This work presents detailed analysis from the GEEC2 spectroscopic survey of galaxy groups at 0.8<z<1. This deep survey, which has a magnitude limit of r_AB<24.75, had previously found a population of intermediate ('green') galaxies between the star-forming ('blue') and quiescent ('red') sequences. GMOS-S spectroscopy for the 11 X-ray selected galaxy groups was obtained and is highly complete ( > 66 per cent) for eight of the eleven groups. Using an optical-NIR colour-colour diagram, the galaxies in the sample are separated with a dust insensitive method into the three categories, star-forming, quiescent, and intermediate. The strongest environmental dependence is observed in the fraction of quiescent galaxies, which is higher inside groups than in the field for all stellar masses. While intermediate galaxies represent ~15-20 per cent of the star-forming population in both the group and field, the average specific star formation rates (sSFR) of the group population is lower by a factor of ~3. The intermediate population also does not show the strong Hδ absorption that is characteristic of starburst galaxies. Inside groups, only 4.4-6.7 per cent of star-forming galaxies are starbursts, which gives additional validity to the assumption that the quenching of star-formation is the primary process in the transition from the star-forming to the quiescent state. With the use of stellar synthesis models, two possible scenarios for the origins of the intermediate population are investigated, including the quenching of star-forming galaxies via environmental processes and the rejuvenation of star formation in early-type galaxies via mixed mergers. To model the quenching scenario, we have tested the use of different exponential quenching timescales (τ_2) and different types of delays between satellite accretion and the onset of quenching. We found that the fraction of intermediate galaxies depends most strongly on the value of τ_2. The relative fractions of galaxies rule out both the no-delay scenario, which would require a long τ_2 that over-produces intermediate galaxies, as well as the constant 3 Gyr delay model, which does not produce a sufficient number of quiescent galaxies. The observed fractions are best matched with a model that includes a dynamical delay time and a τ_2=0.25 Gyr, but this model also predicts intermediate galaxies Hδ strength higher than that observed. For the rejuvenation scenario, we found that the time visible in the intermediate region is directly related to the size of the second 'burst' of star-formation, which can then be further constrained by the Hδ strength for the intermediate population. The observations are best matched to a burst size of ~1 per cent, at a rate of ~3 times per Gyr. In order to properly distinguish between the two scenarios, we will need to both increase the signal-to-noise ratio for the Hδ measurements and conduct a deeper survey of satellite galaxies both inside groups and in the field.

This work presents detailed analysis from the GEEC2 spectroscopic survey of galaxy groups at 0.8<z<1. This deep survey, which has a magnitude limit of r_AB<24.75, had previously found a population of intermediate ('green') galaxies between the star-forming ('blue') and quiescent ('red') sequences. GMOS-S spectroscopy for the 11 X-ray selected galaxy groups was obtained and is highly complete ( > 66 per cent) for eight of the eleven groups. Using an optical-NIR colour-colour diagram, the galaxies in the sample are separated with a dust insensitive method into the three categories, star-forming, quiescent, and intermediate. The strongest environmental dependence is observed in the fraction of quiescent galaxies, which is higher inside groups than in the field for all stellar masses. While intermediate galaxies represent ~15-20 per cent of the star-forming population in both the group and field, the average specific star formation rates (sSFR) of the group population is lower by a factor of ~3. The intermediate population also does not show the strong Hδ absorption that is characteristic of starburst galaxies. Inside groups, only 4.4-6.7 per cent of star-forming galaxies are starbursts, which gives additional validity to the assumption that the quenching of star-formation is the primary process in the transition from the star-forming to the quiescent state. With the use of stellar synthesis models, two possible scenarios for the origins of the intermediate population are investigated, including the quenching of star-forming galaxies via environmental processes and the rejuvenation of star formation in early-type galaxies via mixed mergers. To model the quenching scenario, we have tested the use of different exponential quenching timescales (τ_2) and different types of delays between satellite accretion and the onset of quenching. We found that the fraction of intermediate galaxies depends most strongly on the value of τ_2. The relative fractions of galaxies rule out both the no-delay scenario, which would require a long τ_2 that over-produces intermediate galaxies, as well as the constant 3 Gyr delay model, which does not produce a sufficient number of quiescent galaxies. The observed fractions are best matched with a model that includes a dynamical delay time and a τ_2=0.25 Gyr, but this model also predicts intermediate galaxies Hδ strength higher than that observed. For the rejuvenation scenario, we found that the time visible in the intermediate region is directly related to the size of the second 'burst' of star-formation, which can then be further constrained by the Hδ strength for the intermediate population. The observations are best matched to a burst size of ~1 per cent, at a rate of ~3 times per Gyr. In order to properly distinguish between the two scenarios, we will need to both increase the signal-to-noise ratio for the Hδ measurements and conduct a deeper survey of satellite galaxies both inside groups and in the field.

