Final Report Summary - DRKFRCS (Disambiguating the Invisible Universe)
The aim of Project DRKFRCS "Disambiguating the Dark Side of the Universe" was to develop theoretical tools which will allow the scientific community to determine the nature of dark energy.
Dark energy is the name for the until-now unknown mechanism which is causing the expansion of the universe to become ever faster, instead of slowing down as we expect from the simple property that gravity is attractive. Since its detection in 1997, this acceleration has been confirmed by numerous observations. The underlying mechanism is still unknown and the proposed models range from a non-dynamical property of empty space (the cosmological constant), through to a new exotic form of matter to modifications of general relativity. The generation of cosmological observatories for the next decade (such as the European Euclid satellite) is being designed to differentiate between these models and determine the nature of dark energy. The work in this project was aimed at delivering new theoretical tools for the community which would make it possible to interpret the data from these future missions much more conclusively.
The main thrust of the project was to develop a unified method for predicting the allowed behaviour of very general models of dynamical dark energy and modified gravity. Up until now, each model of dark energy required a completely separate data analysis. More significantly, approximate methods available up until now were so general they included effects which are impossible to obtain in healthy models and thus reduced the discriminatory power of the data. The new methods are based on an Effective Field Theory (EFT) approach. This was developed by a number of groups concurrently. In particular, a version was part of my research proposal but I published it just prior to starting this Marie Curie project. This version has become the de facto standard approach in this field.
This approach to dark energy has allowed for one of the important results of Project DRKFRCS: until recently essentially all predictions in this field were based on a simpler approach called the quasi-static approximation (QS). I have conclusively shown in ref. [2] that this approximation is not going to be good enough for the upcoming surveys and missions in cosmology and the full EFT-like approach must be used.
New data-analysis tools must be created for this EFT-like approach. Thus a large part of the research effort was expended to develop a computer code which would allow to test data for evidence of deviations from standard cosmology using this method. Together with collaborators, we have modified one of the standard computer codes used to predict what experiments should observe to include these new mechanisms for dark energy. This code was publicly released in June 2016 and is available to any cosmologist in the world to test their data for evidence of dark energy (see ref. [1] and http://www.hiclass-code.net). In particular, it is going to become one of the standard codes used by the Euclid Consortium in the data analysis (see ref. [3])and has already been used by some groups to constrain the models with the data already available.
A significant discovery made by me during this project (ref. [7], published in PRL) is the until now unknown connection between models of dark energy which modify general relativity also modify gravitational waves (GW). I showed that in all models where the propagation of gravitational waves is altered (e.g. they move at a speed different from that of light), the gravitational field around matter is also modified in an observable manner (they have gravitational slip). This for the first time allows us to use a completely different probe (i.e. the detection of gravitational waves) to say something about dark energy. Fortuitously, this has become highly relevant, since LIGO detected GW for the first time in February 2016. Currently the constraints on the properties of GW are very weak (see ref. [6]) but they will get rapidly better as new sources are detected. Until recently GW physics and cosmology were completely separate fields, but now as a result of my work for this project there is a competition to see which teams might be able to constrain dark energy models first.
I opened a third research direction, which is to study whether it is possible that some sort of non-standard behaviour of dark matter might destroy the sensitivity that data sets have to dynamical dark energy. Typically, when investigating dark energy and comparing what the models predict to data, one considers dark matter to be a massive and slow-moving particle (cold dark matter) and one varies only the properties of dark energy. It is, however, possible that dark matter is more complicated (e.g. has a temperature or a mass that is not extremely large). In refs [4,5], I investigated whether by adding some non-standard properties to dark matter it is possible to create the sort of phenomena that dark energy is typically responsible for. The conclusions were luckily negative: we showed that the sort of masses and temperatures that are allowed by the cosmic microwave data can in no way modify the observable properties of the universe on the scales at which dark energy is important. This means that future experiments will be able to constrain properties of dark energy and dark matter separately.
Overall, during the 24 months of research funded my the Marie Curie Action, I have been able to significantly improve the preparedness that we have as a community for the data that will be produced by next-generation observatories. In particular, my hi_class code is already in use by multiple groups and will allow them to correctly build in the effects of dark energy into their predictions and thus measure the various properties of the universe and dark energy without bias.
