Suzaku X-ray Follow-up Observation of Weak-lensing-detected Halos in the Field around ZwCl0823.2+0425 Eri Watanabe1, Motokazu Takizawa2, Kazuhiro Nakazawa3, Nobuhiro Okabe4,5, 1 Madoka Kawaharada6, Arif Babul7, Alexis Finoguenov8,9, Graham P. Smith10, James 1 0 E. Taylor11 2 1School of Science and Engineering, Yamagata University, Kojirakawa-machi 1-4-12, Yamagata n a 990-8560 J 6 2Department of Physics, Yamagata University, Kojirakawa-machi 1-4-12, Yamagata 990-8560 [email protected] ] O 3Department of Physics, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033 C 4Institute of Astronomy and Astrophysics, Academia Sinica, P.O.Box 23-141, Taipei 106, Taiwan . h p 5Astronomical Institute, Tohoku University, Aramaki, Aoba-ku, Sendai, 980-8578 o- 6Institute of Space and Astronautical Science/JAXA, 3-1-1 Yoshinodai, Chuo-ku, Sagamihara, r t Kanagawa 252-5210 s a 7Department of Physics and Astronomy, University of Victoria, Victoria, BC, V8P 1A1, Canada [ 8Max-Planck-Institut fu¨r extraterrestrische Physik, Giessenbachstraße, 85748 Garching, Germany 1 v 9University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250, USA 3 3 10School of Physics and Astronomy, University of Birmingham, Edgbaston, Birmingham, B15 2TT, 3 UK 1 1. 11Department of Physics and Astronomy, University of Waterloo, 200 University Avenue West, 0 Waterloo, ON N2L 3G1, Canada 1 1 : v (Received 2010 August 20; accepted 2010 December 27) i X Abstract r a We present the results of Suzaku X-ray follow-up observation of weak-lensing- detected halos in the field around galaxy cluster ZwCl0823.2+0425. We clearly detected X-ray emission associated with most of these halos and determined their detailed physical parameters such as X-ray luminosity, temperature, and metal abun- dance, for the first time. We find that the X-ray luminosity - temperature relation for these halos agrees with former typical results. With mass determined from the weak gravitational lensing data, the mass-temperature relationfor them is also investigated and found to be consistent with the prediction from a simple self-similar model and results of the previous studies with both lensing and X-ray data. We would like to 1 emphasize that the self-similar scaling relation of mass and temperature is shown here for the first time using a weak-lensing selected sample, whereas previous stud- ies of the mass scaling relation used X-ray-selected samples of clusters. Therefore, our study demonstrates importance of X-ray follow-up observations of shear-selected clusters, and shows that a joint X-ray and lensing analysis will be crucial for clusters discovered by the forthcoming weak-lensing surveys, such as the one planned with Subaru/Hyper-Suprime-Cam. Key words: galaxies: clusters: individual (ZwCl0823.2+0425) — X-rays: galax- ies: clusters — gravitational lensing 1. Introduction In the standard theory of structure formation in the cold dark matter (CDM) universe, larger structures such as dark halos corresponding to rich clusters of galaxies form through absorption of smaller dark halos and accretion of surrounding matter, which preferably occur along filaments associated with large scale structures in the universe. Although baryonic com- ponents such as diffuse hot gas and galaxies are visible in X-ray and optical bands, dark matter, which is a dominant component in mass, cannot be directly seen in any kind of radiation, and we recognized its presence only indirectly through dynamical properties of galaxies and hot gas. In recent years, however, this situation has been changed by the development of gravita- tional lensing techniques, which enable us to reveal matter distribution in the universe directly. We do not require any assumptions of the dynamical status and mass-luminosity ratio of the system to get the mass distribution (Schneider et al. 2006). Whereas strong lensing technique is essentially model dependent, weak gravitational lensing enables us to obtain mass distribution in the plane of the sky without any geometrical assumptions and measures directly galaxy clus- ter mass distribution out to the virial radius (Kaiser & Squires 1993; Bartelmann & Schneider 2001). However, it is true that we