Esta tese apresenta resultados sobre a estrutura, relações de escala e cinemática para 48 galáxias em 22 grupos compactos de Hickson, sendo que a apresentação de mapas de velocidades, monocromáticos (na linha H alpha) e de dispersão de velocidades são feitos pela primeira vez para 35 galáxias em 12 dos grupos. A partir dos mapas de velocidades e imagens óticas, foi possível obter os parâmetros cinemáticos, morfológicos e as curvas de rotação das galáxias da presente amostra. Usando as velocidades máximas de rotação para cada galáxia (derivadas das curvas de rotação) e as luminosidades óticas, infravermelhas, as massas estelares e bariônicas, foram estudadas as diferentes relações de Tuly-Fisher (TF) para as galáxias dos grupos compactos. Comparando esses resultados com os apresentados por uma amostra de galáxias de campo, foi encontrado que as galáxias de grupos compactos seguem a relação de TF definida pelas galáxias em ambientes menos densos, no entanto algumas galáxias de baixa massa apresentam altas luminosidades para as suas velocidades de rotação. Surtos de formação estelar e atividade nuclear parecem ser os principais fatores que fazem com que as galáxias de baixas massas dos grupos compactos não estejam na relação de TF definida pelas galáxias do campo. Este resultado indica que as velocidades máximas de rotação não são alteradas em galáxias em interação e portato não há um stripping de massa significativo nas galáxias de grupos compactos, dentro de R(25). O uso das curvas de rotação para estudar a distribuição de massas nestas galáxias revelou que estas curvas apresentam um alto grau de assimetria, a qual seria produzida em eventos de interação galáxia-galáxia. Esses eventos, além de perturbar as curvas de rotação, conseguem expulsar parte do gás neutro das galáxias ao meio intra grupo. Usando dados ultravioleta, nesta tese foram encontradas vários sistemas estelares jovens no meio intergaláctico de grupos compactos. Esses sistemas podem se converter em galáxias satélites ou simplesmente serem dissolvidos, enriquecendo o meio intragrupo. / This thesis presents results on the kinematics, scaling relations and structures of 48 galaxies in 22 compact groups. For 35 galaxies in 12 compact groups, velocity fields, monochromatic maps (derived from H alpha observations) and velocity dispersion maps are presented for the first time. By using these data, it was possible to determine the kinematic and morphological parameters, the rotation curves and to derive the Tully-Fisher relation for the galaxies in dense environments. By using the maximum rotational velocity for each galaxy (derived from its rotation curve) and its optical and near-infrared luminosity and mass, the different Tully-Fisher relations for galaxies in compact groups were derived. Comparing these results with the results displayed by galaxies in less dense environments, it was found that galaxies in compact groups agrees with the Tully-Fisher relation defined by non-interacting galaxies. However, some of the low-mass galaxies are off the Tully-Fisher relation, having too high luminosities for their maximum rotational velocities. This scenario can be explained by a burst of star formation and/or by nuclear activity. We conclude that the maximum rotational velocities of compact groups galaxies are not affected during galaxy-galaxy interactions which implies that there is no significant mass stripping in galaxies of compact groups inside their optical radius. The mass distribution of galaxies in compact groups indicates that the rotation curves of these galaxies are highly asymmetric. The asymmetry could be produced by interactions between galaxies. These interactions, besides affecting the shape of the rotation curve, can eject some neutral gas from the disk of the interacting galaxies into the intragroup medium. By using ultraviolet data, we find several young star-forming regions in the intragroup medium of compact groups. It is still an open question wether these young stellar systems can survive and become new members of the group or if they will fall back onto their parent galaxies.

