Logo Logo
Hilfe
Kontakt
Switch language to English
The group galaxy population through the cosmic time. study of the properties of galaxies in the most common high density environment
The group galaxy population through the cosmic time. study of the properties of galaxies in the most common high density environment
One of the most fundamental correlations between the properties of galaxies in the local Universe is the so-called morphology-density relation (Dressler 1980). A plethora of studies utilizing multi-wavelength tracers of activity have shown that late type star forming galaxies favour low density regions in the local Universe (e.g. G´omez et al. 2003). In particular, the cores of massive galaxy clusters are galaxy graveyards full of massive spheroids that are dominated by old stellar populations. A variety of physical processes might be effective in suppressing star formation and affecting the morphology of cluster and group galaxies. Broadly speaking, these can be grouped in two big families: (i) interactions with other cluster members and/or with the cluster gravitational potential and (ii) interactions with the hot gas that permeates massive galaxy systems. Galaxy groups are the most common galaxy environment in our Universe, bridging the gap between the low density field and the crowded galaxy clusters. Indeed, as many as 50%-70% of galaxies reside in galaxy groups in the nearby Universe (Huchra & Geller 1982; Eke et al. 2004), while only a few percent are contained in the denser cluster cores. In addition, in the current bottom-up paradigm of structure formation, galaxy groups are the building blocks of more massive systems: they merge to form clusters. As structures grow, galaxies join more and more massive systems, spending most of their life in galaxy groups before entering the cluster environment. Thus, it is plausible to ask if group-related processes may drive the observed relations between galaxy properties and their environment. To shed light on this topic we have built the largest X-ray selected samples of galaxy groups with secure spectroscopic identification on the major blank field surveys. For this purpose, we combine deep X-ray Chandra and XMM data of the four major blank fields (All-wavelength Extended Groth Strip International Survey (AEGIS), the COSMOS field, the Extended Chandra Deep Field South (ECDFS), and the Chandra Deep Field North (CDFN) ). The group catalog in each field is created by associating any X-ray extended emission to a galaxy overdensity in the 3D space. This is feasible given the extremely rich spectroscopic coverage of these fields. Our identification method and the dynamical analysis used to identify the galaxy group members and to estimate the group velocity dispersion is extensively tested on the AEGIS field and with mock catalogs extracted from the Millennium Simulation (Springel et al. 2005). The effect of dynamical complexity, substructure, shape of X-ray emission, different radial and redshift cuts have been explored on the LX −sigma relation. We also discover a high redshift group at z~1.54 in the AEGIS field. This detection illustrates that mega-second Chandra exposures are required for detecting such objects in the volume of deep fields. We provide an accurate measure of the Star Formation Rate (SFR) of galaxies by using the deepest available Herschel PACS and Spitzer MIPS data available for the considered fields. We also provide a well-calibrated estimate of the SFR derived by using the SED fitting technique for undetected sources in mid- and far-infrared observations. Using this unique sample, we conduct a comprehensive analysis of the dependence of the total SFR , total stellar masses and halo occupation distribution (HOD) of massive galaxies (M*>10^10 M_sun) on the halo mass of the groups with rigorous consideration of uncertainties. We observe a clear evolution in the level of star formation (SF) activity in galaxy groups. Indeed, the total star formation activity in high redshift (0.5<z<1.1) groups is higher with respect to the low redshift (0.15<z<0.5) sample at any mass by almost 0.8 ± 0.1 dex. A milder difference (0.35 ± 0.1 dex) is observed between the [0.15-0.5] redshift bin and the groups at z < 0.085. This evolution seems to be much faster than the one observed in the whole galaxy population dominated by lower mass halos. This would imply that the level of SF activity is declining more rapidly since z~1.1 in the more massive halos than in the more common lower mass halos, confirming a “halo downsizing” effect as discussed already in Popesso et al. (2012). The HOD and the total stellar mass-M200 relation are consistent with a linear relation in any redshift bin in the M_200 range considered in our analysis. We do not observe any evolution in the HOD since z~1.1. Similarly we do not observe evolution in the relation between the total stellar mass of the groups and the total mass, in agreement with the results of Giodini et al (2012). The picture emerging from our findings is that massive groups at M_200~10^13−14 M_sun have already accreted the same amount of mass and have the same number of galaxies as the low redshift counterpart, as predicted by Stewart et al. (2008). This implies that the most evident evolution of the galaxy population of the most massive systems acts in terms of quenching their galaxy star formation activity. The analysis of the evolution of the fraction of SF galaxies as a function of halo mass or velocity dispersion show that high mass systems seem to be already evolved at z~1 by showing a fraction of star forming galaxies consistent with the low redshift counterpart at z < 0.085. Given the almost linear relation between the total SFR and M_200 in the high-z sample, this implies that most of the contribution to the total SFR of the most massive systems (M_200~ 10^14 M_sun) is given by few highly star forming galaxies, while in lower mass systems (M_200~10^13 M_sun) is given by many galaxies of average activity. This would be an additional sign of a faster evolution in the more massive systems in terms of star formation activity with respect to lower mass groups. Thus, it would confirm the “halo downsizing” effect. The comparison of our results with the prediction of the Millennium