IMPRS-HD Alumni 2020

Alumni 2020

Aida Ahmadi  (10.1.)  -  Christos Vourellis  (21.1.)  Alexander Hygate (30.1.)  -  Daniel Haydon (3.2.)  -  Karan Molaverdikhani (5.2.)  -  Xudong Gao (6.2.)  -  Zdenek Prudil  (27.4.)  -  Johannes Esser  (18.5.)  -  Ondrej Jaura  (27.5.)  -  Nico Krieger  (28.5.)  -  Paula Sarkis  (12.6.)  -  Ivana Barisic  (19.6.)  -  Asmita Bhandare  (19.6.)  -  Pooja Surajbali (8.7.)  -  Theodoros Anagnos  (8.7.)  -  Manuel Riener  (10.7.)  -  Michael Hanke (17.7.)  -  Arianna Musso Barcucci  (23.7.)  -  Christian Lenz (28.7.)  -  Johanna Coronado (29.7.)  -  Jonas Klueter  (31.7.)  -  Neige Frankel  (14.10.)  -  Georg Winner  (19.10.)  -  Matteo Pais  (19.10.)  -  Marta Reina Campos  (22.10.)  -  Simon Sailer (28.10.)  -  Sarah Jeffreson (28.10)  - Matteo Mazzarini  (4.11.)  -  Miriam Keppler (11.11.)


   Miriam Keppler  (Germany)                                                                                                                             11.11.2020

Giant Planet Formation in Disks: An Observational Perspective    ( thesis access restricted till June 2021 )

This Thesis is dedicated to the characterization of protoplanetary disks – the cradles of planets – and to the search for forming planets embedded in these disks. To this goal, the presented work analyses high-angular resolution observations in the near-infrared and sub-millimeter wavelength regime using the Spectro-Polarimetric High-contrast Exoplanet REsearch instrument (SPHERE) and the Atacama Large Millimeter Array (ALMA).

SPHERE observations of the circumstellar disk around PDS 70 reveal the presence of a planetary-mass companion in the gap of the disk. This discovery is the first unambiguous detection of a planet still caught at formation within its natal disk and thus represents an unique laboratory to observationally study processes of planet-disk interactions. We perform a thorough analysis of the system using follow-up observations with ALMA, which reveals the disk to be highly structured in both dust and gas. Most prominently, gas and dust are strongly depleted around the location of PDS 70 b, in agreement with the predictions from theoretical models of planet-disk interactions. Further, analysis of the gas kinematics shows perturbations in the disk which can account for the large cavity in the dust disk.

Finally, we present an analysis of SPHERE scattered light observations of the circumbinary disk around GG Tau A. An outer dust ring and the immediate circumbinary environment shows various substructures. These are found to be the result of binary-disk interactions by comparison to hydrodynamical simulations. We conclude that the observed cavity size can be explained by the interactions with the central binary on an orbit which is coplanar with the outer disk. This system illustrates that in certain configurations, multiple systems are able to sustain massive, gas-rich disks over several million years, providing potentially favourable conditions for planet formation.

Supervisor:    Thomas Henning  (MPIA)

   Matteo Mazzarini (Italy)                                                                                                                             04.11.2020

A numerical investigation of the Milky Way and of the satellite tidal debris in the Galactic environment    ( thesis pdf, 25 MB )

I address the satellite debris distribution in the Milky Way (MW) environment by means of N-body simulations, combining full N-body MW models with realistic high-resolution N-body satellites and cosmologically motivated initial conditions. For the choice of the code, I perform a benchmark on previous N-body simulations of the MW environment, proving that Gadget-2 performs similar to more modern codes, and that Gadget-4 offers an improved momentum conservation compared to Gadget-2. Then, with Gadget-4 I simulate the satellite debris distribution in the MW environment. Stars are stripped less efficiently than dark matter (DM) from the satellites and larger fractions of stellar debris are found in the MW central regions, where the stellar and DM debris have different orientations and do not change them significantly if the MW disc is initially tilted. I conclude that the satellite initial conditions have more impact than the disc on the local debris orientation, and that the DM and stellar debris are spatially uncorrelated. Finally, I present a study of the bar in an N-body MW model that matches the observational constraints of the Galaxy. The strong bar formed in this simulation is a slow rotator that influences the local disc kinematics and dynamics and does not present significant buckling.

Supervisor:    Andreas Just  (ARI)

Sarah Jeffreson   (Australia)                                                                                                                     28.10.2020

Giant molecular clouds under the influence of the galactic environment    ( thesis pdf, 15 MB )

The vast majority of star formation in galaxies begins in cold, dense, fractally-structured reservoirs of molecular hydrogen known as giant molecular clouds. The instantaneous properties of these clouds and the time-scales on which they evolve can therefore be built up into models of the empirical properties of galactic-scale star formation, and so can be used to understand this process. In this thesis, we first propose a simple analytic framework to quantify the expected variation in the physical properties and lifetimes of giant molecular clouds in response to changes in their galactic-dynamical environments, finding that they vary within a fundamental parameter space spanned by the orbital angular velocity of the host galaxy, the degree of galactic shearing, the gravitational stability, and the mid-plane hydrostatic pressure. We then explore this parameter space using a set of high-resolution numerical simulations of Milky Way-like galaxies. Due to their high densities and pressures relative to the galactic mid-plane, we find that giant molecular clouds in Milky Way-like galaxies are self-gravitating and decoupled from galactic dynamics, by contrast to their lower-density progenitor clouds of atomic gas, which display systematic, galactic-dynamical variations. Finally, we analyse the full evolutionary history of each simulated cloud population as a function of the cloud spatial scale. Across all Milky Way-like environments, we find that the lifetimes of self-gravitating clouds decrease with their spatial scale below the scale-height of the thin gas disc of the galaxy, and converge to the disc crossing time at its scale-height.

Supervisor:   Diederik Kruijssen   (ARI)

Simon Sailer    (Germany)                                                                                                                            28.10.2020

The FlashCam camera for CTA: trigger verification and fluorescence light detection capabilities   ( thesis pdf, 20 MB )

The next generation of Gamma-ray observatory - Cherenkov Telescope Array (CTA) - aims at improving the detection capabilities at a broad energy range of the gamma-ray spectrum by a factor of 10 in sensitivity. The FlashCam camera is one of the camera types mounted on the medium-sized telescopes (MST) of CTA, responsible for the observation of the core energy range between 150 GeV and 5 TeV. The first part of this thesis assumed the task of verifying the trigger system of FlashCam. Studies of the trigger efficiency and night-sky background light trigger response were performed, whilst also improving the Monte-Carlo description of the detector. The second part was dedicated to the research of the fluorescence light detection capabilities of the full 25 FlashCam-MST sub-array using simulations. A trigger logic was developed, which allows the detection of air showers with primary energies higher than 1 PeV through their fluorescence emission. The effective area of this detection method was determined and the angular resolution using a shower axis reconstruction calculated. The combination of these individual studies allowed the estimation of the sensitivity on point-like gamma-ray sources emitting in the energy range above 1 PeV.

