A Deep Chandra View of the NGC 404 Central Engine
Breanna Binder, Benjamin F. Williams, Mike Eracleous, Anil C. Seth, Julianne J. Dalcanton, Evan D. Skillman, Daniel R. Weisz, Scott F. Anderson, Terrance J. Gaetz, Paul P. Plucinsky
aa r X i v : . [ a s t r o - ph . H E ] A ug A Deep
Chandra
View of the NGC 404 Central Engine
B. Binder , B. F. Williams , M. Eracleous , A. C. Seth , J. J. Dalcanton , E. D. Skillman ,D. R. Weisz , , S. F. Anderson , T. J. Gaetz , P. P. Plucinsky ABSTRACT
We present the results of a 100 ks
Chandra observation of the NGC 404nuclear region. The long exposure and excellent spatial resolution of
Chandra has enabled us to critically examine the nuclear environment of NGC 404, whichis known to host a nuclear star cluster and potentially an intermediate-mass blackhole (on the order of a few times 10 M ⊙ ). We find two distinct X-ray sources:a hard, central point source coincident with the optical and radio centers ofthe galaxy, and a soft extended region that is coincident with areas of high H α emission and likely recent star formation. When we fit the 0.3-8 keV spectra ofeach region separately, we find the hard nuclear point source to be dominated bya power law (Γ = 1.88), while the soft off-nuclear region is best fit by a thermalplasma model ( kT = 0.67 keV). We therefore find evidence for both a power lawcomponent and hot gas in the nuclear region of NGC 404. We estimate the 2-10keV luminosity to be 1.3 +0 . − . × erg s − . A low level of diffuse X-ray emissionwas detected out to ∼ ′′ ( ∼ Subject headings:
Galaxies: nuclei – X-rays: general – galaxies: individual (NGC404) University of Washington, Department of Astronomy, Box 351580, Seattle, WA 98195 Department of Astronomy & Astrophysics and Center for Gravitational Wave Physics, The PennsylvaniaState University, 525 Davey Lab, University Park, PA 16802 Harvard-Smithsonian Center for Astrophysics, 60 Garden Street Cambridge, MA 02138, USA University of Minnesota, Astronomy Department, 116 Church St. SE, Minneapolis, MN 55455
1. Introduction
It is well established that the nuclei of active galaxies exhibit a broad range of lumi-nosities, from the most energetic quasars to more modest Seyferts. Active galactic nuclei(AGN) with nuclear X-ray luminosities below ∼ erg s − are classified as low luminosityAGNs (LLAGNs; Koratkar et al. 1995), and are the most common variety of AGN observedin the local universe. About 30% of all nearby bright galaxies exhibit LLAGN activity,and many additionally host low-ionization nuclear emission-line regions (LINERs; Ho et al.1997). Many recent studies have successfully linked LINERs to AGN activity; for example,the detections of X-ray cores (Dudik et al. 2005; Flohic et al. 2006; Gonz´alez-Mart´ın et al.2006; Zhang et al. 2009; Gonz´alez-Mart´ın et al. 2009), radio cores (Nagar et al. 2005), andmid-IR coronal lines (Satyapal et al. 2004) are all indicative of an accreting black hole (BH)energy source. At low nuclear luminosities, however, several alternative explanations for thepower source have been investigated, such as circumnuclear starbursts (Gonzales Delgado etal. 2004; Colina et al. 2002), shock heating by supernovae (SNe) in a high-density environ-ment (Alonso-Herrero et al. 2000; Olsson et al. 2007), and photoionization by very hot Ostars (Terashima et al. 2000). Moreover, a number of studies of the SEDs of weak AGNs inLINERs find that the AGN does not produce enough photons to power the emission lines(e.g., Ho 2008; Eracleous et al. 2010, and references therein).At a distance of 3.1 Mpc (Karachentsev et al. 2004), NGC 404 is the nearest S0 typegalaxy to the Milky Way and the closest galactic nucleus to be classified as a LINER by Hoet al. (1997). While the observed nuclear X-ray luminosity is low, only a few times 10 ergs − (Eracleous et al. 2002), the presence of a LLAGN in NGC 404 remains ambiguous. TheX-ray luminosity is consistent with that of a single high mass X-ray binary (HMXB) or agiant star-forming region such as 30 Doradus (Wang & Helfand 1991), and the soft X-rayemission is consistent with a hot gas origin, potentially blown out by a compact starburstor SNe. No radio core has been observed at 15 GHz (to a limiting flux of 1.4 mJy; Nagaret al. 2005), however an unnresolved 3 mJy continuum source is detected at 1.4 GHz by delR´ıo et al. (2004), comparable in luminosity to the Crab Nebula. A compact X-ray sourcewas previously detected in the central region of NGC 404 (Lira et al. 2000; Eracleous etal. 2002), but its low luminosity and soft thermal spectrum indicate a possible starburstevent origin. Mid-IR observations of the NGC 404 nuclear region show high ionization linesconsistent with AGNs (Satyapal et al. 2004); however, the [Ne V] lines (a more reliableindicator of AGN activity) are not detected (Abel & Satyapal 2008).Additionally, optical Hubble Space Telescope (HST) observations show H α emission oc-curring in both a compact source 0 . ′′
