Periodic Reporting for period 4 - GravityLS (Large Scale Structure Constraints of General Relativity)
Reporting period: 2021-03-01 to 2022-08-31
Current observations of the large scale structure of the Universe - of the distribution of galaxies and of the relic radiation left over from the Big Bang - have been used to place constraints on key properties of the universe, such as its geometry, the density of different types of matter and forms of energy and even on the initial conditions. A key assumption is the the dominant force that drives the evolution of the Universe is gravity and that it is perfectly described by Einstein's General Theory of Relativity. The purpose of this project is to develop the tool that will allow us to test this assumption and in doing so, either greatly strengthen constraints on existing fundamental physics or allow us to unambiguously detect evidence for new physics, i.e. modifications to general relativity, on cosmological scales.
Thus far, we have made steady progress on a number of projects. For a start we have developed a mathematical framework that allows us to describe all possible deviations from General Relativity on large scales. This also allows us to figure out if there are interesting observational effects to look for in the data. We have also developed a publicly available piece of software that implements a great part of this framework so that we will be able to constrain it with observations. A key aspect has been to cross-check our codes with over a dozen other codes - this has been useful to correct errors in all the codes so that they are completely trustworthy.
An important development has been our links with observations which are not cosmological. In particular, the recent detection of a binary neutron star merger allowed us to place constraints on a vast swathe of different theories, allowing us to focus on the modifications to general relativity that really matter. We have also understood how to use measurements of gravitational wave signals to place constraints in a completely different regime (where the gravitational field is much stronger) but which can then be combined with cosmological measurements to place even tighter constraints.
We expect the next decade to bring us a wealth of cosmological data which will allow us to place even more stringent constraints on gravity. As preparation for this new wave of information, we have been exploring the various techniques that may be applied to, in some sense a forecast of what we might expect. But we have also taken stock and analyzed on of the current most comprehensive data sets, KIDS-450, and combined it with other data sets to place state of the art constraints on cosmological gravity.
Understanding the fundamental laws that operate on largest scales and placing stringent constraints on them will allow us to be confident about our model of the origin and evolution of the Universe.
Thus far, we have made steady progress on a number of projects. For a start we have developed a mathematical framework that allows us to describe all possible deviations from General Relativity on large scales. This also allows us to figure out if there are interesting observational effects to look for in the data. We have also developed a publicly available piece of software that implements a great part of this framework so that we will be able to constrain it with observations. A key aspect has been to cross-check our codes with over a dozen other codes - this has been useful to correct errors in all the codes so that they are completely trustworthy.
An important development has been our links with observations which are not cosmological. In particular, the recent detection of a binary neutron star merger allowed us to place constraints on a vast swathe of different theories, allowing us to focus on the modifications to general relativity that really matter. We have also understood how to use measurements of gravitational wave signals to place constraints in a completely different regime (where the gravitational field is much stronger) but which can then be combined with cosmological measurements to place even tighter constraints.
We expect the next decade to bring us a wealth of cosmological data which will allow us to place even more stringent constraints on gravity. As preparation for this new wave of information, we have been exploring the various techniques that may be applied to, in some sense a forecast of what we might expect. But we have also taken stock and analyzed on of the current most comprehensive data sets, KIDS-450, and combined it with other data sets to place state of the art constraints on cosmological gravity.
Understanding the fundamental laws that operate on largest scales and placing stringent constraints on them will allow us to be confident about our model of the origin and evolution of the Universe.
The project has been able to construct a consistent framework for constraining general relativity and cosmological physics with large scale structure and astrophysical data. The project was constructed along three main strands. The A strand focussed on developments in theory, the B strand on developments in phenomenology and methods and the C strand on data analysis.
Strand A of the project was constructed to develop the theoretical framework of the programme and build connections with other methods for constraining gravity. The first main subtheme was to construct a formalism that could completely describe all types of gravitational physics on (linear) cosmological scales. The team was able to achieve this goal successfully with an approach which truly comprehensive. A second subtheme was to connect constraints along a vast range of scales. Unexpectedly, this was one of the most successful achievements of this project. In late 2017, a binary neutron star was detected, GW170817, allowing the team to place the tightest constraints on cosmological gravity to date. This led us to explore the connection between cosmological gravity and black hole physics and gravitational wave propagation. It also led us to the accretion of cosmological scalar fields around black holes. Finally, we proposed a completely new approach for incorporating priors assumptions in the analysis of cosmological theories of gravity
Strand B of the project was oriented to developing the algorithms and methods which would be used in the data analysis. We released Hi_Class, a code for solving the linear regime for general scalar tensor theories as well as generalized parametrized theories. Along with EFT Camb, this means that there are now general purpose codes for the type of analysis we advocate. We also lead a world-wide code comparison of over a dozen such solvers which focusd on specific theories. An important strand was the non-linear regime. It has become apparent to the wider community that there are serious impediments to achieving the level of accuracy and generality desired due to a number of effects. We opted to get a full understanding of the impact that non-linearities will have on cosmological and other observables and to construct a halo model code, calibrated against fast N-body codes which could approximate the non-linear regime with the desired accuracy.
