See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/41708749
Identification and study of systems of galaxies in the Shapley supercluster
Article in Astronomy and Astrophysics · January 2006 DOI: 10.1051/0004-6361:20053623
CITATIONS READS 11 41
5 authors, including:
Andreas Reisenegger Hernan Quintana Pontifical Catholic University of Chile Pontifical Catholic University of Chile
111 PUBLICATIONS 1,691 CITATIONS 226 PUBLICATIONS 5,059 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Abell and Group Galaxy Studies View project
PUC Astronomy Group - Cluster of galaxies View project
All content following this page was uploaded by Hernan Quintana on 20 July 2014.
The user has requested enhancement of the downloaded file. A&A 445, 819–825 (2006) Astronomy DOI: 10.1051/0004-6361:20053623 & c ESO 2006 Astrophysics
Identification and study of systems of galaxies in the Shapley supercluster
C. J. Ragone1,2, H. Muriel1,2, D. Proust3, A. Reisenegger4, and H. Quintana4
1 Grupo de Investigaciones en Astronomía Teórica y Experimental, IATE, Observatorio Astronómico, Laprida 854, Córdoba, Argentina 2 Consejo de Investigaciones Científicas y Técnicas de la República Argentina e-mail: [cin;hernan]@oac.uncor.edu 3 GEPI, Observatoire de Paris-Meudon, 92195 Meudon Cedex, France e-mail: [email protected] 4 Departamento de Astronomía y Astrofísica, Pontificia Universidad Católica de Chile, Casilla 306, Santiago 22, Chile e-mail: [areisene;hquintana]@astro.puc.cl
Received 13 June 2005 / Accepted 7 September 2005
ABSTRACT
Based on the largest compilation of galaxies with redshift in the region of the Shapley Supercluster (Proust et al. 2004, P2004), we identified 122 galaxy systems, 60 of which are new systems. Using the SuperCOSMOS catalogue, we have assigned b j magnitudes to each galaxy in our 13 15 −1 compilation. The sample of galaxy systems was used to estimate the mass function of systems in the range 10 to 10 M h . We computed a lower value to the total mass in the region of the Shapley Supercluster with this mass function. Using 15 mock catalogues we derived the mean mass that these kinds of systems have before comparing it with that obtained from the real data.
Key words. galaxies: clusters: general
1. Introduction representing an overdensity range ρ/ρc ∼ 3−20. In this work we explore an alternative way to estimate a lower limit to the to- The Shapley Supercluster (hereafter SSC) has the highest con- tal mass of the SSC. In a high mass concentration like the SSC, centration of clusters and groups of galaxies in the local uni- a significant fraction of galaxies are supposed to belong to verse. It is centrally dominated by three Abell clusters (A3556, groups or clusters of galaxies and, therefore, the mass func- A3558, and A3562) and contains more than 20 Abell clusters. tion of these systems can be used to provide a lower limit to ∼ This supercluster has an extent of 20 Mpc, or even larger, and the total mass of the supercluster. In order to select a sample has not yet virialized. Kull & Böhringer (1999) analyzed a re- of galaxy systems that is as complete as possible, we used the × gion of 6 3 degrees and found evidence of X-ray emission resent catalogue by Proust et al. (2004, hereafter P2004). These connecting the three central clusters of the SSC, indicating that authors compiled the largest velocity catalogue in the region of the inter-cluster gas is being warmed up. Therefore, the core of the SSC. It contains radial velocities for 8341 galaxies and rep- the supercluster is in the procces of virialization. resents a powerful tool for identifying and studying the galaxy A mass estimate of the SSC is particularly interesting due systems in the SSC. to its proximity to the local group. Moreover, it can be used to constrain theoretical models that should be able to predict this Large redshift surveys like 2dFGRS (2 degree Field Galaxy type of mass concentrations. Nevertheless, the fact that super- Redshift Survey) and SDSS (Sloan Digital Sky Survey) have clusters are not virialized systems seriously limitats the pos- stimulated the identification of galaxy groups based on the sible methods that can be used to estimate the mass. Bardelli friends-of-friends algorithm (FOFA). The original algorithm et al. (2000) used galaxy counts to suggest that the SSC has introduced by Huchra & Geller (1982) was modified by a significantly higher mean density than the cosmic average. Merchán & Zandivarez (2002) to take the sky coverage Reisenegger et al. (2000) (R2000 hereafter) applied the spheri- of 2dFGRS into account. Using mock catalogues, this tech- cal collapse model and found that the mass enclosed by a radius nique has been extensively tested and has demonstrated its ro- −1 15 16 −1 bustness in identifying galaxy groups in redshift space (see for of 8 h Mpc is in the range 2.0 × 10 to 1.3 × 10 M h , instance Nolthenius & White 1987). Tables 1 and 2 are only available in electronic form at The aim of this work was to identify and study galaxy sys- http://www.edpsciences.org tems in a 12 × 15 degrees region of the SSC, based on the
