Alumni 2023
Hitesh Lala (India) 15.11.2023
Cepheids and RR Lyrae Stars in the Milky Way and the Magellanic Clouds ( thesis pdf, 40 MB)
RR Lyrae stars and Cepheids play a pivotal role in the study of the Universe. From Milky Way archaeology to cosmology, these variable stars contribute significantly to a myriad of fields. Many large-scale photometric surveys have classified and characterized these stars into various subtypes. However, no single survey is complete, and many of them suffer from varying degrees of contamination. The main goal of this study is to generate a homogeneous, pure, and complete (to date) catalog of Cepheids and RR Lyrae stars in the Milky Way and in the Magellanic Clouds. To this end, we have homogenized RR Lyrae and Cepheid (classical, Type-II, and anomalous) catalogs from eight wide-field optical surveys. We developed a new Bhattacharyya distance-based algorithm for a probabilistic cross-match of these catalogs, taking into account the full astrometry covariance matrix and variability periods. We compared all surveys against each other and numerous literature catalogs to validate their periods and classification. We complemented our compilation by cross-matching with cleaned catalogs of more than 20 other photometric and spectroscopic surveys. These data were used to compute about 200 newly-defined, highly-precise and accurate period-Wesenheit relations. These relations enable the computation of distances at over 98 percent accuracy and better than two percent precision. We release a catalog of about 330\,000 Cepheids and RR~Lyrae stars with more than 1000 features, including periods, subtype classification, optical and infrared photometry, line-of-sight velocities, and distances. This catalog, named RRLCep, holds significant potential in the context of Milky Way archaeology and near-field cosmology.
Supervisors: Eva Grebel (ARI)
Timmy Delage (France) 08.11.2023
The interplay of gas, dust, and magnetorotational instability in magnetized protoplanetary disks ( thesis pdf, 15 MB)
The rich diversity of exoplanets discovered in various physical environments clearly shows that planet formation is an efficient process with multiple outcomes. To un- derstand the emergence of newborn planets, one can rewind the clock of planetary systems by investigating the formation and evolution of their natal environment, the so-called protoplanetary disks. In the core accretion scenario, rocky planets such as the Earth are thought to be formed from cosmic dust particles that grow into pebbles and planetesimals, the building blocks of planets, later assembling to- gether. An intricate puzzle in this theory is how exactly these building blocks are formed and kept long enough in the natal protoplanetary disk. Protoplanetary disks are weakly magnetized accretion disks that are subject to the magnetorotational instability (MRI). It is to date one of the main candidates for explaining their turbulence and angular momentum transport. The nonideal magnetohydrodynamic effects prevent the MRI from operating everywhere in the protoplanetary disk, leading to MRI active regions with high turbulence and non- MRI regions with low turbulence. It has been hypothesized that these variations in the disk turbulence can lead to pressure maxima where dust particles can be trapped. In these so-called dust traps, dust particles can grow efficiently into peb- bles and potentially planetesimals. Yet, it is still an open question how this MRI- powered mechanism shapes the secular evolution of protoplanetary disks, and how it is involved in the first steps of planet formation. It is because the interplay of gas evolution, dust evolution (dynamics and grain growth processes combined) and MRI-driven turbulence over millions of years has never been investigated. The central goal of this thesis is to bridge the gap in the core accretion scenario of planet formation by building the very first unified disk evolution framework that captures self-consistently this interplay. The unique approach adopted in this thesis leads to an exciting new pathway for the generation of spontaneous dust traps everywhere in the protoplanetary disk, which can be potential birth-sites for planets by forming and keeping their necessary building blocks.
Supervisors: Paola Pinilla (MPIA)
Riccardo Franceschi (Italy) 06.11.2023
The Fundamental Properties of Planet-Forming Disks: a Model-Data Comparison ( thesis pdf, 20 MB)
This thesis presents a comprehensive investigation of the properties of protoplanetary disks, the birthplaces of planets. This work is motivated by the rapidly increasing quality of observational data, which necessitates better theories and ways to compare them with what we observe. Firstly, we showcase how the size of the millimeter continuum emitting region constrains disk masses. We assess the method’s efficacy by conducting detailed dust evolution calculations and considering diverse disk properties, particularly in disks exhibiting radial substructures. We then constrain the disk dust grain size distribution by observing only the largest and smallest grains within it. Merging physical dust models with radiative transfer calculations, we compare the model outcomes with the observations of the IM Lup disk, revealing a segregated grain size distribution and providing new insight into grain formation mechanisms and potential triggers for planetesimal formation. Additionally, we characterize the inner disk of the very low-mass star 2MASSJ16053215-1933159 through atomic and molecular hydrogen lines. Using the observational capabilities of the James Webb Space Telescope (JWST), we measure gas temperature, mass, and the stellar accretion rate while showcasing JWST potential in the characterization of protoplanetary disk structures. Lastly, we investigate the gas density and temperature across Class II protoplanetary disks in Taurus using the NOEMA instrument. Employing machine learning-enhanced chemistry and analyzing optically thin and thick CO isotopologue emission, we successfully constrain disk masses and temperatures.