16 north of the nucleus and in structures reminiscientof supernova remnants (Pogge et al. 2000), and [O III] emission originates from a double- 3 –lobed structure along the major axis of the galaxy (Plana et al. 1998) with a higher velocitydispersion than the central H α emission (Bouchard et al. 2010). While the UV spectrumof the nucleus reveals signatures of O stars, the dilution of the lines suggest that ∼
60% ofthe UV flux may originate from a nonthermal source (Maoz et al. 1998). The observed levelof UV variability (the UV emission declined by a factor of 3 between 1993 and 2002; Maozet al. 2005) provides the strongest evidence for the existence of an accreting, massive blackhole in the NGC 404 nucleus.Analysis of NICMOS data by Ravindranath et al. (2001) reveal a nuclear star cluster(NSC) within the central arcsecond of NGC 404. NSCs are present in ∼
70% of galaxies(Graham & Guzman 2003; Mu˜noz Mar´ın et al. 2009), independent of the host galaxymorphology (B¨oker et al. 2002; Carollo et al. 1997; Cˆot´e et al. 2006), and are consideredto be the foundation of circumnuclear starbursts and supernovae that could drive LINERactivity (Meurer et al. 1995; Tremonti et al. 2001; Chandar et al. 2005). However, if NGC404 is dominated by star formation, the rate is exceptionally low, with only two to six Ostars being sufficient to explain the observed luminosity (del R´ıo et al. 2004). Dynamicalmodeling of stellar and gas kinematics in the nucleus by Seth et al. (2010) provide mixedevidence for the presence of a SMBH. They derive a firm upper limit of ∼ M ⊙ , and abest fitting gas dynamical mass of 4 . +3 . − . × M ⊙ (3 σ errors). Although other low-massgalaxies have been identified as candidate IMBH hosts through reverberation mapping (i.e.,NGC 4395; Peterson et al. 2005) and indirect mass measurements (i.e., by narrow opticalline measurements; see Greene & Ho 2007), NGC 404 is potentially the lowest-mass centralBH ever dynamically detected in the center of a galaxy.With both a NSC and possible IMBH, the nuclear region of NGC 404 is a complicatedenvironment. However, NGC 404 provides an ideal test case to address several key questionsrelating to LLAGN activity: is there an intrinsic lower limit to the luminosity of the AGNphenomenon, and what fraction of LINERs are powered by stellar processes n versus thosethat host a dwarf version of more powerful Seyferts and quasars? Deep, high spatial resolu-tion X-ray observations can potentially resolve many ambiguities surrounding the NGC 404nucleus, such as the morphology of the X-ray emission, the shape of the X-ray spectrum,and the variability properties of the source.In this paper, we present the analysis of a new, 100 ks Chandra observation of the NGC404 nuclear region. In §
2, we provide a description of the observations and our data analysisprocedures, including our imaging analysis, timing analysis, and spectral modeling. Ourresults are presented in §
3, and we give a discussion of our results in §
4. A summary of ourwork is given in §
5. 4 –
2. Observations and Data Analysis
NGC 404 was observed with
Chandra
ACIS-S on 2010 October 21-22 for a total useabletime of 97 ks. The optical center of the NGC 404 nucleus centered on the S3 chip anddetected at α J2000 = 01 h m s
99 and δ J2000 = +35 ◦ ′ ′′ ∼ ′′
02) in good agreement (within 0. ′′
4) with the position reported in SIMBAD (Cottonet al. 1999) and previous Chandra detections of the NGC 404 nucleus. The backgroundcount rate was estimated using an annular region devoid of any obvious sources, centered atthe nucleus with an inner and outer radius of 25 ′′ and 35 ′′ , respectively. We find the 0.3-8keV background count rate to be low, ∼ × − ct s − arcsec − , throughout the observation.We additionally obtained two archival Chandra
ACIS-S observations of NGC 404 from 2000Aug 30 (2 ks) and 1999 Dec 19-20 (24 ks). These observations are summarized in Table 1.All observations were reduced using the X-ray data analysis package CIAO version 4.3and using standard reduction procedures. We created exposure maps for the images usingthe CIAO script merge all . Point sources, including the NGC 404 nucleus, were identifiedusing the CIAO task wavdetect . We compared our X-ray point source detections to a 3.6 µ Spitzer
IRAC image of NGC 404 (aligned with the USNO-B1.0 star catalog) and find oneX-ray source other than the NGC 404 nucleus to be coincident with a likely IR counterpart(with a positional offset of 0. ′′ ′′ psextract , and spectral fitting was performed in XSPEC (Arnaud 1996) v.12.6.0q, and spectral models were fit to the unbinned spectra (binnedspectra are shown for display purposes only). All models include a column of neutral absorp-tion fixed at the Galactic value, n H , Gal = 5 . × cm − (Kalberla et al. 2005), estimatedusing the HEASARC n H calculator.We use C -statistics in lieu of traditional χ statistics, due to the low number of X-ray See http://simbad.u-strasbg.fr/. See http://cxc.harvard.edu/ciao/ahelp/merge all.html. See http://cxc.harvard.edu/ciao/ahelp/wavdetect.html. See http://cxc.harvard.edu/ciao/ahelp/psectract.html C -statistic and degrees of freedom ( C/d.o.f. ) for each model, we used the
