Alumni 2018

Bekdaulet Shukirgaliev   (KazakhsTan)                                                                                                                                   29.11.2018

The  life of star clusters, from birth to dissolution: a new approach   ( thesis pdf, 17 MB )

We study the evolution of star clusters, starting from their birth in molecular gas clumps until their complete dissolution in the Galactic tidal field. We have combined the “local-density-driven cluster formation” model of Parmentier and Pfalzner (2013) with direct N-body simulations of star clusters following instantaneous expulsion of their residual star-forming gas. Our model clusters are formed with a centrally peaked star-formation efficiency (SFE) profile, that is, the residual gas has a shallower density profile than stars. We build a large grid of simulations covering the parameter space of global SFEs, cluster masses, sizes and galactocentric distances. We study the survivability of our model clusters in the solar neighborhood after instantaneous gas expulsion and find that a minimum global SFE of 15 percent is sufficient to produce a bound cluster. Then studying their long-term evolution we find that our simulations are able to reproduce the cluster dissolution time observed for the solar neighborhood, provided that the cluster population is dominated by those formed with a low global SFE (about 15%). Finally, we find that the cluster survivability after instantaneous gas expulsion, as measured by cluster bound mass fraction at the end of violent relaxation, is independent of the Galactic tidal field impact.

Supervisor:  Genevieve Parmentier (ARI)

Kseniia Sysoliatina   (Ukraine)                                                                                                                                                 21.11.2018

The structure and kinematics of the Milky Way disk in a view of a semi-analytic chemo-dynamic model   ( thesis pdf, 8 MB )

In this work we study the spatial structure as well as the chemical and kinematic properties of the Milky Way disk on the basis of a semi-analytic chemo-dynamical model from Just and Jahreiß (2010) (JJ model). Assuming inside-out formation and a constant thickness of the MW disk, we generalise the local JJ model to Galactocentric distances of R = 4-12 kpc. At each radius we assume a star formation rate (SFR) with a peak shifting to younger ages for the outer disk and use the four-slope broken power-law initial mass function (IMF) from Rybizki and Just (2015). The age-velocity dispersion and age-metallicity relations (AVR and AMR) are then obtained self-consistently; the latter is constrained by metallicity distributions of the Red Clump stars from the Apache Point Observatory Galactic Evolution Experiment (APOGEE, Eisenstein et al., 2011). Within a forward modelling framework, we validate the local JJ model in the solar cylinder by testing it against the samples from the Radial Velocity Experiment (RAVE, Steinmetz et al., 2006) and the first and second Gaia data releases (DR1 and DR2, Lindegren et al., 2016; Gaia Collaboration et al., 2018). We find an overall mismatch in star counts up to ~9%, with a statistically significant discrepancy identified for the dynamically cold populations. We also develop a new treatment of the asymmetric drift and apply it in three metallicity bins to the RAVE local sample and G-dwarfs from the Sloan Extension for Galactic Understanding and Exploration (SEGUE, Yanny et al., 2009). The tangential component of the solar peculiar motion obtained from the RAVE sample is V = 4.47 ± 0.8 km/s. The rotation curve reconstructed from the SEGUE G-dwarfs in a range of distances R = 7-10 kpc has a near-zero slope of 0.033 ± 0.034. The thick disk G-dwarfs are found to be kinematically homogeneous with a scale length of 2.05 ± 0.22 kpc, which is in agreement with values from the literature.

Supervisor: Andreas Just (ARI)

Michael Rugel (Germany)                                                                                                                                                           21.11.2018

On the formation and destruction of molecular clouds with the Galactic plane survey THOR   ( thesis pdf, 13 MB )

This thesis investigates the properties of molecular clouds with THOR (The HI, OH and Radio Recombination Line (RRL) survey of the Milky Way). We analyze OH absorption at 18 cm within THOR and follow-up observations. We derive the abundance with respect to molecular hydrogen and the total number of hydrogen nuclei: 1) We find a decreasing OH abundance with increasing column density of molecular hydrogen. 2) Due to significant column densities of atomic hydrogen at low N_OH, the OH abundance with respect to N_H is approximately constant. 3) We detect OH components which are associated with gas that is not predominantly molecular or even CO-dark. We conclude that OH is a potential tracer for diffuse gas. Regarding the impact of star clusters on molecular clouds, we detect signatures of feedback in RRL emission in the star forming region W49A. A comparison to the WARPFIELD models (one-dimensional models of feedback-driven shells) indicates that feedback is not yet strong enough to disperse its molecular cloud and that the shell is either in process of re-collapsing to initiate a new event of star formation or has already re-collapsed. This suggests that at least parts of the star formation in W49A is regulated by feedback.