Supervisors: Thomas Henning (MPIA)
Siddhant Deshmukh (Germany) 26.10.2023
Modelling Non-Equilibrium Molecular Formation and Dissociation for the Spectroscopic Analysis of Cool Stellar Atmospheres ( thesis pdf, 30 MB)
Modelling techniques for stellar atmospheres are undergoing continuous improvement. In this thesis, I showcase how these methods are used for spectroscopic analysis and for modelling time-dependent molecular formation and dissociation. I first use CO5BOLD model atmospheres with the LINFOR3D spectrum synthesis code to determine the photospheric solar silicon abundance of 7.57 ± 0.04. This work also revealed some issues present in the cutting-edge methods, such as synthesised lines being overly broadened. Next, I constructed a chemical reaction network in order to model the time-dependent evolution of molecular species in (carbon-enhanced) metal-poor dwarf and red giant atmospheres, again using CO5 BOLD. This was to test if the assumption of chemical equi librium, widely assumed in spectroscopic studies, was still vaild in the photospheres of metal-poor stars. Indeed, the mean deviations from chemical equilibrium are below 0.2 dex across the spectroscopically relevant regions of the atmosphere, though deviations increase with height. Finally, I implemented machine learning methods in order to remove noise and line blends from spectra, as well as to predict the equilibrium state of a chemical reaction network. The methods used and developed in this thesis illustrate the importance of both conventional and machine learning modelling techniques, and merge them to further improve accuracy, precision, and efficiency.
Supervisors: Hans-Guenter Ludwig (LSW)
Arvind Hughes (australia) 20.10.2023
Statistical Methods in the Era of Large Astronomical Surveys ( thesis pdf, 20 MB)
Statistical methods play a crucial role in modern astronomical research. The development and understanding of these methods will be of fundamental importance to future work on large astronomical surveys. In this thesis I showcase three different statistical approaches to survey data. I first apply a semi-supervised dimensionality reduction technique to cluster similar high resolution spectra from the GALAH survey to identify 54 candidate extremely metal-poor stars. The approach shows promising potential for implementation in future large-scale stellar spectroscopic surveys. Next, I employ a method to classify sources in the Gaia survey as stars, galaxies or quasars, making use of additional infrared photometry from CatWISE2020 and discussing the importance of applying adjusted priors to probabilistic classification. Lastly, I utilise a method to estimate the rotational parameters of star clusters in Gaia, with an application to open clusters. This is done by considering the rotation of a cluster as a 3D solid body, and finding the best fitting parameters by sampling constructed likelihood functions. The methods developed in this thesis underscore the significant contributions statistical methodologies make to astronomy, and illustrate how the development and application of statistical methods will be essential for extracting meaningful insights from future large scale astronomical surveys.
Supervisors: Coryn Bailer-Jones (MPIA), Daniel Zucker (Macquarie University)
Sepideh Sadegi (Iran) 26.07.2023
Investigating the Activity of M Dwarfs: Spectroscopic, Photometric, and Modelling Analyses ( thesis pdf, 100 MB)
Stellar activity, characterised by various phenomena such as active regions, flares, coronal mass ejections (CME), etc., can be observed in low-mass stars, such as M-dwarfs. Understanding the activity of M-dwarfs is of paramount importance due to their suitability as hosts for exoplanets, including potentially habitable ones. By comprehending the activity of these stars, we not only enhance the accuracy of exoplanet detection but also gain crucial insights into the habitability of these associated exoplanets. Furthermore, studying activity in low-mass stars provides invaluable knowledge about their internal structure and sheds light on the dynamo processes that generate their magnetic fields. The primary objective of this PhD dissertation is to examine the impact of stellar activity on high-precision spectroscopic and photometric data. In pursuit of this goal, I utilised data from two highly successful exoplanet surveys, namely CARMENES and TESS.
For the first project, a sample of 58 highly active M dwarfs was carefully selected from the CARMENES survey. The aim was to examine the effect of active regions on the radial velocity (RV) measurements in these stars, using the chromatic index (CRX) as a robust parameter capable of capturing the wavelength-dependent RV signatures. The first part of this research focused on assessing the significance of the CRX-RV correlation in these stars, and its potential relationship with key stellar parameters such as mass, rotation period, v sin i, and others. In the second part, the investigation delved into the identification and analysis of periodic RV signals associated with the rotation period (or its harmonics) across the échelle orders of the visible-light (VIS) and near-infrared (NIR) channels of the CARMENES spectrograph.