Simulation semi-analytical model confirms the known problem of the models. We confirm the strong bias due to the “satellite overquenching” problem in suppressing significantly the SF activity of group galaxies (more than an order of magnitude) at any redshift with respect to observations. The HOD predicted by the simulations is remarkably in agreement with the observations. But due to the low SF activity of galaxies in massive halos, the models predict also a lower total stellar mass in groups with respect to the observed one at any redshift. In order to compare the SF activity level of galaxies in different environment, we also define a sample of field galaxies and “filament-like” galaxies. This is done by using the galaxy density field to find isolated galaxies (field) and galaxies in high density region but not associated to any group or more generically to an X-ray extended emission. These two classes of environment in addition to the galaxy group sample are used to study the location of galaxies in SFR-mass plane since z~1.1 as a function of the environment. Indeed, several studies have already shown there is a tight correlation between the SFR and the stellar masses of the bulk of the star forming galaxy population at least over the past 10 Gyr. Quiescent galaxies are mainly located under this main sequence (MS) and in a more scattered cloud. Our analysis shows that the Main Sequence of star forming galaxies in the two redshift bins considered (0.15 < z < 0.5 and 0.5 < z < 1.1) is not a linear relation but it shows a flattening towards higher masses (M* > 10^10.4−10.6 M_sun). Above this limit, the galaxy SFR has a very weak dependence on the stellar mass. This flattening, to different extent, is present in all environments. At low redshift, group galaxies tend to deviate more from the mean MS towards the region of quiescence with respect to isolated and filament-like galaxies. This environment dependent location of low redshift group galaxies with respect to the mean MS causes the increase of the dispersion of the distribution of galaxies around the MS as a function of the stellar mass. At high redshift we do not find significant evidence for a differential location of galaxies with respect to the MS as a function of the environment. Indeed, in this case we do not observe a significant increase of the dispersion of the distribution of galaxies around the MS as a function of the stellar mass. We do not find evidence for a differential distribution in the morphological type of MS galaxies in different environments. Instead, we observe a much stronger dependence of the mean S´ersic index on the stellar mass. These results suggest that star formation quenching in group galaxies is not due to galaxy structural transformations. It also suggests that while morphology of MS galaxies is more stellar mass dependent, star formation quenching is mostly environment dependent. We conclude that the membership to a massive halo is a key ingredient in the galaxy evolution and that this acts in terms of star formation quenching in group sized halos.
Not available
Erfanianfar, Ghazaleh
2014
Englisch
Universitätsbibliothek der Ludwig-Maximilians-Universität München
Erfanianfar, Ghazaleh (2014): The group galaxy population through the cosmic time: study of the properties of galaxies in the most common high density environment. Dissertation, LMU München: Fakultät für Physik
[thumbnail of Erfanianfar_Ghazaleh.pdf]
Vorschau
PDF
Erfanianfar_Ghazaleh.pdf

18MB

Abstract

One of the most fundamental correlations between the properties of galaxies in the local Universe is the so-called morphology-density relation (Dressler 1980). A plethora of studies utilizing multi-wavelength tracers of activity have shown that late type star forming galaxies favour low density regions in the local Universe (e.g. G´omez et al. 2003). In particular, the cores of massive galaxy clusters are galaxy graveyards full of massive spheroids that are dominated by old stellar populations. A variety of physical processes might be effective in suppressing star formation and affecting the morphology of cluster and group galaxies. Broadly speaking, these can be grouped in two big families: (i) interactions with other cluster members and/or with the cluster gravitational potential and (ii) interactions with the hot gas that permeates massive galaxy systems. Galaxy groups are the most common galaxy environment in our Universe, bridging the gap between the low density field and the crowded galaxy clusters. Indeed, as many as 50%-70% of galaxies reside in galaxy groups in the nearby Universe (Huchra & Geller 1982; Eke et al. 2004), while only a few percent are contained in the denser cluster cores. In addition, in the current bottom-up paradigm of structure formation, galaxy groups are the building blocks of more massive systems: they merge to form clusters. As structures grow, galaxies join more and more massive systems, spending most of their life in galaxy groups before entering the cluster environment. Thus, it is plausible to ask if group-related processes may drive the observed relations between galaxy properties and their environment. To shed light on this topic we have built the largest X-ray selected samples of galaxy groups with secure spectroscopic identification on the major blank field surveys. For this purpose, we combine deep X-ray Chandra and XMM data of the four major blank fields (All-wavelength Extended Groth Strip International Survey (AEGIS), the COSMOS field, the Extended Chandra Deep Field South (ECDFS), and the Chandra Deep Field North (CDFN) ). The group catalog in each field is created by associating any X-ray extended emission to a galaxy overdensity in the 3D space. This is feasible given the extremely rich spectroscopic coverage of these fields. Our identification method and the dynamical analysis used to identify the galaxy group members and to estimate the group velocity dispersion is extensively tested on the AEGIS field and with mock catalogs extracted from the Millennium Simulation (Springel et al. 2005). The effect of dynamical complexity, substructure, shape of X-ray emission, different radial and redshift cuts have been explored on the LX −sigma relation. We also discover a high redshift group