cannot get detailed properties of baryonic components asso- ciated with dark matter halos by gravitational lensing alone. Therefore, joint studies of X-ray, optical and weak lensing datasets are very useful and provide us with important information of the interplay between baryonic and dark matter (Okabe & Umetsu 2008; Kawaharada et al. 2010; Umetsu et al. 2010). Inthe very near future, largesurveys invariouswavebands (e.g.Subaru/Hyper-Suprime- Cam, eROSITA, SPT, and ACT) will provide us with huge datasets of galaxy clusters. In such future surveys, the weak gravitational lensing technique is one of the most powerful methods to find galaxy clusters, irrespective of the physical status of the baryonic components (Wittman et al. 2001; Miyazaki et al. 2002b; Miyazaki et al. 2007). Its detection efficiency of clusters/groups is a function of masses as well as the geometry of the universe. A typical procedure is as follows; 2 the lensing convergence field is reconstructed from galaxy’s ellipticities convolved with a suit- able filter (Kaiser & Squires 1993; Seitz & Schneider 2001), and lensing signals above the noise of intrinsic ellipticities are identified as mass structures (Wittman et al. 2001; Miyazaki et al. 2002b; Miyazaki et al. 2007). However, there are still some uncertainties in this method such as contaminations of member galaxies in shear catalogs, spurious peaks induced by intrinsic ellipticities, and the projection effects due to superpositions of structures at different redshifts (Miyazaki et al. 2002b; Hamana et al. 2009; Okabe et al. 2010a). Owing to not only these uncertainties characteristic of the weak lensing surveys but also the systematic difference of detection efficiency among weak lensing, X-ray, and Sunyaev-Zel’dovich effect (SZE) surveys, it is still unclear whether halo candidates detected by weak lensing always contain hot baryons. Thus, X-ray and/or SZE follow-up observations of halo candidates discovered by weak lens- ing analysis are very important as a pilot study before forthcoming large surveys in order to understand systematic differences among different wavelength surveys and the correlations of physical properties between hot baryons and total mass. In particular, a study of the scaling relations between lensing mass and baryonic observables is of prime importance for the cluster cosmology as well as understanding cluster baryonic evolution, such as radiative cooling, en- ergy feedback, and cluster mergers (Stanek et al. 2010; Vikhlinin et al. 2009a; Vikhlinin et al. 2009b; Okabe et al. 2010c). We conducted Suzaku X-ray follow-up observation of weak-lensing-detected halo can- didates. X-ray Imaging Spectrometer (XIS) aboard Suzaku satellite (Mitsuda et al. 2007) is suitable for observing low surface brightness diffuse sources thanks to its low and stable back- ground (Koyama et al. 2007). Indeed, physical properties of the intracluster medium (ICM) around the virial radius, where X-ray emission is too faint for Chandra and XMM-Newton to get meaningful results, have been investigated for several clusters (Fujita et al. 2008; Bautz et al. 2009; George et al. 2009; Reiprich et al. 2009; Kawaharada et al. 2010; Hoshino et al. 2010). This feature is also very useful to investigate diffuse X-ray emission associated with mass structures, even less massive objects, found by weak lensing surveys. As a pilot study for the above-mentioned purpose, we select a field around galaxy cluster ZwCl0823.2+0425. Several dark halos are found around the cluster by weak lensing study (Okabe et al. 2010b) as a collaboration of the Local Cluster Substructure Survey (LoCuSS; PI: Graham P. Smith). Figure 1 shows optical image of the field overlaid with the mass contours derived from the weak lensing analysis of Okabe et al. (2010b). Besides the central halo corresponding to the ZwCl0823.2+0425, four significant mass clumps areseen inthe north, north-west, north-east, and south-east of it. We refer them as ZwCl0823.2+0425 (or C), N, NW, NE, and SE hereafter. The mass peak corresponding to Abell 664 is also found near the boundary. Two major red-sequence galaxy populations are found in this field, whose spatial distributions