XSPEC task goodness to perform Monte Carlo simulations of the spectra using each best-fit model. The proce-dure returns the percentage of the simulated spectra that had a fit statistic less than thatobtained from the fit to the real data. A value of 50% indicates the best-fitting model is agood representation of the data; values much less than this indicate that the data are overpa-rameterized by the model (i.e., random statistical fluctuations in the majority of simulatedspectra are not able to produce a fit statistic as low as that obtained from the real data),and values much higher than this indicate the model is a poor fit to the data (i.e., a largemajority of simulated spectra have a fit statistic less than that obtained from the real data).We perform 5 × realizations for each model, and we denote the resulting percentage as M C .
3. Results3.1. X-Ray Imaging and Hardness Ratio Maps
We co-added the available
Chandra observations to investigate the morphology of theNGC 404 X-ray emission. Earlier work by Eracleous et al. (2002) saw evidence for extendedsoft X-ray emission as far out as 10 ′′ ( ∼ Chandra image into three energy bands: soft (0.3-1 keV), medium (1-2keV), and hard (2-7 keV). Each image was adaptively smoothed using csmooth . Figure 1shows a smoothed RGB rendering of the NGC 404 nucleus. The image shows a hard core,coincident with the optical and radio center of the galaxy, and soft surrounding emission.Extended X-ray emission is detected out to ∼ ′′ ( ∼ ∼ ′′ northeast of the nucleus. For comparison, theNSC extends out to 0. ′′ ∼
10 pc; Seth et al. 2010), and the disk scale length of NGC 404is ∼ ′′ ( ∼ See http://cxc.harvard.edu/ciao/ahelp/csmooth.html α - I color map of the NGC 404 nuclear region (Seth et al. 2010),with contours from our HR map overlaid. Dark regions on the H α - I color map correspondto regions with high H α flux, likely associated with young stars. We find one such region iscoincident with soft, extended X-ray emission seen in our RGB rendering. We find no evidence for long-term ( ∼
10 year) variability; our best-fit 0.5-2 keV lumi-nosity ( ∼ erg s − , see next section) agree with that found by Eracleous et al. (2002).We generated light curves in four energy ranges (the total 0.3-10 keV band, and the soft,medium, and hard bands described above) to search for short-term variability over the courseof our 97 ks observation. We generated cumulative arrival time distributions of counts foreach light curve, then ran a two-sided KS test against the expected cumulative arrival timedistrbution for a constant count rate.Figure 5 shows the cumulative arrival time distributions for our total 0.3-8 keV lightcurve and the hard 2-7 keV light curve. We ran a two-sided KS test for each of our cumulativearrival time distributions against a constant count rate. As summarized in Table 2, we foundno evidence from our KS tests for variability at energies softer than 2 keV; however, our hardlight curve yields a KS chance probability of 5.8 × − . Both AGNs and XRBs exhibit strongand rapid variability; the apparent detection of hard X-ray variability on time scales of ∼ Hard X-ray spectra of LLAGNs in LINERs are typically well represented by a two-component model: a power-law component plus soft thermal emission (Terashima et al.2000), and the H α luminosoties of LINERs are positively correlated with the X-ray lumi-nosities in the 2-10 keV band (Ho 2001). To test whether the NGC 404 nucleus is consistentwith being a LLAGN, we use our observation with archival Chandra
ACIS-S observationsto perform spectral fitting in the 0.3-8 keV energy band. The total nuclear region ( < ′′ ) of 7 –NGC 404 contained ∼ α - I color map, we extracted spectrausing elliptical regions for each of the two distinct sources, with an elliptical area of 11square arcseconds for the hard point source and 27 square arcseconds for the soft extendedsource. The hard nuclear point source contained ∼
500 counts, and our soft, extended regioncontained ∼
90 counts.All our models fix an absorption component due to the Galactic column and an intrinsicabsorption inferred from optical extinction from Schlegel et al. (1998). We find no evidencein our spectral fitting for additional absorption, and we find no evidence for the presence ofa reflection component in any of the 0.3-8 keV spectra. We use the mass-metallicity relationderived in Tremonti et al. (2004) to estimate the NGC 404 metallicity to be 12 + log(O/H) ∼ APEC ; Smith et al. 2001). We find the thermal plasma model providesthe best fit, with
C/d.o.f. = 5.5/4 and
M C = 51% for kT = 0.67 ± .