The second project focused on studying the TESS light curves of the same group of stars, if available, to explore the potential correlation between their flare properties, stellar parameters, and chromospheric activity indices obtained from CARMENES. Given the well-established association between sunspots and flares, it was important to investigate whether a similar relationship exists among these magnetically connected phenomena on other stars. To achieve this, a semi-automated flare detection algorithm was developed to extract key flare characteristics such as amplitude, equivalent duration, and bolometric energy. To investigate the distribution of flares emitted at different energy levels, flare frequency diagrams (FFDs) were generated and parameterised to facilitate the observation of potential relationships.
The final project involved a comprehensive case study that focused on the detailed modelling of the stellar surface of V388 Cas. This was achieved by utilising high-resolution spectra obtained from CARMENES in conjunction with simultaneous TESS photometry. To accomplish this, we employed the powerful modelling tool, StarSim, to effectively simulate a rotating photosphere with spotted regions. The first step involved using the inverse method to retrieve the active regions of the star based on the simultaneous observations. Subsequently, we predicted the RV values across different wavelength ranges, matching the echelle spectrographs’ orders, and compared them to the observed RV values. This comparison allowed us to identify wavelength regions that were more susceptible to spot-induced signals in the absence of atmospheric and instrumental noise. In addition to this analysis, the TESS observations of V388 Cas were leveraged to conduct a study on the evolution of flares. Furthermore, we explored the possibility of a CME occurrence on the star by examining a blue asymmetry in the chromospheric lines of a spectroscopic observation taken 12 hours after a superflare event.
Supervisor: Sabine Reffert / Andreas Quirrenbach (LSW)
Matthew Gent (UK) 17.07.2023
Validation and Application of The Stellar Abundances and atmospheric Parameter Pipeline to derive fundamental parameters of stars in the era of large-scale stellar surveys ( thesis pdf, 16 MB)
SAPP is a pipeline designed to determine accurate parameters of stars in large surveys like Gaia-ESO and Gaia. It combines various observations, including spectra, photometry, astrometry, and asteroseismic data, using Bayesian inference. Validated with benchmark stars, SAPP breaks degeneracies between parameters, yielding precise results. For effective temperature, the typical error is about 100 K, with spectroscopic models dominating uncertainty. Log(g) uncertainty depends on observables, ranging from 0.03 dex to 0.06 dex. Metallicities are recovered with a precision of 0.03 dex for PLATO targets, improved by seismic priors. SAPP also employs an iterative scheme using nu_max = f(Teff, log g) relation, yielding robust results with small differences in temperature and metallicity. It provides fundamental parameters accurate within 1%, meeting PLATO’s goals and enabling exploration of the Galactic structure, including Radial Migration and Age Metallicity Relation. SAPP is used to investigate the alpha-poor and alpha-rich populations in the Galactic disc using Gaia-ESO spectra, Gaia EDR3 astrometry, and photometry. Non-Local Thermodynamic Equilibrium models determine parameters and abundances. A cold metal-poor alpha-poor disc is found in local distributions, suggesting co-evolution of the thick and thin disc. These distributions show well-defined trends in age and kinematic space (Vφ ). SAPP’s accurate age and abundance estimations contribute to understanding Galactic characteristics such as Radial Gradient Measurements.
Supervisor: Maria Bergemann (MPIA)
Jonas Syed (Germany) 13.07.2023
Atomic cloud formation processes - Constraining the properties of the atomic interstellar medium by means of HI emission and HI self-absorption ( thesis pdf, 90 MB)
Atomic hydrogen gas (HI) is an integral constituent of the interstellar medium (ISM) and thus plays a critical role in the assembly of molecular clouds, the sites of star formation. An important physical agent in controlling the transition from atomic to molecular gas is cold HI. HI emission, that traces a wide range of hydrogen properties, is found throughout the Galactic plane and exhibits complex kinematic signatures that are imposed by the Galactic rotation. The unfavorable viewing geometry from our vantage point in the Galaxy requires a new set of tools that allows us to disentangle HI structures along our lines of sight. In this thesis, I constrain the properties of the interstellar hydrogen in our Milky Way galaxy and present a way to isolate cold hydrogen from HI emission using HI self-absorption (HISA). The outer Galaxy offers a less confused view on HI emission, and we exploit this circumstance with unprecedented detail using the high-angular resolution data of The HI/OH/Recombination line survey of the inner Milky Way (THOR).
