Modeling beam chromaticity for high-resolution CMB analyses
We investigate the impact of beam chromaticity, i.e., the frequency dependence of the beam window function, on cosmological and astrophysical parameter constraints from CMB power spectrum observations. We show that for future high-resolution CMB measurements it is necessary to include a color-corrected beam for each sky component with a distinct spectral energy distribution. We introduce a formalism able to easily implement the beam chromaticity in CMB power spectrum likelihood analyses and run a case study using a Simons Observatory (SO) Large Aperture Telescope-like experimental setup and within the public SO software stack. To quantify the impact, we assume that beam chromaticity is present in simulated spectra but omitted in the likelihood analysis. We find that, for passbands of fractional width $\Delta \nu/\nu \sim 0.2$, neglecting this effect leads to significant biases, with astrophysical foreground parameters shifting by more than $2\sigma$ and cosmological parameters by significant fractions of the error.