Periodic Reporting for period 3 - CODEX (The Final 21-cm Cosmology Frontier)
Période du rapport: 2023-12-01 au 2025-05-31
The early Universe holds essential information about its first stars, galaxies, black holes, and fundamental physics and can be probed through the (redshifted) 21-cm line emission from neutral hydrogen. Detecting this faint "21-cm signal" has been challenging due to bright radio emission from various sources, interference signals, and telescope imperfections. In recent years, significant progress has been made using advanced radio telescopes, such as the Low-Frequency Array (LOFAR), and innovative data analysis techniques. The CoDEX program aims to bring together these efforts to significantly improve our understanding of the early Universe by processing massive amounts of data and enhancing our 21-cm signal detection capabilities. This research is crucial before the next-generation instruments like the Square Kilometre Array become available.
As part of the CoDEX program, significant achievements include the installation of a cutting-edge GPU-based data processing cluster at the University of Groningen, enhancing data processing speed. The development of the "NextLEAP" processing pipeline integrates a decade's worth of codes into an automated system, ensuring scalable and reproducible workflows with standardized outputs. We've advanced calibration codes, including direction-dependent gain calibration, currently undergoing testing by dedicated postdocs. A forward-simulation project within CoDEX is underway, testing previously unexplored aspects and incorporating a full ionospheric simulation. Our PhD students have processed a full night of data from the new NenuFAR radio telescope, revealing new insights into signal distortions during the Cosmic Dawn. Simultaneously, the CoDEX team is finalizing the reprocessing of 140 hours of LOFAR data, expanding our understanding of the redshift regime. The LOFAR-AARTFAAC system's data processing has commenced, resulting in publications on transient RFI, with ongoing work on the full data processing pipeline. Additionally, a PhD student has developed a simulation pipeline for space-based instruments, enabling comprehensive testing of instrument specifications and their impact on measuring the 21-cm signal from the Dark Ages.
What is essentially novel is the inclusion, for the very first time, of realistic ionospheric effects in our forward data simulations. This allows us, for the very first time, to investigate its impact on data and data processing. Another breakthrough has been the identification of several likely causes of the elusive "excess variance" seen in the power spectra of the residual data, the inclusion of beam models in our calibration codes and the machine-learning extensions of the foreground data modelling via Gaussian Processes. Similarly, we have advanced our knowledge of how inferences on the ground plane of the radio array impact the 21-cm signal power spectrum and how to model it. Until the end of the project we expect to remove most of the excess variance and be able to place limits on the "21-cm signal" of neutral hydrogen from the universe's first billion year allowing us to start testing astrophysical models