Alumni 2017

Xiaona Sun ( China )                                                                                                                                                           20.12.2017

Nonthermal Processes of Fast Outflows from Astrophysical objects    (thesis pdf, 4 MB)

Outflows are ubiquitous phenomena in the universe. They have been widely observed in both Galactic and extragalactic objects. In this thesis, we analyze three individual sources. Firstly, we re-analyze the high energy γ-ray data of Fermi-LAT on the giant lobes of Centaurus A. We utilize for the first time the Planck observations to derive the fluxes of the lobes. The multiwavelength SEDs reveal a possible leptonic+hadronic origin of the non-thermal emission. Secondly, we re-analyze Chandra observations of the M87 jet with a total exposure time of 1500 kiloseconds to explore the X-ray emission characteristics along the jet. The variabilities of the nucleus and HST-1, as well as the photon spectra for all knots, are investigated. Fitting the radio to X- ray SEDs assuming a synchrotron origin, we show that a broken power-law electron spectrum allows a satisfactory description of the SEDs for most knots except for B, C and D, for which an additional component is needed. We discuss the implications and suggest that a stratified jet model may account for the differences. Finally, we derive the energy distribution of the ultrarelativistic electron in G1.9+0.3, which is the youngest known SNR in the Galaxy, under the assumption that the detected X-ray with Chandra and NuSTAR are of entirely synchrotron origin. The electron acceleration is found to be an order of magnitude slower than the maximum rate provided by the shock acceleration in the nominal Bohm diffusion regime. We discuss the resultant implications in the context of the contribution of SNRs to the Galactic Cosmic Rays at PeV energies.

Maria Jesus Jimenez Donaire ( Spain )                                                                                                                              01.12.2017

Dense gas and interstellar medium in nearby galaxies: the interplay between dense gas and the galactic environment (thesis pdf, 30 MB)

Galaxies act like engines converting gas into stars, which in turn produce the matter around us, laying the foundations of life. Studying the composition of this interstellar gas informs our understanding of how star formation proceeds, while also providing insight into the structure and evolution of our own Galaxy and the Universe. Research has found a strong connection between star formation and the molecular gas within galaxies. However, the most easily accessed molecular observables, such as carbon monoxide (CO) emission, only probe low-density gas in extragalactic systems, allowing us to scratch the surface of the star-forming structures.

Molecules which trace denser gas, such as hydrogen cyanide (HCN), are more challenging to observe but probe the immediate sites of star formation. In this thesis I analyze data from the first survey mapping the entire star forming disk of a sample of nearby galaxies in a suite of dense gas tracers (EMPIRE), to understand the interplay between dense gas and a wide range of galactic environments, distinct from the Milky Way. By studying the content of this dense gas in nearby galaxies I find surprising results: systematic variations of the dense gas fraction and its efficiency to form stars within and among galaxies. While more dense gas is located in regions of high interstellar pressure, this dense gas is less efficient to form stars.

To characterize the dense gas, I also explore how changes in the optical depth can affect the effective gas densities where the dense gas tracers emit, a new measurement in the disks normal star-forming galaxies. To better understand the bulk, lower-density molecular gas out of which the dense gas eventually forms, I also analyze CO line ratios to constrain carbon isotope abundances. I find the abundance to vary systematically within galaxy disks, likely due to strong fractionation effects. The results obtained in this thesis support a model where star formation depends strongly on host galaxy and the local galactic environment in the disk.

Wilma Trick ( Germany )                                                                                                                                                      27.11.2017

Action-based Dynamical Modeling for the Milky Way Disks (thesis pdf, 50 MB)

Understanding the Milky Way’s present structure and assembly history constitutes a crucial constraint on galaxy formation and evolution theory. Galactic surveys like the Gaia mission provide high-precision measurements of positions, velocities, and chemical abundances of soon millions of stars in the Milky Way disk. Exploiting these high quality data requires sophisticated modeling tools

This PhD thesis is dedicated to the development, characterization, and application of RoadMapping, a dynamical modeling machinery aiming to constrain the Galactic gravitational potential and chemo-orbital distribution function (DF) of the stellar disk. RoadMapping proceeds by modeling the observed discrete 6D phase-space positions of stellar mono-abundance populations (MAPs) by an axisymmetric parameterized potential model and an axisymmetric action-based orbit DF in a full-likelihood Bayesian framework. RoadMapping takes into account the survey’s selection function (SF) and measurement uncertainties. RoadMapping builds on previous work by Bovy & Rix (2013), Binney & McMillan (2011), and Binney (2012a).

