Skip to main content

From the Milky Way to the cosmic large-scale structure

Final Report Summary - COSMIWAY (From the Milky Way to the cosmic large-scale structure)

The goals of the COSMIWAY project were to test the validity of the current paradigm for the formation of cosmological structure, the ΛCDM model, and, in so doing, to advance our understanding of the nature of the main constituents of our Universe (dark matter and dark energy) and of the processes of galaxy formation and evolution. The project focused on two critical cosmological diagnostics: the Milky Way and the large-scale distribution of galaxies. The Milky Way informs us about the nature of the dark matter, whereas the galaxy distribution informs us about the nature of the dark energy. An additional goal was to publish data from cosmological simulations and synthetic (or “mock”) catalogues constructed from them.

COSMIWAY had two major strands: a theoretical one based on cosmological simulations and an observational one on based on the Pan-STARRS1 (PS1) survey of galaxies, Milky Way stars and quasars. PS1 used novel technology for high-quality panoramic imaging over a very large field of view using the largest CCD camera ever built for civilian use at the time. By nature this was a high-risk element of the project. Although the telescope was eventually fully operational, it suffered delays during commissioning which limited the use of the data for the project. The survey was completed and a large fraction of the data - a unique legacy dataset of tremendous value for the astronomical community – has now been made publicly available through the Space Telescope Science Institute in Baltimore.

Due to the delays with PS1, the project focused primarily on the theoretical and simulations side. Here we not only achieved all our goals and milestones but significantly exceeded them. We carried out the
largest high-resolution N-body simulations of cosmological volumes to date - the “Copernicus Complexio” (COCO) project - for two different assumptions about the nature of the dark matter, the standard one of cold dark matter (CDM) and an attractive viable candidate, warm dark matter (WDM). The simulations were populated with stars using a novel “particle tagging” technique. We were able to rule out a large region of the parameter space of WDM models. We also showed that CDM and WDM
models can, in future, be distinguished using strong gravitational lensing observations.

A major highlight of our programme was a series of cosmological simulations which follow not just the evolution of dark matter, but also the evolution of baryons or ordinary matter. The EAGLE (“Evolution and Assembly of Galaxies and their Local Environment”) simulation followed the formation and evolution of the galaxy population over a large representative volume of the Universe. It reproduced a host of important properties of the galaxy population and intergalactic gas at various epochs with remarkable fidelity. Using the same code we carried out the ``APOSTLE'' simulations of the Milky Way and its local environment, the ``Local Group'', which also includes the Andromeda galaxy. We showed that two problems that are often deemed to be fatal for the ΛCDM model on subgalactic scales - the ``satellites” and``too-big-to-fail” problems - are neatly solved when the effects of galaxy formation are taken into account.

To study the large-scale distribution of galaxies, we carried out two of the largest high-resolution cosmological N-body simulations to date, the MXXL and the P-millennium. These dark matter simulations were populated with galaxies calculated with a ``semi-analytic'' model of galaxy formation called GALFORM. Both our small-scale and large-scale sets of simulations were used to make mock catalogues, of stars in the Milky Way in the first case, and of galaxies and quasars in the second case. The mock catalogues are very useful in the design and analysis of astronomical surveys and provide the all-important link between observations and cosmological theory. We have made the mocks (and the original simulation data) public using advanced web-based techniques.

The COSMIWAY project was extraordinarily successful as evidenced by the 195 refereed papers stemming from the project that have been published in the scientific literature, including the most cited paper in the whole of astronomy published in 2015.