High Resolution X-Ray Imaging of the Center of IC342
aa r X i v : . [ a s t r o - ph ] J un HIGH RESOLUTION X-RAY IMAGING OF THE CENTER OF IC342
Daisy S.Y. Mak, Chun.S.J. Pun Department of Physics, University of Hong Kong, Pokfulam Road, Hong Kong, PR China
Albert K.H. Kong
Institute of Astronomy and Department of Physics, National Tsing Hua University, 101, Section 2,Kuang-Fu Road, Hsinchu, Taiwan 30013, R.O.C.Kavli Institute for Astrophysics and Space Research, Massachusetts Institute of Technology,Cambridge, MA 02139
ABSTRACT
We present the results of a high resolution ( q FWHM ∼ . ′′ ) 12 ksec Chandra
HRC-I observation of the starburst galaxy IC342 taken on 2 April 2006. We identify 23X-ray sources within the central 30 ′ × ′ region of IC342 and resolve the historicalultraluminous X-ray source (ULX), X3, near the nucleus into 2 sources, namely C12and C13. The brighter source C12, with L . −
10 keV = ( . ± . ) × erg s − , isspatially extended ( ≈
82 pc ×
127 pc). From the astrometric registration of the X-rayimage, C12 is at R.A. = 03h:46m:48.43s, decl. = +68d05m47.45s, and is closer to thenucleus than C13. Thus we conclude that source is not an ULX and must instead beassociated with the nucleus. The fainter source C13, with L . −
10 keV = ( . ± . ) × erg s − is consistent with a point source located at 6 . ′′ (P.A. ≈ ◦ ) of C12.We also analyzed astrometrically corrected optical Hubble Space Telescope andradio Very Large Array images; a comparison with the X-ray image showed simi-larities in their morphologies. Regions of star formation within the central region ofIC342 are clearly visible in
HST H a image which contains 3 optical star clusters andour detected X-ray source C12. We find that the observed X-ray luminosity of C12is very close to the predicted X-ray emission from a starburst, suggesting that the nu-clear X-ray emission of IC342 is dominated by a starburst. Furthermore, we discussthe possibility of AGN emission in the nucleus of IC342, which we can not prove nordiscard with our data. Subject headings: galaxies: individual (IC342) — X-rays: galaxies — galaxies: star-burst corresponding email: [email protected]
1. INTRODUCTION
IC342 is a nearby (1.8Mpc; see Buta & McCall 1999 for a review), almost face-on ( i = ◦ ± ◦ ; Newton 1980) late-type Sc/Scd galaxy in the Maffei Group which is one of the closest groups toour Galaxy with well developed spiral arms. Infrared and radio observations indicated moderatelystrong nuclear starburst activities (Becklin et al. 1980; Rickard & Harvey 1984) in IC342, whereasX-ray observations with Einstein (Fabbiano & Trinchieri 1987),
ROSAT (Bregman, Cox, & Tomisaka1993), and
ASCA (Okada et al. 1998; Kubota et al. 2001) indicated that the majority of the X-ray emission came from a few ultraluminous X-ray sources (ULXs). ULXs are defined as off-nuclear X-ray sources whose luminosities are greater than 2 . × erg s − , the Eddington lu-minosity of a 20 M ⊙ black hole. One of the ULX identified, X3 (based on the designationsfrom Fabbiano & Trinchieri 1987) was located near the galactic center, and was reported to becoincident with the center of IC342 at the resolutions of the ASCA (FWHM > ′ ) and ROSAT (FWHM ≈ ′′ ) data.Optical and near infrared observations of the inner region of IC342 revealed the presence ofa nuclear star cluster (B¨oker, van der Marel, & Vacca 1999). Photometrically distinct, luminous,and compact stellar clusters were often found in the centers of late-type spiral galaxies, such asNGC6240 (Lira et al. 2002) and NGC1808 (Jim´enez-Bail´on et al. 2005). Schinnerer, B¨oker, & Meier(2003) studied high spatial resolution (FWHM ≈ . ′′ ) millimeter interferometric observations ofthe CO ( − ) emission of IC342 and concluded that gaseous matter were accumulating at thenucleus through streaming motions. This suggested that star cluster formation could be a repetitiveprocess and this process could have provided the fuel for starburst activities. As a result, if X3 wereindeed associated with the galactic center, its emission may not be from a single ULX, but couldbe associated with the nuclear star cluster. However, the early identification of X3 as an ULX wasuncertain due to limited resolution of the early X-ray data.The higher light collecting power and lower background level of XMM-Newton provided aclearer picture of the nuclear X-ray source X3. A 10 ksec observation of IC342 was taken withthe PN-CCD camera and two MOS-CCD cameras using the medium filter on 2001 February 11.Two independent studies by Kong (2003) (hereafter K03) and Bauer, Brandt, & Lehmer (2003)(hereafter B03) on the data set yielded similar results, with a total of about 35 sources detected.The slope of the X-ray luminosity function was determined to be ∼ .
5, which was consistentwith other starburst galaxies and Galactic High Mass X-ray Binaries (HMXBs). However, therewere differences in the two studies about the nature and properties of the nuclear source. K03found that the nuclear X-ray source was consistent with the instrumental PSF of a point sourcelocated ∼ ′′ from the galactic center, suggesting a off-center scenario. K03 fit the nuclear sourcewith a power-law + blackbody spectral model ( kT = .
11 keV, N H = . × cm − , photon index a =
2) and found the disk temperature was about 5 to 10 times lower than expected. Such cool disk 3 –system might be evidence of IMBHs (Miller et al. 2003; Miller, Fabian, & Miller 2004). However,we could not ignore the scenario of a stellar mass black hole with inner disk outflows during highaccretion rate period (King & Pounds 2003), and that of magnetic corona atop a cool accretion diskin which X-rays were reprocessed to give soft photons (Beloborodov 1999; Miller et al. 2006). Theblack hole mass of the nuclear X-ray source was estimated to be between 220 M ⊙ and 15000 M ⊙ (K03). Together with the unphysical spectral fit with the broken power-law model, K03 ruled outthe possibility of an accreting stellar mass black hole with beamed relativistic jet emission. Onthe other hand, B03 suggested that the nuclear source was extended, mainly in the soft band, andits emission could be fit by a model of a point source plus a uniform disk with a radius of ≈ ′′ ,which was consistent with a supernova heated hot gas. Moreover, the nuclear source spectrumwas fit by B03 with a MEKAL + power-law model ( kT = . a = . Z = . Z ⊙ , N H = . × cm − ), including contribution from the Fe L-shell emission complexes at ≈ . ≈ . ≈ . ′′ ) Chandra X-ray Observatory ob-servation of the starburst galaxy IC342. The observation and the data reduction procedures of theX-ray data, and the associated optical data for comparison, are presented in Section 2. In Section 3,the analysis of the data, including source detection, photometry, and astrometric registration of theX-ray sources are described. In Section 4, we present the results of the associated counterparts ofthe nuclear X-ray sources in optical and radio images and the properties of the nuclear sources. Wediscuss the nature of nuclear X-ray emission of IC342 in Section 5 and we present a summary ofour findings in Section 6.