The first part of this work was published as Trick et al. (2016a) and gives an overview of the RoadMapping machinery. Its characteristics are studied by analyzing a large suite of axisymmetric mock data sets. It is found that RoadMapping constraints on the gravitational potential are robust against minor imperfections in the knowledge of the optimal potential or DF model family, selection effects, or velocity measurement uncertainties, as long as the distance uncertainties of the stars are better than 10%.

The second part is based on Trick et al. (2017) and investigates RoadMapping in the presence of spiral arms by modeling data drawn from an N-body simulation snapshot of a disk-dominated galaxy with strong spiral arms by D'Onghia et al. (2013). This provides a realistic test scenario for RoadMapping to model non-axisymmetric data with axisymmetric models. It is found that RoadMapping always recovers a good average model for the gravitational forces at the location of the stars that entered the analysis.

The third part applies RoadMapping to real data in the Milky Way. It combines measurements by Gaia-TGAS (Lindegren et al. 2016), RAVE (Kunder et al. 2017), and RAVE-on (Casey et al. 2017). Red clump stars are selected and photometric distances are assigned to them following Bovy et al. (2014). A strategy is devised to setup an SF for this sample that can be used in RoadMapping. The sample consists of 16 MAPs in the low-alpha disk. All MAPs provide independent and consistent constraints on the Milky Way’s gravitational potential, measuring the disk scale length and circular velocity at the Sun to high precision, R_s,disk = 3.01 +/- 0.05 kpc and v_circ(R) = 231.4 +/- 0.7 km/s. The total surface mass density at the Sun that is recovered is, with Sigma_tot,1.1kpc = 98 +/- 3 M_sun/pc^2, larger than previous estimates in the literature, which is attributed, however, to the data.

Kalyan Kumar Radhakrishnan Santhakumari ( India )                                                                                                       22.11.2017

Maximising the Science Returns of the LINC-NIRVANA Multi-Conjugated Adaptive Optics System   (thesis pdf, 40 MB)

Earth’s fully turbulent atmosphere prevents ground-based optical and near-infrared telescopes (larger than 30 cm) to reach their full potential, i.e. their diffraction-limited capabilities. In order to overcome this limitation, real-time correction of the aberrations caused by the atmosphere is essential. This is done using a technique called Adaptive Optics (AO). Achieving wide-field correction requires an extension to this technique known as Multi-Conjugated Adaptive Optics (MCAO). My PhD concentrated on maximising the scientific return of one such MCAO system, the LINC-NIRVANA (LN) instrument currently undergoing commissioning at the Large Binocular Telescope. Starting from alignment and calibration in the lab to the on-going commissioning, I have contributed to the optical assembly, integration, verification, and software development of the LN MCAO system. I also solved a particular challenge faced by the MCAO systems, namely the “partial illumination issue”. In addition, I also developed a concept that can improve the AO performance, which we call as the “wind predictive control”. Finally, to understand the astrophysical capabilities of an AO system, I studied a pre-main sequence star system, the T Tauri, using observations from two instruments with advanced AO systems capable of providing high-contrast
high-resolution near-infrared imagery.

Anna Schauer ( Germany )                                                                                                                                                       25.10.2017

Delaying the formation of the first stars The impact of streaming velocities and Lyman-Werner radiation in cosmological hydrodynamical simulations   (thesis pdf, 10MB)

In this thesis, we study which effects can delay the formation of the first stars in the Universe. These Population III stars were the first luminous objects emerging after recombination, starting to end the dark ages. They form in minihaloes or first galaxies, and the minimum mass of these objects is strongly related to the number of Population III stars that formed. In some regions of the Universe, there is an offset velocity between baryons and dark matter, the so-called streaming velocity. In this thesis, we show that in regions with large streaming velocities, the halo masses necessary to form the first stars increase. These regions are the perfect environment for the formation of direct collapse black holes. Feedback in the form of Lyman-Werner radiation emitted by the first stars can destroy molecular hydrogen and prevent the primordial gas cloud from cooling and collapsing. This also leads to an offset in time and halo mass for Population III star formation. To account for the strength of the Lyman-Werner background, we performed simulations that study the escape fraction from minihaloes and the first galaxies. With these results, we provide a piece of the jigsaw of the star formation history of the Universe.

