Periodic Reporting for period 4 - COSMIC LENS (Delivering on the Promise of Measuring Dark Energy from Cosmic Lensing)
Reporting period: 2020-11-01 to 2021-04-30
Arguably the biggest mystery in cosmology today is the source of the apparent accelerated expansion of the Universe, the discovery of which won the Nobel Prize in 2011. Cosmic lensing has been shown to have the greatest potential of all probes to discover the nature of the proposed dark energy that might cause this acceleration. In addition cosmic lensing yields statistical properties of the dark matter distribution, which also allows investigation of fundamental questions about massive neutrinos and gravity itself. Several experiments have been funded to measure cosmic lensing, the largest such current project is the Dark Energy Survey (DES).
Our goal was to measure the density and clumpiness of the dark matter, and to use this to measure the equation of state of dark energy. This is important to society because of mankind’s innate curiosity about the contents of the Universe, and our desire to understand the missing 95% of the Universe. The overall objective was to use the data from the Dark Energy Survey and other probes to put the definitive constraints on dark energy for the present decade, and to use this to inform future cosmological constraints from surveys such as the Large Synoptic Survey Telescope, which will dominate the cosmological landscape of the decade to come.
As part of a large US-led project, we made the largest ever map of the dark matter, using the largest ever number of galaxies used for cosmic lensing (>100 million) and over the largest area of sky used for cosmic lensing (over 4000 square degrees, or one tenth of the full sky). We measured the clustering of the dark matter and the equation of state of dark energy, making the tightest constraints to-date using cosmic lensing. We compared the results to those from previous projects, finding excellent agreement, in contrast with earlier measurements which had suggested a tension.
Our goal was to measure the density and clumpiness of the dark matter, and to use this to measure the equation of state of dark energy. This is important to society because of mankind’s innate curiosity about the contents of the Universe, and our desire to understand the missing 95% of the Universe. The overall objective was to use the data from the Dark Energy Survey and other probes to put the definitive constraints on dark energy for the present decade, and to use this to inform future cosmological constraints from surveys such as the Large Synoptic Survey Telescope, which will dominate the cosmological landscape of the decade to come.
As part of a large US-led project, we made the largest ever map of the dark matter, using the largest ever number of galaxies used for cosmic lensing (>100 million) and over the largest area of sky used for cosmic lensing (over 4000 square degrees, or one tenth of the full sky). We measured the clustering of the dark matter and the equation of state of dark energy, making the tightest constraints to-date using cosmic lensing. We compared the results to those from previous projects, finding excellent agreement, in contrast with earlier measurements which had suggested a tension.
We played a major role in developing techniques for the analysis of cosmic lensing data, and understanding its implications in the context of the wider cosmological framework.
Our work in this project spanned three main areas, corresponding to the key challenges in interpreting large scale imaging data to produce cosmological conclusions from cosmic lensing.
1. Extremely accurate measurement of galaxy shapes. We developed galaxy shape measurement methodologies and carried out large simulations to test and calibrate the results, providing one of the two galaxy shape catalogues used in the first year (Y1) cosmology results from the Dark Energy Survey (DES). We took a fresh look at shape measurement methodology and proposed a novel approach. We wrote the majority of the code used for the massive simulations used to calibrate the DES third year results (Y3), and contributed to tests of the quality of the results.
2. Measurement of galaxy clustering and shape two-point functions. Cross-correlating the positions and shapes of galaxies provides a powerful window onto the dark universe, but only if confounding effects are taken into account at sufficient accuracy. We led the galaxy clustering results from the first year of DES data, evaluating and removing the main potential contaminants. We showed how combination of the clustering and shape information could provide a key to help understanding the distribution of galaxies as a function of estimate distance from the observer (redshift). For the first time, we developed a way for uncertainties in galaxy redshifts to be incorporated into the standard cosmological parameter estimation framework (MCMC) - hyperrank, demonstrating that this would not be necessary for DES Y3 but paving the way for future years.
3. We contributed to the final cosmology constraints from DES Y1 and DES Y3 through our work in the above areas and by creating a fast pipeline which we used to evaluate the quality of the MCMC samplers and configurations, thereby influencing the choice of method, as well as developing and testing to the modified gravity constraint methodology. We used this work and our work on other key datasets including Planck, to put constraints on wider classes of cosmological model, and investigate tensions between different datasets.