Esta tese apresenta resultados sobre a estrutura, relações de escala e cinemática para 48 galáxias em 22 grupos compactos de Hickson, sendo que a apresentação de mapas de velocidades, monocromáticos (na linha H alpha) e de dispersão de velocidades são feitos pela primeira vez para 35 galáxias em 12 dos grupos. A partir dos mapas de velocidades e imagens óticas, foi possível obter os parâmetros cinemáticos, morfológicos e as curvas de rotação das galáxias da presente amostra. Usando as velocidades máximas de rotação para cada galáxia (derivadas das curvas de rotação) e as luminosidades óticas, infravermelhas, as massas estelares e bariônicas, foram estudadas as diferentes relações de Tuly-Fisher (TF) para as galáxias dos grupos compactos. Comparando esses resultados com os apresentados por uma amostra de galáxias de campo, foi encontrado que as galáxias de grupos compactos seguem a relação de TF definida pelas galáxias em ambientes menos densos, no entanto algumas galáxias de baixa massa apresentam altas luminosidades para as suas velocidades de rotação. Surtos de formação estelar e atividade nuclear parecem ser os principais fatores que fazem com que as galáxias de baixas massas dos grupos compactos não estejam na relação de TF definida pelas galáxias do campo. Este resultado indica que as velocidades máximas de rotação não são alteradas em galáxias em interação e portato não há um stripping de massa significativo nas galáxias de grupos compactos, dentro de R(25). O uso das curvas de rotação para estudar a distribuição de massas nestas galáxias revelou que estas curvas apresentam um alto grau de assimetria, a qual seria produzida em eventos de interação galáxia-galáxia. Esses eventos, além de perturbar as curvas de rotação, conseguem expulsar parte do gás neutro das galáxias ao meio intra grupo. Usando dados ultravioleta, nesta tese foram encontradas vários sistemas estelares jovens no meio intergaláctico de grupos compactos. Esses sistemas podem se converter em galáxias satélites ou simplesmente serem dissolvidos, enriquecendo o meio intragrupo. / This thesis presents results on the kinematics, scaling relations and structures of 48 galaxies in 22 compact groups. For 35 galaxies in 12 compact groups, velocity fields, monochromatic maps (derived from H alpha observations) and velocity dispersion maps are presented for the first time. By using these data, it was possible to determine the kinematic and morphological parameters, the rotation curves and to derive the Tully-Fisher relation for the galaxies in dense environments. By using the maximum rotational velocity for each galaxy (derived from its rotation curve) and its optical and near-infrared luminosity and mass, the different Tully-Fisher relations for galaxies in compact groups were derived. Comparing these results with the results displayed by galaxies in less dense environments, it was found that galaxies in compact groups agrees with the Tully-Fisher relation defined by non-interacting galaxies. However, some of the low-mass galaxies are off the Tully-Fisher relation, having too high luminosities for their maximum rotational velocities. This scenario can be explained by a burst of star formation and/or by nuclear activity. We conclude that the maximum rotational velocities of compact groups galaxies are not affected during galaxy-galaxy interactions which implies that there is no significant mass stripping in galaxies of compact groups inside their optical radius. The mass distribution of galaxies in compact groups indicates that the rotation curves of these galaxies are highly asymmetric. The asymmetry could be produced by interactions between galaxies. These interactions, besides affecting the shape of the rotation curve, can eject some neutral gas from the disk of the interacting galaxies into the intragroup medium. By using ultraviolet data, we find several young star-forming regions in the intragroup medium of compact groups. It is still an open question wether these young stellar systems can survive and become new members of the group or if they will fall back onto their parent galaxies.

