Studio e Lavoro > Tesi di laurea

As part of a large international collaboration, the project will focus on the Euclid Deep Field South (EDFS), a key 23 square degree region for multiwavelength studies. It will exploit a recently completed 1.2 GHz radio survey conducted with the MeerKAT telescope, complemented by Euclid optical data and eROSITA X-ray observations, including cluster catalogs and host galaxy properties. This unique dataset will enable a comprehensive characterization of radio AGN environments and their connection to the underlying large-scale structure.

The project will primarily involve data analysis and interpretation, with the opportunity to perform radio imaging using state-of-the-art software. Through this work, the student will gain hands-on experience with some of the most recent multiwavelength datasets, developing practical skills in handling and analyzing large astronomical surveys.

This thesis aims to investigate the consistency between star formation rates (SFRs) derived from Hα line emission (optical) and Far-Ultraviolet (FUV) continuum in spiral galaxies, through a spatially resolved (i.e., pixel-by-pixel) analysis. The project will combine proprietary Hα maps obtained with the Integral Field Unit (IFU) at the 2.7m Harlan Smith Telescope (McDonald Observatory, Texas) as part of the Metal-THINGS survey, with public FUV maps from the GALEX satellite. The student will compare the spatially resolved results with previous studies based on integrated data. The goal is to analyze how the Hα/FUV ratio varies across different galactic environments (e.g., spiral arms vs. inter-arm regions) and to recalibrate SFR prescriptions (see review Kennicutt & Evans 2012) for spiral systems, accounting for systematic discrepancies in different star-forming regimes.

Within this context, we propose a Master Thesis aiming to explore how polycyclic aromatic hydrocarbons (PAHs), emitting in the mid-IR band and commonly used as indicators of star formation activity, can be used as a proxy to estimate molecular gas content in nearby galaxies. This thesis will be based on the DustPedia project, which is definitively characterizing the dust properties in the Local Universe by exploiting a database of multi-wavelength imagery and photometry that greatly exceeds the scope (in terms of wavelength coverage and number of galaxies) of any similar survey (Davies et al. 2017; Casasola et al. 2017, 2020, 2022). The DustPedia sample consists of 875 extended (D₂₅ > 1′) galaxies of all morphological types, at distances < 40 Mpc (z < 0.01). The rich DustPedia database contains maps and photometry from UV to sub-mm bands (41 bands, on average, per galaxy), in addition to millimeter CO data tracing molecular gas.

By exploiting the DustPedia sample and database, the student will explore global and spatially resolved relationships between PAH (traced at 8 and 12 micron) and CO emission, taking into consideration their mutual correlation with star-formation rate, with the ultimate aim of offering new, improved empirical predictors of molecular gas content based on PAH emission and star formation activity. This topic is especially timely in light of the enormous improvements in mid-IR science that are emerging from the revolutionary James Webb Space Telescope.

Our overall project consist in the systematic, multi-scale, multi-phase analysis of a volume-limited sample of typical local RGs. For the Thesis, the student will specifically focus on the analysis of the cold gas component in some sample sources using recently-acquired HI data from the Australia Telescope Compact Array (ATCA) and/or molecular gas (CO) data from the Atacama millimetre/sub-millimetre array (ALMA). In all the cases, the goal will be to study the cold gas distribution and kinematics, with a special focus on looking for signatures of external gas accretion with HI data and investigating inflows and outflows at pc scales with CO data.

By carrying out this project, the student will learn how to properly use softwares for the reduction and analysis of interferometric data (e.g. CASA) and tools for studying the gas distribution and kinematics (e.g. 3DBarolo, KinMS), as well as deal with multi-wavelength data from some of the most powerful facilities currently operating at mm and radio wavelengths. All the data will be provided fully calibrated, with also some basic imaging performed. For this work, basic knowledge in radio interferometry and Python would be desirable.

Although this framework successfully explains the observed properties of radio halos in nearby clusters, their existence at high redshift remains one of the major open questions in cluster astrophysics. At increasing redshift, relativistic electrons suffer dramatically stronger radiative losses due to Inverse Compton Scattering, whose efficiency scales as (1+z)^4.

This naturally leads to a fundamental physical question:

Does a critical redshift exist beyond which radio halos become physically difficult to form, or can the larger energy released during mergers at early cosmic epochs, together with stronger turbulence and magnetic field amplification, naturally compensate for these increasing losses?

The goal of this project is to answer this question by studying how the physical conditions for radio halo formation evolve with cosmic time. The student will investigate the redshift evolution of merger energetics, turbulent energy injection, magnetic field amplification, and particle acceleration efficiency, in a semi-analytical framework, integrated with the use of cosmological simulations, and state-of-the-art turbulent re-acceleration models.

By comparing particle acceleration rates with radiative losses, the project will determine whether a critical epoch for radio halo formation exists, or whether massive clusters can sustain synchrotron emission up to z ~ 1–1.5. The results will provide direct predictions for current and future radio observations with LOFAR, MeerKAT, and SKA.

The student will use 1-4 GHz observations of the CIZA J2242.8+5301, the famous cluster hosting the Sausage radio relic, performed with the Jansky Very Large Array (JVLA). The total intensity and polarized emission of cluster radio galaxies will be modeled using the QU-fitting and the RM-synthesis approach. Subsequently, the obtained information will be compared with semi-analytical and/or cosmological simulations of galaxy clusters in order to derive the magnetic field properties for this famous galaxy cluster. The student will develop important coding skills using different programming languages (python and IDL).