Supervisor:   Jim Hinton  (MPIK)

Marta Reina Campos   (Spain)                                                                                                                        22.10.2020

Formation and evolution of globular clusters in a cosmological context    ( thesis pdf, 24 MB )

Stellar clusters are observed in a variety of galactic environments, from their current formation sites like the disks of the Antennae galaxies to the old globular cluster populations which in the Milky Way mostly reside in the halo. This suggests that the evolution of these puzzling objects may be linked to that of their host galaxy. This thesis explores the formation and evolution of stellar clusters in a cosmological context. For that, analytical models are developed to describe the role of the galactic environment in shaping their demographics. A suite of cosmological, hydrodynamical simulations of Milky Way-mass galaxies from the E-MOSAICS project are used to study when stellar clusters form and how they evolve over cosmic history. These simulations are also employed to estimate the contribution of stellar clusters to the build-up of stellar haloes. Finally, the EMP-Pathfinder simulations are presented, which represent the next generation of simulations of the co-formation and evolution of stellar clusters alongside their host galaxies in a cold, dense cosmic environment. The conclusion drawn from these studies is that stellar clusters are tightly linked to their host cosmic environments. This leads to exciting new future directions that are briefly discussed.

Supervisor:   Diederik Kruijssen   (ARI)

Georg Winner    (Germany)                                                                                                                            19.10.2020 

On the Evolution and Observational Signatures of Cosmic Ray Electrons in Magnetohydrodynamical Simulations    ( thesis pdf, 10 MB )

Cosmic ray (CR) electrons are ubiquitous in the universe and reveal key insights into the non-thermal physics. Their synchrotron radiation, bremsstrahlung, and inverse Compton emission shed light on the interstellar medium, on galaxies and galactic outflows, on galaxy clusters, and on active galactic nuclei. Models of these emission processes can be tested against observations with magnetohydrodynamical (MHD) simulations. While dynamically important CR protons are often included in MHD simulations, CR electrons are rarely treated because their treatment requires a detailed numerical modeling of their spectra due to complex hysteresis effects. Within the scope of this work, I have developed the efficient post-processing code CREST that evolves spatially and temporally resolved CR electron spectra. This code allows a comparison of CR electron emission signatures to observations by means of multi-frequency spectra and morphology. Further, CREST enables to validate models of CR electron acceleration and to explore whether observations can be explained by leptonic or hadronic interactions. Hence, this work opens up a new capability for MHD simulations of dynamical systems that can be compared to radio, X-ray, and γ-ray observations. I apply CREST to three-dimensional MHD simulations of the supernova remnant of SN 1006. These show that a mixed leptonic-hadronic model explains best the observed γ-ray emission and that CR electrons, similar to CR protons, are preferentially accelerated in environments where the shock direction is quasi-parallel to the upstream magnetic field. In addition, the simulations show that a magnetic field amplification by a volume-filling turbulent dynamo is required and that the Bell amplification needs to be spatially confined to the shock. Both findings are necessary in order to match simultaneously radio, X-ray, and γ-ray observations.

Supervisor:   Christoph Pfrommer   (HITS/AIP)

Matteo Pais    (Italy)                                                                                                                                       19.10.2020 

Magneto-hydrodynamical simulation with cosmic ray acceleration at supernova remnants    ( thesis pdf, 15 MB )

Supernova remnants shocks are considered the best sites for the production of Galactic cosmic rays. The interactions of cosmic rays produced at supernova shocks with photon fields and the interstellar medium generate a multi-wavelength spectrum from radio to gamma rays. In particular, TeV gamma-ray emission may originate from both hadronic and leptonic interactions. Recent results from kinetic simulations suggest that the acceleration of cosmic ray ions strongly depends on the relative angle between the shock normal and the local magnetic field orientation. This means that the underlying topology of the interstellar magnetic field in which the supernova remnant expands determines the emission morphology.

Using 3D magneto-hydrodynamical simulation with the code \AREPO, we study the effect of the obliquity dependent shock acceleration on the emission morphology of bright supernova remnants. We apply the results of idealized cases to well-known bright supernova remnants assuming a hadronic model to reproduce different emission morphologies. From the TeV gamma-ray morphology we predict the local composition of the interstellar medium and the coherence scale of the interstellar magnetic field. Furthermore we study the impact of different interstellar environments, such as the case of a clumpy medium for core-collapse supernova remnants. We show that the hadronic model provides a good match for both the observed morphologies and the spectra.

Supervisor:   Christoph Pfrommer   (HITS/AIP)

Neige Frankel     (France)                                                                                                                                 14.10.2020 

Forward Modeling the Secular Evolution of the Milky Way Disk    ( thesis pdf, 15 MB )

We know precisely the position of the Sun in our Galaxy. Yet, like for most stars, we cannot tell where it was born. Stars undergo dynamical memory loss: their orbits evolve, because the Milky Way, like many galaxies, has non-axisymmetric  structures (e.g. bar, spirals) that shuffle stellar orbits. My thesis quantifies the  strength of that process to answer: How (much) do stars change orbit? Can we still infer their birth places, to constrain the formation of the Milky Way disk? I have combined data from the large stellar surveys APOGEE and Gaia, and developed a method to extract the information they contain on the Galactic disk evolution. I forward-modelled the formation of the stellar disk, the stars’ elemental abundances and their subsquent orbital diffusion, which then informs us about their birth radii through ’weak chemical tagging’. I have found that stars can change orbits by large amounts, and most of this evolution is cold (the orbits stay near- circular). Secular evolution determines how the Milky Way disk is structued. If the Milky Way is typical this explain what drives disk galaxies in general to their typical exponential disk density profiles.

Supervisor:   Hans-Walter Rix    (MPIA)

Klüter, Jonas     (Germany)                                                                                                                              31.07.2020

On the use of Gaia for astrometric microlensing    ( thesis pdf, 30 MB )

Astrometric microlensing is a unique tool to directly determine the mass of an individual star(“lens”). By measuring the astrometric shift of a background source in combination with precise predictions of its unlensed position as well as of the lens position, it is possible to determine the mass of the lens with an uncertainty of a few per cent. In this thesis, the prediction of such astrometric microlensing events using the second data release of Gaia is presented, and the possibility of measuring the deflection of these events with Gaia is discussed. In the first part, it was possible to predict 3914 microlensing events between 2010 and 2065 with an expected astrometric shift larger than 0.1 mas. Of these events, 640 have a date of the closest approach between 2020 and 2030. Furthermore, 127 events could be found, which might lead to a photometric magnification larger than 1 mmag. Since the typical timescales of these events are of the order of a few months to years, it might be possible for Gaia to detect the deflections, and to determinethe masses of the lenses. This is investigated in the second part of this thesis. For that purpose, the individual Gaia measurements for 501 events during the Gaia era (2014.5 - 2024.5) were simulated. It is shown that Gaia can detect the astrometric deflection for 114 events by simultaneously fitting the motions of lens and source stars. Furthermore, for 13 and for 34 events Gaia can determine the mass of the lens with a precision better than 15% and 30%, respectively. The results presented in this thesis allow the optimal selection of targets for future observational campaigns.

