During the last two-and-a-half years, the REA-managed BeyondPlanck project has developed a radically new approach to detecting faint signals in the cosmic microwave background (CMB), an echo from the Big Bang. During a highly attended conference on 18-20 November, the project consortium released and discussed its ground-breaking results on the one single global parametric model for CMB analysis.
The BeyondPlanck project was funded under the H2020-COMPET-2017 call, under the ‘Scientific data exploitation’ topic. The project, which started in March 2018, set out to develop the world's first integrated end-to-end data analysis pipeline for CMB observations, and apply this to data collected by the European Space Agency’s (ESA) Planck satellite mission. These observations provide a unique window into the early universe and the physics of the Big Bang.
The importance of this work lies in the mitigation of instrumental systematic errors. Specifically, as CMB telescopes have become increasingly powerful over the last few decades, they are more sensitive to subtle instrumental uncertainties. Indeed, these uncertainties have become the single most important challenge in the field today, as the race towards the first detection of primordial gravitational waves created during the Big Bang has intensified. This cosmological signal, generated by the gravitational waves, is expected to be tiny - if it exists at all - so even very subtle systematic effects could confuse it.
The BeyondPlanck project consortium has addressed this issue by developing the first CMB model that accounts for the entirety of the analysis problem: from raw satellite observations to final cosmological parameters - all within one coherent statistical framework. This global model is explored using state-of-the-art Bayesian Monte Carlo Markov Chain (MCMC) statistical methods. This results not only in better estimates of each free parameter, but also in uniquely reliable uncertainty estimates, which are essential when searching for weak signals such as gravitational waves.
In a real world demonstration of these methods, the BeyondPlanck project has applied their new framework to existing data from the Planck Low Frequency Instrument (LFI). The results are currently being presented to the cosmological community in a series of 18 research papers. The various scientific highlights of the results include a fundamentally new understanding of the LFI instrument itself that accounts for time variability; a new calibration model that shows consistency between ESA's Planck and NASA's Wilkinson Microwave Anisotropy Probe (WMAP) satellite data; new state-of-the-art maps of the microwave sky between 30 and 70 GHz; and a new model of the Milky Way.
While the BeyondPlanck project in its current form ends on 30 November 2020, this data release marks only the beginning of global CMB analysis. The BeyondPlanck software has been released under an Open Source License, and the entire CMB community is invited to extend and apply it to their own experiments. Furthermore, the software will serve as the algorithmic framework for the recently started European Research Council Executive Agency managed project Cosmoglobe. This project aims to establish a global model of the microwave sky that incorporates data from all available experiments in a community-wide effort.
BeyondPlanck’s consortium partners include the University of Oslo (project coordinator), the University of Milano, INAF - Istituto Nazionale di Astrofisica, the University of Helsinki and Planetek Hellas.
23 Noviembre 2020