & Blakeslee, J. P. 2003, ApJ, 583, 712
Kennicutt, Jr., R. C. et al. 2003, PASP, 115, 928
Kilgard, R. E. et al. 2005, ApJS, 159, 214
Kormendy, J., & Richstone, D. 1995, ARA&A, 33, 581
Kotanyi, C. G. 1980, A&AS, 41, 421
Laor, A. 2003, ApJ, 590, 86
Larsen, S. S. 2004, A&A, 416, 537
Levenson, N. A., Heckman, T. M., Krolik, J. H., Weaver, K. A., & Zycki,˙ P. T. 2006, ApJ, 648, 111
Maoz, D., Filippenko, A. V., Ho, L. C., Rix, H.-W., Bahcall, J. N., Schneider, D. P., & Macchetto, F. D. 1995, ApJ, 440, 91
Marconi, A., & Hunt, L. K. 2003, ApJ, 589, L21
Marconi, A., Risaliti, G., Gilli, R., Hunt, L. K., Maiolino, R., & Salvati, M. 2004, MNRAS, 351, 169
Martini, P., Kelson, D. D., Mulchaey, J. S., & Trager, S. C. 2002, ApJ, 576, L109
McElroy, D. B. 1995, ApJS, 100, 105
McHardy, I. M., Papadakis, I. E., Uttley, P., Page, M. J., & Mason, K. O. 2004, MNRAS, 348, 783
152 McLure, R. J., & Dunlop, J. S. 2001, MNRAS, 327, 199
McLure, R. J., Willott, C. J., Jarvis, M. J., Rawlings, S., Hill, G. J., Mitchell, E., Dunlop, J. S., & Wold, M. 2004, MNRAS, 351, 347
Merloni, A. 2004, MNRAS, 353, 1035
Meurer, G. R., Heckman, T. M., Leitherer, C., Kinney, A., Robert, C., & Garnett, D. R. 1995, AJ, 110, 2665
Mouri, H., & Taniguchi, Y. 1992, ApJ, 386, 68
Nagar, N. M., Falcke, H., & Wilson, A. S. 2005, A&A, 435, 521
Nagar, N. M., Falcke, H., Wilson, A. S., & Ulvestad, J. S. 2002, A&A, 392, 53
Neff, S. G., & Ulvestad, J. S. 2000, AJ, 120, 670
Nicastro, F. 2000, ApJ, 530, L65
Nicastro, F., Martocchia, A., & Matt, G. 2003, ApJ, 589, L13
Park, T., Kashyap, V. L., Siemiginowska, A., van Dyk, D. A., Zezas, A., Heinke, C., & Wargelin, B. J. 2006, ApJ, 652, 610
Paturel, G., Petit, C., Prugniel, P., Theureau, G., Rousseau, J., Brouty, M., Dubois, P., & Cambr´esy, L. 2003, A&A, 412, 45
Pence, W. D., Snowden, S. L., Mukai, K., & Kuntz, K. D. 2001, ApJ, 561, 189
Peterson, B. M. et al. 2005, ApJ, 632, 799
Phillips, A. C., Illingworth, G. D., MacKenty, J. W., & Franx, M. 1996, AJ, 111, 1566
Pogge, R. W. 1988, ApJ, 332, 702
Pounds, K. A., Reeves, J. N., Page, K. L., & O’Brien, P. T. 2004, ApJ, 605, 670
Prestwich, A. H., Irwin, J. A., Kilgard, R. E., Krauss, M. I., Zezas, A., Primini, F., Kaaret, P., & Boroson, B. 2003, ApJ, 595, 719
Quillen, A. C., Shaked, S., Alonso-Herrero, A., McDonald, C., Lee, A., Rieke, M. J., & Rieke, G. H. 2000, ApJ, 532, L17
Ranalli, P., Comastri, A., & Setti, G. 2003, A&A, 399, 39
Randall, S. W., Sarazin, C. L., & Irwin, J. A. 2006, ApJ, 636, 200
153 Rice, W., Lonsdale, C. J., Soifer, B. T., Neugebauer, G., Kopan, E. L., Lloyd, L. A., de Jong, T., & Habing, H. J. 1988, ApJS, 68, 91