Supervisor:    Joachim Wambsganss   (ARI)

Johanna Coronado    (Chile)                                                                                                                               29.07.2020

Small scale structure of the Milky Way's stellar orbit distribution    ( thesis pdf, 20 MB )

The exact processes behind the formation and evolution of galaxies are interesting puzzles in modern astrophysics. Our Galaxy offers us the unique opportunity to be studied in detail, as we can obtain the 3D positions, 3D velocities and also the chemical information on a star-by-star basis. Different Galactic surveys have advanced in the effort of studying the Milky Way. The Gaia mission in particular provides the full 6D stellar position-velocity phase-space measurements for millions of its stars. By combining Gaia with chemical information from spectroscopic surveys, we can obtain a detailed physical picture of our Galaxy. In this thesis, we set out to investigate the stellar orbit distribution of the Milky Way, while also adding their chemical information ([Fe/H]) in a chemical tagging generalization approach. We first make use of the spectroscopic information from LAMOST, in combination with parallaxes and proper motions from Gaia. We develop a method to obtain improved spectrophotometric distances (with errors less than 6%) for 150 000 main sequence stars. With more precise distances at hand, we investigate the small-scale structure in the orbit distribution of the Galactic disc for ∼ 600 000 main sequence stars in LAMOST × Gaia. Most stars disperse from their birth sites and siblings, in orbit and orbital phase, becoming ‘field stars’. We explore and provide direct observational evidence for this process in the Milky Way disc, by quantifying the probability that orbit similarity among stars implies indistinguishable metallicity. We define the orbit similarity among pairs of stars through their distance in action-angle space ∆(J, θ) and their abundance similarity by ∆[Fe/H]. By grouping such star pairs into associations with a friend-of-friends algorithm linked by ∆(J,θ), we find that hundreds of mono-abundance groups –some clusters, some spread across the sky– are over an order-of-magnitude more abundant than expected for a smooth phase-space distribution, suggesting that we are witnessing the ‘dissolution’ of stellar birth associations into the field. We finally explore a significantly larger sample of 6.2 million stars with radial velocities in Gaia, which is not limited to main sequence stars. Although this sample does not have [Fe/H] information, we are able to recover the same major groups found in the previous sample in both action and angle space. Moreover, we are able to identify other known associations by simple inspection, opening up the possibility for this method to be applied to further characterize dissolving associations across the Galaxy.

Supervisor:    Hans-Walter Rix   (MPIA)

Christian Lenz    (Germany)                                                                                                                                   28.07.2020

Semi-analytical Modeling of Planetesimal Formation. Implications for Planet Formation and the Solar Nebula    ( thesis pdf, 6 MB )

Planetesimals are the hypothetical building blocks of planets, halfway between dust aggregation and the formation of planetary embryos. The typical diameter of newborn planetesimals was found to be around 100 km. The timedependent production of these planetesimals and their radial distribution in disks around young stars is still unclear. This thesis proposes a semi-analytical model for the planetesimal formation rate that is regulated by the radial pebble flux. The model is implemented into a code that solves the evolution of gas as well as the growth and radial motion of grains. Within this model, planetesimals form as soon as micron-sized dust has grown to pebble-size (typically ~ mm–cm) and a critical pebble flux is reached. The resulting spatial planetesimal profile is steeper compared to the initial dust and gas distribution. E.g., for a temperature profile T∝r^-0.5, the planetesimal profile is expected to follow Σ_p∝r^-2.25 in the inner disk regions. The maximum local planetesimal production is reached for a planetesimal formation efficiency that allows the planetesimal formation timescale and the pebbles drift timescale to be equal. A disk parameter study is performed which enables to set limits on possible parameters for the Solar Nebula by comparing the produced planetesimal profiles with mass constraints for initial planetesimals. This thesis shows that the Solar Nebula was not too large, enclosing most of the mass within 50 au. Outside of 50 au, particle traps needed several hundreds of orbits to form or never formed there. Compared to the mass constraints, the most appealing case that is analyzed in this thesis has a disk mass of around 0.1 solar masses, a fragmentation speed of particles of 2 m/s, and moderate to weak turbulence (α = 3⋅10^-4). The model introduced in this thesis does not require fine tuning in order to meet mass constraints for the Solar Nebula which stresses the applicability of the proposed parameterization to models of planet formation. By sorting pebbles by their origins, this thesis shows that a significant amount of pebble mass passed major ice lines before forming planetesimals in the inner Solar Nebula or before they were accreted by planetary embryos at the current positions of the asteroid belt and Earth. The relative contribution to planetesimals from regions of different particle origins changes for different times of planetesimal formation. This thesis concludes with the importance of pebble transport and the planetesimal formation efficiency for shaping the spatial distribution of planetesimals. The presented planetesimal formation rate model can be used to bridge the gap between the phases of dust growth and the formation of planetary embryos.

Supervisor:    Hubert Klahr   (MPIA)

Arianna Musso barcucci     (Italy)                                                                                                                        23.07.2020

The  relation between discs and young companions - observational studies   ( thesis pdf, 10 MB )

The direct imaging technique brings advantages with respect to other, indirect methods of detecting planets. It is sensitive to larger separations, it can detect companions on a variety of orbital configurations, and it allows to simultaneously image both a companion and the circumstellar disc it resides in, thus being the perfect tool to study companion-disc interactions. Direct observations of Hα emission from young planetary and low-mass stellar companions can also shed light on the early gas accretion phase of planet formation. In this Thesis I use the direct imaging technique to study various aspects of planet-disc interaction and planet formation and evolution. I present the detection of a previously unknown low-mass stellar companion around HD 193571, observed as part of the NaCo Imaging Survey for Planets around Young Stars (ISPY). The companion appears to reside within the gap between the host star and its surrounding disc, making this the third low-mass stellar companion discovered within a debris disc. This system is thus the perfect laboratory where to study the relative importance between self- and companion-stirring models in discs.
I also present the detection of Hα emission from the known substellar companion around the young star PZ Tel. The derived Hα luminosity, combined with age and disc information, indicates that the emission is likely due to chromospheric activity of the companion. This detection further proves the capability of using high-contrast imaging instruments and techniques to detect Hα signatures from companions around young stars. On a larger scale, I present the L’ band Imaging Survey to find Exoplanets in the North (LIStEN), which targeted ∼30 nearby stars with known and well characterised circumstellar discs. LIStEN focuses on characterising the population of wide-orbit giant planets around disc-hosting stars, as well as studying the intricacies of companion-disc interactions. I present the survey’s scientific goals, data selection and observational strategy, as well as the data reduction and analysis. No new planetary companions were detected, and the mass detection limits derived from the observations are combined with information on the disc size and morphology to constrain the presence of unseen planetary and low-mass stellar companion around these disc-hosting stars.