Supervisor: Henrik Beuther (MPIA)

Sara Rezaei Khoshbakht (Iran)                                                                                                                                                 09.11.2018

3D map of the dust distribution in the Milky Way ( thesis pdf, 10MB )

In this thesis, I present a new non-parametric model for inferring the three-dimensional (3D) distribution of dust density in the Milky Way. Our approach uses the extinction measured towards stars at different locations in the Galaxy at known distances. Each extinction measurement is proportional to the integrated dust density along its line of sight (l.o.s). Making simple assumptions about the spatial correlation of the dust density, we infer the most probable 3D distribution of dust across the entire observed region, including along sight lines which were not observed. This is possible because our model employs a Gaussian process to connect all l.o.s. The result is a smooth, 3D map of the dust density, which is the local property of the interstellar medium (ISM) rather than an integrated quantity. Owing to our smoothness constraint and its isotropy, the method provides one of the first maps without “fingers of God” artefact. I then present the first continuous map of the dust distribution in the Galactic disk out to 7 kpc within 100 pc of the Galactic midplane, using red giant stars from SDSS APOGEE DR14. The resulting map traces some features of the local Galactic spiral arms, even though the model contains no prior suggestion of spiral arms, nor any underlying model for the Galactic structure. This is the first time that such evident arm structures have been captured by a dust density map in the Milky Way. Our resulting map also traces some of the known giant molecular clouds in the Galaxy and puts some constraints on their distances, some of which were hitherto relatively uncertain. I also demonstrate a map of the 3D distribution of dust in the Orion complex. Orion is the closest site of high-mass star formation, making it an excellent laboratory for studying the ISM and star formation. We use data from the Gaia-TGAS catalogue combined with photometry from 2MASS and WISE to get the distances and extinctions of individual stars in the vicinity of the Orion complex. We find that the distance and depth of the cloud are compatible with other recent works, which show that the method can be applied to local molecular clouds to map their 3D dust distribution. We also use data from the recent second Gaia data release (GDR2) to update the map that shows complex dust clouds in the Orion region. I finally show a 3D map of hydrogen density in the local ISM. The hydrogen equivalent column densities were obtained from the Exploring the X-ray Transient and variable Sky project (EXTRAS), which provides equivalent NH values from X-ray spectral fits of observations within the XMM-Newton Data Release. A cross-correlation between the EXTRAS catalogue and the first Gaia Data Release was performed in order to obtain accurate parallax and distance measurements. The resulting map shows small-scale density structures which can not be modelled using analytic density profiles.

Supervisor: Coryn Bailer-Jones (MPIA)

Sabina Puerckhauer (Germany)                                                                                                                                                 07.11.2018

Characterising light concentrators for CTA and optimising the data selection to improve angular resolution and sensitivity ( thesis pdf)

The Cherenkov Telescope Array (CTA) is the next generation ground-based observatory for gamma-ray astronomy that will reach a performance unprecedented in the field. This thesis focuses on optimising this performance in the core energy range of CTA around 1 TeV. In a first part, the wavelength-dependent angular efficiency of light concentrators for the camera system FlashCam, proposed for the medium size telescopes of CTA, is determined with a dedicated test system. For hexagonal concentrators with three reflective coatings a signal-to-noise ratio enhancement of 2.2 compared to a camera without concentrators is observed. By varying the distance between concentrators and photo-sensors, a fine-adjustment of the angular efficiency increases this ratio by another 3%-5%. In a second part, the angular resolution and the sensitivity of CTA are studied by means of simulations. By optimising different quality selection cuts on telescope data, sensitivity enhancements of 20%-40% compared to the CTA requirements are reached and 30%-40% for the angular resolution. With the optimised cuts, spatially extended emission models of the radio galaxy Centaurus A are investigated and it is found that the optimised angular resolution of CTA allows for different theoretical emission models to be discriminated based on the predicted emission regions.

Supervisor: James Hinton (MPIK)

Jolanta Zjupa  (Poland)                                                                                                                                                             26.10.2018

The impact of feedback on galactic and extra-galactic scales  ( thesis pdf, 7 MB)

Recent cosmological hydrodynamical simulations were for the first time able to produce galaxy populations with realistic sizes and morphologies. This success can be attributed to the inclusion of subgrid models for supernovae winds and active galactic nuclei (AGN) feedback. In this thesis, we investigate the impact of feedback driven galactic outflows. First, the expulsion of gas proves to be crucial for the rotational support of haloes hosting realistic galaxies. We employ the state-of-the-art hydrodynamical simulation suites Illustris and IllustrisTNG to characterise the amount of specific angular momentum in the baryonic component of haloes. We find the baryonic spin at z = 0 to be a factor of ∼ 2 higher than the dark matter spin, which is due to the transfer of a constant cumulative spin of Δλ = 0.0013 by z = 0 from dark matter to the gas during mergers, and to the preferential expulsion of low angular momentum gas by mostly AGN feedback. Second, galactic outflows impact the state of the diffuse gas on large scales. We employ the Lyman-α forest to examine the feedback induced changes in the inter-galactic medium (IGM) that serves as gas reservoir for accretion onto galaxies. For a clean comparison, we have run a suite of simulations with both galaxy formation physics and with the Quick Lyman-α (QLA) technique yielding an unperturbed IGM. We find the Lyman-α flux power spectrum to exhibit increasingly more power at large scales and correspondingly less power at small scales in the presence of outflows, as well as the IGM to be generally hotter. Employing IllustrisTNG we investigate the excess Lyman-α absorption as a function of impact parameter for haloes exhibiting strong and weak feedback and find significant differences that can largely be explained by the higher temperature of the perturbed gas.

