THE MASSIVE AND DISTANT CLUSTERS OF WISE SURVEY: MOO J1142+1527, A 10¹⁵ M_⊙ GALAXY CLUSTER AT z = 1.19

In the past few years we have entered a new era of wide-area surveys capable of detecting galaxy clusters at z > 1. The previous generation of high-redshift cluster searches was the first to yield large samples of galaxy clusters at this epoch (e.g., Gladders & Yee 2005; Eisenhardt et al. 2008; Muzzin et al. 2009; Fassbender et al. 2011); however, these programs typically probed less than 100 deg². Consequently, while these surveys have been effective in generating statistical samples of distant galaxy clusters, they have lacked the comoving volume to discover significant numbers of massive clusters (M_500c ≳ 3 × 10¹⁴ M_⊙). The high mass tail of the galaxy cluster population is of significant interest for both galaxy evolution and cosmology. One open question is the extent to which the star formation, active galactic nuclei (AGNs) activity, and assembly histories of cluster galaxies depend upon the mass of the cluster in which they reside (e.g., Brodwin et al. 2013; Ehlert et al. 2015; Ma et al. 2015). For this science, samples of high-mass clusters close to the epochs of assembly and star formation, coupled with existing lower mass samples, provide the necessary dynamical range to quantitatively address this question. For cosmology, massive, high-redshift clusters remain competitive probes of dark energy via a number of methods (e.g., Allen et al. 2011), including evolution in the cluster mass function (Vikhlinin et al. 2009; Bocquet et al. 2015), the clustering of galaxy clusters (e.g., Sereno et al. 2015), and through application of the f_gas test (Mantz et al. 2014). The high mass tail of the galaxy cluster mass function is also a sensitive indicator of primordial non-Gaussianity (Chen 2010; Williamson et al. 2011; Shandera et al. 2013).

In the last several years, the South Pole Telescope (SPT) and Atacama Cosmology Telescope (ACT) have each completed wide-area millimeter surveys to identify galaxy clusters via the Sunyaev–Zel'dovich effect, publishing cluster catalogs drawn from 2500 deg² for the SPT survey (Bleem et al. 2015) and 504 deg² for the ACT survey (Hasselfield et al. 2013). Together, these programs have published nearly 50 massive clusters at z > 1. The upcoming generation of optical, galaxy-based cluster searches will also extend into the wide-area, high-redshift region of parameter space, complementing these millimeter surveys. When complete, the Dark Energy Survey (Flaugher 2005; Sánchez & DES Collaboration 2010) is expected to result in a cluster catalog extending to z ∼ 1, covering a ∼5000 deg² footprint that includes much of the SPT and ACT survey areas.

The Massive and Distant Clusters of WISE Survey (MaDCoWS), which is designed to detect the most massive galaxy clusters at z ≈ 1, offers the largest survey area among current high-redshift cluster searches. The first phase of MaDCoWS covered ∼10,000 deg² within the SDSS footprint; subsequent phases of the program are now extending the search over the full extragalactic sky. In previous papers we presented the first cluster discovered in this survey (Gettings et al. 2012), the redshift distribution of the first 20 clusters (0.75 < z < 1.3, Stanford et al. 2014), and Sunyaev–Zel'dovich masses for five clusters (Brodwin et al. 2015). In this paper we present the discovery and confirmation of the most massive cluster yet identified within the MaDCoWS catalog, which is among the ∼5 most massive clusters expected to exist over the entire sky at z ≳ 1.19. Throughout the paper we use Vega-based magnitudes and assume a WMAP9 cosmology (H₀ = 69.7 km s⁻¹, Ω_m = 0.2821, Ω_Λ = 0.7181, σ₈ = 0.817, n_s = 0.9646; Hinshaw et al. 2013) unless otherwise specified. For cluster masses and radii we include a c or m subscript to denote whether the values are relative to the critical or mean density.