of both optical luminosity and number density are pretty similar to that of mass (Okabe et al. 2010b). The galaxies in the red-sequence at the lower redshift are apparently 3 associated with the C and NW halos. On the other hand, those in the higher redshift red- sequence seem to be related with the N and NE halos. The SDSS spectroscopic data are available for a few galaxies located in each halo center. We found that the mean redshift of galaxies associated with the C and NW halos is z = 0.2248, and that of N and NE halos is z=0.472. The SE halo is likely to be associated with a bright galaxy at z=0.103. However, a few galaxies in the background possibly contribute the lensing signal. This possible projection effect will be discussed in section 5. Although ZwCl0823.2+0425 is recognized as an X-ray source with the flux of 1.8×10−12 erg s−1 cm−2 in Bright Source Catalog of ROSAT All Sky Survey (Ebeling et al. 1998; Ebeling et al. 2000; B¨ohringer et al. 2004), its detailed spectral properties are not known. Moreover, X-ray emission associated with the other halos are not clearly detected. Therefore, deeper X-ray observation for this field is highly desirable to explore physical properties of the hot gas associated with the halos. In this paper, we present Suzaku X-ray follow-up observation of the field around ZwCl0823.2+0425 to investigate the physical properties of the hot gas in the dark matter halos found via gravitational lensing analysis. The rest of this paper is organized as follows. In section 2 we describe the observation and data reduction. In section 3 we present spectral analysis results. In section 4 we describe the imaging simulation to check the contamination of each halo’s spectrum from the others. In section 5 we discuss the results and their implica- tions. In section 6 we summarize the results. Canonical cosmological parameters of H = 70 0 Mpc−1 km s−1, Ω = 0.27, and Λ = 0.73 are used in this paper. Unless otherwise stated, all 0 0 uncertainties are given at the 90% confidence level. 2. Observation and Data Reduction We observed the field around ZwCl0823.2+0425 with Suzaku on 2008 May 17-18. The field of view (FOV) of Suzaku XIS is shown in a Subaru optical image overlaid with the mass contours derived from the weak lensing analysis (Okabe et al. 2010b) in figure 1. The observation was performed at XIS nominal pointing. The XIS was operated in the normal full-frame clocking mode. The edit mode was 3×3 and 5×5, and we used combined data of both modes. The spaced-row charge injection was adopted for XIS. All data were processed with Suzaku pipeline processing, version 2.2. We employed calibration data files (20090925). TheXISdatawereprocessedthroughstandardcriteriaasfollows. EventswithaGRADE of 0, 2, 3, 4, 6 and STATUS with 0:524287 were extracted. We excluded data obtained at the South Atlantic Anomaly (SAA), within 436s after the passage of SAA, and at low elevation angles from an Earth rim of < 5◦ and a Sun-lit Earth rim of < 20◦. As a result, effective exposure time was 41.3 ks. Non X-ray background (NXB) spectra and images of XIS were generated using the ftool “xisnxbgen” (Tawa et al. 2008). Figure 2 represents an 0.5-8.0 keV XIS image combined from those of the front illuminated (FI) CCDs (XIS0, XIS3), overlaid with the mass contours. The image was corrected for exposure and vignetting effects after 4 Fig. 1. OpticalimageofthefieldaroundZwCl0823.2+0425overlaidwiththemasscontoursderivedfrom the weak lensing analysis. The field of view of Suzaku XIS is also shownin a black square. In addition to the centralmass peak correspondingto ZwCl0823.2+0425,four mass peaks areseenaroundit. Hereafter, we refer them as C, N, NW, NE, and SE, as shown in this figure. The mass peak corresponding to Abell 664 is also found outside of the XIS field of view. The contours are spaced in a unit of 1σ reconstruction error, δκ=0.02243. subtracting NXB, and smoothed by a Gaussian kernel with σ=0′.17. Enhanced X-ray emission from the ZwCl0823.2+0425, N and NE halo regions are clearly apparent in figure 2. 