Supervisor: Volker Springel (MPIK)

Svenja Jacob (Germany)                                                                                                                                                             23.10.2018

Cosmic ray feedback in galaxy formation and a numerical model for turbulence ( thesis pdf, 6 MB)

Feedback processes play an important role in galaxy formation since they regulate star formation both in low mass galaxies and in massive galaxy clusters. Which mechanisms dominate and how the feedback couples to the surrounding medium, are still open questions. In this thesis, we study the feedback from cosmic rays in different environments in more detail. We develop steady state models for a sample of galaxy clusters, in which cosmic ray heating together with thermal conduction prevents large cooling flows. Observational constraints reveal that cosmic ray heating is only viable in clusters that do not show signatures of enhanced cooling. This might indicate a self-regulated feedback cycle. On galactic scales, cosmic rays can drive winds if they are allowed to diffuse or stream out of the galaxy. We demonstrate in simulations of isolated galaxies that cosmic rays are able to regulate star formation in low mass galaxies but the wind efficiency drops rapidly with increasing galaxy mass. Furthermore, almost all astrophysical flows are highly turbulent. This is a challenge for numerical simulations, which cannot resolve all scales of the turbulent cascade. Therefore, we implement a model for turbulence on subgrid scales into the hydrodynamics code AREPO. We validate our model in idealized test cases and apply it to simulations of turbulent boxes.

Supervisor: Volker Springel (MPIK)

Tobias Buck (Germany)                                                                                                                                                               19.10.2018

On the formation of the Milky Way system in cosmological context - A numerical study ( thesis pdf, 300 MB)

State-of-the-art cosmological hydrodynamical simulations have succeeded in modelling realistic Milky Way (MW) type galaxies with spatial resolution of the order of a few hundred parsec, similar to the scale-height of MW's stellar disc and the half-light radius of classical satellite galaxies. I divide the present study into two parts, the build-up of MW's stellar disc and bulge and the formation and evolution of its satellites and dwarf galaxies.
In the first part I show that observed clumpy stellar discs in the early phases of the formation of the Galaxy are dynamically unimportant for its further evolution. This confirms recent observational results where a non linear mapping between stellar mass and light causes stellar discs to appear clumpy. I turn then to explore the formation mechanism of a peanut bulge in cosmological context. I study the kinematical properties of the central stellar populations of a model galaxy using a kinematical decomposition technique and find that the observed kinematic features of the (MW) bulge can only be explained if it consists of both, a peanut bulge and a spherically symmetric bulge both formed via disc instabilities. Observing and disentangling both components will soon be possible thanks to large scale Galactic surveys like Gaia.
In the second part I study the dwarf galaxy population of (MW) mass galaxies. The simulations presented here are among the first to be able to study the formation of dwarf satellite galaxies in a realistic cosmological environment. The employed sub-grid models of the simulations reconcile simulated and observed Local Group satellite mass functions and produce dwarf galaxies whose central stellar velocity dispersion agrees with observations. Using the dwarf galaxies, I test the observational prospects of identifying tidally affected dwarfs in the Local Group using three observables: their distance, line-of-sight velocity and central velocity dispersion. Finally, I investigate the evolution of planes-of-satellites in the framework of the Cold Dark Matter model with a cosmological constant (ΛCDM). These planes quickly dissolve because they consist of a large fraction of chance aligned satellites as recently confirmed with the proper motions of the classical satellite galaxies derived from Gaia data.

Supervisor: Andrea Maccio (MPIA, NYAD)

Michael Walther (Germany)                                                                                                                                                       18.10.2018

Monitoring Thermal Evolution in the Intergalactic Medium over 12 Billion Years ( thesis pdf, 25 MB)

The thermal state of the intergalactic medium (IGM) is an important probe of physical properties for the bulk of gas in the universe. Here, we perform a new measurement of the thermal state for redshift z ≤ 5.4 covering 12 billion years from the endstage of reionization to the present day. For this purpose we measure the Lyman-α forest flux power spectrum based on high resolution quasar spectra from different ground- and space-based spectrographs, combine this analysis with archival measurements of percent level precision, analyze hydrodynamical simulations, use powerful statistical techniques for interpolation, and perform Bayesian inference via Markov chain monte carlo. We observe a rise in the temperature at mean density from 6000 K at z = 5.4 towards 14 000 K at z = 3.4 followed by a cooldown phase reaching 6000 K at z = 0.03. This evolution is provides conclusive evidence for photoionization heating due to reionization of He II, as well as the subsequent cooling of the IGM due to an expanding universe in concordance with model predictions. The agreement with previous measurements is good as well, but our analysis supercedes those by accounting for additional parameters that we marginalize over, and by the vast cosmological timespan our measurement spans. At the highest redshifts z > 5 we infer lower temperatures than expected from the standard picture of IGM heating allowing leaving little room for additional smoothing due to warm dark matter free streaming. Additionally, our measurement for z < 0.5 allows additional constraints on the ultraviolet background in contradiction to previous claims of a UV underproduction crisis.