Supervisor:    Thomas Henning  (MPIA)

Michael Hanke     (Germany)                                                                                                                                      17.07.2020

Probing the early Milky Way with stellar spectroscopy   ( thesis pdf, 40 MB )

Stars preserve the fossil records of the kinematical and chemical evolution of individual building blocks of the Milky Way. In its efforts to excavate this information, the astronomical community has recently seen the advent of massive astrometric and spectroscopic observing campaigns that are dedicated to gather extensive data for millions of stars. The exploration of these vast datasets is at the heart of the present thesis. First, I introduce ATHOS, a data-driven tool that employs spectral flux ratios for the determination of the fundamental stellar parameters effective temperature, surface gravity, and metallicity, upon which all higher-order parameters like detailed chemical abundances critically rely. ATHOS’ robustness and widespread applicability is not only showcased in a comparison to large-scale spectroscopic surveys and their dedicated pipelines, but it is also demonstrated to be able to compete with highly specialized parameterization methods that are tailored to high-quality data in the realm of studies with low target numbers. An in-depth study of the latter kind is outlined in the second part of this thesis, where I present a chemical abundance investigation of the metal-poor Galactic halo star HD 20. Using spectra and photometric time series of utmost quality in combination with modern asteroseismic and spectroscopic analysis techniques, I deduce a comprehensive, highly accurate, and precise chemical pattern that proves HD 20 worthy of being added to the short list of metal-poor benchmark stars, both for nuclear astrophysics and in terms of stellar parameters. The decomposition of the chemical pattern shows an imprint from s-process nucleosynthesis on top of the already in itself rarely encountered enhancement of r-process elements. In the absence of a companion that could act as polluter, this poses a striking finding that points towards fast and efficient mixing in the early interstellar medium prior to HD 20’s formation. In the third and last part, spectroscopic data from the SDSS/SEGUE surveys are combined with astrometry from the Gaia mission to form a sample of several hundred thousand chemodynamically characterized halo stars that is scrutinized to establish links between globular clusters and the general halo field star population. Based on the identified sample of probable cluster escapees that includes both first-generation and second-generation (former) cluster stars, I provide important observational constraints on the overall cluster contribution to the buildup of the Galactic halo. A highly interesting – yet tentative – finding is that for those populations of stars that were lost early on, the first-generation fraction appears higher compared to groups that are currently being stripped or still bound to clusters. This observation could indicate either a dominant contribution from since dissolved low-mass clusters or that early cluster mass loss preferentially affected first-generation stars.

Supervisor:    Eva Grebel  (ARI)

Manuel Riener     (Austria)                                                                                                                                      10.07.2020

The detailed velocity structure and distribution of 13CO emission in the Galactic plane   ( thesis pdf, 30 MB )

Studying the detailed velocity structure of molecular gas in our Galaxy is of fundamental importance for understanding structure formation in the interstellar medium. Knowledge about the detailed gas kinematics is moreover essential to map the
distribution and dynamics of the molecular gas in the Milky Way. In this thesis I use the method of spectral decomposition to analyse the 13CO (1–0) observations of the Galactic Ring Survey (GRS). I developed the GaussPy+ package,
specifically designed for the fully automated decomposition of large Galactic plane surveys, to fit the ∼ 2.3 million spectra of this large emission line data set. After extensive validation of the algorithm using synthetic spectra and a GRS test field, I use Gauss Py+ to fit the entire data set of the GRS, resulting in ∼ 4.6 million Gaussian fit components.
These decomposition results provide a new way to analyse the dynamics of the molecular gas over a wide extent of the Galactic plane and study how its velocity structure looks like and varies at Galactic to sub-cloud scales. I find that the velocity dispersion of the gas is increased in the midplane and towards the inner Galaxy, and establish that the integrated emission of the velocity components correlates well with the complexity of the gas emission and the amount of dust emission along the line of sight. Moreover, I uncover qualitatively similar fluctuations in the centroid velocities of the gas components throughout the entire GRS data set, and demonstrate how the fitted linewidths enable the separation of blended gas emission features that originate from nearby regions and far distances.
Finally, I use a Bayesian approach to obtain the current best assessment of the Galactic distribution of 13CO. As prior information, I use the presently most precise knowledge about the structure and kinematics of the Milky Way and an extensive compilation of distances from literature. I perform two different distance calculations that either include or exclude a prior for a model of Galactic features, which allows me to characterise possible biases of the distance estimates and establish more reliable limits on the 13CO distribution. I establish that the majority (76% to 84%) of the 13CO emission is associated with spiral arm features. However, I do not find significant differences between the gas emission properties associated with spiral arm and interarm features.
I conclude that the decomposition results provide a wealth of data enabling new and unexplored ways to interpret the detailed gas velocity structure of large Galactic plane surveys. The methodology and results presented in this thesis allowed for a homogeneous study of the dynamics and distribution of the molecular gas over a large fraction of the Galactic disk. As demonstrated in this work, the information extracted from the detailed gas kinematics and its combination with complementary tracers of the interstellar medium has enormous potential to further our knowledge about the physical processes and mechanisms shaping the interstellar medium.

Supervisor:    Jouni Kainulainen  (MPIA, Chalmers)

Theodoros Anagnos     (Greece)                                                                                                                              08.07.2020

Using astrophotonics to design new components for future telescopes   ( thesis pdf, 20 MB )

With the Extremely Large Telescopes (ELTs) currently under construction we are entering a new era of challenging requirements, which drive spectrograph designs towards techniques that more efficiently use a facility’s light feed. If the spectrograph can operate close to the diffraction limit, this reduces the footprint of the instrument compared to a conventional high resolution spectrograph and mitigates problems and cost issues caused by the use of large optics. By using adaptive optics (AO) to address the wavefront distortions caused by the Earth’s atmospheric turbulence, we can provide diffraction-limited starlight to the telescope’s focal plane. Using astrophotonic spatial reformatters and custom optical fibers to manage the AO output, we can in-
crease the starlight coupled into the instrument. In the first part of the thesis, simulation models are compared to manufactured and on-sky tested astrophotonic reformatters. Re-designing of the structures allowed their simulated performance to be further optimised. This is complemented by the laboratory characterisation of multiple different reformatters.

In the second part of the thesis, everything discussed thus far is combined, leading to the design, manufacture and on-sky test of a novel instrument concept. This new instrument is composed of a multi-core fiber (MCF) with 3D printed micro-optics on its cores, which increase the coupling of light into them. The custom fiber is used to feed starlight from the Subaru Coronagraphic Extreme Adaptive Optics (SCExAO) instrument at the 8.2 m Subaru telescope in Hawaii to a diffraction-limited high resolution spectrograph. The results are promising and highlight the instrument’s potential to change the paradigm with which high resolution spectrographs are built, in particular in the near infrared (NIR), for telescopes equipped with powerful AO systems. This study complements recent work in the field and provides crucial insight for optimising future astrophotonic devices.