2. OBSERVATIONS AND DATA REDUCTION2.1. X-ray
IC342 was observed on 2006 April 2 starting at 15:15:44 (UT) with the
Chandra
High Res-olution Camera (HRC-I) for a total exposure time of 12.16 ksec, with a photon energy range of0.08 – 10 keV. The nominal pointing of the instrument was near the center of the galaxy andthe total region covered by the observation was 30 ′ × ′ which covered most of structure of thegalaxy, including the spiral arms. The field of view of the Chandra observation is presented in 4 –Figure 1 (left), overlaid with the Digital Sky Survey (DSS) image of IC342. The central 5 ′ × ′ of IC342 is magnified in Figure 1 (right) for reference. The X-ray data was analyzed using theChandra Interactive Analysis of Observations software package (CIAO) v.3.3 and Chandra Cali-bration Database (CALDB) v.3.2.1 software applied on the level-2 event fits file. The Chandra data had gone through Standard Data Processing (SDP, or the pipeline) with updated calibrationprocedures, including instrumental corrections and filtering good time intervals, applied.There were 4 other archival Chandra
IC342 data sets: three AXAF CCD Imaging Spectrom-eter (ACIS-S) observations (each 10 ksec) in sub-array mode centered on one of the ULX (X1),and one 3 ksec HRC-I observation. For the 3 ACIS-S data, the detectors did not cover the nuclearregion of IC342 as the sub-array mode was used to reduce pile-up. The 3 ksec HRC-I observation,on the other hand, had insufficient signal for source detection at the galactic center. Therefore onlythe 12 ksec HRC-I image was analyzed in this work.IC342 was also observed 4 times between 2001 and 2005 with
XMM-Newton . Among these4 observations, the data taken in year 2001 were studied by B03 and K03. We have retrieved the2001 EPIC data for the analysis of X-ray spectrum of the nuclear sources. The event files werereprocessed and filtered with the
XMM-Newton
Science Analysis Software (SAS v7.1.0) and weused the HEAsoft v6.2 and XSPEC v12.3.1 packages to perform spectral analysis.
We also analyzed archival
Hubble Space Telescope ( HST ) broadband and emission-line im-ages of IC342 for comparisons with the X-ray data. We searched the
HST
Data Archive for imageswith pointings that included the central part of the galaxy. The calibrated and data-quality imageswere retrieved, with preliminary CCD processing, including bias and dark subtraction, flat-fielding,and shutter-shading correction, automatically applied using calibration frames closest to the timeof the observations. The fields were inspected to ensure that the galactic nucleus did not fall on theboundaries on any of the CCD chips. We found that most of the archival images were saturatedat the central region of the galaxy. The only unsaturated images were the Wide Field PlanetaryCamera 2 (WFPC2) images taken on 1996 January 7 in broadband V (F555W), I (F814W), and innarrow-band H a (F656N), and they were selected for comparison with our Chandra
HRC-I data.The exposure time of each frame was 260 sec each for V and B band and 500 sec for H a . Mosaicswere produced using the STSDAS task WMOSAIC which removed known relative rotations andgeometric distortions of the individual WPFC2 chips. Astrometry of the calibrated HST images http://cxc.harvard.edu/ciao/dictionary/sdp.html
3. ANALYSIS3.1. X-ray Source Detection
We performed source detection on the X-ray image using the WAVDETECT algorithm (Freeman et al.2002) which employs a wavelet method to detect point sources. This algorithm was selected be-cause it is able to effectively identify extended sources with properly chosen wavelet scales; thisis important for the analysis of our
Chandra
IC342 image, which has an extended source near thegalactic center. Source detection results from the WAVDETECT algorithm depend on two param-eters, the significance threshold ( sigthresh ) and the image scale. We set the wavelet radii, in pixels,to be ‘1.0 1.414 2.0 2.828 4.0 5.657 8.0 11.314 16.0’ and the parameter sigthresh to be 2 × − for all runs. This detection threshold corresponds to less than one false detection in the Chandra
HRC-I field due to statistical fluctuations in the background (Freeman et al. 2002).The runtime of WAVDETECT depends heavily on the selected resolution of the image, anda resolution of 1024 × Chandra mirror and detector, especially at the edges ofthe field. We found the list of sources detected depended on the binning factor applied. A binningfactor of 4 was used in the final data reduction as a compromise between spatial resolution andfaint source detection. The
Chandra field was divided into 4 equal quadrants with overlapping atthe central regions before being recombined to the full-frame. The final image scale with a binningfactor of 4 was 0 . ′′ /pixel.To generate the final source list, we applied two additional selection criteria for the sourcesidentified by the WAVDETECT algorithm:1. The source should be detected in at least 2 different binning images, with one of them beingthe one with a binning factor of 4.2. The S/N of the source had to be greater than 3 in the binning factor 4 image.With these selection criteria we detected a total of 23 X-ray sources in our Chandra
HRC-I IC342 6 –image. Properties of the detected sources are summarized in Table 1. A total of 14 and 8 of oursources were also previously detected in the
XMM-Newton (B03; K03) and the
ROSAT obser-vations (Bregman et al. 1993) respectively. The alternate
XMM-Newton and
ROSAT source IDsare listed in Table 1. The locations of all sources detected are shown overlaid on the DSS im-age in Figure 1. A majority of the identified sources are located in the spiral arms of the galaxy,as previously noted by B03 and K03 with
XMM-Newton observations. Using the Chandra DeepField data (Brandt et al. 2001), we estimated that there could be 9–10 background objects withinour 30 ′ × ′ Chandra frame, implying that ∼
40% of the sources currently detected could bebackground objects.With the high resolution
Chandra
HRC-I image, we were able to resolve for the first timeX-ray emission from the center of IC342 into two distinct components, namely C12, the brighterand extended component near the center of the galaxy, and C13, a much fainter component located6 . ′′ from C12 at P.A. ≈ ◦ . The two sources could be identified by the WAVEDETECTalgorithm when the binning factor were set to be either 2 or 4, suggesting that they were genuinedetections. Images of the central 25 ′′ × ′′ of the Chandra
HRC-I image of IC342 are shown inFigure 2, together with the positions of the two nuclear sources C12 and C13, and the X-ray fluxcontours. Details of the X-ray emission morphology will be discussed in Section 4.1.1.
We extracted the integrated counts of the 23 detected X-ray sources of IC342 using the CIAOtool DMEXTRACT, and listed the results in Table 1. The source regions were defined in DS9 usingcircular apertures with different radii based on the relative brightness of the source compared to thelocal background and their off-axis angles. All the background regions were extracted using circu-lar annuli centered on individual sources. The only exception was the nuclear source C12, in whichan elliptical aperture was used because of its extended nature. Moreover, the background regionsof the two nuclear sources C12 and C13 (6 . ′′ apart) had to be modified to avoid overlapping.We used WebPIMMS to convert the count rates for all detected X-ray sources into unabsorbed0.08–10 keV luminosity by assuming an absorbed power-law model with N H = × cm − and a =
2, adopted from the average parameters derived from spectral fitting of the
XMM-Newton databy K03. These results are also presented in Table 1. This set of average parameters was simi-lar to that derived from the same
XMM-Newton data set by B03, with N H = . × cm − and a = .
63. The difference in 0.08–10 keV luminosities obtained with these two sets of spectralparameters is about 10%, with slightly lower values for the parameters from K03.For C12, B03 and K03 derived the X-ray flux by fitting the source spectrum with a simpleabsorbed power-law model (B03: N H = . × cm − and a = .