We discover the “Maggie” filament, one of the largest, almost purely atomic filaments in the Milky Way. Maggie has a length of 1.2 kpc and is located on the far side of the Galaxy at a distance of 12 kpc from the Galactic center. Optical depth measurements suggest that Maggie is in a largely cold HI phase and molecular gas is only found on the smallest spatial scales. When targeting molecular clouds in the inner Galactic plane, we detect pronounced HISA as a tracer of cold hydrogen. While the kinematic correlation is less surprising due to the selection bias of our sample, the spatial distribution of cold HI gas is also found to be well correlated with that of the molecular gas. The detection of HISA furthermore extends to the whole of our survey. We frequently find absorption signatures without any associated molecular counterpart. While cold atomic gas is rendered vitally important to the formation process of molecular clouds, the cold phase of atomic hydrogen marks a distinct gas phase in the ISM that exists throughout the Galaxy, independent of the occurrence of molecular gas.
Supervisor: Henrik Beuther (MPIA)
Nicholas Kurtovic (Chile) 12.07.2023
Origin and characterization of disks substructures, and their relation to stellar hosts ( thesis pdf, 30 MB)
Planets are formed from the gas and dust content available in planet-forming disks around young stars, creating substructures in their density, thermal, and chemical distribution. Characterizing those substructures can provide constraints on the planet-formation potential of each disk. To improve our understanding of how planets are formed around the stars that are the most common in our galaxy, very low mass stars and binary stars, I studied high spatial resolution observations of dust and gas emission from these objects. To maximize information recovery, I analyzed these datasets with visibility-based methods.
The results demonstrate that substructured emission in the dust continuum is present in all spatially resolved disks around very low mass stars, which could be explained by ongoing planet formation. In circumbinary disks, the combination of hydro-models and observations suggest that measuring the eccentricity gradient as a function of radii can be used as a tracer for the presence of Saturn-like planets embedded in the disks. On the other hand, for multiple disk systems, I showed the feasibility of recovering the orbital motion of young objects through the relative movement of their disks, which is crucial to interpreting the emission substructures.
Supervisor: Paola Pinilla (MPIA, MSSL/UCL)
Emily Hunt (UK) 12.07.2023
Improving the census of open clusters in the Milky Way with data from Gaia ( thesis pdf, 20 MB)
For over a century, open clusters have been a key tool for understanding stellar and galactic evolution. Now, thanks to groundbreaking new astrometric and photometric data from the European Space Agency's Gaia satellite, it is possible to study open clusters to never before seen levels of accuracy and precision. In this thesis, I develop and apply new methodologies to improve the census of open clusters with data from Gaia. I focus on using modern, efficient, and statistically rigorous techniques, aiming to maximise the reliability and usefulness of the open cluster census despite the many challenges of working with the billion-star dataset of Gaia. Firstly, I conducted a comparative study of clustering algorithms for retrieving open clusters blindly from Gaia data. I found that a previously untrialed algorithm, HDBSCAN, is the most sensitive algorithm for open cluster recovery. Next, using this methodology, I used Gaia DR3 data to create the largest homogeneous catalogue of open clusters to date, recovering a total of 7167 clusters -- 2387 of which are candidate new objects. I developed an approximate Bayesian neural network for classifying the reliability of the colour-magnitude diagrams of the clusters in the census. Additionally, I used a modification of this network to infer parameters such as the age and extinction of these clusters. Finally, since many of the objects in my catalogue appeared more compatible with moving groups, I measured accurate masses, Jacobi radii, and velocity dispersions for these clusters, thus creating the largest catalogue of these parameters for open clusters to date. Using said parameters, I showed that no more than 5619 of the clusters in my catalogue are compatible with bound open clusters. I used my mass estimates to derive an approximate completeness estimate for the Gaia DR3 open cluster census, finding that the approximate 100% completeness limit depends strongly on cluster mass. The results of this thesis show that it is possible to reliably create a catalogue of open clusters with a single blind search, in addition to measuring parameters for these objects. The methods developed in this thesis will be applicable to future data releases from Gaia and other sources.