11 keV and abundancesfixed at their solar values. The 0.3-10 keV luminosity of the soft X-ray emission is (2.5 ± . × erg s − , with an estimated 2-10 keV luminosity of 8.8 +0 . − . × erg s − . Our data areconsistent with the idea of gas being ejected from the central region in a superbubble.We attempted to model the hard, nuclear point-source as a simple power law, but wereunable to obtain an acceptable fit ( C/d.o.f. = 33/25 and
M C = 76%). Single-temperaturethermal plasma models, with abundances either fixed at their solar values or allowed tovary, additionally did not result in acceptable fits (
C/d.o.f. = 68/25 for
Z/Z ⊙ fixed at solar,and C/d.o.f. = 42/24 with
Z/Z ⊙ < M C = 0.4% with
C/d.o.f. = 11/23 for our model with
Z/Z ⊙ = 1, and M C = 0.3%with
C/d.o.f. = 11/22 for
Z/Z ⊙ < kT = 0.78 keV, with a power law photon index Γ = 1.88 +0 . − . contributing to ∼
72% of the total 0.3-8 keV photon flux and
C/d.o.f. = 21/24. We find a 8 –0.3-10 keV luminosity of 2.4 +0 . − . × erg s − and a 2-10 keV luminosity of 1.2 +0 . − . × erg s − . Although the fit moderately overparameterizes the data ( M C =21%), this modelproduces our best fitting parameters and is consistent with our imaging analysis.Finally, we applied the results of our spectral fitting to the soft, diffuse emission andthe hard, nuclear point source to model the entire nuclear region of NGC 404 as a powerlaw contaminated by thermal plasma emission. We assume the best-fit thermal plasmatemperature (0.67 keV) from the soft, diffuse emission and the best-fit power law photonindex (Γ = 1.88) from the hard nuclear point source. We find the power law componentcontributes ∼
55% of the 0.3-8 keV photon flux, with
C/d.o.f. = 124/126, with
M C ∼ +0 . − . using this approach.We estimate the 0.3-10 keV luminosity to be 3.0 +0 . − . × erg s − and a 2-10 keV luminosityof 1.3 +0 . − . × erg s − .The results of our spectral fitting are summarized in Table 3. Figure 6 shows the 0.3-8keV spectra for the soft, diffuse emission, the hard nuclear point source, and the total nuclearregion (with our best-fit models superimposed).The presence of a power law component in the NGC 404 nucleus, in addition to vari-ability in the 2-7 keV emission, lends support to the idea that the NGC 404 nucleus hostsan accreting black hole, but its low luminosity, on the order of a few times 10 erg s − , iscomparable to that of a single XRB.
4. Discussion
By modeling the hard nuclear point source and diffuse soft emission found in the NGC404 nucleus separately, we are able to resolve the ambiguity of the X-ray emission: a hardpoint source provides a power law component, and extended, diffuse gas supplies the thermalplasma emission.The Eddington luminosity is defined as L Edd = 1.3 × ( M BH / M ⊙ ) erg s − , and theEddington ratio ξ is commonly defined as ξ = log ( L/L
Edd ), where the bolometric luminos-ity L is typically estimated as L/L . − = 16 for AGNs (Ho 2008), whereas the bolometriccorrection factor for XRBs is roughly 2-5 times lower than for AGN (Wu & Gu 2008; here-after, WG08). The relationship between the X-ray power law photon index Γ and Eddingtonratio has been investigated for XRBs (WG08) and low luminosity AGNs (LLAGNs; Con-stantin et al. 2009; hereafter C+09). An anticorrelation is found for LLAGNs, whereas a 9 –positive correlation is observed for XRBs and luminous AGNs (Wang et al. 2004; Shemmeret al. 2006).We estimate ξ for NGC 404 assuming the black hole is an XRB, with M BH ∼ M ⊙ , andan IMBH AGN, with M BH ∼ M ⊙ . In Figure 7, we use our best-fit photon index for thehard nuclear point source and estimates of ξ to compare our NGC 404 data to the observedrelationships for both XRBs and LLAGNs from WG08 and C+09, respectively. The errorsin ξ indicate a factor of 3 change in BH mass (i.e., an AGN ranging from 3.3 × M ⊙ to3 × M ⊙ and an XRB ranging from 3.3 M ⊙ to 30 M ⊙ ). We find that while our data fallwell below