67; K03: N H = . × cm − a = . ′′ =
87 pc), identical to that in Boker et al. (1997, 1999) andK03, and similar to that (2 Mpc) assumed by Meier & Turner (2005). Using an elliptical (4 . ′′ and9 . ′′ for semiminor and semimajor axis) integrating window to include all the diffused emission,we found the stronger central nuclear source C12 has 255 ±
17 net counts, corresponding to anunabsorbed X-ray luminosity of L . −
10 keV = ( . ± . ) × erg s − ; the weaker source C13was one of the weakest source detected in our image, with a net count of 15 ± L . −
10 keV = ( . ± . ) × erg s − . One of the main objectives of our study is to obtain accurate positions of X-ray sources nearthe center of IC342. The
Chandra
HRC-I images have nominal 90% and 99% confidence astro-metric accuracy of 0 . ′′ and 0 . ′′ respectively, while the astrometric accuracies of the HST
WFPC2data is about 1 ′′ (Ptak et al. 2006). In order to identify optical counterparts of our Chandra
X-raysources in the
HST data, we first registered their positions to the same reference frame for compar-ison. We used the USNO-B1.0 Catalog (Monet et al. 2003) to improve the astrometric accuraciesfor both the
Chandra and
HST images. The advantage of using this is that proper motions arealready included in the catalog, and hence can be corrected for. The nominal RMS accuracy of theUSNO B1.0 catalog is 0 . ′′ .The astrometry of our Chandra image can be improved when one or more X-ray sourcesare associated with optical or radio sources of known positions. As seen in Figure 1, one of thebrightest sources, C5, is clearly associated with a bright stellar object within 1 ′′ in the DSS image.This source was also previously identified to be associated with a star in the ROSAT (Source 4in Bregman et al. 1993) and the
XMM-Newton observation (X12 in K03). The spectrum of X12is soft and can be fitted with a blackbody model ( kT = 0.17 keV). B03 determined the X-ray tooptical flux ratio log ( f x / f opt ) of X12 to be ≤ -3, which was consistent with that expected from aforeground star of log ( f x / f opt ) ≤ -1. B03 and K03 also identified it to be a foreground star. Usingthe USNO-B1.0 Catalog, we found that the source C5 was offset by 0 . ′′ (RA offset by 0 . ′′ and DEC offset by 0 . ′′ ) from this foreground star, which was within the expected 0 . ′′ pointing 8 –accuracy of the Chandra data. While there are other X-ray sources which could be associated withpoint sources in the DSS image, their positional offsets are either ≥ . ′′ or their S/N ratios weretoo low ( ≤
3) to make firm registrations with their optical counterparts. Thus we corrected thepointing of the
Chandra image using the CIAO command WCSUPDATE by linearly shifting ituntil C5 and the star coincided. This astrometric reference was applied to all the sources and theirshifted positions were tabulated in Table 1.The astrometry of the
HST images were corrected using the IRAF task CCMAP. With manybright stars within the WFPC2 frame, we could visually identify stars from the catalog in the
HST data products and compare them with the USNO-B1.0 Catalog to improve the astrometry. A totalof 7 stars were used for astrometric registration for the V and I band data, and 3 stars for the H a data. The RMS positional errors returned by CCMAP were 0 . ′′ for V , 0 . ′′ for I , and 0 . ′′ for H a . These registrations are similar to that of our Chandra
HRC-I images, hence allowingmeaningful spatial comparison of the sources in different wavelengths.
4. RESULTS4.1. Morphology of the IC342 center
The high spatial resolution of the
Chandra
HRC-I image ( q FWHM ∼ . ′′ ) provides an un-precedented detailed look into the X-ray emitting structure of the IC342 galaxy center. Detailedstructures of the X-ray emission from the IC342 center are illustrated in Figure 2 with the central25 ′′ × ′′ of the Chandra image. The two nuclear sources, C12 and C13, are separated from otherX-ray sources and stand out from the low diffuse X-ray background. The brighter source, C12, wasobserved to have at least two components. The first component is the core emission of radius ∼ ′′ at the center, roughly consistent with that bounded by the X-ray contour line of 1.4 counts. Thesecond component is the diffuse emission out to a radius of 5 ′′ , with an asymmetric extension alongthe north-south direction. The spatial extent, derived from the X-ray image at 99.5% linear scale,is 14 . ′′ in the north-south and 9 . ′′ in the east-west directions (127 pc ×
82 pc assuming a distance1.8 Mpc), resulting in a major-to-minor axis ratio of ∼ .
6. The observed north-south elongationis similar to that observed in the near infrared wavelength by Boker, Forster-Schreiber, & Genzel(1997), who detected a diffused emission with a major-axis of 110 pc and a high surface brightnesssource at the center. The authors interpreted the results as evidence for a small-scale nuclear stellarbar and a central concentration of stars.We determined the X-ray luminosity of the core and the circumnuclear diffuse regions of C12 9 –by extracting respectively the source counts of the inner 4 ′′ region with a circular aperture, and thesource counts of an annulus of r = ′′ − ′′ . The resulting X-ray luminosities (assuming the samespectral and distance parameters as listed in Section 3.2) are L core0 . −
10 keV = . × erg s − and L diffuse0 . −
10 keV = . × erg s − , implying roughly equal flux contributions in X-ray from thesetwo regions.To analyze the structure of emission from C12, its radial flux profile was extracted and nor-malized at the center. The results are shown in Figure 3 (top-left). As a comparison, the normal-ized radial flux profile of the Chandra
PSF at the location of C12 derived from the
Chandra
PSFsimulator, the
Chandra
Ray Tracer (ChaRT) , is also shown ( dash ). The emission from the corecomponent of C12 at the central r < ′′ is consistent with that from a point source. However, thediffused emission component at radius r ≥ ′′ does not decrease as steeply as that of the PSF. Wealso computed the radial flux profile of a point source plus a uniform r = ′′ circular disk at thelocation of C12 computed from the simulation task MARX . The results ( dot-dash ), also shownin Figure 3 (top-left), are generally consistent with that observed for C12, with a hint of slowly-decreasing surface brightness in the range r ∼ ′′ − ′′ for C12. A Kolmogorov-Smirnov test (K-Stest) was performed to measure the difference between radial profile of C12 and the models. TheK-S test statistics D , the maximum vertical deviation between the two curves, is 0.78 (confidenceprobability < . ′′ disk. The surface brightness profiles for selected bright X-ray sourcesC3, C6, and C21, are also plotted in Figure 3 (top-right), 3 (bottom-left), and 3 (bottom-right),respectively for comparison. The K-S statistic D of the reference sources C3, C6, and C21 are ina much lower range of 0.17–0.33 (corresponding to confidence probabilities 2.2%–76.0%). Theseresults reinforce the conclusion that the emission from C12 is not consistent with that from a pointsource; the X-ray emission of C12 might be due to multiple sources or diffuse emission or a com-bination of both. However, we cannot conclude whether the emission is consistent with the variouscomponents proposed by Boker et al. (1997) due to the limited resolution of the X-ray data.The newly detected X-ray source C13 is one of the faintest source in our Chandra observation,with L C13 / L C12 ≈ http://cxc.harvard.edu/chart/ http://space.mit.edu/ASC/MARX/
10 –
We compared the
Chandra image of the galaxy center of IC342 with the optical observations.The registered positions of the X-ray nuclear sources C12 and C13 are overlaid on the
HST V band and H a images of the central 17 ′′ × ′′ region of the IC342 in Figures 4 (top) and 5 (top)respectively. The spatial extent of the C12 core emission is also shown by a 4 ′′ circle ( dash ) in thefigures. The HST I band image was also analyzed but was found to be similar to the V data andthus not shown. In the HST images, there are many star forming regions at the center of IC342 withobscured emissions and bright knots. Both