at z~1.54 in the AEGIS field. This detection illustrates that mega-second Chandra exposures are required for detecting such objects in the volume of deep fields. We provide an accurate measure of the Star Formation Rate (SFR) of galaxies by using the deepest available Herschel PACS and Spitzer MIPS data available for the considered fields. We also provide a well-calibrated estimate of the SFR derived by using the SED fitting technique for undetected sources in mid- and far-infrared observations. Using this unique sample, we conduct a comprehensive analysis of the dependence of the total SFR , total stellar masses and halo occupation distribution (HOD) of massive galaxies (M*>10^10 M_sun) on the halo mass of the groups with rigorous consideration of uncertainties. We observe a clear evolution in the level of star formation (SF) activity in galaxy groups. Indeed, the total star formation activity in high redshift (0.5<z<1.1) groups is higher with respect to the low redshift (0.15<z<0.5) sample at any mass by almost 0.8 ± 0.1 dex. A milder difference (0.35 ± 0.1 dex) is observed between the [0.15-0.5] redshift bin and the groups at z < 0.085. This evolution seems to be much faster than the one observed in the whole galaxy population dominated by lower mass halos. This would imply that the level of SF activity is declining more rapidly since z~1.1 in the more massive halos than in the more common lower mass halos, confirming a “halo downsizing” effect as discussed already in Popesso et al. (2012). The HOD and the total stellar mass-M200 relation are consistent with a linear relation in any redshift bin in the M_200 range considered in our analysis. We do not observe any evolution in the HOD since z~1.1. Similarly we do not observe evolution in the relation between the total stellar mass of the groups and the total mass, in agreement with the results of Giodini et al (2012). The picture emerging from our findings is that massive groups at M_200~10^13−14 M_sun have already accreted the same amount of mass and have the same number of galaxies as the low redshift counterpart, as predicted by Stewart et al. (2008). This implies that the most evident evolution of the galaxy population of the most massive systems acts in terms of quenching their galaxy star formation activity. The analysis of the evolution of the fraction of SF galaxies as a function of halo mass or velocity dispersion show that high mass systems seem to be already evolved at z~1 by showing a fraction of star forming galaxies consistent with the low redshift counterpart at z < 0.085. Given the almost linear relation between the total SFR and M_200 in the high-z sample, this implies that most of the contribution to the total SFR of the most massive systems (M_200~ 10^14 M_sun) is given by few highly star forming galaxies, while in lower mass systems (M_200~10^13 M_sun) is given by many galaxies of average activity. This would be an additional sign of a faster evolution in the more massive systems in terms of star formation activity with respect to lower mass groups. Thus, it would confirm the “halo downsizing” effect. The comparison of our results with the prediction of the Millennium Simulation semi-analytical model confirms the known problem of the models. We confirm the strong bias due to the “satellite overquenching” problem in suppressing significantly the SF activity of group galaxies (more than an order of magnitude) at any redshift with respect to observations. The HOD predicted by the simulations is remarkably in agreement with the observations. But due to the low SF activity of galaxies in massive halos, the models predict also a lower total stellar mass in groups with respect to the observed one at any redshift. In order to compare the SF activity level of galaxies in different environment, we also define a sample of field galaxies and “filament-like” galaxies. This is done by using the galaxy density field to find isolated galaxies (field) and galaxies in high density region but not associated to any group or more generically to an X-ray extended emission. These two classes of environment in addition to the galaxy group sample are used to study the location of galaxies in SFR-mass plane since z~1.1 as a function of the environment. Indeed, several studies have already shown there is a tight correlation between the SFR and the stellar masses of the bulk of the star forming galaxy population at least over the past 10 Gyr. Quiescent galaxies are mainly located under this main sequence (MS) and in a more scattered cloud. Our analysis shows that the Main Sequence of star forming galaxies in the two redshift bins considered (0.15 < z < 0.5 and 0.5 < z < 1.1) is not a linear relation but it shows a flattening towards higher masses (M* > 10^10.4−10.6 M_sun). Above this limit, the galaxy SFR has a very weak dependence on the stellar mass. This flattening, to different extent, is present in all environments. At low redshift, group galaxies tend to deviate more from the mean MS towards the region of quiescence with respect to isolated and filament-like galaxies. This environment dependent location of low redshift group galaxies with respect to the mean MS causes the increase of the dispersion of the distribution of galaxies around the MS as a function of the stellar mass. At high redshift we do not find significant evidence for a differential location of galaxies with respect to the MS as a function of the environment. Indeed, in this case we do not observe a significant increase of the dispersion of the distribution of galaxies around the MS as a function of the stellar mass. We do not find evidence for a differential distribution in the morphological type of MS galaxies in different environments. Instead, we observe a much stronger dependence of the mean S´ersic index on the stellar mass. These results suggest that star formation quenching in group galaxies is not due to galaxy structural transformations. It also suggests that while morphology of MS galaxies is more stellar mass dependent, star formation quenching is mostly environment dependent. We conclude that the membership to a massive halo is a key ingredient in the galaxy evolution and that this acts in terms of star formation quenching in group sized halos.