the WG08 relationship for a high/soft state XRB, our data are consistent withthe anticorrelation found for the C+09 LLAGN sample. However, due to the large scatterobserved in the LLAGN sample and the errors in our observed photon index, we cannotdecisively rule out the possibility that the NGC 404 central engine is powered by an XRBin the low/hard state. Our data moderately favor the IMBH AGN interpretation. Althoughthe current work moderately favors the low-mass AGN interpretation, the UV spectrum andnuclear star cluster still allow the possibility of an XRB component.In Table 4, we summarize the observed multiwavelength properties of NGC 404, foundin both the literature and presented in this work, and indicate if the origin is likely to bean AGN or XRB. Additionally, the upper limits on the radio core (Nagar et al. 2005) andthe detection of an unresolved radio continuum (del R´ıo et al. 2004) can be combined withour deep X-ray observations to place upper limits on the central BH mass of NGC 404 – acorrelation has been established relating the radio luminosity L R and X-ray luminosity L X (Corbel et al. 2003; Gallo, Fender & Pooley 2003; Gallo et al. 2006) cover many ordersof magnitude in BH mass and luminosity, forming the “fundamental plane of black holeactivity.” We use the best-fit BH “fundamental plane” recently presented by Bell et al.(2011) and the upper limit on the radio core flux of NGC 404 to estimate M BH < × M ⊙ . If the unresolved 3 mJy continuum source is indeed powered by an accreting BH, itwould imply M BH ∼ × M ⊙ .Additionally, we use the following relation between radio luminosity and star formationrate (Condon 1992), (cid:18) L N W Hz − (cid:19) ∼ . × (cid:16) ν GHz (cid:17) − α (cid:20) SF R ( M ≥ M ⊙ ) M ⊙ yr − (cid:21) , (1)where α ∼ . SF R is the star formation rate (in M ⊙ yr − ), to test whether the observed radio luminosity is consistent with the observed lowstar formation rate. Using the SF R upper limit ( ∼ − M ⊙ yr − ) estimated in Seth etal. (2010), we predict an upper limit on the radio luminosity at 1.4 GHz to be ∼ ×
10 –erg s − . This upper limit is roughly eight orders of magnitude below the observed 1.4 GHzupper limit for the NGC 404 nucleus. We therefore conclude the observed radio luminositycannot be explained by star formation alone, and is evidence for the presence of an AGN inthe NGC 404 nucleus.The X-ray luminosity of the AGN in NGC 404 bears directly on the question of whetheraccretion power can account for the observed luminosities of the optical emission lines, whoserelative intensities are the defining characteristic of LINERs. In a recent study of the energybudgets of three dozen LINERs, including NGC 404, Eracleous et al. (2010) found that inthe majority of cases the weak AGN does not provide enough ionizing photons to accountfor the observed luminosities of the hydrogen recombination lines. This conclusion is ingeneral agreement with previous studies, as discussed in Eracleous et al. (2010). In theparticular case of NGC 404 the number of ionizing photons was found to be deficient by afactor of ≈
60. The X-ray luminosity of NGC 404 measured here is only ≈
25% higher thanthat measured by Eracleous et al. (2002), after accounting for the different distance usedin that paper, and is consistent (within errors) with the result obtained here. Therefore,the situation regarding the ionizing photon output of the AGN remains the same. However,Maoz et al. (1998) have estimated the ionizing luminosity of hot stars in the nucleus ofNGC 404 based on their measurements of the UV spectrum with the
HST and found it to beadequate to power the emission lines. This conclusion was re-iterated by Seth et al. (2010),who also noted that this ionizing luminosity could be provided by a relatively small numberof O stars in the nuclear star cluster. Therefore, the LINER in NGC 404 appears to bepowered by stellar processes.