Supervisor: Joe Hennawi (MPIA, UCSB)

Fabian Klein (Germany)                                                                                                                                                              17.10.2018

Simulations of an accretion disk surrounding a supermassive black hole and its interaction with a nuclear star cluster ( thesis pdf, 34 MB)

We investigate the time evolution of an AD surrounding a SMBH in an active galactic nucleus (AGN) and its dynamical interactions with a the nuclear star cluster (NSC). The AD is important in these interactions because of its dissipative force acting on the NSC stars, resulting in an increased mass flow to the SMBH and asymmetries in the phase space distribution due to its rotation. As the StarDisk project (Just et al., Kennedy et al.) only treated a static AD, viscous hydrodynamical simulations including gravity and self-gravity are used in this thesis to take dissipative feedback and lifetime checks of the AD into account. These simulations were performed using the PLUTO code along with additional modules written by Rolf Kuiper and equilibrium initial condition similar to Shakura & Sunyayev. The results were a quasi-static state as well as the confirmation of the scale-height assumptions from Kennedy et al. and the estimation of the accretion rate reproducing the expected result from Shakura & Sunyayev. Furthermore, the obtained data was used to interpolate the dissipative forces in the direct N-body code NBODY6++GPU and carry out a first test. The inclusion of more physics into the hydrodynamics as well as the advancement of the NBODY6++GPU project to real applications are both tasks for future research.

Supervisor: Rainer Spurzem (ARI)

Gustavo Morales (Chile)                                                                                                                                                            25.07.2018

Stellar Tidal Streams as Cosmological Diagnostics: Comparing data and simulations at low galactic scales ( thesis pdf, 7 MB)

In hierarchical models of galaxy formation, stellar tidal streams are expected around most galaxies. Although these features may provide useful diagnostics of the LCDM model, their observational properties remain poorly constrained. Statistical analysis of the counts and properties of such features is of interest for a direct comparison against results from numerical simulations. In this work, we aim to study systematically the frequency of occurrence and other observational properties of tidal features around nearby galaxies. The approach featured here is based on a visual classification of diffuse features around a sample of nearby galaxies, using a post-processing of optical survey imaging optimized for the detection of low-surfacebrightness stellar structure. At the limiting surface brightness of this sample, 14 − 17% of the galaxies exhibit evidence of diffuse features likely to have arisen from minor merging events. For simulated images, the frequency is 16 − 19%. Our technique recovers all previously known streams in the selected sample and yields a number of new candidates. We conclude that this methodology provides a reliable foundation for the statistical analysis of diffuse circumgalactic features in wide-area imaging surveys, and for the identification of targets for follow-up studies.

Supervisor: David Martinez-Delgado, Eva Grebel (ARI)

Chiara Mazzucchelli (Italy)                                                                                                                                                      12.07.2018

The Physical Properties and Cosmic Environments of Quasars in the First Gyr of the Universe ( thesis pdf, 10 MB)

Luminous quasars at redshift z > 6, i.e. < 1 Gyr after the Big Bang, are formidable probes of the early universe, at the edge of the Epoch of Reionization. These sources are predicted to be found in high–density peaks of the dark matter distribution at that time, surrounded by overdensities of galaxies. In this thesis, we present a search for and study of the most distant quasars, from the properties of their innermost regions, to those of their host galaxies and of their Mpc–scale environments. We search for the highest redshift quasars in the Panoramic Survey Telescope and Rapid Response System 1 (Pan-STARRS1, PS1), discovering six new objects at z >6.5. Using optical/near–infrared spectroscopic data, we perform a homogeneous analysis of the properties of 15 quasars at z> 6.5. In short : 1) The majority of z >6.5 show large blueshifts of the broad CIV 1549 Å emission line, suggesting the presence of strong winds/outflows; 2) They already host supermassive black holes (M_BH = 0.3 - 5 x 10^9 M_sun) in their centers, which are accreting at a rate comparable to a luminosity–matched sample at z =1-3) No evolution of the Fe II/MgII abundance ratio with cosmic time is observed; 4) The sizes of their surrounding ionized bubbles weakly decrease with redshift. We present new millimeter observations of the dust continuum and of the [CII] 158 mm emission line (one of the main coolant of the intergalactic medium) in the host galaxies of four quasars, providing new accurate redshifts and [CII]/infrared luminosities. We study the Mpc–scale environment of a z =5.7 quasar, via observations with broad– and narrow–band filters. We recover no overdensities of galaxies. Among the potential explanations for these findings, are that the ionizing radiation from the quasar prevents galaxy formation, the sources in the fields are dust–obscured, or quasars do not live in the most massive dark matter halos. Finally, we report sensitive optical/near–infrared follow–up observations of gas–rich companion galaxies to four quasars at z >6, firstly detected with the Atacama Large Millimeter Array (ALMA).With the exception of one source, we detect no emission from the stellar population of these galaxies. Our limits on their stellar masses (< 10^10 M_sun) and unobscured star formation rates.