Supervisor: Ralf Klessen (ITA)

Daniele Sorini ( Italy )                                                                                                                                                            17.10.2017

Constraining the physics of the intergalactic and circumgalactic media with Lyman-alpha absorption (thesis pdf, 23MB)

Lyman-alpha (Ly-alpha) absorption features detected in quasar spectra in the redshift range 0 < z < 6 are a powerful tool to probe the intergalactic and circumgalactic media (IGM and CGM) and, consequently, to constrain models of galaxy formation and cosmology. In the first part of this thesis, I overcome certain numerical challenges posed by cosmological hydrodynamic simulations by developing a novel semi-analytic technique to predict various statistics of the Ly-alpha absorption in the IGM with large N-body cosmological simulations. The technique developed in this work is more accurate than previous attempts in the literature. More importantly, it can be applied on Gpc-scale N-body simulations, allowing an accurate investigation of the Ly-alpha absorption at unprecedentedly large scales. In the second part of the thesis, I consider observations of Ly-alpha absorption in spectra of background quasars passing at di↵erent transverse separations (between 25 kpc and 17 Mpc) from foreground galaxies, and compare them with the predictions of different state-of-the-art hydrodynamic cosmological simulations. For the first time, I show that combining observations of the Ly-alpha absorption in the IGM and the CGM can tightly constrain the models of sub-resolution physical processes implemented in simulations (e.g., feedback). With near-future high-precision observations of Ly-alpha absorption, the tools developed in this work will set the stage for even stronger constraints on models of galaxy formation and cosmology.

Supervisor: Joe Hennawi (MPIA)

Kirsten Schnuelle ( Germany )                                                                                                                                                    27.07.2017

Studying the Radiative Response of Circumnuclear Dust of AGNs   (thesis pdf, 10 MB)

Paul Molliere ( Germany )                                                                                                                                                           12.07.2017

Modeling of Exoplanet Atmospheres   (thesis pdf, 30 MB)

Spectrally characterizing exoplanet atmospheres will be one of the fastest moving astronomical disciplines in the years to come. Especially the upcoming James Webb Space Telescope (JWST) will provide spectral measurements from the near- to mid-infrared of unprecedented precision. With other next generation instruments on the horizon, it is crucial to possess the tools necessary for interpretating observations. To this end I wrote the petitCODE, which solves for the self-consistent atmospheric structures of exoplanets, assuming chemical and radiative-convective equilibrium. The code includes scattering, and models clouds. The code outputs the planet’s observable emission and transmission spectra. In addition, I constructed a spectral retrieval code, which derives the full posterior probability distribution of atmospheric parameters from observations. I used petitCODE to systematically study the atmospheres of hot jupiters and found, e.g., that their structures depend strongly on the type of their host stars. Moreover, I found that C/O ratios around unity can lead to atmospheric inversions. Next, I produced synthetic observations of prime exoplanet targets for JWST, and studied how well we will be able to distinguish various atmospheric scenarios. Finally, I verified the implementation of my retrieval code using mock JWST observations.

Supervisor: Thomas Henning (MPIA)

Sebastian Stammler ( Germany )                                                                                                                                                   01.06.2017

The Role of Ices in the Process of Planet Formation       (thesis pdf, 6 MB)

When molecular clouds collapse to form stars, protoplanetary disks consisting of gas and dust are often formed as byproducts of star formation. It is assumed that planets are built from this leftover gas and dust in these disks by collisional growth of dust particles and subsequent accretion of gas. The exact mechanism, however, is not well understood. Protoplanetary disks, in general, have temperature profiles with decreasing temperatures with increasing distances from the star. At the location, where the temperature drops below the condensation temperature of a volatile molecular species – the ice line –, the volatile freezes out as ice. This changes the chemical composition of the dust particles depending on their location in the disk. The composition of planets and planetesimals, that are formed from the dust in these disk, is therefore depending on their formation locations relative to the ice lines. We developed a model to investigate the transport of volatile molecular species in protoplanetary disks including dust coagulation and transport, gas advection and diffusion, and evaporation and condensation of volatiles. We found that particles shortly outside of ice lines are enriched in the respective volatile species due to backward diffusion and recondensation of vapor. This recondensation has also a direct effect on the coagulation physics of the dust particles and can create ring-like, axis-symmetric dust features in protoplanetary disks.