HST images show that the optical emission from thecentral d ≈ ′′ of IC342 is concentrated in three star clusters aligned in the north-south direction.The star clusters have also been observed previously in the near-IR CO band by B¨oker et al. (1999).We named the three clusters here as the nuclear star cluster (NSC; middle), SC1 (top), and SC2(bottom). The centroid of C12 lies close to that of the optical maximum of the NSC, with an offsetof 0 . ′′ measured in all the V , I , and H a images. The observed separations were only slightlylarger than the uncertainty of the X-ray bore sight correction of C12 (0 . ′′ ) in our Chandra image.Using the near-IR CO band observations, B¨oker et al. (1999) deduced that the NSC is a rela-tively young cluster with an age of 10 . − . yrs, and a mass of M NSC ≈ × M ⊙ . These resultsfavored an instantaneous burst model rather than a constant star formation rate model for the NSC,meaning the dominant stellar population are of the same age as the NSC. Also, NSC could haveformed as a result of gas flowing to the nucleus initiated by torques from a nuclear stellar bar.The H a image is consistent with this scenario in that it shows gas spiraling around the NSC (seeFigure 4 and 5).The X-ray contours are overlaid on the broadband V and H a images of IC342 in Figures 4 (bot-tom) and 5 (bottom) respectively for a comparison of the diffuse X-ray emission with the detailedstructure of the optical galaxy center. These contour lines encircle the central r ∼ ′′ region ofIC342 and its structure is found to generally agree with those observed in the optical wavelengths,despite the lower angular resolution of the X-ray data. Moreover, the X-ray diffused emission iselongated along the north-south axis, which is similar to the alignment of the three optical nuclearstar clusters. These coincidence suggest that the observed X-ray nuclear emission is associatedwith the ongoing starburst and the three optical star clusters. However, the newly detected X-raynuclear source C13 does not coincide with any bright features in the H a image, suggesting that itis not associated with the star formation activities at the galaxy center. 11 – For comparisons with the X-ray and optical data, we also obtained Very Large Array (VLA)2 cm (15 GHz) and 6 cm (5 GHz) radio images of IC342. The observations and the data reductionprocedures have already been described in Tsai et al. (2006). These images have spatial resolutionsof 0 . ′′ at 6 cm and 0 . ′′ at 2 cm, respectively, with a 0 . ′′ uncertainty in the absolute position ofdetected sources. The positions of all the radio sources detected in the 2 cm and 6 cm are markedin Figures 4 (top) and 5 (top). The optical images of IC342 presented in Tsai et al. (2006) werenot astrometrically registered for comparison with the radio images. However, our current analysispresents improved astrometry on the X-ray and optical data, allowing accurate identifications ofradio counterparts for the X-ray and optical sources. Our revised astrometry explains the slightdifferences in the radio source positions of our Figures 4 and 5 compared with the Figure 2 ofTsai et al. (2006).The brightest radio source, source A using the naming scheme of Tsai et al. (2006), is nearthe optical star cluster SC1 and it was identified as a supernova remnant. One of the radio source,source J, lies very close to the center of the optical emission from NSC at a distance of 0 . ′′ , and adistance 0 . ′′ to the center of the nuclear X-ray source C12. This source is very weak, with only a1 . s detection in 2 cm and was actually not detected in the 6 cm data. Tsai et al. (2006) explainedsource J as a H II region situated between SC1 and NSC, requiring the excitation from a star clusterpowered by at least 70 O7 stars. However, with our corrected astrometry, source J was found to bemuch more likely to be associated with the center of the NSC. We will discuss the possibility ofthe radio emission from a radio-quiet AGN in Section 5.2. It should also be noted that none of theradio sources identified as H II regions by Tsai et al. (2006), except source J, could be associatedwith any star clusters in the optical images. This could be due to the high internal extinction A V in the central regions of IC342, consistent with the presence of high molecular gas concentrations,which could fuel more star formation in the region.The radio contours from the 6 cm VLA image are overlaid on the HST image and
Chandra contours in Figures 4 (bottom) and 5 (bottom). The morphology of the radio emission in the centralregion is very different from both the X-ray and optical images. C12 is encompassed by severalradio sources aligned from northeast to southwest that have been identified as supernova remnantsand H II regions (Tsai et al. 2006). The majority of the bright radio sources and the radio diffusedradio emission are located west of C12, suggesting enhanced star forming activity in that region.On the other hand, the faint X-ray source C13 is about 3 ′′ from a compact H II region (sourceL in Tsai et al. 2006). We noticed that the position of the source L listed in Tsai et al. (2006) wasincorrect and the revised position should be R.A. = 03h:46m:49.13s, decl. = +68d05m46.4s (Tsai2007, private communication), which was used here. With the pointing accuracy of the Chandra data at ≈ . ′′ , we concluded that C13 is unlikely to be associated with any radio emissions. 12 – To investigate the long term variability of the nuclear source C12, we compared the ob-served X-ray luminosities at the center of IC342 over a period of 13 years using data taken by
Chandra (this paper; 2006 April 2),
XMM-Newton (B03, K03; 2001 February 11), and
ROSAT (Bregman et al. 1993; 1991 February 13). We could not directly compare luminosities from theliterature because of the different assumptions made in those works. For example, the luminos-ity values in the earlier
ASCA and
ROSAT publications were computed assuming a distance toIC342 of 4.5 Mpc. We recalculated the luminosities from all the observations by assuming thesame energy range, spectral model, and assumed distance of IC342. Thus instead of quoting fromthe literature, we extracted the fluxes of C12 from the archival data of
XMM-Newton and
ROSAT ,together with our
Chandra observation, and analyzed them with the procedures described in Sec-tion 3.2. The fluxes from the different observatories were converted from source counts using thespectral model derived from
XMM-Newton since it had the highest spectral resolution. We adoptedthe simpler spectral model assumed for C12 (named X21) in K03, i.e., a power-law model with N H = ( . ± . ) × cm − and a = . ± .
16. Furthermore, we scaled the luminositiesand their errors to a distance of 1.8 Mpc and to the same energy range as HRC-I (0.1–10 keV).The scaled luminosities and errors of the nuclear source C12 from the different X-ray missions arelisted Table 2. The results from the two brightest X-ray sources in our
Chandra image, C3 and C6,are also listed for reference. The spectral models used were:1. C3: power-law model with N H = . + . − . × cm − and a = . + . − .
2. C6: disk blackbody model with N H = . + . − . × cm − and kT = . + . − . keVWe note that there was an ASCA observations taken on September 1993. However, the poorspatial resolution of
ASCA (FWHM > ′ ) and the large extraction radius (2 ′ ) used in analyzingthe nuclear source (Okada et al. 1998) implies that the ASCA results might suffer from confusionproblems. In our
Chandra data, a 2 ′ circular region centering at C12 would have included six othersources (C9, C10, C13, C14, C15, and C18). Using the present Chandra photometric information,we estimated that 46% of the measured
ASCA counts of C12 could be due to confusion, implying a0.1–10 keV luminosity of 11 . × erg s − . However, this estimate did not include uncertaintiesdue to long term spectral and flux variability information of all these X-ray sources. As a result, wedid not include the ASCA data point in the present variability study. On the other hand, though thespatial resolution of
ROSAT and
XMM-Newton were lower than
Chandra , the confusion problemin the two observations were not as significant as that in
ASCA . The spatial extent of C12 weresimilar in both
XMM-Newton and
ROSAT , about 8 ′′ − ′′ , and a circular region of this size wouldinclude both C12 and C13. 13 –Nonetheless, the source counts of C13 is only 6% of C12, and thus its contribution is neg-ligible unless it significant varies in flux. Thus, we decided to ignore the confusion effect in the XMM-Newton and the
ROSAT data.This analysis is the first to study the luminosity of C12 over a long period of time. In theenergy range 0.1–10 keV, the observed luminosity of the IC342 nuclear X-ray source C12 variesroughly by a factor of 2 over the 3 observations spanning a period of over 15 years. The observedvariations of the luminosity of C12 is actually the smallest amongst the three X-ray sources in-vestigated. It is worth noting that the flux uncertainties tabulated in Table 2 only include photonstatistics. Uncertainties in the relative calibration of various X-ray detectors have not been in-cluded. However, previous studies indicate that variations of normalized flux in different X-rayinstruments are small, at a level of ±
10% (Snowden 2002), and thus are not expected to accountfor all the flux variability observed. Another uncertainty in the current study is the assumption ofa single spectral model of C12 over the entire period as there was no spectral information from the
Chandra and
ROSAT data.We also studied the short-term variability of C12. We extracted the source and backgroundlightcurves of C12 using DMEXTRACT and high level background intervals were filtered. Theshort-term lightcurves of C12 were studied. However, we were not able to detect any short-termvariability on timescale of the observation because of the observed low count level.