Supervisor: Fabian Walter, Hans-Walter Rix (MPIA)

Jonas Frings (Germany)                                                                                                                                                              11.07.2018

Structure and evolution of simulated dwarf galaxies and Milky Way satellites in Cold and Warm dark matter models ( thesis pdf, 4 MB)

The satellite galaxies and dwarf galaxies in the neighborhood of the Milky Way provide us with detailed observations that can be used to test our standard model of cosmology and structure formation, the $\Lambda$CDM model. I present a sample of 27 cosmological hydrodynamical simulations with virial masses between $5\times 10^8$ and $10^{10}\,\Msun$ that are aimed to study the properties of dwarf galaxies before accretion. The simulated galaxies are able to reproduce observed scaling relations like the dispersion - size and metallicity - stellar mass relation. The stochasticity of merger induced star formation causes a large scatter in the stellar mass - halo mass relation. In galaxies of stellar masses below $10^6 \, \Msun$ stellar feedback is unable to affect the dark matter halo and hence those galaxies retain a cuspy profile. A subsample of 7 halos is used as initial conditions for simulations of satellite - host galaxy interactions in a Milky Way mass halo. The mass removal due to tidal forces creates flat stellar velocity dispersion profiles and efficiently decreases the circular velocity at $0.5\,\mathrm{kpc}$ without stripping a large amount of stars. Additionally the stripping seems to happen in a way that effectively steepens the central dark matter density slope. To investigate the effects of warm dark matter on dwarf galaxies and Milky Way satellites I repeat the study in a $3\,\mathrm{keV}$ warm dark matter scenario. I present the simulations of 21 halos in both CDM and WDM. In WDM the critical halo mass for the onset of star formation is shifted towards higher masses, while the simulations that do produce stars, reproduce the same scaling relations as their CDM counterparts. However, WDM seems to delay the bulk star formation, making galaxies in CDM look about $2\,\mathrm{Gyr}$ older. While halo concentrations are significantly lower in WDM, the central dark matter density slope is slightly steeper for the low mass end. For four halos (in WDM and CDM) I present their evolution as Milky Way satellites. In contrast to the CDM halos, WDM halos are stripped more effective due to their lower concentrations. The survival probability for WDM satellites, on the other hand, is not necessarily lower because of their steeper central slope. Again, the WDM as well as CDM satellites end up with very cuspy profiles after being stripped. I come to the conclusion that the predictions from my simulations do not challenge the $\Lambda$CDM compared to current observational data of dwarf galaxies and Milky Way satellites. An observation of a cored density profile in one of the very low mass objects, however, would force us reconsider the dark matter model. Also WDM would not pose a solution to this problem.

Supervisor: Andrea Maccio (NYU Abu Dhabi, MPIA)

Carsten Littek (Germany)                                                                                                                                                           29.06.2018

Kinetic Field Theory: Momentum-Density Correlations and Fuzzy Dark Matter ( thesis pdf, 1 MB)

Building upon the recent developments of Kinetic Field Theory (KFT) for cosmic structure formation we develop a systematic way to calculate correlation functions of the momentum-density field. We show that these correlators can be calculated from the factorised generating functional after application of partial derivatives with respect to the momentum shift. For visual aid and in order to facilitate an automatic evaluation of corrections by particle interactions we introduce a diagrammatic representation of terms. We employ this formalism to calculate the 2-point momentum-density correlation tensor including initial correlations to quadratic order and completely. A comparison of the results shows that the initial correlations are responsible for the deformation of the power-spectrum on small scales rather than the particle interactions. In the spirit of the Born approximation we use an effective force term to calculate the corrections due to gravity. Our results are in good agreement with previous analytic and simulation results. Recently, Fuzzy Dark Matter models such as Ultra-Light Axions have caught a lot of interest. Their dynamics is described by the classical equations of a condensate. This introduces a quantum potential in the Euler equation and is generally repulsive. We have developed an extension to KFT treating the effects of the quantum potential on the dynamics and on the initial density fluctuation power-spectrum. We find the effects to be largest on scales in the range of 3h/Mpc > k > 0.3h/Mpc, close to the onset of non-linear structures.

Supervisor: Matthias Bartelmann (ITA)

Andreas Schreiber (Germany)                                                                                                                                                    18.05.2018

Diffusion Limited Planetesimal Formation ( thesis pdf, 60 MB)