Supervisor:    Robert Harris, Andreas Quirrenbach (LSW), Christian Schwab (McQuarie)

Pooja Surajbali      (Mauritius)                                                                                                                               08.07.2020

Observing large-scale structures in the gamma-ray sky   ( thesis pdf, 90 MB )

The essence of this doctoral research constitutes the development and application of novel data analysis and modelling techniques to observations from the High Altitude Water Cherenkov (HAWC) gamma-ray observatory. This thesis is organised in three main parts, culminating in the study of extended very-high-energy sources such as the Fermi bubbles. We first develop a novel discriminator to distinguish between gamma-ray-induced and proton-induced atmospheric showers. Our discriminator is independent of core reconstruction and is useful for enhancing the accuracy of the detector simulation. Secondly, we developed a new background model which incorporates the cosmic-ray anisotropy, exploits all statistics available and has fast computation times. Thirdly, we present a profile likelihood approach to calculate the significance and flux from any region of the sky, which allows the combination of data from different shower sizes while consistently accounting for their relative contributions. With the above tools, we perform blind searches for large-scale structures in the TeV gamma-ray sky. We find a candidate source region with significance up to 5.30 sigma at 16 degree integration scale, which could be a TeV halo associated with a pulsar, molecular clouds or a galactic outflow. Finally, with no significant signal from the north Fermi bubble and its base, we compute their integral flux upper limits, at 95% confidence level and present a hadronic model with an estimated proton cut-off energy at 85 TeV.

Supervisor:    James Hinton  (MPIK)

Asmita Bhandare     (India)                                                                                                                                         19.06.2020

Numerical simulations of star and disc formation  ( thesis pdf, 80 MB )

Magnetized, cold, dense molecular cloud cores provide the birth environment for stars, discs, and planets. The multi-scale scenario of low-mass star formation occurs via the formation of two quasi-hydrostatic cores. Furthermore, the conservation of angular momentum can lead to the formation of a disc around the second core (i.e.~the forming protostar). During these early stages of star formation, magnetically driven outflows and jets can be launched from the first and second cores, respectively. Star, disc, and outflow formation involve complex physical processes, which require a robust, self-consistent numerical treatment.

In this thesis, we use numerical simulations to probe the gravitational collapse scenario that involves the transition of an isolated molecular cloud core to a hydrostatic core with a surrounding disc. We use the \emph{PLUTO} code to perform radiation (magneto-)hydrodynamic (MHD) collapse simulations, using one- and two-dimensional (2D) grids. We include the effects of self-gravity and a grey flux-limited diffusion approximation for the radiative transfer. Additionally, we use for the gas equation of state density- and temperature-dependent thermodynamic quantities to account for the dissociation, ionisation, and molecular vibrations and rotations.

Our spherically symmetric simulations span seven orders of magnitude in spatial scale. We survey a wide range of initial low- to high-mass (0.5 -- 100 M_sun) molecular cloud cores, yielding the largest parameter scan so far. Our results highlight the dependence of the first and second hydrostatic core properties on the initial cloud core properties. These simulations indicate that in the high-mass regime, the first hydrostatic cores do not have enough time to form due to large accretion rates.

We further expand our studies to three different sets of 2D simulations using axial and midplane symmetry. First, we perform 2D simulations for non-rotating molecular cloud cores with masses of 1 M_sun, 5 M_sun, 10 M_sun, and 20 M_sun. For each of these cases, we use an unprecedented resolution to model the evolution of the second core for >~100 years after its formation. For the first time, we demonstrate that convection is generated in the outer layers of the second core. This supports the intriguing possibility that dynamo-driven magnetic fields may be generated during the earliest phases of star formation. Following which, for the 1 M_sun case, we analyse the effects of solid-body rotation on the properties of the hydrostatic cores and disc formation. In this model, the first hydrostatic core evolves into a more oblate, pseudo-disc like structure and a sub-au disc starts forming after the formation of the second core. Finally, we explore the effects of ideal and non-ideal (including Ohmic resistivity) MHD during the collapse of rotating molecular cloud cores. We investigate the dependence of molecular outflows and disc formation on the initial cloud core mass, rotation, resistivity, and magnetic field strength. We find the presence of magnetically driven outflows launched from both first and second cores in the resistive models. We also reveal ongoing disc formation in some of our resistive simulations.

In conclusion, we use detailed thermodynamical modelling to quantify the properties of the hydrostatic cores, outflows, and discs for collapse scenarios with a wide range of initial cloud core properties. The models presented herein will serve as the foundation for follow-up studies that link these theoretical insights with observational signatures.

Supervisor:    Thomas Henning    (MPIA)

Ivana Barisic    (Croatia)                                                                                                                                              19.06.2020

Dust Attenuation and Maintenance Mode Feedback in the 6Gyr old Universe and Now  ( thesis pdf, 15 MB )

This thesis addresses two distinct challenges in the galaxy formation and evolution theory. The first one is to accurately measure the dust attenuation. Understanding the global effects of dust on stellar light is crucial in deriving a number of galaxy physical properties. The second one is to explain how galaxies with the most massive halos remain passive. Models have implemented various feedback mechanisms to explain this phenomenon. Quiescence has been linked to the presence of radio-loud AGN through evidence gathered in the local universe. However, such observations were not feasible at high redshift up until recently. The first part of the thesis presents a novel approach to measure the attenuation of individual star forming galaxies at z ~ 0.8 based on deep optical LEGA-C survey spectra and multi-band photometry. A new attenuation curve modeling technique, and a new prescription for a typical attenuation curve of z ~ 0.8 galaxies are introduced. Using this technique, a large diversity among the main attenuation curve features (slope and UV bump strength) is observed, and the variation of those features with global galaxy properties is inspected. The main finding is that geometric effects dominate observed variations in attenuation. The second part of the thesis explores maintenance-mode feedback in quiescent galaxies through the incidence of radio-loud AGN, both in the local and high redshift (z ~ 1) universe. A high incidence rate of radio-loud AGN among round quiescent galaxies is observed, based on a z ~ 0.1 sample drawn from the SDSS/FIRST/NVSS surveys. On the other hand, radio-loud AGN are not seen among flat quiescent galaxies. This finding brings the general validity of the maintenance-mode feedback picture into question, as a different mechanism must be responsible for keeping these galaxies quiescent. This work is extended to z ~ 1 for a radio-loud AGN sample drawn from LEGA-C/VLA. The finding is that radio-loud AGN preferentially reside in galaxies with large stellar velocity dispersions, old stellar populations and round shapes but a larger sample is required to explore the dependence on flattening.

