Periodic Reporting for period 4 - BAHAMAS (A holistic approach to large-scale structure cosmology)
Reporting period: 2022-12-01 to 2023-11-30
The standard model of cosmology, which is the framework upon which our understanding of the evolution of the Universe is based, is remarkably successful at explaining a wide range of observations. However, it is now being subjected to much more stringent tests than ever before, and recent large-scale structure (LSS) measurements appear to be in tension with its predictions. Is this tension signalling that new physics is required? For example, time-varying dark energy, or perhaps a modified theory of gravity? A contribution from massive neutrinos? Before coming to such bold conclusions we must be certain that all of the important systematic errors in the LSS tests have been accounted for.
Presently, the largest source of systematic uncertainty is from the modelling of complicated astrophysical phenomena associated with galaxy formation. In particular, energetic feedback processes associated with star formation and black hole growth can heat and expel gas from collapsed structures and modify the large-scale distribution of matter. Furthermore, the LSS field is presently separated into many sub-fields (each using different models, that usually neglect feedback), preventing a coherent analysis.
Cosmological hydrodynamical simulations (are the only method which) can follow all the relevant matter components and self-consistently capture the effects of feedback. With the “BAHAMAS” programme we are leading the development of large-scale simulations with physically-motivated prescriptions for feedback that are unrivalled in their ability to reproduce the observed properties of massive systems. Our team is exploiting these developments to produce a large suite of simulations designed specifically for LSS cosmology applications with the effects of feedback realistically accounted for and which will allow us to unite the different LSS tests. The primary objective is to perform the first self-consistent comparisons with the full range of LSS cosmology tests, and critically assess the evidence for physics beyond the standard model.
Presently, the largest source of systematic uncertainty is from the modelling of complicated astrophysical phenomena associated with galaxy formation. In particular, energetic feedback processes associated with star formation and black hole growth can heat and expel gas from collapsed structures and modify the large-scale distribution of matter. Furthermore, the LSS field is presently separated into many sub-fields (each using different models, that usually neglect feedback), preventing a coherent analysis.
Cosmological hydrodynamical simulations (are the only method which) can follow all the relevant matter components and self-consistently capture the effects of feedback. With the “BAHAMAS” programme we are leading the development of large-scale simulations with physically-motivated prescriptions for feedback that are unrivalled in their ability to reproduce the observed properties of massive systems. Our team is exploiting these developments to produce a large suite of simulations designed specifically for LSS cosmology applications with the effects of feedback realistically accounted for and which will allow us to unite the different LSS tests. The primary objective is to perform the first self-consistent comparisons with the full range of LSS cosmology tests, and critically assess the evidence for physics beyond the standard model.
The primary objective of the ERC project is to carry out state-of-the-art cosmological hydrodynamical simulations in the context of the standard model of cosmology and beyond and to perform self-consistent comparisons with the full range of large-scale structure (LSS) cosmology tests, assessing the evidence for physics beyond the standard model. The project consisted of 10 work packages. The first half of the programme was focused on the development of cosmological hydrodynamical simulations primarily in the context of the standard model of cosmology. The second half of the project extended our simulations beyond the standard model, including massive neutrino cosmologies, dynamical dark energy cosmologies, and alternative inflationary models. We characterised the impact of processes associated with galaxy and black hole formation (and their uncertainties) on various large-scale structure observables (such as galaxy clustering, cosmic shear, galaxy cluster counts, etc.) and assessed whether claimed tensions between constraints on certain cosmological parameters from large-scale structure observations and those of the cosmic microwave background are significant (i.e. cannot be explained by said galaxy formation processes). To date, our group has published 69 publications in top international, peer-reviewed journals that deliver on the project goals. We have made our scientific results, as well as the underlying data and simulation codes, publicly available where possible. And we have given numerous presentations, both at scientific conferences and events for the general public, on our findings.
Current large-scale structure tests of cosmology rely almost exclusively on so-called collisionless (or “gravity only”) simulations that ignore the important role that baryons and processes associated with galaxy formation play. However, ours and other groups have demonstrated that these effects are likely to be significant and cannot be ignored if one wishes to rigorously test models for the evolution of the Universe.
Part of the problem of including their effects is that we do not yet have a complete theory of galaxy formation and there are significant uncertainties in the modelling of feedback processes in particular. Our ERC-funded project has taken a novel and significant step forward, where instead of using a single model for galaxy formation, we explore the full “feedback landscape”, by varying the parameters associated with the feedback modelling over a wide range of theoretically-plausible values. Because baryonic effects can sometimes mimic extensions beyond the standard model of cosmology (e.g. feedback from supermassive black holes can affect the clustering of matter in a similar way to the free streaming of massive neutrinos) it is important to vary both the parameters characterising the baryonic and cosmological effects simultaneously and to explore their degeneracies. We have produced two new large suites of simulations which achieve these important aims. These simulations, called Antilles and FLAMINGO, have redefined the state of the art and are the first to explore in a simultaneous and systematic way the possible natures of dark matter and dark energy while factoring in key new processes (and their uncertainties), such as energetic feedback from supermassive black holes. These simulations represent a major advance over current theoretical models, which are either based on simulations which neglect these key important “baryonic” processes or that implement them in a very simplistic way that lacks self-consistency and accuracy. Our simulations have, for the first time, achieved the accuracy required to interpret and exploit data from forthcoming “stage IV” surveys of large-scale structure, including cosmic shear and galaxy clustering measurements with Euclid and LSST.
Part of the problem of including their effects is that we do not yet have a complete theory of galaxy formation and there are significant uncertainties in the modelling of feedback processes in particular. Our ERC-funded project has taken a novel and significant step forward, where instead of using a single model for galaxy formation, we explore the full “feedback landscape”, by varying the parameters associated with the feedback modelling over a wide range of theoretically-plausible values. Because baryonic effects can sometimes mimic extensions beyond the standard model of cosmology (e.g. feedback from supermassive black holes can affect the clustering of matter in a similar way to the free streaming of massive neutrinos) it is important to vary both the parameters characterising the baryonic and cosmological effects simultaneously and to explore their degeneracies. We have produced two new large suites of simulations which achieve these important aims. These simulations, called Antilles and FLAMINGO, have redefined the state of the art and are the first to explore in a simultaneous and systematic way the possible natures of dark matter and dark energy while factoring in key new processes (and their uncertainties), such as energetic feedback from supermassive black holes. These simulations represent a major advance over current theoretical models, which are either based on simulations which neglect these key important “baryonic” processes or that implement them in a very simplistic way that lacks self-consistency and accuracy. Our simulations have, for the first time, achieved the accuracy required to interpret and exploit data from forthcoming “stage IV” surveys of large-scale structure, including cosmic shear and galaxy clustering measurements with Euclid and LSST.