Rossa, J., van der Marel, R. P., B¨oker, T., Gerssen, J., Ho, L. C., Rix, H.-W., Shields, J. C., & Walcher, C.-J. 2006, AJ, 132, 1074
Satyapal, S., Boeker, T., Mcalpine, W., Gliozzi, M., Abel, N. P., & Heckman, T. 2009, ArXiv e-prints, 0908.1820
Satyapal, S., Dudik, R. P., O’Halloran, B., & Gliozzi, M. 2005, ApJ, 633, 86
Satyapal, S., Sambruna, R. M., & Dudik, R. P. 2004, A&A, 414, 825
Satyapal, S., Vega, D., Dudik, R. P., Abel, N. P., & Heckman, T. 2008, ApJ, 677, 926
Satyapal, S., Vega, D., Heckman, T., O’Halloran, B., & Dudik, R. 2007, ApJ, 663, L9
Seth, A., Ag¨ueros, M., Lee, D., & Basu-Zych, A. 2008, ApJ, 678, 116
Shankar, F., Salucci, P., Granato, G. L., De Zotti, G., & Danese, L. 2004, MNRAS, 354, 1020
Shields, J. C. et al. 2007, ApJ, 654, 125
Shields, J. C., Walcher, C. J., B¨oker, T., Ho, L. C., Rix, H.-W., & van der Marel, R. P. 2008, ApJ, 682, 104
Sivakoff, G. R. et al. 2007, ApJ, 660, 1246
Skrutskie, M. F. et al. 2006, AJ, 131, 1163
Snellen, I. A. G., Lehnert, M. D., Bremer, M. N., & Schilizzi, R. T. 2003, MNRAS, 342, 889
Soria, R., Baldi, A., Risaliti, G., Fabbiano, G., King, A., La Parola, V., & Zezas, A. 2007, MNRAS, 379, 1313
Soria, R., Fabbiano, G., Graham, A. W., Baldi, A., Elvis, M., Jerjen, H., Pellegrini, S., & Siemiginowska, A. 2006a, ApJ, 640, 126
Soria, R., Graham, A. W., Fabbiano, G., Baldi, A., Elvis, M., Jerjen, H., Pellegrini, S., & Siemiginowska, A. 2006b, ApJ, 640, 143
Soria, R., & Kong, A. K. H. 2002, ApJ, 572, L33
Stern, D. et al. 2005, ApJ, 631, 163
154 Stetson, P. B., et al. 1998, ApJ, 508, 491
Stobbart, A.-M., Roberts, T. P., & Wilms, J. 2006, MNRAS, 368, 397
Terashima, Y., & Wilson, A. S. 2003, ApJ, 583, 145
Tremaine, S. et al. 2002, ApJ, 574, 740
Tully, R. B. 1988, Nearby galaxies catalog (Cambridge and New York, Cambridge University Press, 1988, 221 p.)
Tzanavaris, P., & Georgantopoulos, I. 2007, A&A, 468, 129
Ulvestad, J. S., & Ho, L. C. 2002, ApJ, 581, 925
Veilleux, S., & Osterbrock, D. E. 1987, ApJS, 63, 295
Vila, M. B., Pedlar, A., Davies, R. D., Hummel, E., & Axon, D. J. 1990, MNRAS, 242, 379
Walcher, C. J. et al. 2005, ApJ, 618, 237
Williams, R. J., Mathur, S., & Pogge, R. W. 2004, ApJ, 610, 737
Willis, A. G., Oosterbaan, C. E., & de Ruiter, H. R. 1976, A&AS, 25, 453
Young, A. J., Nowak, M. A., Markoff, S., Marshall, H. L., & Canizares, C. R. 2007, ApJ, 669, 830
Yuan, F., & Cui, W. 2005, ApJ, 629, 408
Zhang, W. M., Soria, R., Zhang, S. N., Swartz, D. A., & Liu, J. 2009, ArXiv e-prints 0904.1091
155