Galaxy properties are known to correlate with their environment, suggesting that environment plays a significant role in galaxy evolution. In particular, blue star forming spiral galaxies are preferentially found in low density regions while red, passive elliptical galaxies are found in the densest clusters. This suggests galaxies falling into groups and clusters experience a decrease in their star formation rate (SFR) and a morphological transformation from spiral to elliptical, but the timescales associated with these changes are not well constrained. This thesis explores the impact of environment on galaxy SFRs and morphologies for a large sample of galaxies from the Sloan Digital Sky Survey. We separate galaxies into two environments (groups and clusters) and use location in projected phase space as an estimate for how long a galaxy has been a part of its current environment. We calculate the timescales associated with the changes in galaxy SFRs and morphologies, and determine SFRs change more quickly than morphology. By comparing to a sample of field galaxies, we find evidence that prior group environments impact current galaxy properties via pre-processing. / Thesis / Master of Science (MSc)

<p>The influence of environment in galaxy evolution has been observed in a wide range of environments, where in general red quiescent early-type galaxies preferentially reside in high-density regions. The aim of this thesis is to understand the role that galaxy groups, and in particular group dynamics, play in galaxy evolution.</p> <p>We examine substructure in rich groups at intermediate redshifts and find that 4/15 groups contain substructure, which is preferentially found on the group outskirts. Galaxies in groups with substructure have properties similar to the field, while the galaxies in groups with no substructure are similar to cluster populations. These results indicate that substructure galaxies do not feel the effects of the environment until well inside the group potential.</p> <p>We then study the evolution of group dynamics to z~1 and find that the fraction of non-Gaussian groups increases with redshift, while the fraction of groups with substructure remains constant. Additionally, we find that the quiescent fraction correlates with galaxy stellar mass, but has little or no dependence on group dynamical mass or state. However, we do observe some correlation between substructure and quiescent fraction for low mass galaxies.</p> <p>Finally, we investigate infalling subhalo galaxies to probe the importance of pre-processing in galaxy evolution. At r200 > 2, galaxies in subhaloes show enhanced quenching, when compared to non-subhalo galaxies. At these radii, the infall population dominates, indicating that enhanced quenching is due to the infalling subhalo population. Additionally, the fraction of groups with subhaloes is a function of halo mass, where more massive systems have a higher fraction of subhaloes. We conclude that for groups, pre-processing is insignificant; however, for the most massive clusters a significant fraction (∼ 25%) of the member galaxies have been pre-processed.</p> / Doctor of Philosophy (PhD)

This thesis focuses on how a galaxy's environment affects its star formation, from the galactic environment of the most luminous IR galaxies in the universe to groups and massive clusters of galaxies. Initially, we studied a class of high-redshift galaxies with extremely red optical-to-mid-IR colors. We used Spitzer spectra and photometry to identify whether the IR outputs of these objects are dominated by AGNs or star formation. In accordance with the expectation that the AGN contribution should increase with IR luminosity, we find most of our very red IR-luminous galaxies to be dominated by an AGN, though a few appear to be star-formation dominated. We then observed how the density of the extraglactic environment plays a role in galaxy evolution. We begin with Spitzer and HST observations of intermediate-redshift groups. Although the environment has clearly changed some properties of its members, group galaxies at a given mass and morphology have comparable amounts of star formation as field galaxies. We conclude the main difference between the two environments is the higher fraction of massive early-type galaxies in groups. Clusters show even more distinct trends. Using three different star-formation indicators, we found the mass--SFR relation for cluster galaxies can look similar to the field (A2029) or have a population of low-star-forming galaxies in addition to the field-like galaxies (Coma). We contribute this to differing merger histories: recently-accreted galaxies would not have time for their star formation to be quenched by the cluster environment (A2029), while an accretion event in the past few Gyr would give galaxies enough time to have their star formation suppressed by the cluster environment. Since these two main quenching mechanisms depend on the density of the intracluster gas, we turn to a group of X-ray under luminous clusters to study how star-forming galaxies have been affected in clusters with lower than expected X-ray emission. We find the distribution of star-forming galaxies with respect to stellar mass varies from cluster to cluster, echoing what we found for Coma and A2029. In other words, while some preprocessing occurs in groups, the cluster environment still contributes to the quenching of star formation.