MaDCoWS is a WISE-based (Wright et al. 2010) search for galaxy clusters at z ≃ 1 that employs color and magnitude selection to identify massive galaxies at z ≳ 0.75, and then uses a wavelet technique to detect galaxy overdensities. A key element of this search approach is the combination of the WISE data with uniform optical photometry. The initial detection of Massive Overdense Object (MOO) J1142+1527 used the WISE All-Sky Data Release (Cutri et al. 2012) and SDSS DR8 (Aihara et al. 2011) to identify candidates within the footprint of the SDSS. In this WISE+SDSS MaDCoWS search, MOO J1142+1527 was identified as one of the 200 highest significance cluster candidates.

We have subsequently refined the search algorithm and transitioned to use of the AllWISE Data Release (Cutri et al. 2013). A detailed description of the MaDCoWS survey and detection algorithm will be provided in a forthcoming paper. Briefly, cluster candidates in the current AllWISE+SDSS search were detected as overdensities of sources with W1 < 16.9, W1–W2 > 0.2, and i_AB > 21.3. The W1 magnitude cut, which corresponds to the 5σ AllWISE limit on the ecliptic, is imposed to maintain uniform selection. The color and i_AB cuts together minimize contamination from sources at z < 0.8. MOO J1142+1527 remains the twenty-seventh highest significance candidate in this more recent, refined version of the catalog, with a position (α, δ) = (11:42:43.9, 15:27:07). In the left panel of Figure 1 we show a WISE [3.6] cutout of the cluster field. The red squares in this panel denote the galaxies that passed the color, magnitude, and quality cuts in this search, highlighting the detected overdensity. Because the WISE magnitude limit and blending of sources in the WISE data result in detection significance being a high scatter richness measure, we have obtained Spitzer/IRAC observations to determine more robust richness estimates.

Zoom In Zoom Out Reset image size

In Spitzer Cycle 9 we were awarded 37.9 hr to obtain IRAC 3.6 μm and 4.5 μm imaging of the 200 highest significance overdensities from our All-Sky search (Program ID 90177; PI Gonzalez). For each cluster the total exposure times were 180 s in each band, obtained using 30 s frame times and 6 positions in a medium scale cycling dither pattern. This exposure time was designed to reach a nominal 5σ depth of 6 μJy (18.7 mag) at [4.5], which is sufficient to identify galaxies more than one magnitude below L* up to z ≃ 1.5.

We reduced and mosaicked the basic calibrated data using the MOPEX package (Makovoz & Khan 2005) and resampled to a pixel scale of 06. The MOPEX outlier (e.g., cosmic ray, bad pixel) rejection was optimized for the regions of deepest coverage in the center of the maps corresponding to the position of the MaDCoWS detection.

We ran SExtractor (Bertin & Arnouts 1996) in dual image mode for source detection and photometry, using the [4.5] frame as the detection image and adopting IRAC-optimized SExtractor parameters from Lacy et al. (2005). Flux densities were measured in 4'' diameter apertures. Following Wylezalek et al. (2013), we then applied aperture corrections to the [3.6] and [4.5] flux densities (factors of 1.42 and 1.45, respectively). We determined a 95% completeness limit of 10 μJy, corresponding to limiting magnitude of [4.5] = 18.2, by comparing number counts to deeper photometry from the Spitzer UKIDDS Ultra Deep Survey (SpUDS) as in Wylezalek et al. (2013, 2014). This completeness limit is adopted as the flux density cut in all subsequent analysis. We show the central 35 × 35 of the IRAC [3.6] image in the right panel of Figure 1. As in the left panel, the red squares denote the positions of WISE sources that contributed to detection of the cluster. In some cases, the initial WISE source resolves into multiple galaxies with IRAC.

To prioritize follow-up of our Cycle 9 IRAC targets, we defined a simple richness estimator based upon the overdensity of galaxies with red [3.6]–[4.5] color within a fixed angular radius. Specifically, we defined the richness as the number of galaxies with [3.6]–[4.5] > 0.1 and [4.5] < 18.2 within 1' of the cluster position measured in the MaDCoWS search. We note that this IRAC color is relatively insensitive to current star formation, selecting both passive and star-forming galaxies in distant clusters. By this measure, MOO J1142+1527 has a richness of 64, which is the ninth highest among the 200 Cycle 9 targets.