5. Nature of the Nuclear X-ray Emission
The high resolution
Chandra
HRC-I observation resolves for the first time the X-ray emissionat the center of IC342, revealing the structure of the nucleus of this starburst galaxy. As seen fromFigure 4 and 5, both the X-ray and the optical emission are confined to a region of radius ∼
100 pcin the center of IC342. The most intense X-ray emission (inner 4 ′′ ) is coincident with the NSCseen in both V band and H a images. There are many regions of star formation within the centralregion of IC342 which were clearly visible in H a . It is this region that contains the 3 optical starclusters and includes our detected Chandra
X-ray source C12. The extension seen in the X-raydiffuse emission of C12 is most probably due to starburst activities induced by star clusters andmassive stars. This region also contains radio sources which aligns from northeast to southwest.As discussed in Tsai et al. (2006), these radio sources could be interpreted as H II regions whichwere small clusters that required excitation by individual O star clusters, each with over 5000 Ostars (Becklin et al. 1980; Turner & Ho 1983). Possible contributors to the X-ray emission from anuclear starburst include supernovae, O stars, High Mass X-ray Binaries (HMXBs), ULXs, and anobscured AGN. Using our Chandra
X-ray data, we attempted to explain in the following the stellarnature in the NSC. 14 –
Many previous infrared observations of IC342 (Boker et al. 1997; Schinnerer et al. 2003;Meier & Turner 2005) have proposed a starburst scenario for the nuclear region. The model ofthe central starburst includes a nuclear star cluster surrounded by a mini spiral arms-ring systemon the scale of ∼ ′′ (e.g., Fig. 10 in Meier & Turner 2005), as supported by observations ofmolecular line emissions. The X-ray image generally agrees with the spatial scales of this model,with the size of the C12 emission ∼ ′′ × ′′ is roughly the size of the proposed central ring( ∼ ′′ ), while the mini spiral arms that extend in the north-south direction to ∼ ′′ would bebeyond the spatial extent of C12 (see Figure 1 of Meier & Turner 2005 for a comparison of theoptical and CO (1-0) infrared emission). This model suggests that most of the star formationoccurs in the central ring region rather than along the spiral arms. Our
Chandra observation isconsistent with this scenario since the X-ray emission is concentrated in the central 100 pc. Thisstar formation structure is smaller in size compared to those in other starburst galaxies, which haveradii of order 1 kpc (Boker et al. 1997).We could study the starburst properties of IC342 using the results from starburst galax-ies and AGN survey Mas-Hesse et al. (1995) with the far-infrared to soft X-ray luminosity ra-tio. The far-infrared flux measured by the
Infrared Astronomical Satellite ( IRAS ) at 60 m m was F m m = ( ± ) Jy in the inner r=16 ′ region of IC342 (Beck & Golla 1988). Tsai et al. (2006)found that the luminosity of the compact H II regions in the nucleus was of the order of ∼ − F m m ( C ) ∼ −
26 Jy. We derived the soft (0 . − . L C120 . − . = ( . ± . ) × erg s − . This led to a flux ratio log ( L C120 . − . / L C1260 m m ) with a range of -3.1 to -3.4. This is in agreement with the value for star formation galaxies, -3.33 ( s = . s = .
81) (Mas-Hesse et al. 1995).Ward (1988) derived a model to predict X-ray emission from starbursts using the Brackett g emission. The model suggests that the X-ray emission arises from the accretion of material inpopulation I binary systems which are formed together with the OB stars. The total X-ray emissionin the starburst is calculated to be L . − = × − L B g L Xbin R ( erg s − ) , (1)where L B g is the observed Brackett g emission, L Xbin is the average X-ray luminosity of binary sys-tems and was assumed to be 10 erg s − , and R is the number of OB stars per massive binary sys-tems and was assumed to be 500 (Fabbiano, Feigelson, & Zamorani 1982). The Brackett g emis-sion from the inner 100 pc of IC342 was measured to be ≈ . × erg s − (Boker et al. 1997), 15 –thus implying a predicted X-ray luminosity from the starburst of L . − = . × erg s − .Our observed 0.3–5 keV X-ray luminosity of C12 is about 20% of this predicted luminosity. Thismight suggest that the X-ray emission from the starburst component in C12 was not dominated bypopulation I binary systems as suggested in the Ward (1988) model. It might instead be due tothermal emission such as coronal from young stars, supernovae/supernova remnants, and the hotinterstellar medium heated by supernovae and stellar winds. This might suggest that the soft X-rayemission in C12 was due to part of starburst component and some other thermal emission. We willdiscuss the possible thermal origin of the diffuse emission from C12 in section 5.3. Another possible origin of the core X-ray emission from the galactic center is an AGN. Us-ing the
XMM-Newton data, B03 proposed that the hard X-ray from C12 could be coming froma hidden AGN, although there was no direct evidence to confirm its existence in IC342. Thestarburst-AGN connection has been a topic that generated great interests but the exact relationshipremains uncertain. The triggering mechanism for both phenomena could be the interacting or themerging of gas-rich galaxies, which generates fast compression of the available gas in the innergalactic regions, causing both the onset of a major starburst and the fueling of a central black hole,hence raising the AGN. Examples of this class of objects include NGC6240 (Lira et al. 2002),NGC4303 (Jim´enez-Bail´on et al. 2003), and NGC1808 (Jim´enez-Bail´on et al. 2005). One particu-larly interesting object for comparison is NGC1808, where a similar core+diffuse X-ray emissionmorphology was observed. The extended X-ray emission of the NGC1808 center ( ∼
850 pc) isdominated by soft radiation ( ≤ a and optical-UVemission, suggesting intense star formation in the region. In contrast to IC342, two X-ray pointsources (with a separation of ∼
65 pc) were resolved in the core of NGC1808. Spectral infor-mation from the
Chandra
ACIS data showed that they were associated respectively with hot gasemission and hard X-ray emission. The nuclear location from the 2MASS data of NGC1808 wascloser to the softer point source (source S1 in Jim´enez-Bail´on et al. 2005). Furthermore, both theX-ray spectra of the centers of IC342 and NGC1808 can be modeled with absorbed power-lawplus MEKAL models, and the Fe K a emission is missing in both spectra. Jim´enez-Bail´on et al.(2005) compared the ratios of the detected lines such as O VIII and Ne IX in NGC1808 with thoseobserved in M82 and concluded a starburst-AGN coexistence in NGC1808.With astrometrically corrected X-ray and radio data, we were able to study the radio loudnessof the possible AGN at the center of IC342. As discussed in Section 4.1.3, radio observations ofIC342 reveal a group of weak sources within 5 ′′ radius of the center, with an integrated L ≈ . erg s − . This is slightly lower than the the range of 10 − erg s − observed from the 48 16 –low-luminosity AGNs (LLAGNs) sampled by Terashima & Wilson (2003). A weak radio source,source J, is near the X-ray center source C12, with F = . ± . F < . s limit), while a stronger source, source A, with F = . ± . F = . ± . . ′′ away from C12 (Tsai et al. 2006). Therefore if we concentrate on the central 2 ′′ radiusregion of IC342, corresponding to the size of the core component of C12, then the integrated radioflux in the region would be ∼ × − erg cm − s − . Together with the hard (2 – 10 keV) X-rayflux of the C12 core determined from WebPIMMS ( F C12core2 −
10 keV = . × − erg cm − s − ), wecould determine the radio loudness parameter R X = L L −
10 keV , as defined by Terashima & Wilson(2003), to be log R X ≈ −
3. This value is comparable to the average for Seyfert galaxies ( − . + . − . )as determined from a sample of 51 AGN candidates (Capetti & Balmaverde 2006).Based on the distribution and flux of the radio sources in the nucleus, we could classify thatIC342 as a radio-quiet type AGN, if there existed one. Radio-quiet AGN included LINERs, Seyfert(1 and 2) galaxies, and QSOs (Capetti & Balmaverde 2006). Assuming the X-ray luminosity ofC12 were due entirely to an AGN, then the hard X-ray luminosity observed L C122 −
10 keV = ( . ± . ) × erg s − would be just below the lower range of other LINERs observed with ROSAT , ASCA , and
Chandra (in the range 10 − erg s − ), but would be low when compared to Seyfert1 galaxies or quasars (typically > erg s − ). Another property to consider is the X-ray fluxvariability. As shown in Table 2, we observed the IC342 center X-ray flux to have varied by afactor of ∼ ∼ XMM-Newton spectral fit, a = . ± .
