Modelling the large-scale structure
One of the key objectives of cosmology is to understand the formation of structure and the evolution of the Universe across cosmic time. This is pursued through tracers of the large-scale structure, such as galaxies, quasars, and the Lyman-α forest, which provide an observable imprint of the underlying matter field.
Detailed numerical simulations have played a fundamental role in this effort. By including gravity together with a broad range of astrophysical processes, they offer invaluable insight into the formation of galaxies and the complexity of the cosmic web. However, these approaches also require many physical prescriptions and parameters, some of which are indispensable for astrophysical realism but are not always central to the cosmological information one seeks to extract.
A complementary route is therefore to model galaxies and other tracers as biased realizations of the large-scale structure, and to construct effective descriptions that capture the relevant physics without resolving every small-scale process in detail. Such an approach is also demanded by the data: the cosmic volumes covered by modern surveys are so vast that it is computationally unfeasible to model them massively with detailed simulations alone.
Effective models are therefore essential. In our work, we develop both fast and accurate effective gravity solvers and effective field-level bias models to describe the large-scale structure of the Universe and its observable tracers.