Figure 2 shows the Spitzer [3.6]–[4.5] color–magnitude diagram for galaxies that lie within 1' of the SZ centroid (see Section 4). These galaxies correspond to the central overdensity of red sources shown in Figure 1. The median color of these galaxies is [3.6]–[4.5] = 0.3, which for a Bruzual & Charlot (2003) passively evolving stellar population corresponds to a galaxy at z ≃ 1.2. We also highlight the spectroscopically confirmed members (green stars), four of which lie within 1' of the SZ centroid, and the non-members (red crosses), which are described in greater detail in the next section.

Zoom In Zoom Out Reset image size

3.1. Redshift Determination

MOO J1142+1527 was observed with the Combined Array for Research in Millimeter-wave Astronomy (CARMA)²⁰ for approximately 5 hr on-source beginning on UT 2014 July 03. The data are centered around a frequency of 31 GHz. For these observations the array was in its most compact "E+SH" configuration. All 23 antennas were correlated across 2 GHz of bandwidth using the CARMA "spectral line" correlator. To maximize sensitivity to the SZ signal, the "wideband" correlator processed 7.5 GHz of bandwidth for the innermost eight 6.1-m antennas. CARMA is optimized for the detection of distant clusters via their SZ signatures in this array and correlator configuration. The data from these baselines achieve a sensitivity of 1.2 mJy per ∼50'' × 90'' beam. The gain calibrator J1224+213 was observed for 3 minutes between 15-minute target observations, and the absolute calibration is derived from Mars via the model of Rudy et al. (1987). Figure 3 (left) shows a CLEAN-deconvolved (Högbom 1974) image of the cluster using all baselines with a Gaussian taper to 10% at 4 kλ, after removal of a point source (see below). The color scale is in units of SNR per beam for this tapered image, in which the noise per beam is 90 μJy (80 μK).

Zoom In Zoom Out Reset image size

Cluster properties were determined using a Markov Chain Monte Carlo method to simultaneously fit an Arnaud et al. (2010) pressure profile model and point source models to the unflagged data. The single point source in the field, NVSS J114247+152711 (Condon et al. 1998), was located using the higher-resolution, long baseline data from the spectral line correlator. This point source was found to have a flux density of 3.4 mJy, and is coincident with the brightest candidate cluster member (purple circle in Figures 2 and 3), but was not targeted in our spectroscopic program. For this fit the reduced χ² = 1.0085 for 44182 degrees of freedom. We next fit a model consisting of only the point source but no cluster. The Δχ² between these two models yields a 13.2σ cluster detection significance. The centroid of the SZ decrement is located at (α, δ) = (11:42:46.6, +15:27:15), with uncertainties (σ_α, σ_δ) = (44, 30). The SZ centroid and WISE position, which are separated by 41'', bracket the peak of the galaxy distribution. This separation, which corresponds to ∼345 kpc, is somewhat larger than the typical separations found in Brodwin et al. (2015), for which the largest separation was 282 kpc. The offset between the peak of the red galaxy distribution in Spitzer and the WISE position corresponds to an offset of less than two pixels in the WISE detection map, and is consistent with expected uncertainties. The BCG lies close to the SZ centroid, but offset from the peak of the galaxy distribution, suggesting that there may also be a physical component to the observed separation between the WISE and SZ centroids.

The combined fit of cluster and point source models gives the spherically integrated Comptonization parameter, ${Y}_{500c}=(9.7\pm 1.3)\times {10}^{-5}$ Mpc². We used the Andersson et al. (2011) M_500c–Y₅₀₀ scaling relation to determine mutually consistent values of M_500c and r_500c and the associated uncertainties, where M_500c$\;=\;(4\pi {r}_{500}^{3}/3)\times (500{\rho }_{c}).$ This procedure results in a cluster mass and radius of M_500c = (6.0 ± 0.9) × 10¹⁴M_⊙ and r_500c = 0.83 ± 0.04 Mpc, respectively. The quoted uncertainties are derived by combining in quadrature the propagated uncertainty and a 12% intrinsic scatter in M_500c at fixed Y_500c from Andersson et al. (2011). For the Duffy et al. (2008) mass–concentration relation, the derived mass corresponds to M_200c = (9.9 ± 1.5) × 10¹⁴ M_⊙, or M_200m = (1.1 ± 0.2) × 10¹⁵ M_⊙.