Supervisor: Sabine Reffert (LSW)
Sofia Rojas Ruiz (Colombia) 28.06.2023
Accreting Supermassive Black Holes in the First Billion Years: Impact on their Environments from Parsecs to Megaparsecs ( thesis pdf, 30 MB)
Studying the environments of quasars in the first Gyr of the universe, or at z > 5.5, is crucial to understand their growth and evolution from such early times. These massive quasars with ∼ 10^8 − 10^9 Solar mass black holes are predicted to be born in the most massive halos of the underlying dark matter distribution and thus would be immersed in protoclusters of galaxies. However, the impact of a quasar’s powerful radiation on the formation and growth of galaxies in its Mpc-scale environment is still debated observationally. Additionally, the different components contributing to the quasar activity from subpc- to kpc-scales can affect the gas and dust used to form stars in its host galaxy, thus impacting the black hole – host galaxy co-evolution. This thesis aims at providing more understanding of the different scales of quasar environments by investigating two exemplary quasars. We investigated the luminous quasar ULAS J1342+0928 at z = 7.54 and looked for galaxies in its ∼ 1 pMpc^2 environment. We found one UV-bright Lyman-break galaxy candidate in addition to a [CII]-emitter associated with the quasar environment and clustered within a projected distance of ≲ 220 pkpc from the quasar. Future observations in the near-IR and mm would be necessary to confirm the redshift and physical properties of these galaxies and assess the number density of one of the earliest quasar large-scale environments yet explored. We also studied extensively the radio-loud quasar P352–15 at z = 5.832; the only source found thus far with evidence of a kpc-scale extended jet. Thus, this quasar is the ideal laboratory to investigate the first stages of black hole–jet–host galaxy co-evolution. We first studied the cold dust and [CII] gas of the host galaxy of P352–15 to explore whether the jet presence affects the host galaxy properties. The results on the inferred star formation rate and far-infrared luminosity were found to be comparable to studies on the radio-quiet quasar population. However, we found evidence of a spectral break in the jet synchrotron emission affecting the cold dust emission of the host galaxy. We further investigated the effect of the synchrotron spectral break by requesting additional Karl G. Jansky Very Large Array (VLA) radio observations. Using these new data, we found the frequency of the spectral break, we were able to calculate the jet age since its time of launch. We compared this jet timescale to the quasar lifetime, or the time since the last black hole accretion event. Both timescales were found to be comparatively young of just ≲ 10^4 yrs within their uncertainties, evidencing a fairly recent quasar activity of P352–15. The work in this thesis shows that studying different scales of quasar environments at z ≳ 6 is essential to understanding the formation and evolution of galaxies and black holes in the early universe.
Supervisor: Eduardo Banados / Frederik Davies (MPIA)
Simon Steinmassl (Germany) 06.06.2023
Probing particle acceleration in stellar binary systems using gamma-ray observations ( thesis pdf, 30 MB)
Only a few gamma-ray sources have been established as proton accelerators over the last decades, among them two extraordinary binary systems, the massive colliding wind binary Eta Carinae (eta Car) and the recurrent nova RS Ophiuchi (RS Oph). In this thesis, the nature of acceleration processes up to TeV energies are probed in these systems using very-high energy (VHE; ≥ 100 GeV) gamma-ray data from the High Energy Stereoscopic System (H.E.S.S.) in conjunction with data from the Fermi Large Area Telescope (LAT). To obtain reliable results from Imaging Atmospheric Cherenkov Telescopes like H.E.S.S., an accurate match between simulations and actual observations is crucial. Thus, in the first part of this thesis the successful validation of the simulations of the full 5-telescope H.E.S.S. array is presented. Based on this, the scientific verification of the monoscopic analysis was achieved using data from the large 28 m-telescope recently upgraded with a FlashCam prototype. The resulting spectrum of the Crab Nebula, the standard candle in the VHE regime, is found to be in good agreement with previous measurements by H.E.S.S. and other instruments. Using this verified analysis configuration, the nova RS Oph was successfully detected during its 2021 outburst, making it the first nova with confirmed VHE emission. A detailed light curve was derived from the highly statistically significant gamma-ray signal observed with the full H.E.S.S. array. The combined properties of the H.E.S.S. measurements with simultaneous Fermi-LAT data clearly favor a common origin of the whole gamma-ray emission, implying efficient acceleration of hadrons at the external shock caused by the eruption. eta Car has been firmly established as a source of gamma-rays by Fermi-LAT and H.E.S.S. over the last decade. With its highly eccentric orbit lasting 5.5 years, the periastron passage of the two stars is extremely close, making it a particularly interesting phase range. The 2020 periastron passage was the first such event to be extensively monitored by H.E.S.S. In this thesis, the detection of a VHE signal from eta Car during the 2020 periastron is presented, making use of a novel time-based image cleaning technique for the monoscopic analysis. In combination with simultaneous Fermi-LAT data, its spectral properties are characterized and together with previous and follow-up observations, for the first time, a VHE light curve spanning a full orbit is derived. At least some fraction of the accelerated particles traced by the gamma-ray emission likely escape from eta Car, potentially interacting with target material on different spatial scales. With the detection of Fermi-LAT excess emission associated to molecular clouds in the Carina Nebula, this hypothesis is tested. Whereas the cosmic ray density profile is indicative of an origin of the interacting cosmic rays from eta Car, a larger escaping flux than predicted by models or a contribution from other cosmic ray sources is needed to match the measured flux.