Article published by EDP Sciences and available at http://www.edpsciences.org/aa or http://dx.doi.org/10.1051/0004-6361:20053623 820 C. J. Ragone et al.: Shapley supercluster redshift compilation of P2004. The resulting catalogue was at reproducing observational data, we chose the ΛCDM sce- used to estimate a lower boundary to the total mass of the SSC. nario with cosmological parameters: Ωm = 0.3, ΩΛ = 0.7, nor- −1 −1 The outline of this paper is as follows. A description of the malization σ8 = 0.9, and H0 = 70 km s Mpc . SSC is given in Sect. 2. Details of the simulation and the mock Due to the large box size, it was possible to find a large catalogue construction are described in Sect. 3. A brief descrip- number of rich systems, so we chose those that represent high tion of the identification process for both observational data density regions. R2000, using estimates of the SSC mass en- −1 and mock catalogues is presented in Sect. 4. In Sect. 5 we anal- closed by 8 Mpc h , find a density range ρ/ρc ∼ 3−20. The yse and estimate the mass enclosed by the SSC, and finally in lower limit of this range corresponds to the Diaferio and Geller Sect. 6 we discuss our main conclusions. method (1997), whilst the upper one is obtained using the spherical collapse model (for more details see R2000). Following these authors, we searched in the simulation for 2. Observational data overdense regions in spheres of a radius of 8 Mpc h−1 and se- The compilation of P2004 contains 8341 galaxies with red- lected 14 of them with overdensities in the mentioned range. −1 shifts taken from Quintana et al. (1995, 1997, 2000), Proust We refer positions to a coordinate system situated 145 Mpc h et al. (2004), Drinkwater et al. (1999, 2004), Watson et al. away from the sphere center in order to mimic the real position (2000), Bardelli et al. (1998, 2000, 2001), Kaldare et al. (2003), of the SSC, and we selected a cone-shaped volume which sub- −1 and data from the NED NASA/IPAC database. The P2004 cat- tended 20 Mpc h of radius at the distance of the sphere cen- ρ/ρ ∼ . alogue covers a region of 12 × 30 degrees on the sky, where ter. Within this volume, overdensities ranged from c 0 5 half of the galaxies are members of the SSC, with velocities to 1.1. Finally, we assigned particles within this volume a lu- −1 −1 minosity function that can be described by a Schechter fit with within 9000 and 18 000 km s s . The main region is centered ∗ − α = − . = − . at v = 14 500 km s 1. parameters 1 19 and Mb 19 79 (Madgwick et al. + After taking the numerous sources involved in the 2002). The adopted k e correction was taken from (Norberg P2004 compilation into account, a non-uniform redshift et al. 2001) and is given by the equation completeness is expected. The P2004 compilation overlaps z + 6z2 the SuperCOSMOS Sky Survey (hereafter SSS) completely. k(z) + e(z) = · (1) + 3 Therefore, in order to test the redshift incompleteness we have 1 20z assigned the corresponding magnitude in the SSS survey to A set of 15 mock catalogues was constructed from this data. each galaxy in the P2004 catalogue. Since none of the sources Galaxies were selected within the described volume by ap- of the P2004 compilation includes faint galaxies, the cross cor- plying the apparent magnitude limit of the SSC survey. The relation with the SSS was performed selecting galaxies brighter redshift incompleteness of the SSC was not reproduced in = than b j 19. Using a matching radius of 10 arcsec, we found the mock catalogues due to the non-uniformity of the sample a positive cross-correlation between P2004 and SSS for more (i.e. high density regions show a higher completeness on av- than 96% of the galaxies. Of these objects, 14% correspond to erage). In the following analysis, we select galaxies brighter multiple identifications. After visually inspecting several im- than mb = 17.5 in both mocks and real data. The mean redshift ages of multiple identifications we introduce several criteria completeness for the SSC subsample is ∼0.5. based on the matching radius and the magnitudes in order to correctly assign the SSS galaxy. We estimated that more than 95% of the Shapley galaxies have been correctly identi- 4. Identification of galaxy groups fied in the SSS. The completeness mask described in Sect. 2 allows us to iden- With the aim of performing a completeness mask, we used tify galaxy groups in the SSC following the procedure de- the HEALPix tool to produce an equal-area pixelisation of the − scribed in Merchán & Zandivarez (2002). According to these sky. For each pixel of 2.6 × 10 4 strd, we counted galaxies up authors, the identification can be performed using a similar to a designed apparent magnitude limit, defining the complete- algorithm similar to the one developed by Huchra & Geller ness in a pixel as the ratio of the number of galaxies in the (1982) but modified in order to take the incompleteness of the SSC to the number of galaxies in the SSS survey. The result- galaxy sample into account by introducing a correction in the ing mean completeness for an apparent magnitude limit of 17.5 linking lengths’ scale factor. The algorithm links pairs of galax- is ∼0.5. The fraction of galaxies with measured redshift signifi- ies according to the following conditions: cantly decreases for fainter magnitudes. It should be noted that the completeness is not uniform, so it shows a tendency to be θ12 V higher in the high density regions of the SSS. D12 = 2sin ≤ DL = D0R (2) 2 H0