3. Spectral Analysis ForaspectralanalysisoftheXISdata,redistributionmatrixfiles(RMFs)weregenerated using the ftool “xisrmfgen”. In addition, ancillary response files (ARFs), which describe the response of X-Ray Telescope aboard Suzaku and the amount of the XIS optical blocking filters contamination, were generated with the ftool “xissimarfgen” (Ishisaki et al. 2007). XIS spectra for each sensor (XIS0, XIS1, XIS3) were fitted simultaneously. For the spectral fitting, we used the energy band of 0.5 – 10.0 keV and 0.5 – 8.0 keV for FI CCDs (XIS0 and XIS3) and back illuminated (BI) one (XIS1), respectively. In the spectral fitting with Galactic absorption, we assumed N =3.18×1020 cm−2 (Dickey & Lockman 1990). The CXB level was estimated in the H same way as Nakazawa et al. (2009) from the Lockman hole observation (Suzaku observation ID, 101002010). The detailed procedure is described in appendix 1 of Nakazawa et al. (2009). We defined the photon flux model as N(E) =9.19×10−4×E−1.4 in photons cm−2 s−1 keV−1 FOV−1, where E is the photon energy in keV. 3.1. The Background Model Before entering the spectral analysis of each halo, we need to construct a background model fromastrophysical origininaddition toNXB. As shown in figure3, we eliminated regions corresponding to each halo and the foreground bright star near the west edge of the XIS FOV, 5 Fig. 2. XIS image of the field around ZwCl0823.2+0425 in the 0.5-8.0 keV band combined from XIS FI CCD images, overlaid with the mass contours. The image was corrected for exposure and vignetting effects after subtracting NXB, and smoothed by a Gaussian kernel with σ=0′17. the rest of which was used for spectral fit to determine the background model. For the analysis of the background model, uniform emission over a circular region with 20’ radius was used as an input image to generate an ARF. The Cosmic X-ray background (CXB), the Galactic halo’s hot gas (GH), and the local hot bubble (LHB) were considered as the background components. Then, the spectrum of the background region was fitted by a model as follows, apec1+wabs×(apec2+apec3+powerlaw) (1) where apec1, apec2+apec3, and powerlaw represent the LHB, GH, and CXB, respectively. The temperature of the LHB was fixed to be 0.08 keV. The metal abundance of the both LHB and GH were also fixed to be solar. The spectrum of the background region fitted with the above-mentioned model is shown in figure 4, where the black, red, and green crosses show the spectrum of XIS0, XIS1, and XIS3, respectively. The total and each component of the best fit model spectra are also plotted as solid and dashed histograms, respectively. The detailed results of the fit are summarized in table 1. At first glance, the higher temperature component of GH may seem to be too high. We tried to fit the background region spectrum by a model with a single temperature GH. However, the results were not so good and we had some residuals around ∼2 keV in the data. This might be because our spectral modeling is too simple for the data. However, our main purpose here is to construct a plausible background spectrum model to investigate the halos’ hot gas. Thus, we do not pursue this issue in more detail. 6 Fig. 3. A Region used to determine the background model (yellow field). We eliminated regions corre- sponding to each halo and the foreground bright star near the west edge of the XIS FOV (green circles and ellipses). Fig. 4. XIS spectrum of the region shown in figure 3 fitted with the background model described in the text. The black, red, and green crosses show the spectrum of XIS0, XIS1, and XIS3, respectively. The total and each component of the best fit model spectra are also plotted as solid and dashed histograms, respectively. 7 Table 1. BestfitparametersfortheXISspectrumofthebackgroundregion. The background model kT ∗ 0.08(fixed) LHB N † 1.19+0.35×10−2 LHB −0.40 kT ∗ 0.34+0.24 GH,low −0.0542 N † 5.76+1.90×10−4 GH,low −2.24 kT ∗ 1.71+0.45 GH,high −0.38 N † 5.18+1.73×10−4 GH,high −2.11 Γ ‡ 1.4(fixed) PL N § 9.19×10−4(fixed) PL χ2/d.o.f. 403.5/336 ∗ Temperatureoftheeachcomponent inkeV. † Normalizationintheapeccodeforeachcomponent. ‡ Photonindexofthepower-lawcomponent. § Normalizationinthepower-lawcomponent. 