Planets are surprisingly abundant in our own solar system, but also in extrasolar systems. It is striking to find no explanation for them, as dust in protoplanetary disks was found to not outgrow metres in size. The growth barrier of dust to km-sized planetesimals thus states a missing link onto their formation mechanisms. It is evident for planetesimals to have been present in the early solar system, as their remnants prowl the solar system today in the form of asteroids, Kuiper belt objects, and comets. Of them, many were found to be pristine, giving a hint on what once populated the early solar nebula. Studying the sizes of these pristine objects revealed for all of them a characteristic diameter of 100 km. It is stunning to find this feature independent of distance from the Sun in most pristine object families, hence this feature has to be an imprint of their formation mechanism. This thesis derives a formation criterion for planetesimals out of particle cloud collapse within protoplanetary disks. The found mechanism is capable of reproducing the characteristic sizes of these pristine objects, as it is to first order independent of radial distance from the star. By comparing collapse timescale with turbulent particle diffusion timescale, a minimum size criterion for a dust cloud to collapse is found and investigated. Naturally, dust cloud collapse happens at high dust-to-gas ratios, thus the streaming instability is a good candidate for this turbulent process. Hence, the streaming instability is studied in 2-d and 3-d simulations at dust-to-gas ratios well above unity and on typical collapse length scales. This study found a new instability, namely the azimuthal streaming instability. It operates in the radial-azimuthal plane and has characteristics similar to the streaming instability, thus its name. Subsequent collapse simulations in 2-d and 3-d proved the diffusion limited planetesimal formation to produce planetesimals right at the expected 100 km diameter. It is the conclusion of this thesis to have shown a fundamental concept to be applied in future studies on planetesimals. It has the prospect to make verifiable predictions which can proof this mechanism to have shaped the solar system as we see it today.

Supervisor: Hubert Klahr (MPIA)

Tim Tugendhat (Germany)                                                                                                                                                            15.05.2018

On the Impact of Intrinsic Alignments of Galaxies on Measurements of Weak Gravitational Lensing ( thesis pdf, 8 MB)

Future weak lensing surveys will be too precise for their own good: due to their impressive statistical precision, their systematics must be treated with extreme care. In particular, intrinsic alignments (IA) of galaxies, which cause galaxy ellipticities to be correlated, are indistinguishable from gravitational lensing. I present two established models for IAs and will link them to the physical properties of two galaxy morphologies, namely spiral and elliptical galaxies. Their amplitudes relative to the lensing spectrum is determined and a realistic mix of galaxies and thus utilised to predict the impact of IAs on a Euclid-like tomographic survey and its inferences on a cosmological parameter set. It will be shown that IAs will give rise to significant systematic errors (biases) in future surveys. Furthermore, I present a method for suppressing the IA signal by using colour information and separating the survey into subsamples of blue and red galaxies. This method can also be used to suppress the lensing signal instead, enabling the treatment of just IAs. Finally, I will show a possible way of constrain deviations from General Relativity (Gravitational Slip) by using the model for elliptical galaxies and leveraging its signal against the one from a known weak lensing signal.

Supervisor: Bjoern Malte Schaefer (ARI)

Thales Gutcke (Germany / USA)                                                                                                                                                 02.05.2018

The quenching of star formation in galaxies ( thesis pdf, 20 MB)

This thesis is concerned with investigating what makes star formation inefficient in galaxies. Cosmological, hydrodynamical simulations of galaxy formation show that the energy produced in stars, supernova explosions and active galactic nuclei must couple back into a galaxy’s interstellar medium to prevent excess star formation. However, the physical processes at work in this feedback loop are not well understood. This thesis unravels the details of the baryon cycle to constrain the strength of feedback. The first part explores a phenomenological model of star formation quenching in massive galaxies, showing that gas starvation is a viable pathway to realistic elliptical galaxies. In the second part, a state-of-the-art implementation of stellar feedback is put to the test by comparing the chemical composition of the circum-galactic medium with the latest observations. The simulations exhibit a deficiency in highly ionized oxygen, indicating that models of thermally coupled feedback may be insufficient. The last part delves into the star formation prescription itself, since this directly affects the resulting stellar feedback cycle. An empirical model of a metallicity-dependent stellar initial mass function reveals the significant uncertainty resulting from the common assumption of its universality. Thus, this analysis links star formation processes with stellar feedback and shows how they affect the baryon cycle of entire galaxy ecosystems.

Supervisor: Andrea Maccio (MPIA/NYAD)

Robert Reischke (Germany)                                                                                                                                                        18.04.2018

Cosmology as a probe of gravity with future surveys ( thesis pdf, 7 MB)

The subject of this thesis are different aspects of cosmology as a probe of the underlying gravitational theory with future surveys. In the first part of this work we discuss the parameter dependence of covariance matrices of the power spectrum estimator of the large-scale structure. Its variation across parameter space is calculated analytically by constructing a suitable basis and is then compared with numerical simulations. The method presented is applicable to any matrix-valued function which is everywhere positive-definite. The second part investigates the influence of tidal gravitational fields on the formation of dark matter halos at peaks in the density field of the large-scale structure. We extend the spherical collapse model to incorporate the influence of shear and rotation by treating them as inhomogeneities in the non-linear evolution equation. We investigate the statistics of the tidal field and how it is inherited to the statistics of the critical over-density δc. It is shown that the collapse in a tidal field will always proceed faster than the collapse in a homogeneous background. The last part investigates the combination of observations of weak gravitational lensing, galaxy clustering and the cosmic microwave background and the cross-correlations between the probes to investigate scalar-tensor theories of gravity. We carry out a Fisher analysis as well as a Monte-Carlo-Markov-chain to estimate the expected statistical errors. The analysis shows that gravitational theories can be constrained very well with future surveys.