5. Summary
We present the results of a 100 ks
Chandra observation of the nearby LINER and S0galaxy NGC 404. The deep exposure has allowed us to critically test several forms forthe 0.3-10 keV spectrum, and the excellent spatial resolution of
Chandra has enabled us toinvestigate the X-ray morphology of the NGC 404 nuclear region. We find the 0.3-10 keVspectrum to be consistent with emission from hot gas plus a power law continuum, and weare specifically able to separate a point source of high energy photons from a diffuse source ofsoft X-ray emission. Additionally, find evidence for variability in the hard 2-10 keV emission.The presence of a power law component and a moderate level of variability in the hardemission is indicative of X-ray emission powered by accretion onto a BH. The estimated0.3-10 keV luminosity ( ∼ × erg s − ) is both comparable to that of a single GalacticXRB and consistent with a IMBH accreting at extremely low levels, on the order of a few 11 –times 10 − M ⊙ yr − . We estimate the Eddington ratio for both scenarios (assuming a 10 M ⊙ XRB and a 10 M ⊙ AGN) and compare our best-fit photon index Γ of the hard nuclear pointsource to the observed trends with ξ for both XRBs and LLAGN. We find the NGC 404 X-ray spectral shape and luminosity to be consistent with observed LLAGNs, and inconsistentwith observed XRBs. We therefore favor the scenario in which the NGC 404 nucleus ispowered by an IMBH, with a mass on the order of 10 M ⊙ as dynamically estimated bySeth et al. (2010). Such a weak AGN does not produce a sufficient quantity of ionizingphotons necessary to power a LINER – we therefore conclude that the LINER in NGC 404is powered by stellar processes.Very low accretion rates are common in nearby galaxies with BH masses less than a fewtimes 10 M ⊙ (e.g., Baganoff et al. 2001; Garcia et al. 2000; Ho et al. 2003). Additional,multiwavelength observations of the NGC 404 nucleus are required to robustly determine themass of the cental BH – for example, resolved stellar populations within the nucleus wouldenable a robust dynamical mass determination, and a radio detection of the compact sourcewould verify the location of NGC 404 on the fundamental plane of BH activity.The authors would like to thank the anonymous referee for helpful comments. B. Binderand B. F. Williams acknowledge support from Chandra grant GO1-12118X. T. Gaetz andP. P. Plucinsky acknowledge support of NASA Contract NAS8-03060.
REFERENCES
Abel, N. P. & Satyapal, S. 2008, ApJ, 678, 686Alonso-Herrero, A., Rieke, M. J., Rieke, G. H., & Shields, J. C. 2000, ApJ, 530, 688Arnaud, K. A. 1996 ASPC, 101, 17Baganoff, F. K. et al. 2001, Nature, 413, 45Baggett, W. E., Baggett, S. M., & Anderson, K. S. J. 1998, AJ, 116, 1626Bell, M. E. et al. 2011, MNRAS, 411, 402B¨oker, T., Laine, S., van der Marel, R. P., Sarzi, M., Rix, H.-W., Ho, L. C., & Shields, J. C.2002, AJ, 123, 1389Bouchard, A., Prugniel, Ph., Koleva, M., & Sharina, M. 2010, A&A, 513, 54Carollo, C. M., Stiavelli, M., de Zeeuw, P. T., & Mack, J. 1997, AJ, 114, 2366Chandar, R., Leitherer, C., Tremonti, C. A., Calzetti, D., Aloisi, A., Meurer, G. R., & deMello, D. 2005 ApJ, 628, 210 12 –Chu, Y.-H. & Mac Low, M.-M. 1990, ApJ, 365, 510Colina, L., Gonz´alez Delgado, R., Mas-Hesse, J. M., Leitherer, C. 2002, ApJ, 579, 545Condon, J. 1992, ARAA, 30, 575Constantin, A., Green, P., Aldcroft, T., Kim, D.