Dark energy is the name for the until-now unknown mechanism which is causing the expansion of the universe to become ever faster, instead of slowing down as we expect from the simple property that gravity is attractive. Since its detection in 1997, this acceleration has been confirmed by numerous observations. The underlying mechanism is still unknown and the proposed models range from a non-dynamical property of empty space (the cosmological constant), through to a new exotic form of matter to modifications of general relativity. The generation of cosmological observatories for the next decade (such as the European Euclid satellite) is being designed to differentiate between these models and determine the nature of dark energy. The work in this project was aimed at delivering new theoretical tools for the community which would make it possible to interpret the data from these future missions much more conclusively.
The main thrust of the project was to develop a unified method for predicting the allowed behaviour of very general models of dynamical dark energy and modified gravity. Up until now, each model of dark energy required a completely separate data analysis. More significantly, approximate methods available up until now were so general they included effects which are impossible to obtain in healthy models and thus reduced the discriminatory power of the data. The new methods are based on an Effective Field Theory (EFT) approach. This was developed by a number of groups concurrently. In particular, a version was part of my research proposal but I published it just prior to starting this Marie Curie project. This version has become the de facto standard approach in this field.
This approach to dark energy has allowed for one of the important results of Project DRKFRCS: until recently essentially all predictions in this field were based on a simpler approach called the quasi-static approximation (QS). I have conclusively shown in ref. [2] that this approximation is not going to be good enough for the upcoming surveys and missions in cosmology and the full EFT-like approach must be used.
New data-analysis tools must be created for this EFT-like approach. Thus a large part of the research effort was expended to develop a computer code which would allow to test data for evidence of deviations from standard cosmology using this method. Together with collaborators, we have modified one of the standard computer codes used to predict what experiments should observe to include these new mechanisms for dark energy. This code was publicly released in June 2016 and is available to any cosmologist in the world to test their data for evidence of dark energy (see ref. [1] and http://www.hiclass-code.net). In particular, it is going to become one of the standard codes used by the Euclid Consortium in the data analysis (see ref. [3])and has already been used by some groups to constrain the models with the data already available.
A significant discovery made by me during this project (ref. [7], published in PRL) is the until now unknown connection between models of dark energy which modify general relativity also modify gravitational waves (GW). I showed that in all models where the propagation of gravitational waves is altered (e.g. they move at a speed different from that of light), the gravitational field around matter is also modified in an observable manner (they have gravitational slip). This for the first time allows us to use a completely different probe (i.e. the detection of gravitational waves) to say something about dark energy. Fortuitously, this has become highly relevant, since LIGO detected GW for the first time in February 2016. Currently the constraints on the properties of GW are very weak (see ref. [6]) but they will get rapidly better as new sources are detected. Until recently GW physics and cosmology were completely separate fields, but now as a result of my work for this project there is a competition to see which teams might be able to constrain dark energy models first.
I opened a third research direction, which is to study whether it is possible that some sort of non-standard behaviour of dark matter might destroy the sensitivity that data sets have to dynamical dark energy. Typically, when investigating dark energy and comparing what the models predict to data, one considers dark matter to be a massive and slow-moving particle (cold dark matter) and one varies only the properties of dark energy. It is, however, possible that dark matter is more complicated (e.g. has a temperature or a mass that is not extremely large). In refs [4,5], I investigated whether by adding some non-standard properties to dark matter it is possible to create the sort of phenomena that dark energy is typically responsible for. The conclusions were luckily negative: we showed that the sort of masses and temperatures that are allowed by the cosmic microwave data can in no way modify the observable properties of the universe on the scales at which dark energy is important. This means that future experiments will be able to constrain properties of dark energy and dark matter separately.
Overall, during the 24 months of research funded my the Marie Curie Action, I have been able to significantly improve the preparedness that we have as a community for the data that will be produced by next-generation observatories. In particular, my hi_class code is already in use by multiple groups and will allow them to correctly build in the effects of dark energy into their predictions and thus measure the various properties of the universe and dark energy without bias.