Supervisor: Bjoern-Malte Schaefer (ARI)

Clio Bertelli Motta (Italy)                                                                                                                                                        20.02.2018

The footprints of stellar evolution on the chemical composition of the Galactic old open cluster M67 ( thesis pdf, 40 MB)

In this work we investigate how stellar evolutionary processes change the surface chemical composition of stars. As a test-bench, we use the old open cluster M67, for whose stars high-resolution spectroscopic data are available in many different evolutionary stages, from the main sequence to the red clump. In particular, we use data retrieved from the archives of large spectroscopic surveys such as APOGEE and Gaia-ESO. First we investigate the effects of the so-called first dredge-up on the surface [C/N] abundance of M67 stars. We then analyse variations in the surface abundances of several elements from the main-sequence to the red-giant phase of M67 stars and discuss how these can be explained by atomic diffusion effects. We also present the results of these investigations in the broader context of Galactic archaeology studies. Furthermore, we investigate the chemical composition of three blue straggler stars and two evolved blue straggler stars in M67 in order to find hints for their formation scenario and discuss the results from the point of view of stellar evolution. Finally, we present an experiment based on TGAS data for the study of the dynamical evolution of OB associations.

Supervisor: Anna Pasquali, Eva Grebel (ARI)

Johannes King (Germany)                                                                                                                                                           09.02.2018

Hochenergetische Gammastrahlung aus dem Galaktischen Zentrum ( thesis pdf, 10 MB)

Multi-wavelength observations of the centre of our Galaxy have revealed a number of energetic processes that are related to the central supermassive black hole. Gamma-ray astronomy contributes to these observations with measurements from satellite and ground based instruments such as Fermi-LAT and HESS. These measurements can make important contributions towards answering some of the key open questions. An analysis of the source HESS J1745-290, potentially linked to acceleration of cosmic rays to PeV-energies, is presented in this thesis. The analysis is performed using the HESS II array, which includes the largest existing Cherenkov telescope, in order to measure photon energies below 100 GeV. Furthermore, the Galactic centre is a promising target for indirect dark matter searches. Such a search is performed in this thesis based on HESS and Fermi-LAT data. Due to the limited angular resolution of current instruments in gamma-ray astronomy, spectral analysis is often the only tool for the investigation of gamma-ray sources. Therefore this thesis includes studies to test which spectral analysis methods can lead to biased parameter estimators.

Supervisor: Werner Hofmann (MPIK)

Carolin Wittmann (Germany)                                                                                                                                                    09.02.2018

Ultra-compact and ultra-diffuse stellar systems in nearby galaxy clusters: signs of environmental influence? ( thesis pdf, 40 MB)

In this thesis we investigate ultra-compact and ultra-diffuse stellar systems in the cores of the nearby Perseus and Fornax galaxy clusters for signs of environmental influences. We search for possible disturbances of their stellar structures by examining their light distributions in deep optical wide field imaging data. In the Fornax cluster we analyse a sample of 355 spectroscopically confirmed compact stellar systems. Our data reveal that many objects show distorted outer structures, although we do not find long tidal streams around any of them. We investigate their spatial and phase-space distributions, and interpret our results in the framework of proposed formation scenarios. In the Perseus cluster we identify a population of 89 diffuse low surface brightness galaxy candidates for which we perform photometry. The majority of the diffuse candidates appear unperturbed based on their stellar structures. We find, however, that galaxies with large sizes seem to be absent in the dense cluster core region. We discuss possible implications for the dark matter content of these systems and compare our sample to faint low surface brightness galaxies in the Coma cluster. Our data reveal a few low-mass galaxies with tidal tails or disturbed morphology, and several diffuse streams and tidal debris. Nevertheless, the number of recent galaxy disruption events seems to be very low in both the Perseus and Fornax galaxy cluster cores, indicating that most of the low-mass galaxy population was probably shaped at earlier epochs.

Supervisor: Thorsten Lisker (ARI)

Valeriy Vasilyev (Russia)                                                                                                                                                         07.02.2018

Dynamical model atmospheres for the abundance analysis of pulsating stars ( thesis pdf, 15 MB)

The chemical composition of Cepheid variables can provide information on the chemo-dynamical evolution of the Galaxy and beyond. The standard method for determining atmospheric parameters and abundances of Cepheids is based on one-dimensional plane-parallel hydrostatic model atmospheres, where convection is treated by Mixing Length Theory. The aim of the thesis is to investigate the impact of the atmospheric dynamics on observable spectroscopic properties. Two approaches are followed: firstly, I construct one-dimensional pulsating atmosphere models implementing a non-local, time-dependent theory of convection, and secondly check the validity of the quasi-static approach against a t wo-dimensional dynamical Cepheid model.
The spectroscopic analysis of the classical Cepheid KQ Scorpii with my one-dimensional model showed that pulsations do not produce strong enough velocity gradients in the line-formation region to explain the estimated microturbulent velocities. The spectroscopic investigation of the two-dimensional Cepheid model allowed to explain the residual line-of-sight velocity of Galactic Cepheids, long known as the “K-term”, by lineshifts of convective origin. Moreover, hydrostatic 1D model atmospheres can provide unbiased estimates of stellar parameters and abundances of Cepheids for particular phases of their pulsations. Summarizing, the main result is a change of paradigm in the context of spectroscopic investigations of Cepheids toward a greater importance of convection than thought previously.