16 (K03), is slightly steeper than the typi-cal value of radio-quiet AGN and LINERs at a = . − . N H ∼ cm − (B03, K03), is much lower than typical AGN values of N H ≥ cm − , together with no intrinsicabsorption in the X-ray spectral fit. There was also no sign of the 6.4 keV Fe K a line emission, typ-ically an indicator of AGN, in the XMM-Newton
EPIC spectrum (B03, K03). However, this line isalso absent in a number of starburst galaxies identified as having hidden AGN (Tzanavaris & Georgantopoulos2007) and thus we cannot exclude the possibility of the presence of an AGN in IC342.The positional coincidence of the X-ray center with the optical star clusters in IC342 couldsuggest a scenario of coexistence of the nuclear star cluster and an AGN, as discussed by otherauthors (Seth et al. 2008; Shields et al. 2008). We attempt to place a lower limit to the black holemass. Assuming that the black hole is accreting at sub-Eddington rate, we could place a lowerlimit of the black hole mass at the center of IC342 at M BH ≥ × M ⊙ given the bolometricluminosity of the galaxy L bol ≈ × erg s − as determined by Becklin et al. (1980). On the 17 –other hand, it had been reported that nuclear star clusters are more massive than the coexistingblack holes and their mass ratio M BH / M NSC ∼ . − M BH ≤ × M ⊙ = M NSC , the nuclear star cluster mass as determined by IR data by B¨oker et al. (1999).Another estimate of the black hole mass comes from the M BH − s bulge relation (Gebhardt et al.2000; Tremaine et al. 2002; Hu 2008), which had been shown to apply even in the massive starcluster regime (Gebhardt et al. 2005; Shields et al. 2008). Using the velocity dispersion measure-ments for the nuclear star cluster ( s NSC = − , B¨oker et al. 1999), and the M BH − s bulge relation of Gebhardt et al. (2000), we derived M BH ≈ × M ⊙ , which is consistent with the lim-its described above. Based on these calculations, the range of black hole mass is suggestive of anintermediate mass black hole (IMBH). However, we should not rule out the alternative possibilityof an inactive supermassive black hole (mass ∼ − M ⊙ ) similar to known Seyfert nucleithat the nuclear emission is due to advection-dominated accretion flows (ADAF; Rees et al. 1982;Narayan & Yi 1995; Fabian & Rees 1995). Further spectral analysis could help to address whetherthis is the case. The spatial extent of the X-ray emission from the nucleus of IC342 could alternatively indi-cate the presence of populations of stellar objects uniformly distributed throughout the core region.These objects could be X-ray binaries, OB stars, or supernova remnants. For example, as pointedout by Stevens et al. (1999), populations of massive X-ray binaries evolved from high mass starscould be formed in starburst events, and such populations would be consistent with models as-sociated with starbursts. It had been suggested that photoionization from these stars might haveaccounted for the the enhanced C + millimeter emission in the central ring of the mini spiral modelof IC342 (Meier & Turner 2005).We considered the possibility of multiple unresolved point sources, consisting of a mixtureof stars and XRBs in the nuclear star cluster, contributing to the observed X-ray structure atthe center of IC342. A burst of star formation usually resulted in the production of large num-ber of OB stars (Stevens et al. 1999). With the typical X-ray luminosity for O stars ≈ − erg s − (Sciortino et al. 1990), and the number of O stars in the nucleus of IC342 at ∼ g emission (Ward 1988), we derived the X-ray luminosity from thesestars at log L X ≈ . − .
6. This is less than 0.1% of the total observed X-ray luminosity ofC12. The more luminous HMXBs, usually associated with young stellar population like O stars,might have also contributed to the observed X-ray emission. This could be inferred from the ra-tio of total X-ray luminosity from HMXB (with L X ≤ erg s − ) to the number of O stars, i.e. ( − ) × (Helfand & Moran 2001). For the nucleus of IC342, L Xbin was estimated to be 18 –about 8 × − erg s − (10%–90% of X-ray luminosity of C12), and the corresponding numberof X-ray binaries was about 10 as inferred from L B g . If these O stars and XRBs were uniformly dis-tributed in a region of radius 50 pc (the central 4 ′′ region of C12), the mean separation ( ≈ . ′′ forHMXBs and ≤ . ′′ for O stars) would be smaller than the spatial resolution of Chandra
HRC-I.As a result, a model with 4000 O stars and ≈
10 HMXBs could account for both the observedmorphology and flux of core component of the X-ray emission of C12.We also considered whether the diffuse component of C12 could also be explained by themodel above. As shown in Figure 5, the nucleus contains many regions of star formation, whichcould be the result of an outflow induced by the stellar winds of the massive stars or X-raybinaries present in the star clusters, as observed in M82 (Strickland, Ponman, & Stevens 1997;Moran & Lehnert 1997). The exact nature of this diffuse emission remains uncertain given thelimited amount of X-ray data available. Further observations with possible spectral analysis ofthis diffuse emission could help to link the starburst activities and the X-ray emission of the stellarobjects in the nuclear star cluster.