Supervisor: Jim Hinton (MPIK)
Albrecht Kamlah (Germany) 25.05.2023
The impact of the stellar evolution of single and binary stars on the global, dynamical evolution of dense star clusters across cosmic time ( thesis pdf, 22 MB)
Star clusters in the Universe represent dense, self-gravitating and typically dynamically collisional environments of thousands to millions of stars. They populate galactic discs, halos and even galactic centres across the cosmos and are postulated to act as a fundamental unit in a hierarchy of cosmic structure formation. Importantly, they are typically much denser than their host galaxy, which makes them incredibly fascinating astronomical objects. Unlike their surroundings, stars and compact objects within in star clusters experience frequent dynamical encounters, form dynamical binary stars, merge by emitting gravitational waves, are ejected due to three-body dynamics and in rare cases even collide directly. As a result, star clusters are factories of all exotic binary stars, from, e.g. Thorne-Zytkow objects and Cataclysmic Variables to compact binary stars such as the elusive black hole-neutron star binaries. Furthermore, for increasing particle number, unique gravitational effects of collisional many-body systems begin to dominate the early cluster evolution that lead to a contracting and increasingly rapidly rotating core of the star clusters containing massive and binary stars, compact objects and an expanding halo of lower mass stars. Star clusters are therefore not only a laboratory for gravitational many-body physics, but also stellar evolution of single, binary and higher-order stars. All of these physical processes cannot be disentangled - they reinforce each other and many happen on similar time-scales. In this thesis, I aim to shed some light on the impact that stellar evolution has on the global dynamics of star clusters using direct gravitational N -body and Hénon-type Monte-Carlo simulations of star clusters. I focus on the evolution of metal-poor stellar populations (population II) that are present in globular clusters and extremely metal-poor stellar populations (population III) found in the oldest star clusters in the Universe.
Supervisor: Rainer Spurzem (ARI)
Philipp Hottinger (Germany) 23.05.2023
Integrated Tip-Tilt Sensing for Single-Mode Fiber Coupling ( thesis pdf, 17 MB)
This thesis presents the development and on-sky tests of the novel Microlens-Ring Tip-Tilt (MLR-TT) sensor. The sensor consists of a micro-lens ring (MLR) that is printed directly on the face of a fiber bundle with a central single-mode fiber (SMF) accepting the light almost unclipped if the beam is aligned. The edge of the beam, however, is refracted by the MLR to couple into six surrounding multi-mode fibers (MMFs). Detecting the flux in these sensor fibers allows reconstruction of the beam position, i.e. the tip and tilt aberrations of the wavefront.
The lenses are manufactured in collaboration with Karlsruhe Institute for Technology (KIT) with state-of-the-art two-proton polymerization, a novel technology that allows the fabrication of very precise and freeform lenses. The sensor is integrated with the instrument’s fiber link and features a small physical size of 380 µm. This novel integration of a sensor into existing components reduced opto-mechanical footprint and complexity, as well as reducing non-common path aberrations (NCPAs) to a bare minimum.
This thesis describes the various steps that were part of this development, starting with designing, optimizing, and characterizing the sensor itself, setting up a corresponding laboratory environment, and developing a control system for on-sky testing. The system is tested on-sky with iLocater fiber coupling front-end (acquisition camera) at the Large Binocular Telescope (LBT). It was found that principle reconstruction is possible but the observed accuracy is ∼0.19 λ/D both for tip and for tilt. With this accuracy, it was not possible to improve the resulting SMF coupling efficiency. A strong correlation between sensor accuracy and the instantaneous Strehl ratio (SR), i.e. residual adaptive optics (AO) aberrations, is found. Additionally, the corresponding power spectral density (PSD) reveals that most of the reconstruction inaccuracy occurs in low temporal frequencies. This suggests that the dominating limitations of the accuracy of the MLR-TT sensor arise from residual AO aberrations and the false signal they introduce in the sensor.
These findings are discussed in detail and the future prospects of further analysis and development are outlined in the context of the most beneficial application environment.