Supervisor: Cornelis Dullemond (ITA)

Christian Arnold ( Germany )                                                                                                                                                        24.05.2017

Hydrodynamical cosmological simulations in f(R) modified gravity   (thesis pdf, 9 MB)

I study the effects of Hu and Sawicki f(R)-gravity on astrophysical processes and cosmological observables. Employing hydrodynamical cosmological simulations carried out with the MG-GADGET and AREPO codes in modified gravity the influences on the Lyman-alpha forest, Milky-Way sized halos, clustering and lensing on large scales as well as on the Sunyaev-Zeldovich effect are investigated. Comparing MG-GADGET to other codes I find that different f(R)-gravity simulation methods agree on a percent-level accuracy for matter power spectra and halo profiles. The f(R) effects are in general smaller at higher redshift, resulting in very small differences between f(R)-gravity and general relativity (GR) for the Lyman-alpha forest. Structural properties of Milky-Way sized halos are however altered by up to 40%. Requiring that the Solar system is screened within the Milky-Way leads to f_R0 = -10^-7 as a constraint on the background parameter. The fifth force is well described by a theoretical approximation in ideal NFW-halos while this estimate is much less accurate for realistic halos. Two point correlation functions and angular power spectra are increased in f(R)-gravity compared to GR. Dark matter halos are in contrast less strongly correlated, leading to a lower halo clustering bias in f(R)-gravity. The angular power in both thermal and kinetic Sunyaev-Zeldovich maps is by a few percent higher in modified gravity compared to a LCDM model.

Supervisor: Volker Springel (HITS)

Richard Hanson ( Germany / UK )                                                                                                                                                 19.05.2017

Mapping 3D Extinction and Structures in the Milky Way      (thesis pdf, 12 MB)

We present a parametric Bayesian method to simultaneously infer interstellar extinction, stellar effective temperature and distance modulus to stars. We create three dimensional maps of extinction towards the Galactic poles using multiband photometry from Sloan Digital Sky Survey and UKIRT Infrared Deep Sky Survey, and maps in the Galactic plane using data from Pan-STARRS1 and Spitzer GLIMPSE surveys. Using optical and near-infrared photometry we train a forward model to emulate the colour change due to properties of stars and the interstellar medium. We predict the probability density function of astrophysical parameters for millions of stars individually and then construct weighted distance profiles in extinction A0 and extinction parameter R0. Furthermore we present a non-parametric model to self-consistently predict the three-dimensional dust density in the Milky Way using a Gaussian process. Using individual extinction and distance measurements to stars and basic assumptions about the spatial correlation of the dust density we infer the most probable density at any point, even if no observations are present along that line of sight. We demonstrate the method’s ability to reliably reconstruct known dust structures with mock data.

Supervisor: Coryn Bailer-Jones (MPIA)

Qian Qian ( China )                                                                                                                                                                        10.05.2017

Accretion and ejection in resistive GR-MHD   (thesis pdf, 10 MB)