6. SUMMARY AND CONCLUSION
This paper aimed to study the nature of the X-ray emission in the nucleus of IC342 usinghigh resolution ( q FWMH ≈ . ′′ ) Chandra
HRC-I observation. The historical ULX at the centerof IC342, X3, was resolved into 2 sources, namely C12 and C13, for the first time since theX-ray telescopes prior to
Chandra did not have enough spatial resolution. The brighter sourceC12, with over 90% of the X-ray nuclear flux, was found to be located very close to the opticalgalactic center. The source C12 was also revealed as an extended source with a complex X-raymorphology, with a core component of diameter 4 ′′ that is roughly consistent with a point sourceof L . −
10 keV = ( . ± . ) × erg s − , and a diffuse component that extends out to a diameterof ∼ ′′ , or ∼
150 pc. Thus, we concluded that C12 is not an ULX. The much weaker source C13,with L . −
10 keV = ( . ± . ) × erg s − , is consistent with a point source situated 6 . ′′ atP.A. = 240 ◦ from C12.Registration of the X-ray image with optical ( HST ) and radio (VLA) data allowed us to findthe counterparts of X-ray sources. The registered position of C12 is R.A. = 03h:46m:48.43s, decl.= +68d05m47.45s, which is located 0 . ′′ from the optical NSC (B¨oker et al. 1999) and 0 . ′′ from a weak radio source (source J in Tsai et al. 2006), and within the boresight correction 0 . ′′ of the Chandra data. Intense star formation was observed in the
HST H a image of the inner 100pc of IC342, where the 3 optical star clusters (SC1, SC2, and NSC) and our Chandra
X-ray sourceC12 are found. Moreover, the diffuse X-ray emission extends preferentially along the north-southdirection, which is the same as the alignment of the star clusters in the optical. This suggests that 19 –the nuclear X-ray emission is associated with star formation.We discussed possible contributors to the X-ray emission in the nuclear region of IC342, witha main focus on the extended source C12. A starburst model for IC342 based on molecular lineobservations (Meier & Turner 2005) that included a mini spiral arms-ring system was found tobe consistent with the spatial extent of C12. Furthermore, the predicted X-ray emission from star-bursts as derived from the Brackett g emission (Ward 1988) is close to the actual observed X-ray lu-minosity of C12, indicating that X-ray from the nucleus is dominated by starburst activities. On theother hand, the hard emission from previous XMM-Newton observation suggested that there couldbe an AGN in IC342 (B03, K03). The luminosity of C12, L . −
10 keV = ( . ± . ) × erg s − ,is consistent with that of a LINER at (typical luminosity 10 − erg s − ). By comparing our cur-rent observation with earlier X-ray missions, we found a possible long term variability factor of 1.8in the IC342 nuclear X-ray emission over 15 years. This amount of variation would be consistentwith that from an AGN. Weak radio emission of L ∼ erg s − in the vicinity of C12 is con-sistent with a radio-quiet AGN. The observed properties (luminosity, absorption column density)are similar to those of type 2 low-luminosity LINERs or Seyfert 2 galaxies. The black hole mass asinferred from the M BH − s bulge relation for the AGN of ≈ × M ⊙ , is consistent with an IMBH,but we should not ignore the possibility of an inactive supermassive black hole. In conclusion, ourcurrent data did not confirm the existence of an AGN in the core of IC342, but it could neither bediscarded.The lack of any spectral information of our Chandra
HRC-I data did not allow for detailedstudies of the surface brightness distribution of the nuclear source C12 in different X-ray energybands. In particular, separate spectral information on the core and diffuse components of the X-ray galactic center emission of IC342 would allow us the opportunity to put a tighter constrainton the co-existence of a low luminosity AGN and a starburst in IC342. With the X-ray emissionfrom the core and diffused components at roughly the same level, further high spectral resolutionobservations with the
Chandra
ACIS could shed new light on how the nucleus of IC342 fit into thestarburst-AGN family.
7. ACKNOWLEDGEMENTS
We thank L. Sjouwerman and C.W. Tsai for generously providing the radio maps and helpfuldiscussions. This work is based on observations obtained with
Chandra , support for this work wasprovided by the National Aeronautics and Space Administration through Chandra Award NumberG06-7110X issued by the Chandra X-ray Observatory Center, which is operated by the Smithso-nian Astrophysical Observatory for and on behalf of the National Aeronautics Space Administra-tion under contract NAS8-03060. S.Y. Mak acknowledges support from HKU under the grant of 20 –Postgraduate Studentship. C.S.J. Pun acknowledges support of a RGC grant from the governmentof the Hong Kong SAR. A.K.H. Kong acknowledges support from the National Science Council,Taiwan, through a grant NSC96-2112-M-007-037-MY3.
REFERENCES
Bauer, F. E., Brandt, W. N., & Lehmer, B. 2003, AJ, 126, 2797Beck, R., & Golla, G. 1988, A&A, 191, L9Becklin, E. E., Gatley, I., Matthews, K., Neugebauer, G., Sellgren, K., Werner, M. W., & Wynn-Williams, C. G. 1980, ApJ, 236, 441Beckmann, V., Barthelmy, S. D., Courvoisier, T. J.-L., Gehrels, N., Soldi, S., Tueller, J., & Wendt,G. 2007, A&A, 475, 827Beloborodov, A. M. 1999, ApJ, 510, L123Boker, T., Forster-Schreiber, N. M., & Genzel, R. 1997, AJ, 114, 1883B¨oker, T., van der Marel, R. P., & Vacca, W. D. 1999, AJ, 118, 831Brandt, W. N. et al. 2001, AJ, 122, 2810Bregman, J. N., Cox, C. V., & Tomisaka, K. 1993, ApJ, 415, L79Buta, R. J., & McCall, M. L. 1999, ApJS, 124, 33Capetti, A., & Balmaverde, B. 2006, A&A, 453, 27Fabbiano, G., Feigelson, E., & Zamorani, G. 1982, ApJ, 256, 397Fabbiano, G., & Trinchieri, G. 1987, ApJ, 315, 46Fabian, A. C., & Rees, M. J. 1995, MNRAS, 277, L55Freeman, P. E., Kashyap, V., Rosner, R., & Lamb, D. Q. 2002, ApJS, 138, 185Gebhardt, K. et al. 2000, ApJ, 539, L13Gebhardt, K., Rich, R. M., & Ho, L. C. 2005, ApJ, 634, 1093Georgantopoulos, I., Panessa, F., Akylas, A., Zezas, A., Cappi, M., & Comastri, A. 2002, A&A,386, 60 21 –George, I. M., Turner, T. J., Yaqoob, T., Netzer, H., Laor, A., Mushotzky, R. F., Nandra, K., &Takahashi, T. 2000, ApJ, 531, 52Helfand, D. J., & Moran, E. C. 2001, ApJ, 554, 27Hu, J. 2008, MNRAS, 475+Jim´enez-Bail´on, E., Santos-Lle´o, M., Dahlem, M., Ehle, M., Mas-Hesse, J. M., Guainazzi, M.,Heckman, T. M., & Weaver, K. A. 2005, A&A, 442, 861Jim´enez-Bail´on, E., Santos-Lle´o, M., Mas-Hesse, J. M., Guainazzi, M., Colina, L., Cervi˜no, M.,& Gonz´alez Delgado, R. M. 2003, ApJ, 593, 127King, A. R., & Pounds, K. A. 2003, MNRAS, 345, 657Kong, A. K. H. 2003, MNRAS, 346, 265Kubota, A., Mizuno, T., Makishima, K., Fukazawa, Y., Kotoku, J., Ohnishi, T., & Tashiro, M.2001, ApJ, 547, L119Lira, P., Ward, M. J., Zezas, A., & Murray, S. S. 2002, MNRAS, 333, 709Mas-Hesse, J. M., Rodriguez-Pascual, P. M., Sanz Fernandez de Cordoba, L., Mirabel, I. F.,Wamsteker, W., Makino, F., & Otani, C. 1995, A&A, 298, 22Meier, D. S., & Turner, J. L. 2005, ApJ, 618, 259Miller, J. M., Fabbiano, G., Miller, M. C., & Fabian, A. C. 2003, ApJ, 585, L37Miller, J. M., Fabian, A. C., & Miller, M. C. 2004, ApJ, 614, L117Miller, J. M., Homan, J., Steeghs, D., Rupen, M., Hunstead, R. W., Wijnands, R., Charles, P. A.,& Fabian, A. C. 2006, ApJ, 653, 525Monet, D. G. et al. 2003, AJ, 125, 984Moran, E. C., & Lehnert, M. D. 1997, ApJ, 478, 172Narayan, R., & Yi, I. 1995, ApJ, 452, 710Newton, K. 1980, MNRAS, 191, 169Okada, K., Dotani, T., Makishima, K., Mitsuda, K., & Mihara, T. 1998, PASJ, 50, 25Ptak, A., Colbert, E., van der Marel, R. P., Roye, E., Heckman, T., & Towne, B. 2006, ApJS, 166,154 22 –Rees, M. J., Begelman, M. C., Blandford, R. D., & Phinney, E. S. 1982, Nature, 295, 17Reeves, J. N., & Turner, M. J. L. 2000, MNRAS, 316, 234Rickard, L. J., & Harvey, P. M. 1984, AJ, 89, 1520Schinnerer, E., B¨oker, T., & Meier, D. S. 2003, ApJ, 591, L115Sciortino, S., Vaiana, G. S., Harnden, Jr., F. R., Ramella, M., Morossi, C., Rosner, R., & Schmitt,J. H. M. M. 1990, ApJ, 361, 621Seth, A., Ag¨ueros, M., Lee, D., & Basu-Zych, A. 2008, ApJ, 678, 116Shields, J. C., Walcher, C. J., Boeker, T., Ho, L. C., Rix, H.-W., & van der Marel, R. P. 2008,ArXiv e-prints/0804.4024, 804Snowden, S. L. 2002, ArXiv Astrophysics e-prints/0203311Stevens, I. R., Forbes, D. A., & Norris, R. P. 1999, MNRAS, 306, 479Strickland, D. K., Ponman, T. J., & Stevens, I. R. 1997, A&A, 320, 378Terashima, Y., & Wilson, A. S. 2003, ApJ, 583, 145Tremaine, S. et al. 2002, ApJ, 574, 740Tsai, C.-W., Turner, J. L., Beck, S. C., Crosthwaite, L. P., Ho, P. T. P., & Meier, D. S. 2006, AJ,132, 2383Turner, J. L., & Ho, P. T. P. 1983, ApJ, 268, L79Tzanavaris, P., & Georgantopoulos, I. 2007, A&A, 468, 129Ward, M. J. 1988, MNRAS, 231, 1P
This preprint was prepared with the AAS L A TEX macros v5.2.