Supervisor: Andreas Quirrenbach (LSW)
Da Eun Kang (Korea) 24.05.2023
Determining physical properties of star-forming regions using conditional invertible neural network ( thesis pdf, 11 MB)
Star formation is one of the most fundamental subjects in astronomy where astronomers have been seeking answers to key questions: how efficiently stars form and how newly born stars affect their surroundings. Our understanding of star formation relies mostly on the observations of star-forming regions. However, it is a non-trivial task to interpret the observations because diverse physical processes are non-linearly coupled so the observational data are highly degenerate. Additionally, the ever-expanding volume of observational data in recent days necessitates a new method that analyses large amounts of data more quickly and effectively.
In this thesis, we introduce deep learning-based tools we have developed to efficiently and effectively interpret massive data of observed star-forming regions. We adopt the conditional invertible neural network (cINN) architecture specialised in solving the inverse problem of degenerate systems. We introduce the cINNs developed for cloud-scale observations and cINNs for individual star-scale observations. Our networks are very useful tools that can consistently and quickly analyse large amounts of data. We evaluate the performance of the networks, demonstrating that our networks predict physical properties accurately and precisely.
Supervisor: Ralf Klessen (ITA)
Sofia Savvidou (Greece) 24.05.2023
How dust shapes protoplanetary discs and the implications to planet formation ( thesis pdf, 35 MB)
Protoplanetary discs are the natal environments of planets and contain the building blocks from which planets form. It is therefore of crucial importance to understand how the dust growth and evolution shapes discs and what the implications are for planet formation. At the same time, our observational capabilities have improved in the recent years, providing us with more constraints that need to be considered in our theoretical studies. The goal of this thesis is to determine through numerical simulations how the dust shapes the (thermal) structure of the protoplanetary disc, how the conditions within the disc affect the growth of planets and how the forming planets affect the dust mass itself. We find that opacity models based only on micrometer-sized dust grains might not be a good approximation to simulate the disc's structure, especially for discs with significant viscous heating. There is a trade-off between the pebble isolation mass and the planetary growth timescale, which is important for the modeling of the growth of super-Earths via pebble accretion. We also find that the most favorable conditions for giant planet formation are high disc mass, early formation, and a large enough disc, however we conclude that their formation is mainly the outcome of a combination of beneficial factors or lack of adverse ones. Our findings strengthen the hypothesis that planet formation has already happened or is ongoing in Class II discs and we show that the assumption of an optically thin emission significantly underestimates the total dust mass in discs, if a giant planet is present that traps dust exterior to its orbit. We conclude that we should use the ever-increasing and improved observational data to better constrain the protoplanetary disc properties and connect the dots better to the observed exoplanets, based on our more sophisticated theoretical models.
Supervisor: Bertram Bitsch (MPIA)
Jae Yeon Kim (South Korea) 17.04.2023
Quantifying the environmental dependence of the molecular cloud lifecycle in 54 main sequence galaxies ( thesis pdf, 14 MB)
Giant molecular clouds (GMCs) form stars with initial conditions set by their local host galaxy environment, which feedback energy and matter into their surroundings, contributing to galaxy evolution. However, the detailed characteristics of these processes between molecular gas and young stars remain elusive, primarily due to a lack of observational constraints. By capitalizing on CO and Halpha observations from PHANGS, we have systematically measured the evolutionary timeline from GMCs to exposed HII regions, across 54 galaxies, the largest and most statistically complete sample to-date. Strong correlations between GMC evolutionary time-scales and the host galaxy properties have been identified, revealing the connection between galactic-scale dynamics and the small-scale GMC lifecycle. Furthermore, in the 5 nearest galaxies of my sample (D<3.5Mpc), we have established that the initial half of the embedded star formation detected in Spitzer 24microns is invisible in Halpha. Finally, using novel JWST observations of NGC628 with 10 times better resolution compared to Spitzer, we have further demonstrated that the embedded phase of star formation can be characterized at a greater distance (D=9.8Mpc), pioneering the way for the systematic determination of the early phases of star formation across the nearby galaxy population (up to 20Mpc) with PHANGS-JWST.
Supervisor: Melanie Chevance (ITA)
Laura Olivera Nieto (Spain) 07.02.2023
Resolving particle acceleration and transport in the jets of the microquasar SS 433 with H.E.S.S. and HAWC ( thesis pdf, access restricted)
The microquasar SS 433 offers a unique laboratory to study the physics of mildly relativistic jets and the associated non-thermal processes. It hosts a compact binary system, from which a pair of counter-propagating jets is observed to emanate. The jets are resolved by observations out to distances of approximately 0.1 pc from the central source, but further out, they remain dark until they abruptly reappear at around 25 pc as bright X-ray sources. These outer jets were recently reported to be sources of TeV gamma-rays by the High Altitude Water Cherenkov (HAWC) observatory. This thesis presents a complete picture of the TeV emission from the jets of SS 433 including new data from the High Energy Stereoscopic System (H.E.S.S.) and the HAWC observatory. To fully exploit the capabilities of the H.E.S.S. observations, a new approach to background rejection is presented. It is based on the detection of Cherenkov light from muons by large Imaging Atmospheric Cherenkov Telescopes (IACTs), such as the telescope located at the center of the H.E.S.S. array. The application of this technique leads to a factor four reduction in background above several tens of TeV in the H.E.S.S. stereoscopic analysis.