Supervisor:    Arjan van der Wel   (MPIA)

Paula Sarkis     (Lebanon)                                                                                                                                              12.06.2020

Transiting Exoplanets: Linking Observations and Theory   ( thesis pdf, 20 MB )

The detection of new exoplanets is an important aspect of the exoplanet field in order to increase our understanding on planet formation and evolution. Within this context, this thesis is divided into two parts, where the first part deals with the characterisation of transiting exoplanets and the second part addresses the inflated radii of hot Jupiters by linking the observed properties to theoretical models.

Supervisor:    Christoph Mordasini  (Univ. of Bern),  Thomas Henning (MPIA)

Nico Krieger    (Germany)                                                                                                                                               28.05.2020

Zooming into the Blast Furnace - A close Look into the Molecular Gas in the NGC253 Starburst with ALMA  ( thesis pdf, 40 MB )

Starburst galaxies are characterized by intense star formation at high star formation rate Surface densities and short gas depletion times. Strong stellar feedback drives galaxy-scale winds and outflows in all gas phases, i.e. ionized, neutral and molecular. The extreme conditions in local starburst galaxies are thought to be similar to those in typical high-redshift star forming galaxies, e.g. at the peak of the cosmic star formation history.

In this thesis, we present and analyze 0.15'' (~2.5pc) resolution ALMA CO(3–2) observations of the nuclear starburst in NGC253. Using this data, we study the molecular outflow in unprecedented detail, zoominto NGC253’s super star clusters (SSCs) and compare the starburst to the similar but more quiescent center of the MilkyWay.

Firstly, we kinematically decompose the molecular gas emission in NGC253 into a disk and non–disk component to then separate out the molecular outflow.We systematically improve on previous measurement and obtain mass outflow rates M ~ 14-39 Msun yr^-1 for the starburst. The kinetic energy and momentum of the molecular outflow dominates over the other gas phases and is consistent with being supplied by the starburst at a few percent efficiency.

Secondly, we study the physical and chemical conditions in the molecular gas in the (proto-)SSCs in NGC253, the places where future outflows will be launched from. The SSCs differ significantly in chemical complexity and show up to 55 lines belonging to 14 different chemical species. Spectral modelling allows us to infer spectral line ratios and physical properties. The molecular gas in the SSCs is hot, consistent with UV photon-dominated chemistry and permeated by intense infrared radiation.

Thirdly, we compare the molecular cloud properties in the starbursting center of NGC253 and the MilkyWay Galactic Center (GC), that shares similar properties as NGC253. Using a structure identification algorithmon resolution-, area- and noise-matched datasets allows for a direct comparison of the kinematic structure. Through common cloud scaling relations, we infer a high external pressure (Pext ~ 10^7-7.5 K cm^-3) in NGC253 and a significant amount of unbound (non-self-gravitating) molecular gas that is characterized by high velocity dispersion.

In summary, in this thesis we could follow the life cycle of a molecular outflow from an actively star forming molecular cloud before the launching of an outflow all the way out to distances that are hundreds of parsecs above the starburst disk where the outflow fades away.

Supervisor:    Fabian Walter   (MPIA)

Ondrej Jaura   (Czech Republic)                                                                                                                                      27.05.2020

Study of radiation feedback during formation of Population III stars in primordial minihalos   ( thesis pdf, 4 MB )

In this thesis, we utilize novel radiation transfer technique called SimpleX to study complex ionization and dissociation processes in Population III star formation around 400 million years after the Big Bang. The first part describes the SimpleX method and its implementation as a radiation transfer code SPRAI in the hydrodynamic code Arepo. We test it on several standard test cases and demonstrate its usability for physically accurate calculations in the astrophysical context. The second part presents our model of first-star formation and discuss our results. We follow the collapse of the primordial gas cloud in the central regions of a minihalo. The collapse of the gas is evolved until the first stars form in the densest regions. Subsequently, we cut out the central four parsecs region of the simulation around the star- forming area, continue the simulation for the next 20 kyrs, and simulate radiation feedback from individual stars. The results show that the effect of the ionizing radiation strongly depends on the starting position of the escaping photons and resolution. Simulations in the previous literature neglect accretion disks within the inner 10 AU around the stars. Our simulations show that the lack of resolution leads to an overestimation of the escaping ionizing photons from the accreting Population III stars. We report the trapping of the ionizing radiation on the scale length of the height of the accretion disk.

Supervisor:    Ralf Klessen   (ITA)

Johannes Esser   (Germany)                                                                                                                                              18.05.2020

Physical properties of the circumnuclear cloud distribution in Active Galactic Nuclei   ( thesis pdf, 5 MB )

In this thesis the structure of the broad line region (BLR) and the inner dust torus of Active Galactic Nuclei (AGN) is studied. In a first project we have carried out a multiwavelength reverberation mapping campaign of hot dust in AGN for a sample of 25 nearby AGN with redshifts below 0.2. Reverberation mapping allows to measure the radius of the dust torus which, in relation to the accretion disk (AD) luminosity, can be used as a cosmological standard candle. Despite the radius the multiwavelength approach allows for the investigation of other dust properties like the temperature. An influence of dust temperatures on the relation between the dust radius and the AD luminosity is expected, as hotter dust should be located closer to the heating source at smaller radii. We were able to determine 14 reliable dust radii and 20 reliable dust temperatures for our AGN. This is the largest sample of homogeneously reverberation mapped inner dust tori using a multiwavelength approach and among the largest dust reverberation mapping samples overall. We got tighter constraints on the luminosity radius relation when a novel temperature normalization was applied. Especially the slope of the relation is only in good agreement with the expected value of 0.5 if the temperature normalization is taken into account. Additionally we can determine the surface area of the dust only when the temperature is known. We found that the radial extend of the dust torus behaves comparable to the delay with respect to luminosity.
In the second project we compare concurrent changes of the dust radius to shape variations of broad emission lines (BELs) for NGC 4151 observed from 2004 to 2006. These simultaneous changes are discussed in a variety of dust and BEL formation schemes. Furthermore we use the shape variations to assess possible (especially azimuthal) cloud distributions, which could be responsible for the observed variations. A dust inflated AD provides the framework best suited to explain our findings. The changes in the BELs suggest that this dusty cloud formation happens in spatially confined areas on rather short timescales.

Supervisor:    Joerg-Uwe Pott   (MPIA)

Zdenek Prudil   (Czech Republic)                                                                                                                                    27.04.2020

RR Lyrae stars as tracers of substructure and Galactic archaeology   ( thesis pdf, 30 MB )

The submitted thesis encompasses several topics linked to the usage of RR Lyrae stars in various astrophysical applications related mainly to Galactic archaeology. In addition, projects related to pulsation and physical properties of RR Lyrae variables are also discussed, e.g., uncertainty in the mass of RR Lyrae stars and distortion of photometric light curves due to the shocks.