Furthermore, we took a proactive approach to influencing the design of future experiments, including incorporating the potential of gravitational waves and standard sirens, and future data from the cosmic microwave background, taking an overview of future datasets and their potential to resolve tensions or discover new physics.
A major legacy of this project is the SkyPy open-source Python package for simulating the astrophysical sky. It comprises a library of physical and empirical models across a range of observables and a command line script to run end-to-end simulations. The library provides functions that sample realisations of sources and their associated properties from probability distributions. Simulation pipelines are constructed from these models using a YAML-based configuration syntax, while task scheduling and data dependencies are handled internally and the modular design allows users to interface with external software.
The results from the Dark Energy Survey have been widely disseminated, including via press releases. We have given a number of talks at seminars and conferences about the DES results and about the SkyPy project, as well as on our broader results on cosmological constraints.
Our work in this project spanned three main areas, corresponding to the key challenges in interpreting large scale imaging data to produce cosmological conclusions from cosmic lensing.
1. Extremely accurate measurement of galaxy shapes. We developed galaxy shape measurement methodologies and carried out large simulations to test and calibrate the results, providing one of the two galaxy shape catalogues used in the first year (Y1) cosmology results from the Dark Energy Survey (DES). We took a fresh look at shape measurement methodology and proposed a novel approach. We wrote the majority of the code used for the massive simulations used to calibrate the DES third year results (Y3), and contributed to tests of the quality of the results.
2. Measurement of galaxy clustering and shape two-point functions. Cross-correlating the positions and shapes of galaxies provides a powerful window onto the dark universe, but only if confounding effects are taken into account at sufficient accuracy. We led the galaxy clustering results from the first year of DES data, evaluating and removing the main potential contaminants. We showed how combination of the clustering and shape information could provide a key to help understanding the distribution of galaxies as a function of estimate distance from the observer (redshift). For the first time, we developed a way for uncertainties in galaxy redshifts to be incorporated into the standard cosmological parameter estimation framework (MCMC) - hyperrank, demonstrating that this would not be necessary for DES Y3 but paving the way for future years.
3. We contributed to the final cosmology constraints from DES Y1 and DES Y3 through our work in the above areas and by creating a fast pipeline which we used to evaluate the quality of the MCMC samplers and configurations, thereby influencing the choice of method, as well as developing and testing to the modified gravity constraint methodology. We used this work and our work on other key datasets including Planck, to put constraints on wider classes of cosmological model, and investigate tensions between different datasets.
Furthermore, we took a proactive approach to influencing the design of future experiments, including incorporating the potential of gravitational waves and standard sirens, and future data from the cosmic microwave background, taking an overview of future datasets and their potential to resolve tensions or discover new physics.
A major legacy of this project is the SkyPy open-source Python package for simulating the astrophysical sky. It comprises a library of physical and empirical models across a range of observables and a command line script to run end-to-end simulations. The library provides functions that sample realisations of sources and their associated properties from probability distributions. Simulation pipelines are constructed from these models using a YAML-based configuration syntax, while task scheduling and data dependencies are handled internally and the modular design allows users to interface with external software.
The results from the Dark Energy Survey have been widely disseminated, including via press releases. We have given a number of talks at seminars and conferences about the DES results and about the SkyPy project, as well as on our broader results on cosmological constraints.
The Dark Energy Survey results, which we contributed to as described above, represent major progress beyond the state of the art. The cosmological constraining power of this largest cosmic lensing survey offered great potential for the most precise constraints ever from this method. However, in order to deliver on this promise it was necessary to rewrite the methodologies to mitigate confounding effects that arise at this level of precision. We identified ways that galaxy clustering could interfere with galaxy shape measurement, and demonstrated the necessity of large simulations to calibrate the effect. It was therefore necessary to write the most detailed ever galaxy image simulation code, which we contributed to, and this was used to calibrate the largest ever cosmic lensing galaxy catalogue. We devised new tests and methods for calculating the cross-correlation between galaxy positions and shapes. We tested a key assumption in how matter clustering is calculated assuming equal times for each of the two spatial positions, and found it adequate for present integrated data. We developed a new methodology for propagating uncertainties in galaxy distance estimate distributions through to cosmological constraints. Working as part of a large international collaboration, these innovations contributed to the tightest ever cosmological constraints on the galaxy clustering amplitude from cosmic lensing.