From the Large-Scale Structure to the Milky Way halo

The growing significance of structure detection in astronomy, driven by large simulations and observational surveys necessitates the development of efficient structure mining tools that can operate on large data sets.  In her thesis, Petra Awad focuses on the detection and analysis of filamentary structure on various astrophysical scales spanning cosmic filaments of the Large-Scale Structure and stellar streams within the Milky Way halo.

Awad combined five Machine Learning algorithms into a coherent toolbox termed 1-DREAM. The novel methodologies utilize concepts from swarm-intelligence, evolutionary computation, and probabilistic modeling designed to detect, extract, and model 1-dimensional components of large data sets. Awad highlighted her methods’ effectiveness in studying cosmic web simulations, competing with and sometimes outperforming other tools that are widely used in the field. 

After applying these tools on simulated data, Awad moved on to observational surveys. She traced cosmic filaments in the Fornax-Eridanus Complex, revealing a large network of filaments connecting its galaxy groups. She also studied the impact of these environments on galaxy morphology. Awad then moved on to galactic archeology, investigating the Jhelum stellar stream. She discovered a relation between the seemingly double morphology of this stream and two overdensities present in its dynamical space, allowing her to propose a scenario involving a globular cluster within a dwarf galaxy that could explain Jhelum’s formation. Finally, Awad explored the globular clusters NGC 1261 and NGC 1904 and offers new insights into the formation of their tidal tails based on their current phase of orbit.

Dissertation: https://hdl.handle.net/11370/11b17bc6-f5ed-449c-a088-9993f4d82cae