23 –Table 1.
Chandra
HRC source list of IC342
ID RA DEC RA DEC Net Counts S/N L X ( . −
10 keV) RemarkJ2000.0 J2000.0 error ( ′′ ) error ( ′′ ) (10 erg s − )(1) (2) (3) (4) (5) (6) (7) (8) (9)C1 3:45:10.62 68:02:31.95 1.1 0.7 83 . ± .
89 5.60 2 . ± . . ± .
02 4.93 1 . ± . ROSAT . ± .
24 27.17 26 . ± .
98 IC342 X1,
ROSAT . ± .
64 3.81 1 . ± .
32 XMM X9C5 3:46:06.58 68:07:05.70 0.5 0.4 82 . ± .
59 7.14 2 . ± . ROSAT B =12.8 R =11.82C6 3:46:15.79 68:11:12.98 0.3 0.4 404 . ± .
57 18.76 13 . ± .
73 IC342 X2,
ROSAT . ± .
29 3.03 0 . ± .
21 XMM X14C8 3:46:27.36 68:04:10.83 0.5 0.4 17 . ± .
52 3.79 0 . ± . . ± .
44 7.60 2 . ± . ROSAT . ± .
41 2.46 0 . ± . . ± .
32 3.09 0 . ± . ROSAT . ± .
37 14.66 8 . ± .
59 nucleus, IC342 X3,
ROSAT . ± .
00 3.77 0 . ± . . ± .
54 2.40 0 . ± .
12 XMM X24C15 3:46:57.49 68:06:19.51 0.2 0.4 108 . ± .
13 9.78 3 . ± . ROSAT . ± .
50 3.73 0 . ± . . ± .
35 2.89 0 . ± . . ± .
64 3.02 0 . ± . . ± .
14 2.86 0 . ± . . ± .
62 8.88 10 . ± . B =11.7 R =10.0C21 3:47:18.73 68:11:29.79 0.8 0.5 88 . ± .
01 7.36 2 . ± .
40 XMM X30C22 3:47:22.91 68:08:59.81 0.4 0.4 72 . ± .
62 7.49 2 . ± .
32 XMM X31C23 3:48:06.93 68:04:54.32 0.8 0.7 28 . ± .
10 3.54 0 . ± . Chandra . Column (2) and (3) list the X-ray sourcepositions using the astrometric reference as discussed in Section 3.3. Column (4) and (5) list the errors of RA and DEC in arc seconds, including theerrors returned by the WAVDETECT task and the shift used for correcting the astrometry. Column (6) lists the the background subtracted countsand the errors, as discussed in Section 3.2. Column (7) lists the signal to noise ratios of X-ray sources. Column (8) lists the unabsorbed 0.08-10keV luminosities and their errors in unit of 10 erg s − , assuming the average power-law spectrum of N rmH = × cm − and a = B and R magnitudes taken fromthe USNO-B1.0 Catalog.
24 –
E N5’
E N
13 12 789141518 (a) (b)Fig. 1.— Digitized Sky Survey (DSS) blue band image of the field of view (30 ′ × ′ ) of Chandra (left) and the central 5 ′ × ′ (right) of IC342, with detected Chandra
X-ray sources overlaid . Theradius of the source circles is 15 ′′ (left image) and 5 ′′ (right image). The dotted-line rectangle atthe center in the left image showed the region enlarged in the right image. The field of view of the XMM-Newton observation (26 ′ × ′ ) was also shown as the solid rectangular region in the left. 25 – D E C ( J2000 . ) D E C ( J2000 . ) C13
Fig. 2.—
Top : The smoothed
Chandra
HRC-I image of central 25 ′′ × ′′ region of IC342. The image was smoothed with a Gaussian kernel radius=3 and displayed with linear scaleat 99.5%. The centroid of C12 at R.A. = 03h:46m:48.43s, decl. = +68d05m47.45s is marked with a blue cross at the center of the image. The dimmer source C13 is located 6 . ′′ from C12 atP.A. ≈ ◦ and its position is also marked by a blue cross. Bottom : The
Chandra
HRC-I image (same as above) with X-ray contours (blue) superposed. The data have been smoothed with gaussianfunction, and contours are at 0.3, 0.6, 0.9, 1.2, and 1.5 counts. The dashed line circle showed a 4 ′′ region at the center. Details of the X-ray morphology are discussed in Section 4.1.1.
26 –Fig. 3.— The observed radial profiles of X-ray sources C12 (top-left), C3 (top-right), C6 (bottom-left), and C21 (bottom-right) (solid square) and the calculated profiles of the PSF from the ChaRTprogram (black dashed line) at the location of the sources. The errors on the ChaRT PSF are negli-gible. For the radial profile of C12, the radial profile of a model with one point source overlappedon a circular disk of r = ′′ (red dashed line) was also shown. The bump in the radial profile ofC12 between 6 ′′ and 7 ′′ is from C13. 27 – C E N5’’C13 S N S C S1 J A L K IH G F ED CB E N5’’
Fig. 4.—
HST V band image of the central 19 ′′ × ′′ region of IC342 displayed in square root scale to bring out the features. Top : Locationsof the 3 optical star clusters are labeled by black circles. The centroids of the X-ray center sources C12 and C13 (blue cross), and a 4 ′′ diameterregion centered around of C12 (blue dashed circle), are also shown. Radio sources detected in the VLA 2 cm and 6 cm maps are marked and labeled(red cross) according to the naming convention by Tsai et al. (2006). Bottom : Overlay of
Chandra
X-ray (blue; same as figure 2) and radio 6 cmcontours (red) on the optical data. Contour levels of radio 6 cm flux are 0.37, 0.93, 1.49, 2.05, 2.61 mJy beam − .
28 – C E N5’’C13 S N S C S1 J A L K IH G F ED CB E N5’’
Fig. 5.— Similar to figure 4, except that the background image is
HST H a . 29 –Table 2. Comparison of luminosities of the 3 brightest X-ray sources in IC342 in differentobservations Luminosity in 0.1-10 keV (10 erg s − )Source Chandra XMM-Newton ROSAT (2006 April 2) (2001 February 11) (1991 February 13)(1) (2) (3) (4)C3 18 . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . s uncertainties of the X-ray sources in observations by Chandra (this paper),
XMM-Newton , and