In our first study, we employed several classes of variable stars (including RR Lyrae variables) to examined a small overdensity found north from the Small Magellanic Cloud – SMCNOD. Using variable stars spatially associated with the SMCNOD we linked this overdensity with the Small Mag- ellanic Cloud. We also speculated about its origin from the Small Magellanic Cloud, due to one of the interactions with the Large Magellanic Cloud.

Two subsequent projects focused on the spatial and kinematical study of the Galactic bulge using the RR Lyrae stars. In the central region of the Milky Way, we found two distinct groups of RR Lyrae stars that differ in their pulsation properties. We associated them with the Oosterhoff groups pre- viously found in the Milky Way’s globular clusters. Both populations are evenly distributed in the Galactic bulge and neither of them can be spatialy and kinematically associated with the Galactic bar.

The RR Lyrae stars in the solar neighbourhood and their association with the Galactic disc is discussed in the following project. We found a small population of the local RR Lyrae stars (up to ≈ 3 kpc distance from the Sun) that kinematically and chemically resembles the thin disc population of stars. Our finding challenges our understanding of the Galactic disc formation, which if these RR Lyrae stars are truly members of the Galactic thin disc, happened more than 10 Gyr ago.

The last two studies focus on the properties of RR Lyrae stars as variable objects and expand our current knowledge about their pulsation and physical properties. First, we expanded the current number of candidates for binary systems among RR Lyrae stars. We analyzed the variation in their ephemerides and estimated the physical parameters of a possible binary companion. Second, for the first time, we provided an extensive photometric study of the atmospheric shocks among RR Lyrae stars. We link the prominence of shocks in RR Lyrae phased light curves with their pulsation properties and discuss their selective behavior inside the Instability strip.

Supervisor:    Eva Grebel   (ARI)

Xudong Gao   (China)                                                                                                                                                          06.02.2020

Low-mass Stellar Evolution Traced with Non-LTE Abundances   ( thesis pdf, 5 MB )

The detailed chemical composition of stellar atmospheres can reveal the structure and evolution of the stellar interiors, otherwise hidden from direct site, as well as the structure and evolution of our entire Galaxy. The advent of several large-scale stellar spectroscopic surveys promises breakthroughs in our understanding of the physical processes that shape stellar surface abundances. However, the full potential of these extremely large and precise surveys is not yet being reached, as standard elemental abundance determinations today are based on the simplifying and incorrect assumption that the stellar atmosphere is in local thermodynamic equilibrium (LTE).

In this thesis I have employed non-LTE radiative transfer methods to tackle two outstanding astrophysical problems. The first problem is related to the chemical homogeneity in the open clusters, which for example is very important to understand how disrupted clusters have formed the Galactic disk and pinpoint the birth location of field stars. Abundance trends with stellar effective temperature have been found in all the analysed elements, indicating that the chemical abundance varies along with evolutionary phase past the turn-off. The overall agreement between our measured abundance patterns and the predictions by the stellar models with atomic diffusion and mixing, implies that the process of atomic diffusion poses a non-negligible effects during the main-sequence phase, which leads to the inhomogeneities in the abundances of open clusters.

The second problem is related to lithium evolution in low-mass main-sequence stars. The primordial elemental abundances predicted by Standard Big Bang nucleosynthesis (SBBN) generally show good agreement with observations. However, a glaring exception is the cosmic abundance of lithium, which SBBN estimates to be three times higher than what is observed in the atmospheres of metal-poor stars in the Galactic halo (i.e. stars on the so-called Spite Plateau). This long-recognized discrepancy has become known as the Cosmological Lithium Problem. In this thesis, I present observational evidence, based on a state-of-the-art non-LTE spectroscopic analysis of more than 100,000 stars from the large-scale spectroscopic “Galactic Archaeology with HERMES"(GALAH) survey, that the surface lithium abundances of these Spite Plateau do not in fact reflect their initial (SBBN) lithium abundances; rather, they have been depleted by a factor of three. This further strengthens the case for an astrophysical solution to the cosmological problem, reconciling tension with predictions of the SBBN.

Supervisor:    Karin Lind   (MPIA)

Karan Molaverdikhani    (Iran)                                                                                                                                            05.02.2020

Characterization of Planetary Atmospheres ( thesis pdf, 65 MB )

After the discovery of the first exoplanet in 1990’s and a fast growing number of discoveries since then, there have been many attempts to observe and characterize their atmospheres. In particular, water and methane have been the focus of many investigations due to their relevance to the origin of life and habitability, as well as their major roles to shape the structure of planetary atmospheres. Abundances retrieved for these species can be also used as a tracer of carbon-to-oxygen ratio (C/O) and metallicity of these atmospheres; hence potentially linking the formation scenarios with the observations. Water’s spectral signature is everywhere, but despite many efforts, there has been only one robust detection of methane and only recently. The question is, “where is methane?”. By applying a hierarchical modelling approach (utilising more than 177,000 thermochemical equilibrium cloud-free, disequilibrium cloud-free, and thermochemical equilibrium cloudy models) we predict that there are four classes of irradiated gaseous planets; two of them (Class-I and Class-II; Teff<1650 K) likely to show signatures of CH4 in their transmission spectra, if cloudy-free and C/O above a certain threshold (aka the “Methane Valley”). The effect of disequilibrium processes on the classification found to be modest with a more continuous transition between Class-II and III planets. Clouds, however, heat-up the deeper parts of Class-I and Class-II planets; removing CH4 from the photosphere. Simultaneously, clouds obscure any molecular features; hence making the observation of methane even more challenging.

Supervisor:    Thomas Henning  (MPIA)

Daniel Haydon    (UK)                                                                                                                                                            03.02.2020

Visualising the Synthetic Universe ( thesis pdf, 45 MB )

Star formation cannot truly be understood from observational data alone;only with simulations is it possible to assemble the complete picture. Observations guide the physics we build into our simulations, yet the impact of different star formation and feedback models can only be investigated with simulations. Synthetic observations allow us to make are alistic comparison to true observations as well as teach us about the emission tracers we depend upon. Through coupling the stellar population synthesis code slug2 to galaxy simulations,we can generate synthetic star formation rate tracer maps. These maps assume differentstellarmetallicities,starformationratesurfacedensities,andsuffer fromvariedamountsofextinction. Thisallowsustoexploreandconstrainthe environmental effects on the characteristic emission lifetimes—the duration for which a tracer is visible. With these emission lifetimes and inconjunction with a new statistical method, the ‘uncertainty principle for star formation’, constraintscanbeplaceduponthedurationsofdifferentevolutionaryphases ofthestarformationprocess,allowingustobetterunderstandthephysicsof starformation and feedback on sub-galacticscales. Studying the interstellar medium can also reveal information about stellar feedback: the gas density structure is altered as a result of the injected energy,momentum,and matter. Surveys of the COemission in galaxies can tell us how the properties of this medium have evolved over cosmic time. Using despotic to model CO line emission of gas found within the Illustris TNG50cosmological simulation,we produce an equivalent synthetic survey. This synthetic survey can be used as a basis for comparison and predictor of observational trends.