In this paper we have presented confirmation of a massive galaxy cluster at z = 1.19. Originally identified by the MaDCoWS project, the cluster MOO J1142+1527 has a mass of M_500c = (6.0 ± 0.9) × 10¹⁴M_⊙ [M_200m = (1.1 ± 0.2) × 10¹⁵M_⊙], making it the most massive confirmed galaxy cluster at z ≥ 1.15 identified by any technique. Figure 4 illustrates the position of this cluster in the mass—redshift plane compared to a selection of recent wide-area cluster surveys. The solid black curve in this figure is a curve of constant co-moving number density for a Tinker et al. (2008) mass function, highlighting that there are few clusters over this entire redshift interval as rare as MOO J1142+1527. The only more massive cluster known at z > 1 is SPT-CL J2106–5844 (z = 1.13, M_200m$\;=\;(1.27\pm 0.21)\times {10}^{15}{h}_{70}^{-1}$ M_⊙; Foley et al. 2011). We also include in this figure IDCS J1426.5+3508 (z = 1.75), as it is the closest progenitor analog to MOO J1142+1527 at z > 1.5.

Zoom In Zoom Out Reset image size

The existence of MOO J1142+1527 is not in tension with the ΛCDM paradigm, but such clusters are expected to be extremely rare. We use the halo mass function code hmf from Murray et al. (2013)²¹ with a Tinker et al. (2008) mass function to calculate the expected number of such clusters. For WMAP9 and Planck (Planck Collaboration et al. 2014a) cosmologies, there are predicted to only be ∼3 or ∼7 clusters this massive over the full sky at z ≳ 1.19, respectively, and only ∼1–2 within our SDSS survey area. The discovery of this cluster highlights the potential of wide area cluster surveys like MaDCoWS to identify such extreme systems, which are natural targets for a range of cosmological and evolutionary investigations. Our ongoing Cycle 11 Spitzer program (PID 11080, PI Gonzalez), which targets ∼1750 additional MaDCoWS candidates drawn from the full extragalactic sky, promises to enable construction of a sample of comparably rich galaxy clusters at this epoch.

We thank the anonymous referee for comments that improved the quality of this paper. Support for this research was provided by NASA through Spitzer GO program 90177, ADAP grant NNX12AE15G, and NASA Exoplanet Science Institute grants 1461527 and 1486927. The work by SAS at LLNL was performed under the auspices of the U. S. Department of Energy under Contract No. W-7405-ENG-48.

Support for CARMA construction was derived from the Gordon and Betty Moore Foundation; the Kenneth T. and Eileen L. Norris Foundation; the James S. McDonnell Foundation; the Associates of the California Institute of Technology; the University of Chicago; the states of California, Illinois, and Maryland; and the National Science Foundation. CARMA development and operations were supported by NSF under a cooperative agreement and by the CARMA partner universities; the work at Chicago was supported by NSF grant AST-1140019. Additional support was provided by PHY-0114422. This publication makes use of data products from the Wide-field Infrared Survey Explorer, a joint project of the University of California, Los Angeles, and the Jet Propulsion Laboratory/California Institute of Technology, funded by NASA. This work is based in part on observations made with the Spitzer Space Telescope, which is operated by the Jet Propulsion Laboratory, California Institute of Technology under a contract with NASA. This work is based in part on data obtained at the W. M. Keck and Gemini Observatories. The authors wish to recognize and acknowledge the very significant cultural role and reverence that the summit of Mauna Kea has always had within the indigenous Hawaiian community. We are most fortunate to have the opportunity to conduct observations from this mountain.

Facilities: WISE - Wide-field Infrared Survey Explorer, Spitzer (IRAC) - Spitzer Space Telescope satellite, CARMA - Combined Array for Research in Millimeter-Wave Astronomy, Keck:I (MOSFIRE) - , Keck:II (DEIMOS) - , Gemini:Gillett (GMOS) - .