The student will analyze observations taken with the Hydrogen Epoch of Reionization Array, a custom-built radio interferometric arrays that observes the 21 cm line in the 6 < z < 30 range. The student will work on the data calibration and foreground separation with the ultimate goal of improving the current 21 cm power spectrum, providing a better constraint on the physical properties of the first stars and galaxies – e.g., the average mass of the dark matter halo where they formed, their X-ray properties, how their star formation process started and evolve with time.

The thesis will consist in the analysis of the available data from Large-aperture Experiment to Detect the Dark Age (LEDA), located in Owens Valley in California, and/or simulations of the Radio Experiment for the Analysis of Cosmic Hydrogen (REACH), deployed in the Karoo radio reserve in South Africa. Both experiments aim to measure the sky-averaged 21 cm signal in the 12 < z < 30 range.

The analysis will include the development of techniques to disentangle the signal from the much stronger Galactic and extra-galactic foregrounds and the characterization of the impact of the systematic effects.

As IM observations trace the underlying matter distribution of the Universe on large scales, a 21 cm detection would place tight constraints on cosmological parameters.

The student will take an active part in the MeerKLASS IM survey of the MeerKAT radio telescope.  The MeerKLASS data cover a  300~deg^2 sky patch with the goal to observe IM up to z ~ 0.4 and new data will be available soon. Thesis activities will include (but not necessarily limited to) data analysis, foreground separation, and simulations of the expected cosmological signal.

A novel interpretation has been suggested that several of these radio sources reside in double systems in different phases of their merging evolution with rapidly precessing jets which are unable to escape the inner region of the host galaxy.

The implication for this scenario is a transverse motion of the hot-spots respect the center of activity, in contrast with the reported evidence of proper motions of the hot spots expanding outwards. Some transverse motions have been already detected, and many of the reported expansion speeds need to be revised.

The thesis will consist in data reduction and analysis of VLBI data taken in various epochs for selected radio sources, to find new evidence for transverse motions of the hot-spots, and bring support to the precessing jet scenario.

This thesis project focuses on FBOT AT2018cow, which is thus far the nearest (60 Mpc) and best studied target with a rich multi-wavelength dataset. In particular, the student will work on a new deep X-ray observation taken about 6 yr after the transient discovery, with the goal to: (i) map the long-term evolution and spectrum in X-rays of this enigmatic source and (i) shed light on the possible connection of FBOTs with manifestations of super-Eddington accretion on BHs.

This thesis builds on the results of a previous UniBO master’s thesis, in which a Software to calculate the geometric coefficients for telescope pointing was developed. The proposed work will complete and extend that study by focusing on instrumental calibration.

The overall work consists of the following main activities:

• Study calibration procedures for interferometers and beamformers;
• Implementation and optimization of instrumental calibration software in Python;
• Comparison with the currently used MATLAB code;
• Perform observational tests to validate algorithms and characterize the array beam.

For this work, basic to intermediate programming skills in Python and MATLAB are required.

With the recent upgrades of the North-South Arm and the upcoming ones of the East-West Arm, it is expected a very high candidate rate during FRB search campaigns, so it becomes of crucial importance the improvement and optimization of the currently used machine learning (ML) algorithm for candidate classification.

The overall work consists of the following main activities:

• Study of the ML model currently used in the detection pipeline
• Optimization with specific training of the model based on simulated and real data
• Performance evaluation.

For this work, it is required programming skills in Python and familiarity with ML/DL models.

The resources identified will be labelled according to their content, but also according to the teaching methodology used and the cognitive processes engaged. It is also essential to break down the activities and resources into shorter segments or micro-themes, to allow the new proposals to be tailored to specific needs (outreach contexts, school level, diverse cultural backgrounds, etc.). The data collected during the thesis will be used as a training dataset for AI experimentation based on the use of a Large Language Model (LLM). The resources selected, validated, labelled and broken down into micro-processes and micro-content can be integrated into retrieval-augmented generation (RAG) procedures. In this way, the LLM will not rely solely on its own general knowledge, but will be guided by specific, reliable educational and outreach materials that are classified and consistent with specific educational methodologies or science communication practices. The aim is not to produce a mature artificial intelligence system, but to assess whether the identified resources can effectively guide an AI model in producing educational and outreach resources that are coherent, methodologically grounded and scientifically reliable.

The search for content to train the LLM prototype will begin with a basic use of commercially available AI systems, the results of which will then be assessed and examined. The student undertaking this thesis will therefore also gain experience and practice in the critical and competent use of AI in research, as well as in everyday life.

Thesis activity detail:

– Accurate search for sources in datasets of educational activities and resources produced by recognized institutions and organizations in the field of astrophysics education and outreach research and practice.

– Analysis and validation of material from additional sources.

– Identification and tagging of resources (or parts of them) and the used methodologies.

– Breakdown into micro-processes and micro-content.

– Development and validation of a prototype AI system, based on RAG and LLM, aimed at the AI-assisted generation of educational and outreach activities in the field of astrophysics.