3.2. Hot Gas in Each Halo Weperformed spectral analysis foreach halousing thebackground model inthe previous subsection. Figure 5 shows regions used in the analysis of each halos, where the center of each region was determined through the each halo’s mass peak in the weak lensing data of Okabe et al. (2010b). The radii of the regions are 2.5’, 2.0’, 1.8’, 1.8’, and 2.0’ for C, N, NE, NW, and SE, respectively. It is true that these radii should be as large as the expected virial radii of the corresponding halo. However, because each halo partially overlaps with one another and the spatial resolution of Suzaku is moderate, the radii have to be small enough to reduce contamination from the others. No core removal was made in the spectral extraction. We adopted model images of the β-model profile with core radius 0.2 Mpc and β = 0.66 at each halo’s redshift as an input image to generate ARFs, considering that Suzaku does not have spatial resolution good enough to resolve inner structures of each halo and that there is no X-ray image of this field deep enough for our purpose by other instruments. Systematic errors of both CXB and NXB were taken into account in the following anal- ysis. It is well-known that the CXB fluctuations can be modeled as σ /I ∝ Ω−0.5S0.25, CXB CXB e c where Ω and S are the effective solid angle and upper cutoff flux of a point source, respec- e c tively. From the HEAO-1 A2 results, σ /I =2.8% with Ω =15.8 deg2 and S =8×10−11 CXB CXB e c erg s−1 cm−2 (Shafer 1983). We adopted upper cutoff flux S ∼1.0×10−14 erg s−1 cm−2 taking c into account the fact that point sources with similar flux were clearly detected in the Lockman hole observation (Suzaku observation ID, 101002010). Thus, the CXB fluctuations for C, N, NE, NW are expected to be 26, 33, 36, and 36 % at the 90% confidence level. Tawa et al. 8 Fig. 5. Regions used in the spectral analysis of each halo. (2008) reported that reproducibility of NXB was 6.0% and 12.5% for XIS FI and BI at the 90% confidence level, respectively. We did not take account of systematic errors of LHB and GH, considering that Yoshino et al. (2009) reported that prominent spatial fluctuations in a scale smaller than that of XIS FOV were hardly seen in LHB and GH. In the spectrum analysis, NXB and CXB components were fluctuated at the 90 % confidence level of the systematic uncertainty mentioned above, which caused the changes of the best-fit parameters in the fits and gave us the systematic errors. Each halo’s spectrum was fitted by the photoabsorbed single temperature apec model (wabs×apec) with the background model described in the previous subsection. The redshift of each halo was fixed to be a value obtained from the SDSS data. We fixed the metal abundance for NE, NW, and SW to be 0.3 solar because we do not have enough photons for these regions to determine it. Figure 6, 7, 8, 9, and 10 represent the XIS spectrum fitted with the above- mentioned model of C, N, NE, NW, and SE, respectively. In each figure, the black, red, and green crosses show the spectrum of XIS0, XIS1, and XIS3, respectively. The best fit model spectra are also plotted as solid histograms. In general, all data except for SE are fitted well by the adopted model. Unfortunately, however, the statics of the SE spectrum is not good enough to obtain meaningful fitting results. This is consistent with the fact that SE lensing signal in the mass map is due to the projection effect, which is mainly associated with a galaxy at low redshift z ∼0.1, as described in section 5. The detailed fitting results are presented in table 2. The luminosity in 0.5−10.0 keV of each halo is calculated from the fitting results and summarized in table 3. We used model images to generate ARFs because Suzaku has only moderate spatial resolution (∼ 2′), and because we do not have appropriate X-ray images with better spatial resolution by other instrument for this field. However, it should be noted that an effective area 9 Fig. 6. XIS spectrum of the ZwCl0823.2+0425 fitted with the model described in the text. The black, red, and green crosses show the spectrum of XIS0, XIS1, and XIS3, respectively. The best fit model spectra are plotted as solid histograms. Fig. 7. Same as figure 6, but for the N halo. Fig. 8. Same as figure 6, but for the NE halo. 10