V = |V − V |≤V = V R (3) 3. Mock catalogues 12 1 2 L 0
In this study we used one of the publicly available Very Large where θ12 and V are the angular separation and the mean ra- Simulations (VLS) (Yoshida et al. 2001). Briefly, this simula- dial velocity, respectively; D0 and V0 are the transversal and −1 3 tion models a box of 479 Mpc h in size, with 512 particles, radial linking cut-off at the fiducial velocity Vf. Scale factor R is 10 −1 each of mass 6.86×10 M h . Given the success of the model introduced to compensate for the decline in the selection C. J. Ragone et al.: Shapley supercluster 821 function with distance. In this case the scaling factor is com- higher completeness in high density regions, where most of the puted as galaxy systems are present. ⎡ ⎤− 1 The following are median values of the properties listed be- ⎢ M12 φ ⎥ 3 ⎢ ∞ (M)dM (C + C )⎥ fore for both mock and real samples, median dynamical prop- R = ⎢ 1 2 ⎥ (4) ⎣ Mlim ⎦ erties of SSC groups are higher given the lack of small systems φ(M)dM 2 ∞ mentioned above: where M and M are the absolute magnitude of the bright- −1 −1 lim 12 VrSSC = 14 597.8kms Vrmock = 14 370.9kms est galaxy visible at distances V /H and V/H , respectively. −1 −1 f 0 0 σSSC = 344.6kms σmock = 200.8kms The (C + C )/2 factor in the previous equation corresponds −1 −1 1 2 Rvir−SSC = 0.78 Mpc h Rvir−mock = 0.57 Mpc h to the correction introduced to the original Huchra & Geller 13 −1 13 −1 Mvir−SSC = 4.6×10 M h Mvir−mock = 1.4×10 M h . (1982) algorithm to take the incompleteness of the sample into Table 1 shows the different properties of the galaxy systems account, C and C are the completeness values in the direc- 1 2 identified in the SSC sample. As noted before, if we re- tion of each galaxy. Finally D and V are the corresponding 0 0 strict the sample to the Shapley Supercluster region (9000 < linking lengths at the fiducial velocity V and φ(M)isthelumi- f v < 18 000), the resulting number of systems is 73. We per- nosity function of the sample. r formed a cross correlation between our catalogue and the NED In this particular case, given the irregularity of the sky cov- NASA/IPAC database, using a search radius of 20 arcsec and erage, the remarkably rich concentration of galaxies and the − a ∆v ≤ 2000 km s 1. The last restriction was only applied large mass system range in the SSC, we need to take special r to those systems in the NED with known redshift. We found care override misidentifications and spurious merging. In order that 60 of the 122 are new galaxy systems. Of the 62 already to optimize the identification of systems, we took into account known systems, 37 were identified as clusters, althoug, many the X-ray flux contour plot of the Shapley Supercluster core of them were also identified as group. (34 are Abell clus- by Kull & Böhringer (1999). We performed a series of system ters). The remaining 25 were only identified as groups. Table 2 identifications in this same region using different linking length shows the name of the systems (Col. 2), the type of system, the values V and D ,whereD is related to the number density 0 0 0 mean radial velocity, and the difference in the central position contour surrounding a group relative to the mean number den- (Cols. 3−5, respectively). sity (δρ/ρ). We found that standard values of δρ/ρ (∼80) caused the two central groups (SC1329-313 and SC1327-312) in the Kull 5. Mass statistics & Böhringer (1999) plot to merge into a single system. In order In this section we compute the mass function of groups in to find a fair number of groups and to ensure that most of them the SSC and use it to estimate a lower limit to the total mass are single dynamical systems, we selected the lowest overden- of this system. We used the mock catalogues and the simu- sity that does not merge those X-ray systems that appear as lation described in Sect. 3 in order to compute the real mass single systems. Despite this, this choice yields a high value of enclosed in a volume like the SSC one and to compare with δρ/ρ ∼ 400, which ensures small groups will be identified in a the inferred mass. The group mass function was derived us- correctly. ing the 1/Vmax method, which has the advantage of being a Figure 1 shows the best correlation between the identified non-parametric procedure, i.e. it does not assume a shape of systems and the X-ray contours of Kull & Böhringer (1999). the mass function. Moreover, it not only gives shape but also Identification of these systems corresponds to a transversal and normalization at the same time. The differential mass function −1 a line-of-sight linking lengths of D0 = 0.14 Mpc h and V0 = − can be computed using the equation 200 km s 1, respectively. dN 1 As a result of the identification process we found 122 sys- (M) = (5) dM Vmax(Mi) tems of galaxies containing at least 4 members. The number |Mi−M|≤dM/2 of systems drops to 73 when considering groups within the where M and M are the group mass and absolute magnitude