Supervisor: Hans-Guenther Luwdig, Norbert Christlieb (LSW)

Adriana Pohl (Germany)                                                                                                                                                            23.01.2018

Structure of Planet-forming Disks: Multi-wavelength Polarization Diagnostics ( thesis pdf, 10 MB)

The study of dynamic processes that drive the evolution of planet-forming disks is fundamental to understand the origin and diversity of planetary systems. This requires observations at high spatial resolution and sensitivity, which nowadays typically reveal intriguing disk substructures including gaps, rings, spirals, and shadows. This thesis investigates the capability of polarization observations at multiple wavelengths to trace the earliest stages of planet formation. In-depth radiative transfer calculations are carried out in order to link numerical simulations of dust and gas evolution in disks with their observational indicators. This approach demonstrates that measuring polarization is a powerful tool to identify the shaping effects that possible embedded planets have on the density distribution of different dust grain sizes. On the observational part, this work presents several case studies of individual planet-forming disks that were observed with polarimetric imaging by the VLT/SPHERE instrument and subsequently modeled to quantify their structure. A particular focus is the characterization of spiral and ring/gap structures in the context of dust growth, planet-disk interactions, and dust dynamics near ice lines. Furthermore, a modeling study of marginally gravitationally unstable disks is presented to study the influence of the disk self-gravity on the shape and contrast of planet-induced spiral arms in scattered light images. Additionally, it is demonstrated that polarized emission of disks at millimeter wavelengths can be caused by self-scattered thermal dust emission. It is shown that the latter is a viable method to constrain grain properties and identify dust concentrations of different origin. New ALMA observations are presented that offer the first look at a dust trap in polarized scattered light in the sub-millimeter range.

Supervisor: Thomas Henning (MPIA)

Roxana Chira (Germany)                                                                                                                                                           22.01.2018

On Filaments within Molecular Clouds and their Connection to Star Formation ( thesis pdf, 15 MB)

In recent years, there have been many studies on the omnipresence and structures of filaments in star-forming regions, as well as the role of their fragmentation in the process of star formation. However, only a few comprehensive studies have analysed the evolution of filaments and their distribution with the Galactic disk where the filaments form self-consistently as part of large-scale molecular cloud evolution. In this thesis, I study the effect of inclination on dust observations of filaments to evaluate whether the variations would enable the identification of further filaments in existing dust surveys. I address the early evolution of pc-scale filaments that form within individual clouds and focus on the questions how and when the filaments fragment, and how the fragmentation relates to typically used observables of the filaments. I perform dust radiative transfer calculations on models of cylinders and reconstructions of observed star-forming regions. For evaluating the equilibrium state of filaments and the nature of their fragmentation I examine three simulated molecular clouds formed in kpc-scale numerical simulations modelling a self-gravitating, magnetised, stratified, supernova-driven interstellar medium. I find that the observables of filaments in dust emission are on average on small scales influenced by inclination; yet the variations strongly depend on the structure of the object. The first fragments appear when the line masses of the simulated filaments lie well below the critical line mass of Ostriker’s isolated hydrostatic equilibrium solution. This indicate that, although the turbulence of the entire clouds is mostly driven by gravitational contraction, fragmentation does not occur do to gravitational instability, but is supported by colliding flow motions. I conclude that there is no single quantity in my analysis that can uniquely trace the inclination and 3D structure of a filament based on dust observations alone. A simple model of an isolated, isothermal cylinder may not provide a good approach for fragmentation analysis, independently of the dominant driving source of the parental cloud.

Supervisor: Thomas Henning (MPIA)

Rainer Weinberger (Germany)                                                                                                                                                   17.01.2018

Supermassive black holes and their feedback effects in galaxy formation ( thesis pdf, 9 MB)

Supermassive black holes play a key role in modern galaxy formation research. They are conjectured to be present in almost all massive galaxies, and through the release of enormous amounts of energy triggered by gas accretion, they are able to substantially change the properties of the host galaxy. To which extent and how the interaction mechanisms work is an open question. In this thesis, I review the current state of galaxy formation research with a focus on cosmological simulations of structure formation as well as the basic theories of supermassive black holes as far as they are important for galaxy formation. Subsequently, I discuss a new model for black hole growth and feedback in cosmological simulations, along with its application in large cosmological volume simulations. I show how supermassive black holes affect the formation and evolution of their host galaxy as well as their own growth. Furthermore, I present a model for supermassive black hole jets in a galaxy cluster environment. Applying this model, I study the coupling between the jet and the surrounding intra-cluster gas.

Supervisor: Volker Springel (HITS)