No Arabic abstract
We study the dwarf galaxy population in the central ~700 arcmin^2 of the Coma cluster, the majority of which are early-type dwarf elliptical (dE) galaxies. Analysis of the statistically-decontaminated dE galaxy sequence in the color-magnitude diagram reveals a highly significant trend of color with magnitude (Delta (B-R)/Delta R = -0.056pm0.002 mag), in the sense that fainter dEs are bluer and thus presumably more metal-poor. The mean color of the faintest dEs in our sample is (B-R)~1.15 mag, consistent with a color measurement of the diffuse intracluster light in the Coma core. This intracluster light could then have originated from the tidal disruption of faint dEs in the cluster core. The total galaxy luminosity function (LF) is well modeled as the sum of a log-normal distribution for the giant galaxies, and a Schechter function for the dE galaxies with a faint-end slope alpha = -1.41pm0.05. This value of alpha is consistent with those measured for the Virgo and Fornax clusters. The spatial distribution of the faint dE galaxies (19.0 < R le 22.5 mag) has R_c = 22.15 arcmin (~0.46h^{-1} Mpc), significantly larger than the R_c = 13.71 arcmin (~0.29h^{-1} Mpc) found for the cluster giants and the brighter dEs (R le 19.0 mag), consistent with tidal disruption of the faint dEs. Finally, we find that most dEs belong to the general Coma cluster potential rather than as satellites of individual giant galaxies: An analysis of the number counts around 10 cluster giants reveals that they each have on average 4pm 1 dE companions within a projected radius of 13.9h^{-1} kpc. (Abridged)
This is the second paper in a series studying the photometric and spectroscopic properties of galaxies of different luminosities in the Coma cluster. The sample selection, spectroscopic observations and completeness functions are presented here. To study the spectral properties of galaxies as a function of their local environment, two fields were selected for spectroscopic observations to cover both the core (Coma1) and outskirts (ie. south-west of the core and centered on NGC4839)- (Coma3) of the cluster. Medium resolution spectroscopy (6-9 AA) was carried out for a total of 490 galaxies in both fields (302 in Coma1 and 188 in Coma3), using the WYFFOS multi-fiber spectrograph on the William Herschel Telescope. The galaxies cover a range of $12 < R < 20$, corresponding to -23 < M_R < -15 (H0=65 km/sec/Mpc). The redshifts are measured with an accuracy of 100 km/sec. The spectral line strengths and equivalent widths are also measured for the same galaxies and analysed in Poggianti et al (2001- paper III). A total of 189 (Coma1) and 90 (Coma3) galaxies are identified as members of the Coma cluster. An analysis of the colors show that only two members of the Coma cluster in our sample have B-R > 2. The completeness functions for the spectroscopic sample is presented.
We present the study of a large sample of early-type dwarf galaxies in the Coma cluster observed with DEIMOS on the Keck II to determine their internal velocity dispersion. We focus on a subsample of 41 member dwarf elliptical galaxies for which the velocity dispersion can be reliably measured, 26 of which were studied for the first time. The magnitude range of our sample is $-21<M_R<-15$ mag. This paper (paper I) focuses on the measurement of the velocity dispersion and their error estimates. The measurements were performed using {it pPXF (penalised PiXel Fitting)} and using the Calcium triplet absorption lines. We use Monte Carlo bootstrapping to study various sources of uncertainty in our measurements, namely statistical uncertainty, template mismatch and other systematics. We find that the main source of uncertainty is the template mismatch effect which is reduced by using templates with a range of spectral types. Combining our measurements with those from the literature, we study the Faber-Jackson relation ($Lproptosigma^alpha$) and find that the slope of the relation is $alpha=1.99pm0.14$ for galaxies brighter than $M_Rsimeq-16$ mag. A comprehensive analysis of the results combined with the photometric properties of these galaxies is reported in paper II.
Deep B- and R-band CCD images of the central ~700 arcmin^2 of the Coma cluster core have been used to measure the dwarf-galaxy population in Coma. In this paper, we describe a newly developed code for automated detection, photometry and classification of faint objects of arbitrary shape and size on digital images. Intensity-weighted moments are used to compute the positions, radial structures, ellipticities, and integrated magnitudes of detected objects. We demonstrate that Kron-type 2r_1 aperture aperture magnitudes and surface brightnesses are well suited to faint-galaxy photometry of the type described here. Discrimination between starlike and extended (galaxy) objects is performed interactively through parameter-space culling in several possible parameters, including the radial moments, surface brightness, and integrated color versus magnitude. Our code is tested and characterized with artificial CCD images of star and galaxy fields; it is demonstrated to be accurate, robust and versatile. Using these analysis techniques, we detect a large population of dE galaxies in the Coma cluster core. These dEs stand out as a tight sequence in the R, (B-R) color-magnitude diagram.
We have carried out surface photometry and an isophotal analysis for a sample of 25 early-type dwarf (dE and dS0) galaxies in the Virgo cluster based on CCD images taken at the VLT with FORS1 and FORS2. For each galaxy we present $B$ and $R$-band surface brightness profiles, as well as the radial colour ($B-R$) profile. We give total apparent $BR$ magnitudes, effective radii, effective surface brightnesses and total colour indices. The light profiles have been fitted with Sersic models and the corresponding parameters are compared to the ones for other classes of objects. The observed profiles of the brightest cluster dwarfs show significant deviations from a simple Sersic model, indicating that there is more inner structure than just a nucleus. In addition, we find a relation between the effective surface brightness, at a given luminosity, and the strength of the offset of the galaxys nucleus with respect to the center of the isophotes. Dwarfs with large nuclear offsets also tend to have stronger isophotal twists. In sum, our findings suggest the presence of substructure in most, and preferentially in the less compact, bright early-type dwarfs. The physical (dynamical) meaning of this has yet to be explored. (abridged)
Intrinsic color profiles of a sample of nine Brightest Cluster Galaxies (BCGs) are recovered from the observed color profiles by evaluating spurious gradients introduced by errors in the determination of the sky levels and by different seeing conditions between the observations. Isophote shapes and surface brightness profiles are presented for the four newly observed BCGs. Three out of nine BCGs show color gradients of the order of 0.10 mag per decade in radius. Five BCGs do not possess any color gradient larger than 0.01 mag per decade in radius. We do not see any correlation between the presence (or the sign) of a color gradient and the BCG morphology (slope and shape of its surface brightness profile), or the cluster richness. This argues against a strong and recent influence of the environment on the BCG morphology. The slope of the BCG surface brightness profile is correlated to the cluster richness, posing a constraint on the initial conditions, or on the physical mechanism that is responsible for the present BCG morphology. Finally, only one BCG can be a cD despite visual classification as such of all studied BCGs.