We investigate the uses of gravitational lensing for analysing the dark matter haloes around galaxies, comparing galaxies within groups and clusters to those in the field. We consider two cases: when only photometric redshift data is available, and when spectroscopic redshift data is available for a sufficiently large sample of galaxies. For the case of data with photometric redshifts, we analyse the CFHTLenS dataset. This dataset is derived from the CFHTLS-Wide survey, and encompasses 154 deg^2 of high-quality shape data. Using the photometric redshifts to estimate local density, we divide the sample of lens galaxies with stellar masses in the range 10^9 Msun to 10^10.5 Msun into those likely to lie in high-density environments (HDE) and those likely to lie in low-density environments (LDE). Through comparison with galaxy catalogues extracted from the Millennium Simulation, we show that the sample of HDE galaxies should primarily (~61%) consist of satellite galaxies in groups, while the sample of LDE galaxies should consist of mostly (~87%) non-satellite (field and central) galaxies. Comparing the lensing signals around samples of HDE and LDE galaxies matched in stellar mass, we show that the subhaloes of HDE galaxies are less massive than those around LDE galaxies by a factor 0.65+/-0.12, significant at the 2.9 sigma level. A natural explanation is that the haloes of satellite galaxies are stripped through tidal effects in the group environment. Our results are consistent with a typical tidal truncation radius of ~40 kpc. For the case of data with spectroscopic redshifts, we analyse the GAMA-I and the ongoing GAMA-II surveys. We demonstrate the possibility of detecting tidal stripping of dark matter subhaloes within galaxy groups using weak gravitational lensing. We have run ray-tracing simulations on galaxy catalogues from the Millennium Simulation to generate mock shape catalogues. The ray-tracing catalogues assume a halo model for galaxies and groups, using various models with different distributions of mass between galaxy and group haloes to simulate different stages of group evolution. Using these mock catalogues, we forecast the lensing signals that will be detected around galaxy groups and satellite galaxies, as well as test two different methods for isolating the satellites' lensing signals. A key challenge is to determine the accuracy to which group centres can be identified. We show that with current and ongoing surveys, it will possible to detect stripping in groups of mass 10^12 Msun to 10^15 Msun.

Entender como as galáxias se formam e evoluem ao longo do tempo é um dos maiores desafio da cosmologia moderna. Vários processos estão presentes na formação de galáxias, tais como o feedback de supernovas e núcleos galácticos ativos, evolução química e dinâmica, e também efeitos ambientais. Esta tese abrange estes processos, a partir de um ponto de vista observacional. A Via Láctea tem um papel fundamental na compreensão dos vários processos envolvidos na formação de uma galáxia, e começamos nosso projeto estudando nossa própria galáxia. Diferentes processos deixam assinaturas típicas na distribuição de velocidades e metalicidades das estrelas. Por esta razão, combinando cinemática e abundâncias químicas, foi possível determinar a origem de uma amostra de estrelas velhas e ricas em metais. Compreender como e onde essas estrelas se formaram está intimamente relacionado com mecanismos presentes na evolução do disco Galáctico. Apesar de não podermos observar estrelas individuais em galáxias distantes, somos capazes de inferir a história de formação destas galáxias combinando modelos de população estelar simples, de forma a reproduzir o espectro observado. Usando esta metodologia, foi possível traçar a história de formação estelar de galáxias elípticas, e dessa forma restringir os mecanismos de feedback que regulam a formação de estrelas em halos. No cenário Lambda-CDM, as estruturas menores são formadas primeiro, e então elas se agrupam, formando assim estruturas cada vez maiores. As galáxias, ao serem incorporadas à sistemas maiores, sofrem os efeitos de diversos processos que atuam em ambientes de alta densidade, mudando assim suas propriedades. Desta forma, a evolução das galáxias e a formação de estruturas em grande escala andam lado a lado, como mostramos em nosso estudo de propriedades de galáxias em grupos. Exploramos a distribuição espacial das galáxias na vizinhança de grupos, e também usamos a distribuição de velocidades das galáxias para determinar o estágio evolutivo do grupo. Foram encontradas correlações importantes entre o estágio evolutivo do grupo e as populações de galáxias que nestes residem. / Understanding the