-W., Haggard, D., Barkhouse, W., & An-derson, S. F. 2009, ApJ, 705, 1336Cotton, W. D., Cotton, J. J., & Arbizzani, E. 1999, ApJS, 125, 409Corbel, S., Nowak, M. A., Fender, R. P., Tzioumis, A. K., & Markoff, S. 2003, A&A, 400,1007Cˆot´e, P. et al. 2006, ApJS, 165, 57del R´ıo, M. S., Brinks, E., & Cepa, J. 2004, AJ, 128, 89Dudik, R. P., Satyapal, S., Gliozzi, M., & Sambruna, R. M. 2005, ApJ, 620, 113Eracleous, M., Shields, J. C., Chartas, G., & Moran, E. C. 2002, ApJ, 565, 108Eracleous, M., Hwang, J. A. & Flohic, H. M. L. G. 2010, ApJ, 711, 796Flohic, H. M. L. G., Eracleous, M., Chartas, G., Shields, J. C., & Moran, E. C. 2006, ApJ,647, 140Gallo, E., Fender, R. P., & Pooley, G. G. 2003, MNRAS, 344, 60Gallo, E., Fender, R. P., Miller-Jones, J. C. A., Merloni, A., Jonker, P. G., Heinz, S.,Maccarone, T. J., & van der Klis, M. 2006, MNRAS, 370, 1351Garcia, M. R., Murray, S. S., Primini, F. A., Forman, W. R., McClintock, J. E., & Jones,C. 2000, ApJ, 537, 23Graham, A. W. & Guzm´an, R. 2003, ApJ, 125, 2936Gonz´alez Delgado, R. M., Cid Fernandes, R.; P´erez, E., Martins, L. P., Storchi-Bergmann,T., Schmitt, H., Heckman, T., & Leitherer, C. 2004, ApJ, 605, 127Gonz´alez-Mart´ın, O., Masegosa, J., M´arquez, I., Guerrero, M. A., & Dultzin-Hacyan, D.2006, A&A, 460, 45Gonz´alez-Mart´ın, O., Masegosa, J., M´arquez, I., Guainazzi, M., & Jim´enez-Bail´on, E. 2009,A&A, 506, 1107Greene, J. E. & Ho, L. C. 2007, ApJ, 670, 92Ho, L. C., Filippenko, A. V., & Sargent, W. L. W., 1997, ApJS, 112, 315Ho, L. C. 2001, ApJ, 549, 5Ho, L.C., Terashima, Y., & Ulvestad, J. S. 2003, 589, 783 13 –Ho, L. C. 2008, ARA&A, 46, 475Kalberla, P. M. W. et al. 2005, A&A, 440, 775Karachentsev, I. D., Karachentseva, V. E., Huchtmeier, W. K., & Makarov, D. I. 2004, AJ,127, 2031Koratkar, A., Deustua, S. E., Heckman, T., Filippenko, A. V., Ho, L. C., & Rao, M. 1995,ApJ, 440, 132Lira, P., Lawrence, A., & Johnson, R. A. 2000, MNRAS, 319, 17Maoz, D., Koratkar, A., Shields, J. C., Ho, L. C., Filippenko, A. V., & Sternberg, A. 1998,AJ, 116, 55Maoz, D., Nagar, N. M., Falcke, H., & Wilson, A. S. 2005, ApJ, 625, 699Meurer, G. R., Heckman, T. M., Leitherer, C., Kinney, A., Robert, C., & Garnett, D. R.1995, AJ, 110, 2665Mu˜noz Mar´ın, V. M., Gonz´alez Delgado, R. M., Schmitt, H. R., Cid Fernandes, R., P´erez,E. 2009, Ap&SS, 324, 253Nagar, N. M., Falcke, H., & Wilson, A. S. 2005, A&A, 435, 521Olsson, E., Aalto, S., Thomasson, M., Beswick, R., & H¨uttemeister, S. 2007, A&A, 473, 389Plana, H., Boulesteix, J., Amram, Ph., Carignan, C., & Mendes de Oliveira, C. 1998, A&AS,128, 75Peterson, B. M. et al. 2005, ApJ, 632, 799Pogge, R. W., Maoz, D., Ho, L. C., & Eracleous, M. 2000, ApJ, 532, 323Ravindranath, S., Ho, L. C., Peng, C. Y., Filippenko, A. V., & Sargent, W. L. W. 2001, AJ,122, 653Satyapal, S., Sambruna, R. M., & Dudik, R. P. 2004, A&A, 414, 825Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525Seth, A.C. et al. 2010, ApJ, 714, 713Shemmer, O., Brandt, W. N., Netzer, H., Maiolino, R., Kaspi, S. 2006, ApJ, 646, 29Smith, R. K., Brickhouse, N. S., Liedahl, D. A., & Raymond, J. C. 2001, ApJ, 556, 91Terashima, Y., Ho, L. C., Ptak, A. F., Mushotzky, R. F., Serlemitsos, P. J., Yaqoob, T.,Kunieda, H. 2000, ApJ, 533, 729Tremonti, C. A., Calzetti, D., Leitherer, C., & Heckman, T. M. 2001, ApJ, 555, 322Tremonti, C. A. et al. 2004, ApJ, 613, 898 14 –van Paradijs, J. 1981, A&A, 103, 140Wang, Q., & Helfand, D. J. 1991, ApJ, 370, 541Wang, J.-M., Watarai, K.-Y. & Mineshige, S. 2004, ApJ, 607, 107Wu, Q. & Gu, M. 2008, ApJ, 682, 212Zhang, W. M., Soria, R., Zhang, S. N., Swartz, D. A., & Liu, J. F. 2009, ApJ, 699, 281Table 1:
Chandra
ACIS-S Observations
Date ObsID R.A. Decl. Exposure(J2000) (J2000) (ks)(1) (2) (3) (4) (5)2010 Oct 22 12239 01 09 26.99 +35 43 05.1 972000 Aug 30 384 01 09 27.06 +35 43 05.0 21999 Dec 19-20 870 01 09 26.94 +35 43 05.5 24
Table 2: Temporal Variability as a Function of Energy
Energy range (keV) Net Counts K-S chance probability(1) (2) (3)0.3-10 1237 0.7560.3-1 487 0.8111-2 261 0.3652-7 220 5.8 × − This preprint was prepared with the AAS L A TEX macros v5.2.
