Galaxies are the fundamental building blocks of the Universe. Massive galaxies can be subdivided into two major morphological groups. Elliptical galaxies are characterised by a dominant spheroidal stellar component with stars moving on random orbits. Spiral galaxies, on the other hand, show a dominant disk component made of stars and gas, including a central spheroidal bulge component. In both cases, massive, extended dark matter halos appear to surround the visible components of the galaxies. While massive galaxies contain most of the mass in the Universe, by far the most frequent galaxy type are dwarf galaxies which often have complex, irregular gas distributions as even a few supernovae can strongly perturb or even destroy any disk component. Dwarfs are often found as satellites orbiting more massive galaxies, providing interesting information about the hierarchical building blocks that merged into giant galaxies. High-resolution photometry in different wavelength regimes and spectroscopic data now gives very detailed insight into the internal structure of galaxies. Many observations are puzzling and contain information about the complex formation history of galaxies and their internal dynamics. Questions that we are currently investigating are:
The Peculiar Structure of High-Redshift Disk Galaxies
Young disk galaxies have recently been observed at high cosmological redshifts of z=2, corresponding to epochs where the Universe was just 1/3rd its present age. This epoche is especially interesting as it corresponds to the peak in the star formation rate of the Cosmos and the time when the morphologies of the galaxies were determined. High-redshift galaxies have recently been detected and observed with the larges telescopes e.g. of ESO. It soon became clear that these obejcts are very different compared with present-day disk galaxies. For example, they form stars with enormous rates that are a factor of 10 - 100 higher than in nowadays. The star formation is concentrated in gigantic clumps of molecular gas, about 1000 times larger than present-day molecular clouds, and as big as dwarf galaxies. In addition, the gas in the disks is highly turbulent, driven by energetic sources that are not well understood up to now. In collaboration with the observational infrared/submillimeter group at the Max-Planck-Institute for Extraterrestrial Physics in Garching we are investigating the origin and structure of the observed giant gas clumps in young galactic disks as well as their evolution and star formation history. Are these clumps the sites for the formation of the old globular clusters and supermassive black holes that are found today in galaxies as relics of an earlier time of violent star formation? Is there evidence of massive gas accretion onto these young galactic disks from the surrounding cosmic web which serves as a reservoir of gas? Into which present-day galaxy type do these disks evolve? How important are these galaxies for the chemical enrichment of the cosmos with heavy elements?
The Structure of Elliptical Galaxies and their Relation to Supermassive Black holes
In recent years observations have revealed a strong correlation between the central supermassive black holes (BHs) and their host galaxies as manifested in the relation between the BH mass and the bulge velocity dispersion, M_BH-sigma and the bulge stellar mass M_BH-M_bulge. More massive BHs typically reside in more massive galaxies, with the observed relations thus indicating a link between the formation of the BHs and their host galaxies. Energetic feedback from BHs during the quasar phase can potentially drive out gas from the galaxies thus setting a maximum scale for galaxy masses. In addition, this AGN feedback might quench star formation in the most massive galaxies explaining why massive galaxies are typically non-starforming ellipticals We have implemented a simple Bondi-Hoyle based BH accretion model into our numerical code. In this model 0.5% of the accreted rest mass energy is coupled as thermal energy to the surrounding gas. Expanding on previous work we have showed that the observed relations can be produced in both equal- and unequal-mass merger simulations of gas-rich disks and early-type galaxies. In the Figure below we show a comparison of our simulated sample with the local observed relations and demonstrate that both equal- and unequal-disk mergers reproduce the observed results equally well. In addition we have studied the evolution of the scaling relations during the merger of the BHs as it is not obvious if the scaling relations are valid during the merging process and at what stage the galaxies evolve onto the relations. Finally, we have studied in detail the termination of star formation by energetic BH feedback during galaxy mergers as this is a viable mechanism for reproducing the observed population of dead and red elliptical galaxies. The image shows the M_BH-sigma (left) and M_BH-M_* (right) relations for our complete sample of 1:1 (circles) and 3:1 (triangles) mergers. The black, green and red symbols show the effect of varying the initial gas mass fraction, whereas the blue symbols show the variation caused by the orbit and initial geometry for a fixed gas fraction. The lines show the observed relations with errors by Tremaine et al. (2002) and Haring&Rix (2004), respectively (Johansson et al. 2009).
