Abstract

CMB polarization patterns hold the key to unveiling physical processes at play in the instants directly following the birth of the universe; primordial B-modes, in particular, are highly sought after as their observation would allow to determine the energy scale of inflation. The Simons Observatory (SO), currently in construction in the Atacama Desert and due to start operations in 2023, is expected set significantly tighter constraints on the tensor-to-scalar ratio r than any previous CMB experiment. Reaching SO’s target precision requires a thorough understanding and mitigation of other large-scale B-mode sources such as Galactic foregrounds and weak gravitational lensing. In this talk, I will present a first demonstration of simultaneous foreground cleaning and delensing on SO-like simulations, using a cross-spectral component separation algorithm adapted to include a lensing B-mode template. The improvement of after delensing will be assessed in the presence of increasingly complex foregrounds, and prospects for further optimization will be outlined.