Supervisor:    Diederik Kruijssen  (ARI)

Alexander Hygate    (UK)                                                                                                                                                            31.01.2020

The Physics of Cloud-scale Star Formation and Feedback Across Cosmic Time   ( thesis pdf, 60 MB )

Stars are an important visible and massive constituent of galaxies. They form out of cool, dense molecular gas regions known as molecular clouds and in turn impact this gas by emitting energy and mass known as "stellar feedback". Therefore, understanding the formation of stars and the feedback they generate is crucial for understanding galaxy formation. As a result of modern telescope facilities, high sensitivity, high resolution (cloud-scale) imaging of molecular gas is becoming available in an increasing number of galaxies. Analysis of this data with matched resolution observations of recently-formed, massive stars allows the characterisation of the star formation process on the cloud scale. The "uncertainty principle for star formation" is a statistical method for measuring the relative duration and spatial distribution of evolutionary phases of the star formation process. When applied to observational images that trace molecular clouds and regions of young stars, the method measures the duration of molecular cloud lifetimes, the timescale of their destruction by stellar feedback and the mean separation length between star forming regions. In this thesis, I investigate the physics of star formation and feedback on the cloud scale and present contributions to the development of methods for this analysis. First, I assess the impact of noise, astrometric offsets and diffuse emission on measurements made with the "uncertainty principle for star formation". I present a physically motivated method for separating emission from compact structures and diffuse extended structure in an image. The method separates diffuse and compact emission via filtering in Fourier space, with a filter defined by the mean separation length between star forming regions. This method enables the determination of the molecular cloud lifecycle and the mean separation between star forming regions with the "uncertainty principle for star formation" in data containing a diffuse background component. Second, I present the application of the "uncertainty principle for star formation" to determine the lifecycles of molecular clouds in the nearby flocculent galaxy M33. These measurements indicate that clouds in M33 have lifetimes approximately once or twice the timescale for their collapse due to gravitational freefall. Subsequently clouds are dispersed by stellar feedback over a timescale that could allow the earliest supernovae to explode whilst still embedded in their natal clouds. Third, I present the decomposition of tracer images of the molecular and ionised gas in nine nearby galaxies into compact and diffuse components and thus determine the fraction of emission coming from these components. I then present a correlation analysis between these emission fractions and a number of parameters characterising the galaxies in the sample. Last, I summarise the work of the thesis and present some future prospects for extending analyses such as the work presented in this thesis to other galactic environments in the Nearby Universe and further out into cosmic history.

Supervisor:    Diederik Kruijssen (ARI), Fabian Walter (MPIA)

Christos Vourellis    (Greece)                                                                                                                                              21.01.2020

GRMHD Launching of Resistive and Dynamo Active Disks  ( thesis pdf, 60 MB, corrigdm )

Astrophysical jets appear as linear collimated objects of high speed that are typically found in young stellar objects, X-Ray binaries, gamma-ray bursts, or active galactic nuclei. The physical procedures that lead to the development of these jets have been studied extensively in the past years. We believe that the launching of highly relativistic jets requires the existence of an accretion disk threaded by a strong magnetic field that rotates around a black hole. We perform general relativistic magnetohydrodynamic simulations of outflow launching from thin accretion disks. As in the nonrelativistic case, resistivity is essential for the mass loading of the disk wind. We implemented resistivity in the ideal GRMHD code HARM3D, which allows us to run simulations with larger physical grids, higher spatial resolution, and longer simulation time. We present the numerical details of the code and we show numerical test in the resistive regime that prove the robustness of the code. As a reference simulation, we consider an initially thin, resistive disk orbiting the black hole, threaded by a large-scale magnetic flux. As the system evolves, outflows are launched from the black hole magnetosphere and the disk surface. We mainly focus on disk outflows, investigating their MHD structure and energy output in comparison with the Poynting-dominated black hole jet. The disk wind encloses two components -- a fast component dominated by the toroidal magnetic field and a slower component dominated by the poloidal field. The disk wind transitions from sub- to super-Alfvenic speed, reaching velocities approximately 0.1c. We provide parameter studies varying spin parameter and resistivity level and measure the respective mass and energy fluxes. A higher spin strengthens the disk wind dominated by the toroidal component of the magnetic field along the inner jet. We disentangle a critical resistivity level that leads to a maximum matter and energy output for both, resulting from the interplay between reconnection and diffusion, which in combination govern the magnetic flux and the mass loading. For counterrotating black holes the outflow structure shows a magnetic field reversal. We also show the structure and direction of the electric field and its connection with the velocity and magnetic field vectors. Finally, we present the first fully dynamical simulation of dynamo generated poloidal magnetic field in a GRMHD environment. We simulate cases of both accretion tori and disks and we find induced magnetic field with both dipolar and quadrupolar structure. We follow the evolution of the field structure and strength and we show the launching of outflows from the torus/disk surface and the black hole magnetosphere.

Supervisor:    Christian Fendt  (MPIA)

Aida Ahmadi   (Canada/Iran)                                                                                                                                                  10.01.2020


This thesis is dedicated to the search and characterization of disks in high-mass star formation. The work presented is part of the CORE survey, a large observational program making use of interferometric observations from the NOrthern Extended Millimetre Array (NOEMA)for a sample of 20 high-mass protostellar objects in the 1.3 millimetre wavelength regime. An in-depth analysis of the W3(H2O) star forming region examines its fragmentation into two hot cores, separated by 2300 au and engulfed in a rotating circumbinary envelope of dense gas. Higher resolution observations reveal that embedded within each core is a rotating disk-like structure with outflows being ejected along the disk rotation axes. Studying the stability of the disk-like structures confirms that they are gravitationally unstable and prone to disk fragmentation. In an effort to understand the uncertainties involved, we created synthetic observations of a high-resolution 3D radiation-hydrodynamic simulation that leads to the fragmentation of a massive disk at different inclinations and distances. We find that the kinematics of differentially rotating disks resemble rigid-body-like rotation in poorly resolved observations, leading to overestimation of protostellar masses. Despite the lack of resolution, we find that the stability analysis correctly predicts disk fragmentation regardless of the uncertainties. Studying the kinematics of the full CORE sample, we find rotational signatures in dense gas perpendicular to bipolar molecular outflows in most regions. Modelling the level populations of various rotational transitions of the dense gas tracer CH3CN, we find the disk candidates to be on average warm (200 K). Applying the robust stability analysis, we find that most high-mass young stellar objects are prone to disk fragmentation early in their formation due to high disk to stellar mass ratio. Since most high-mass stars are found to have companions, disk fragmentation seems to be an important mechanism by which such systems may be formed.

Supervisor:    Henrik Beuther   (MPIA)

loading content
Go to Editor View