The Outer Structure of Elliptical Galaxies and their Dark Matter Halos
The outer haloes of elliptical galaxies have been studied by observations for several years to adress the questions of the nature of the Dark Matter halo they are embedded in and the mechanisms that drive the formation of elliptical galaxies. While the extended cold gas disks in spiral galaxies can be used to measure the rotation curves and the line-of-sight velocity dispersions (LOSVD), for ellipticals we need to use other tracers for the outher haloes since ellipticals do not contain any cold gas anymore. Instead, planetary nebulae have been used as tracers for the LOSVD. Those measurements revealed a huge variety of different profiles, some flattening to nearly constant values, some increasing again in the outer part and some decreasing fastly, showing nearly no evidence of a dark matter halo. While observations only show us a snapshot in the life of the galaxy, with simulations we can look at evolution scenarios. Currently there are two different scenarios discussed: The build up of an elliptical by several minor mergers spread over time, and the build up through a single major merger (a merger with a mass ratio less than 3:1). Also from obeservations we know that especially very massive elliptical galaxies are most frequent in very dense environments, like in galaxy clusters, and those environments provide best conditions for both merger scenarios. We study a sample of simulated ellipticals with different formation scenarios, from major merger scenarios to minor merger build ups, and different environments, from dense cluster environments to isolated field environments, to explain the origin of these different profiles and understand what we can learn from them.
The Origin and Structure of Magnetic Fields in present-day Galaxies and Galaxy Groups
The effect of nonthermal pressure components, i.e. magnetic fields and high energy particles, on the evolution of galaxies and structure formation in the Universe, is still not well understood. Contrary to the thermal pressure, nonthermal components are not subject to cooling processes. Thus, once present, a nonthermal pressure affects the environment over long timescales. This influence might impede the collapse of gas and thus lower the star formation rate. Along with the reduction of star formation, the injection of cosmic rays as well as the associated turbulence is also lowered. Turbulence, however, is believed to be the main source of magnetic field amplification. This correlation suggests a self-regulated state and thus equipartition between all corresponding pressure components. This theoretical scenario, which is supported also by the Radio-FIR-correlation, is the framework of our research. Magnetic fields pervade intracluster and interstellar material. Even more astonishing, strong (10-100 micro Gauss) magnetic fields have been observed in very young objects like damed Ly-alpha systems, an observation, which is up to now not satisfactory explained. As a first step towards a more complete understanding of the evolution of magnetic fields in the Universe, we want to shed light on its evolution in isolated and interacting present-day galaxies. Our numerical investigations of the magnetic field evolution in interacting and merging galaxies suggest equipartition between the magnetic and turbulent pressure, as expected from theory. Thereby, even small intitial magnetic fields are amplified efficiently up to the equipartition level during the interaction. This result is particularly interesting in the framework of hierarchical structure formation, within which the formation of galaxies is characterized by more or less intense merging of smaller galactic subunits. Given our numerical results, the merging of these subunits might have been accompanied by a significant amplification of the magnetic field, thus explaining the high magneitc field values obserbed in damped Ly-alpha systems. Artificial radio maps derived from our simulations also compare well with observations.
Supernova Feedback and Galactic Halo Enrichment
Star formation in galaxies causes supernova explosions, and thereby both turbulence and, if strong enough, winds or galactic fountains. Galactic winds are found in nearby starburst galaxies as well as in high-redshift Lyman-break systems. High star formation rates, outflow features and turbulence are prime characteristics for high-redshift galaxies. Phenomenologically, the stars form in clusters. The supernova feedback quickly evacuates these sites by forming hot bubbles. The star formation is hence quickly terminated, the bubbles overlap and form the wind. The outflowing ISM forms filaments that are observed in optical emission lines, but also in soft X-rays. Previous work suggests that the hot gas carried away by the wind is highly enriched with metals, comparable to solar metallicity, which is due to direct enrichment from supernovae. Winds are typically strong enough to propagate far into the intergalactic medium, suggesting that in combination with supernova feedback they likely play an important role in metal enrichment of galactic haloes. Here, we investigate thie phenomenon of galactic outflows with 3D hydrodynamics simulations, in particular with the aim to find a logical connection between star formation, turbulence and outflow characteristics. Both wind and convective solutions shall be investigated. Up to now, it is still poorly understood which mechanisms are most significant to launch a galactic outflow. Our simulations yield evidence that ISM turbulence generated by SN feedback is insufficient to give rise to an outflow, and a medium comprising several gas phases of different temperatures and densities is required in any case. Yet, buoyancy of SN bubbles might already become significant in a few cases, though it is likely not the main wind driver. This leaves the thermal pressure within superbubbles as the most promising source of energy strong enough to provoke galactic-scale outflows. It turned out, however, that under a wide range of conditions our simulations fail to blow a wind. To clarify in detail, which processes favour the development of convective or wind-like outflows, we propose a systematic parameter study.