Strand C was oriented towards data analysis and observations. We devoted a substantial amount of time to defining the road-map for understanding the effects of baryons on observables. From our work in strand A on black hole accretion we were able to model gravitational dynamical friction on black hole orbits. In parallel we constructed the first fully consistent model for the noise for the anisotropic gravitational wave background arising from large scale structure. A key achievement was the complete analysis of the base scalar-tensor theory, Jordan-Brans-Dicke theory in a way which allowed us to explore the interplay between the main cosmological parameters, systematic effects and gravity. In parallel, we focussed on tomographic data leading to the tightest constraints on the evolution of the growth rate in terms of the largest compilation of tomographic data to date. Members of my team were able to obtain constraints from ACT, KIDS and BOSS on general relativity. Finally we have produced some of the most comprehensive forecasts for the next generation of surveys for scalar-tensor theories and are finishing a forecast for future Euclid data.
Strand A of the project was constructed to develop the theoretical framework of the programme and build connections with other methods for constraining gravity. The first main subtheme was to construct a formalism that could completely describe all types of gravitational physics on (linear) cosmological scales. The team was able to achieve this goal successfully with an approach which truly comprehensive. A second subtheme was to connect constraints along a vast range of scales. Unexpectedly, this was one of the most successful achievements of this project. In late 2017, a binary neutron star was detected, GW170817, allowing the team to place the tightest constraints on cosmological gravity to date. This led us to explore the connection between cosmological gravity and black hole physics and gravitational wave propagation. It also led us to the accretion of cosmological scalar fields around black holes. Finally, we proposed a completely new approach for incorporating priors assumptions in the analysis of cosmological theories of gravity
Strand B of the project was oriented to developing the algorithms and methods which would be used in the data analysis. We released Hi_Class, a code for solving the linear regime for general scalar tensor theories as well as generalized parametrized theories. Along with EFT Camb, this means that there are now general purpose codes for the type of analysis we advocate. We also lead a world-wide code comparison of over a dozen such solvers which focusd on specific theories. An important strand was the non-linear regime. It has become apparent to the wider community that there are serious impediments to achieving the level of accuracy and generality desired due to a number of effects. We opted to get a full understanding of the impact that non-linearities will have on cosmological and other observables and to construct a halo model code, calibrated against fast N-body codes which could approximate the non-linear regime with the desired accuracy.
Strand C was oriented towards data analysis and observations. We devoted a substantial amount of time to defining the road-map for understanding the effects of baryons on observables. From our work in strand A on black hole accretion we were able to model gravitational dynamical friction on black hole orbits. In parallel we constructed the first fully consistent model for the noise for the anisotropic gravitational wave background arising from large scale structure. A key achievement was the complete analysis of the base scalar-tensor theory, Jordan-Brans-Dicke theory in a way which allowed us to explore the interplay between the main cosmological parameters, systematic effects and gravity. In parallel, we focussed on tomographic data leading to the tightest constraints on the evolution of the growth rate in terms of the largest compilation of tomographic data to date. Members of my team were able to obtain constraints from ACT, KIDS and BOSS on general relativity. Finally we have produced some of the most comprehensive forecasts for the next generation of surveys for scalar-tensor theories and are finishing a forecast for future Euclid data.
We have, by now, a firm understanding of the theoretical panorama of gravitational theories - we know how to model their effects on larges scales at the linear level, with great accuracy and, with the constraints from the binary neutron star event GW170817, we have been able to severely restrict the space of allowed theories. We have also, in place, highly accurate and calibrated codes with which we can calculate cosmological observables on linear scales. Finally, we have been able to establish important links with black holes and gravitational wave physics. We have been able to construct a powerful analysis pipeline which combines all the different aspects that can affect the science interpretation of data - theoretical modelling, statistical and systematic effects - which allows us to constrain cosmological gravity to a degree that has not yet been done. This will also allow us to prepare for the next generation of surveys which will come online in the next few years.