way galaxies form and evolve throughout the cosmic time remains one of the greatest challenges of modern cosmology. Several processes are known to play a role in the formation of galaxies, such as feedback from supernovae and active galactic nuclei, chemical and dynamical evolution and environmental effects. This thesis encompasses these processes, from an observational point of view. The Milky Way plays a pivotal role in understanding the various processes involved in the formation of a galaxy, and we start our understanding program by studying our own Galaxy. Different formation processes leave typical signatures in the velocity and metallicity distribution of stars. For this reason, we were able to trace the origin of old and metal-rich stars by combining their kinematics and chemical abundances. Understanding how and where these stars were formed is closely related to mechanisms driving the evolution of the Galactic disk. Although we cannot observe individual stars in distant galaxies, only the integrated spectra, we are able to infer the mass assembly history of galaxies by combining single stellar population (SSP) models that reproduce the observed spectrum. Using this methodology, we traced the star formation history of elliptical galaxies and, by studying the signatures left in the star formation history, we were able to constrain the feedback mechanisms regulating the star formation within halos. In the LCDM scenario, small scale structures are formed first, and then they merge forming larger and larger structures. Therefore, galaxies grow into more and more massive systems, and processes operating in these high-density environments change their properties. For this reason, galaxy evolution and formation of large-scale structures go hand in hand, as we show in our study of properties of galaxies in groups. We explored the spatial distribution of galaxies within and in the surrounding of groups, and we also used the velocity distribution of galaxies as a probe of the evolutionary stage of the group. We found important correlations between the evolutionary stage of the group and the population of galaxies residing within it.

We investigate the uses of gravitational lensing for analysing the dark matter haloes around galaxies, comparing galaxies within groups and clusters to those in the field. We consider two cases: when only photometric redshift data is available, and when spectroscopic redshift data is available for a sufficiently large sample of galaxies. For the case of data with photometric redshifts, we analyse the CFHTLenS dataset. This dataset is derived from the CFHTLS-Wide survey, and encompasses 154 deg^2 of high-quality shape data. Using the photometric redshifts to estimate local density, we divide the sample of lens galaxies with stellar masses in the range 10^9 Msun to 10^10.5 Msun into those likely to lie in high-density environments (HDE) and those likely to lie in low-density environments (LDE). Through comparison with galaxy catalogues extracted from the Millennium Simulation, we show that the sample of HDE galaxies should primarily (~61%) consist of satellite galaxies in groups, while the sample of LDE galaxies should consist of mostly (~87%) non-satellite (field and central) galaxies. Comparing the lensing signals around samples of HDE and LDE galaxies matched in stellar mass, we show that the subhaloes of HDE galaxies are less massive than those around LDE galaxies by a factor 0.65+/-0.12, significant at the 2.9 sigma level. A natural explanation is that the haloes of satellite galaxies are stripped through tidal effects in the group environment. Our results are consistent with a typical tidal truncation radius of ~40 kpc. For the case of data with spectroscopic redshifts, we analyse the GAMA-I and the ongoing GAMA-II surveys. We demonstrate the possibility of detecting tidal stripping of dark matter subhaloes within galaxy groups using weak gravitational lensing. We have run ray-tracing simulations on galaxy catalogues from the Millennium Simulation to generate mock shape catalogues. The ray-tracing catalogues assume a halo model for galaxies and groups, using various models with different distributions of mass between galaxy and group haloes to simulate different stages of group evolution. Using these mock catalogues, we forecast the lensing signals that will be detected around galaxy groups and satellite galaxies, as well as test two different methods for isolating the satellites' lensing signals. A key challenge is to determine the accuracy to which group centres can be identified. We show that with current and ongoing surveys, it will possible to detect stripping in groups of mass 10^12 Msun to 10^15 Msun.