In this thesis, the accretion and ejection processes from a black hole accretion system is investigated by means of resistive general relativistic magnetohydrodynamic simulations. As a supplement to the results from prior research with non-relativistic simulations, my results confirm that the winds and outflows originated from thin accretion disks can also be observed in general relativistic simulations. In the first part, the execution of the implementation of resistivity, namely magnetic diffusivity, into the existing non-resistive general relativistic magnetohydrodynamic code HARM is illustrated. The test simulations of the new code rHARM include the comparison with analytical solution of the diffusion equation and a classic shock tube test. rHARM shows reliable performances in these tests. In the second part, rHARM is applied to investigate the evolution of magnetized tori. The results show that the existence of resistivity leads to inefficient accretion of matter from tori onto black holes by weakening the magnetorotational instability inside the tori. An indication for a critical magnetic diffusivity in this simulation setup is found beyond which no magnetorotational instability develops in the linear regime. In the third part, as the main purpose of this PhD project, rHARM is used to perform simulations of magnetically diffusive thin accretion disks that are threaded by a large-scale poloidal magnetic field around non-rotating and rotating black holes. These long-term simulations last 3000 code time units, which are about 195 rotation periods at the disk inner boundary, correspondingly. Their computational domains extend from black hole horizon to 80 Schwarzschild radii. Outflows driven from the accretion disk are clearly seen. These outflows have the typical radial velocity of 0.1 speed of light. In my analyses, I argue that these outflows are driven by the magnetic pressure gradient from the toroidal magnetic field generated by the rotation of the disk. The small ratios of the poloidal field strengths to the toroidal field strengths suggest the interpretation of the outflows as “tower jet,” rather than centrifugally driven winds (Blandford-Payne effect). Furthermore, I find direct evidence of the growths of magnetorotational instabilities inside the accretion disks, which are suppressed by the increasing levels of magnetic diffusivity. This suppression leads to inefficient accretion and ejection processes of the accretion system. Finally, the influence of rotating black holes on the accretion systems are explored. The results show an suppression effect on the black hole spin on the accretion and ejection processes in the system. The tangled field lines within the ergosphere induced by the black hole rotation produce magnetic pressure that pushes against the accreting matter from the disk. In the simulations with large spin parameters, energy extraction from the black hole (Blandford-Znajek effect) is observed, which is, nevertheless, ∼10^2 times smaller than the energy production from the disk outflow.

Supervisor: Christian Fendt (MPIA)

Jakob Herpich ( Germany )                                                                                                                                                                 08.05.2017

On the Physical Origin of Radial Surface Density Profiles in Disk Galaxies    (thesis pdf, 2 MB)

Observations have long established that the radial stellar surface density profiles in disk galaxies are nearly exponential (Type-I profiles). Stellar disks in numerical simulations also tend to approach an exponential profile. Deep imaging has revealed systematic deviations in the profile at large galactocentric radii. Beyond a break the profile may continue with a steeper (Type-II) or shallower (Type-III) exponential profile. In this thesis, I present numerical and analytical models that aim towards a physical understanding of how such profiles come about. I carried out numerical simulations designed to give extensive control over the physical conditions of disk galaxy formation. On this basis, I argue that the type of profile correlates with the initial spin of a galaxy’s host dark matter halo: Type-II/III disks are hosted by high-/low-spin halos. Type-I disks occur at intermediate spins. The formation mechanism for the Type-II disks is consistent with previous results in the literature. Through a very detailed analysis of the low-spin simulations I show that the formation of Type-III profiles can be linked to the formation of a strong bar in low-spin halos. Observational predictions are provided to test the presented hypotheses. The evolution of the radial disk structure can be interpreted as shuffling of the individual stars’ angular momenta. Maximizing a suitably defined entropy in stellar angular momentum space yields an analytic prediction for the radial surface density profiles, given any galactic rotation curve and the corresponding stellar mass and angular momentum of the disk. I carefully compare this result with observational data and simulated disks. It gives a fair match to observations and is in very good agreement with those simulations that provide the closest match to the model assumption of perfectly circular stellar orbits.

Supervisor:   Hans-Walter Rix  (MPIA)

Emanuela Giannini ( Italy )                                                                                                                                                                  08.02.2017

MiNDSTEp differential photometry of the gravitationally lensed quasars WFI2033-4723, HE0047-1756 and Q2237+0305   (thesis pdf, 3 MB)