Galaxy formation and evolution
The formation and evolution of galaxies is still unknown in many aspects. Studies of the large scale structure of the universe revealed that galaxies live in several different environments. Some galaxies live in the field, that is they show no signs of interaction with neighbouring galaxies and the distances between the galaxy and its nearest massive neighbours are large compared to their size. Most of the field galaxies are spiral galaxies. In contrast a cluster consists of more than 100 galaxies within 2-10 Mpc and therefore is a very dense region. Galaxies in these environments have very high relative velocities and show signs of interactions. The fraction of elliptical galaxies is high compared to the field. Independent of the environment observations show a decrease of the fraction of star forming galaxies with decreasing redshift. This lead to the assumptions that elliptical galaxies are build by collisions of galaxies, which is a common event in a dense environment, while in a field environment galaxy mergers are uncommon and therefore the galaxies remain undisturbed and are supposed to mostly grow by smooth accretion.
General formation setup for galaxies
In the standard cosmology assumed today, the $\Lambda$ Cold Dark Matter scenario, in the early universe small perturbations collapse first and produce dark matter halos that accumulate baryons at the center. The small structures aggregate successively into larger structures. This process is called the hierarchical growth. Therefore, if we want to study galaxy formation we need to analyze and understand their assembly history. Numerical simulations have shown that there are several processes for the mass assembly: merger events between two nearly equally massive halos, called major merger, merger between a massive halo and a small sattelite halo, called minor merger, and the accretion of material that is not correlated with the assembly of a halo, for example in the course of a flyby event or by accretion of gas along filamental structures followed by in situ star formation. With increasing computational power it is now possible to perform cosmological simulations and to follow the merger and accretion history of halos from high redshifts to z=0.
Elliptical galaxies build up scenarios
For elliptical galaxies there are two main possible scenarios assumed: Major merger and minor merger (or cosmological) growth. Observations tell us that there exists ellipticals already at high redshifts, but we also see major merger currently at work (for example in the Antaennae system and the Mice system), especially in medium dense environments like galaxy groups. Simulations of major mergers between two big spiral galaxies have successfully reproduced elliptical galaxies, but they failed to explain the formation of the most massive ellipticals in the universe, the BCGs (Brightest Cluster Galaxies). Cosmological simulations recently revealed another formation scenario: Here the most massive structures grow through the continuous infall of small galaxies, so called minor mergers. In these scenario the galaxy grows fast at high redshift, where major merger are more common due to the fact that the masses of the galaxies are lower at higher redshifts (hierarchical growth) and the environment is generally denser (less evolved structures), and the accretion slowes down at lower redshifts, dominated by the infall of less massive structures. If enough of these small structures fall into a galaxy, they are able to change the morphology of the galaxy very efficiently. In addition, cosmological simulations show that about 40% of the material of a galaxy is accreted by smooth accretion. Star formation is triggered by major and minor mergers, thus in both scenarios there is (nearly) no cold gas inside the ellipticals left when the elliptical has formed, and the star formation gets to an end, the ellipticals are red and dead, as observed.
Joint evolution of galaxies and black holes
In the current picture of galaxy formation it is now generally accepted that every massive spheroidal galaxy host a supermassive black hole in its center. Moreover, black hole masses are found be tightly correlated to host galaxy properties, as luminosity, bulge mass and velocity dispersion. This suggests that the evolution of galaxies and the growth of black holes is not de-coupled, but goes hand and hand. E.g. feedback from actively accreting black holes is assumed to regulate black hole growth itself and also to influence the evolution of the host galaxy by quenching cooling and further star formation. Furthermore, major mergers and the corresponding infall of large amounts of cold gas, which is then available for star formation and provides also a fuel for black hole growth, are believed to be important events in establishing the observed, local relations between black hole masses and host galaxy properties. However, detailed physical processes are still poorly understood and thus, subject of current research. For example, a puzzling, open question is to explain the anti-hierarchical trend in black hole growth within a hierarchical structure formation model. Different observational studies show that - investigating the number density evolution of active galactic nuclei (AGN) - more luminous AGN peak at higher redshift than less luminous ones. This implies that more massive black holes seem to form already very early in the Universe, whereas less massive objects seem to predominantly evolve only at later times, in contrast to what is expected from our current hierarchical structure formation models. Here, semi-analytic models are appropriate tools in order to study easily the effect of different trigger mechanisms for black hole accretion, the corresponding accretion efficiencies and different prescriptions for AGN feedback in a large statistical sense, in particular the influence on the AGN number density evolution.