15 –Table 3: Best-Fit Spectral Models for the NGC 404 0.3-8 keV Nucleus
Region Best-Fit Model Parameter Best-Fit Value(1) (2) (3) (4)Soft, diffuse thermal kT ± C/d.o.f.
M C L . − (2.5 ± . × erg s − L − +0 . − . × erg s − Hard, nuclear power law + kT +0 . − . frac. PL 72% C/d.o.f.
M C L . − +0 . − . × erg s − L − +0 . − . × erg s − Total nuclear power law + kT +0 . − . frac. PL 55% C/d.o.f.
M C L . − +0 . − . × erg s − L − +0 . − . × erg s − Table 4: AGN vs. XRB Properties Exhibited by the NGC 404 Nucleus
Property or Observation AGN XRB(1) (2) (3)Radio fluxes and upper limits X Soft X-ray emission † X Supernova remnant-like optical H α emission † X Mid-IR high ionization lines; hot dust X UV spectrum
X X
UV variability X Dynamical BH estimates X Nuclear star cluster X X-ray 2-10 keV variability; this work
X X
X-ray power law emission (Γ = 1.88 +0 . − . ); this work X † Observations are not associated with the hard nuclear point source.
16 – . . : : . : . Right ascension D ec li n a ti on Fig. 1.— Adaptively smoothed image of our
Chandra observation of the NGC 404 nuclearregion. This rendering emphasizes the hard nuclear point source and the soft extended regionin the center of the galaxy, as well as a super-soft X-ray source ∼ ′′ to the northeast of thenucleus and low levels of diffuse emission out to ∼ ′′ . Red = 0.3-1 keV, green = 1-2 keV,and blue = 2-7 keV. 17 – . . : : . : . Right ascension D ec li n a ti on . . : : . : . Right ascension D ec li n a ti on Fig. 2.— The left panel shows the smoothed 0.3-1 keV emission, and the right panel showsthe smoothed 2-7 keV emission. The red countours are set to the same levels for both images.The soft, 0.3-1 keV contours show extended, asymmetric emission, while the hard 2-7 keVcontours are consistent with a point source origin. -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 D ec li n a t i o n Fig. 3.— The smoothed hardness map of the NGC404 nuclear X-ray emission. Dark regionsindicate soft emission, while light regions indicate hard emission. A hard point source isclearly visible within a region of extended soft emission. 18 – D ec li n a t i o n Fig. 4.— The HST H α - I color map of the NGC 404 nucleus with HR map contours superim-posed (red). Dark regions in the map are likely associated with areas of younger stars, andappear to coincide with the soft, extended X-ray emission region seen in our HR map andRGB rendering. The center of the radio emission is shown by the yellow cross, and coincideswith the center of the hard nuclear point source. C u m u l a ti v e c oun t s C u m u l a ti v e c oun t s Fig. 5.— Cumulative arrival time of counts during the observation. The dashed lines rep-resent the predicted arrival times assuming a constant count rate. The left panel is for thetotal 0.3-8 keV light curve, and is consistent with a constant count rate. The right panel isfor the hard 2-7 keV light curve, and shows evidence for variability. 19 – −4 −3 C oun t R a t e p e r B i n ( s − k e V − ) Hard Nuclear Point Source10.5 2 511.5 r e s i du a l s / σ Energy (keV) 10 −4 −3 C oun t R a t e p e r B i n ( s − k e V − ) Soft Diffuse Emission10.5 2 50.811.21.41.6 r e s i du a l s / σ Energy (keV) −5 −4 −3 C oun t R a t e p e r B i n ( s − k e V − ) Total Nuclear Region10.5 2 5024 r e s i du a l s / σ Energy (keV)
Fig. 6.—
Top left : The 0.5-8 keV hard nuclear point source spectrum, containing ∼ Top right :The 0.5-8 keV soft, off-nuclear source spectrum, containing ∼
90 counts, with the best-fitthermal plasma model superimposed.
Bottom : Thetotal NGC 404 nucleus, with our best-fitmodel (a thermal plasma diluted by a power law) superimposed. Each model was fit to thetotal, unbinned spectrum – the spectra have been binned here for display purposes only. 20 – -6 -5 -4 -3 -2 -1 0log(L/L
Edd )1234 P ho t on I nd e x LLAGNs (C+09)XRBs (WG08)
Fig. 7.— The observed relationship between power law photon index and Eddington ratiofor LLAGNs (C+09, blue points) and XRBs (WG08, red points). The dashed line showsthe best-fitting anticorrelation to the LLAGN sample, and the dot-dashed line shows thebest-fitting correlation for the XRB sample. The estimated Eddington ratio for NGC 404,assuming a 10 M ⊙ and 10 M ⊙⊙