This work focusses on studying the brightness variation of gravitationally lensed multiply imaged quasars. The main goal is the optimization of the relative differential photometry procedures, which are based on the difference image analysis (DIA) method. Moreover, it aims at isolating uncorrelated flux variations among the quasar images, which can be explained as due to quasar microlensing events, and at the estimation of the time delays of the observed systems from the retrieved light curves. We present V and R photometry of the gravitationally lensed quasars WFI 2033-4723, HE 0047-1756 and Q2237+0305. The analyzed data belong to the MiNDSTEp collaboration and were taken with the 1.54m Danish telescope at ESO/La Silla from 2008 to 2012. The differential photometry is based on the already published method by Alard and Lupton as implemented in the HOTPAnTS package, and additionally uses the GALFIT package for obtaining the quasar photometry. The quasar WFI 2033-4723 shows brightness variations of ~~0.5 mag in V and R during the campaign. The two lensed components of quasar HE 0047-1756 vary by ~ 0.2 to 0.3 mag within five years. We provide for the first time an estimate of the time delay of component B with respect to A of Δt = (7.6+-1.8) days for this object. We also find evidence for a secular evolution of the magnitude difference between components A and B in both filters, which we explain as due to a long-duration microlensing event. We also find that both quasars WFI 2033-4723 and HE 0047-1756 become bluer when brighter, which is consistent with previous studies. The quasar Q2237+0305 shows impressive uncorrelated variations of the four components in both the V and R bands, with brightness variations between ~0.2 and ~1.3 mag. In particular, component D shows flux variations of ~ 1.3 mag in the V band and ~0.8 mag in the R band during the 5-year monitoring campaign, along a caustic-crossing feature of the light curve. We also find that the color of this component becomes redder by ~0.6 mag while it becomes fainter. Image C becomes brighter by ~0.7 mag between the last two monitoring seasons and this again suggests a high-magnification microlensing event.

Supervisor:  Joachim Wambsganss (ARI)

Nina Hernitschek ( Germany )                                                                                                                                                                 26.01.2017

Astrophysical Modeling of Time-Domain Surveys (thesis pdf, 30 MB)

The goal of this work is to develop and apply algorithmic approaches for astrophysical modeling of time- domain surveys. Such approaches are necessary to exploit ongoing and future all-sky time-domain surveys. I focus on quantifying and characterizing source variability based on sparsely and irregularly sampled, non-simultaneous multi-band light curves, with an application to the Pan-STARRS1 (PS1) 3 pi survey: variability amplitudes and timescales are estimated via light curve structure functions. Using PS1 3 pi data on the SDSS "Stripe 82" area whose classification is available, a supervised machine-learning classifier is trained to identify QSOs and RR Lyrae based on their variability and mean colors. This leads to quite complete and pure variability-selected samples of QSO and RR Lyrae (away from the Galactic disk), that are unmatched in their combination of area, depth and fidelity. The sample entails: 4.8 x 10^4 likely RR Lyrae in the Galactic halo, and 3.7 x 10^6 likely QSO. The resulting map of RR Lyrae candidates across 3/4 of the sky reveals targets to 130 kpc, with distances precise to 3%. In particular, the sample leads to an unprecedented map of distance and width of Sagittarius stream, as traced by RR Lyrae. Furthermore, the role of PS1 3 pi as pilot survey for the upcoming LSST survey is discussed.

Supervisor: Hans-Walter Rix (MPIA)

Jorge Abreu Vicente ( Spain )                                                                                                                                                                  25.01.2017

Molecular Cloud Structure at Galactic Scales   (thesis pdf, 60 MB)

Molecular clouds are the sites were stars are born and they play a crucial role in galactic evolution. Despite their main role on star formation and galaxy evolution, physics of molecular clouds are still poorly understood. Particularly, the processes controlling the formation, structure, and evolution of molecular clouds are still a matter of debate and so are the processes that regulate their star–forming activity. Previous to the beginning of this thesis, observational studies of molecular cloud structure and accurate measurements of star–forming activity in molecular clouds existed only for the Solar neighborhood, proving a very limited range of Galactic environments. Extending these studies to larger distances is crucial. This thesis is dedicated to provide the observational assets needed to obtain a Galactic picture of the processes involved in the molecular cloud structure and star– formation. We present the first systematic study of molecular cloud structure and evolution including molecular clouds in nearby spiral arms. We present a census of filamentary–shaped molecular clouds that are thought to be connected to the spiral Galactic structure. Finally, we also develop a new technique that improves the quality of the existing observational data to obtain more accurate observational assets, crucial in the study of molecular cloud structure.