This thesis presents the detection of the SS 433 outer jets for the first time with an IACT array using H.E.S.S.. The superior angular and energy resolution of H.E.S.S. compared to HAWC allow for a detailed study of the emission from the jets, including a measurement of the physical extension of the emission and of the spectra out to tens of TeV. These observations also reveal the presence of striking energy- dependent morphology, ruling out a hadronic origin for the bulk of the gamma-ray emission. Photons above 10 TeV are observed only close to the base of the outer jets, implying efficient particle acceleration to very-high energies at that location. Evidence suggests that the acceleration is due to a shock, thus providing a clue to the long-standing question of the reappearance of the jets.
The observed energy-dependent morphology is modeled as a consequence of the particle cooling times and the advection flow of the jet, which constrains the jet dynamics and, in particular, results in an estimate of the velocity of the outer jets at their base. This solves several issues concerning the non-thermal processes occurring in the jets and their dynamics, but also opens up new questions that highlight our incomplete understanding of the SS 433 system. A joint analysis of the H.E.S.S. and HAWC data would provide insights on the system across the entire range of TeV energies. To make this possible, a tool capable of reading and analyzing the data from both instruments is required. This thesis presents the extension and validation of an existing data format and analysis tool shared among IACTs to the data from particle detector arrays such as the HAWC observatory. This framework is then used to revisit the HAWC observations of the SS 433 region with the inclusion of additional data taken since the first detection was reported. The existence of this framework enables for the first time the joint analysis of the H.E.S.S. and HAWC data, the preliminary results of which are presented
Supervisor: James Hinton (MPIK)
Ekaterina Magg (Russia) 01.02.2023
Constraining stellar physics with the NLTE radiative transfer ( thesis pdf, 120 MB)
All chemical elements in the Universe, except the very few lightest species, are produced in a nuclear fusion inside the stars. Following the stellar life cycle, these chemical elements are expelled into the interstellar medium where they proceed to contribute to the chemical enrich- ment of their surroundings. Spectroscopic observations are currently the only way to infer the chemical make-up of the stars. Combining those with the physical modelling of the radiation in the stellar plasma allows us to detect and measure the number of chemical elements we all are made of. The approach is generally applied to individual stars constituting the larger scale populations, from clusters to galaxies including our Milky Way and beyond. In this thesis I focus on the non-equilibrium modelling of stellar radiation and its influence on the measured chemical abundances of various elements. I provide a general overview of the methods and necessary information used to infer the stellar chemical composition. I then present developments in the non-equilibrium modelling and apply it to the analysis of a star cluster, our host star – the Sun, and eventually a broader Galactic population. I focus on the opportunities our modelling approach presents to observationally constrain the stellar evolution and consequential enrichment of the Galactic populations.
Supervisor: Maria Bergemann (MPIA)
Grigorii Smirnov-Pinchukov (Russia) 12.01.2023
Formylium as a tracer of circumstellar disks physics ( thesis pdf, 26 MB)
There are many different tracers of circumstellar disk physics, most notably, micrometer to millimeter-sized dust, and one of the most abundant molecules, CO. Formylium (HCO+) is another commonly observed species. Its chemistry is more complex than CO chemistry, and more interpretation steps are necessary to build the bridge between the disk structure and observed emission. Its isotopologs DCO+ and H13CO+ complement the picture and allow a more precise understanding of the disk structure. In this thesis, I present my results achieved by combining and developing physical modeling, chemical kinetics, and radiative transfer methods to understand circumstellar disks' physical properties through formylium isotopologs observations. I explain the observed DCO+ increase in the protoplanetary disk gap and use it as proof of the reduced amount of gas in the gaps. I show that HCO+ should be the brightest molecule after CO isotopologs in the gas-rich debris disks, and its brightness would reveal the elemental composition, but a next-generation observatory is needed to detect it. Then I present an application of the machine learning approach to predict the modeled disk chemistry instantaneously based on the pre-computed disk models' data set, allowing the replacement of computationally expensive thermo-chemical models in the fitting pipelines. Finally, I demonstrate the data which will be analyzed using this approach.
Supervisor: Thomas Henning (MPIA)