Supervisor: Thomas Henning (MPIA)

Richard Teague ( UK )                                                                                                                                                                                20.01.2017

Tracing the Earliest Stages of Planet Formation through Modelling and Sub-mm Observations     (thesis pdf, 20 MB)

This thesis explores the utility of molecular line emission as a tool to unravel the physical structures and processes involved in planet formation.
Observations of molecular ions, HCO+ and DCO+, in the disk of DM Tauri allow for a study of the ionization structure. These constraints are an essential ingredient in modelling the physical and chemical evolution of a disk which directly impact the efficiency of planet formation.
We also present the first spatially resolved direct measurement of turbulence in a protoplanetary disk using CO, CN and CS molecular line emission. Such a measurement is vital in identifying the physical mechanisms driving turbulence. In addition, we perform a thorough analysis of all uncertainties involved when determining turbulent velocities in disks and demonstrate that all measurements of turbulence will be ultimately limited by the precision to which the gas temperature can be derived.
Finally, the CS emission profile from TW Hydrae displays a dip-like feature coincident with features observed in scattered light observations of the disk, suggesting a common origin. extensive modelling demonstrates that this may be the first detection of a surface density perturbation through molecular line emission, potentially the manifestation of an embedded protoplanet or a strong magneto-rotational instability.

Supervisor:   Thomas Henning (MPIA)

Peter Zeidler ( Germany )                                                                                                                                                                             18.01.2017

Revealing the secrets of Westerlund 2 - A young massive star cluster observed with the Hubble Space Telescope (thesis pdf, 42 MB)

We present a detailed study of the Galactic young massive star cluster Westerlund 2 (Wd2) using an optical and near-infrared high resolution multi-band survey observed with the Hubble Space Telescope. Images obtained in Hα and Paβ filters allowed us to derive a high-resolution pixel-to-pixel E(B − V)_g gas extinction map. This map helped us to individually deredden the stellar photometry, to identify the cluster population, and to determine the properties of Wd2, such as distance (d = 4.16 kpc), total-to-selective extinction (RV = 3.95), and age (1.04 ± 0.72Myr). We identified 240 bona fide pre-main-sequence Hα excess emitters indicating active disk accretion. A careful analysis of the radial dependence of the Hα excess emission shows a 60% lower mass accretion rate in the cluster center, indicating a more rapid disk dispersal in close proximity to the massive OB-stars. We performed state-of-the-art artificial star tests to study the completeness-corrected spatial distribution of the stars. This revealed that Wd2 consists of two subclusters. Additionally, we determined the present-day mass function (PDMF) with a slope of G = −1.53 ± 0.05, which translates to a total stellar cluster mass of (3.6±0.3)·10^4 M_sun. The spatial analysis of the PDMF and the young age of Wd2 indicates that the cluster is, most likely primordial, mass segregated. A spatially uniformly distributed low-mass (< 0.15 M_sun) population, extending into the gas and dust cloud, as well as a confined region of reddened stars, most likely caused by a foreground CO cloud, suggests that cloud-cloud collision might be the origin of the formation of Wd2.

Supervisor: Eva Grebel (ARI)

Tanya Edwards ( UK )                                                                                                                                                                                 16.01.2017

Separation of gamm-Ray, Electron and Proton induced Air Showers applied to Diffuse Emission Studies with H.E.S.S.   (thesis pdf, 10 MB)

A fundamental issue in ground-based gamma-ray astronomy is the identification of γ-ray events among the overwhelming background of air showers induced by charged cosmic rays. Reconstruction techniques exist to distinguish most of the background of hadrons but an irreducible background of electrons and gamma-like protons still remain. I present here a new technique making use of high-altitude Cherenkov light emitted by the charged primary particle and air shower development properties. This method provides a way to distinguish between electrons and gamma rays on a statistical basis. In addition to this, the remaining proton background can also be identified. The technique was developed, tested and applied to studies using the High Energy Stereoscopic System (H.E.S.S.) located in Namibia. The analysis method is especially important in the detection of diffuse signals and eliminates the necessity of a background region in the field of view. The technique was applied to three scientific studies. The latitude profile of the Galactic diffuse gamma-ray emission was analysed. A width of σ = 0.25±0.05 deg (0.20±0.06 deg) for energies of 380 to 900 GeV (1 to 6 TeV) was determined. The cosmic electron spectrum was measured between 0.38 and 14 TeV and a broken power law was fit to the data. The spectrum steepens from an index of 3.08±0.06 to 3.72±0.12 at a break in energy of 1.11±0.04 TeV. In addition, upper limits on the maximum y-ray contamination from the Isotropic γ-Ray Background was placed at 4 x 10^3 (5 x 10^3) MeV cm^-2 s^-1 sr^-1 for energies of 1 to 6 TeV (380 to 900 GeV).

Supervisor: Werner Hofmann (MPIK)