http://arxiv.org/abs/2304.09959
Accurate detection of the cosmological 21-cm global signal requires galactic foreground models that can fit spectra down to $\sim 20$ mK or less, representing a removal of power over nearly six orders of magnitude. Rarely are such models tested to this level, let alone their dependence upon model inputs like sky temperature maps. We therefore test the ability of seven commonly employed foreground models — including nonlinear and linear forward-models, polynomials, and maximally-smooth polynomials — to fit realistic simulated mock spectra, as well as their dependence upon model inputs. The mock spectra are synthesized from intrinsic foregrounds with realistic spatial and spectral structure, chromatic beams, horizon profiles, and discrete time-sampling. For a single LST bin spectrum, the nonlinear-forward model with 4 parameters is preferred using a KS-test of the noise-normalized residuals, while the linear forward-model fits well with 6-7 parameters. The polynomials and maximally-smooth polynomials, like those employed by the EDGES and SARAS3 experiments, cannot produce good fits with 5 parameters. However, we find that polynomials with 6 parameters pass the KS-test, although a 9 parameter fit produces the highest p-value. When fitting multiple LST bins simultaneously to decrease overlap with global signal models, we find that the linear forward-model outperforms the nonlinear for 2, 5 and 10 LST bins. In addition, the nonlinear forward-model fails to produce good fits to spectra with 10 LST bins, in contrast to the linear. Importantly, the KS-test consistently identifies best-fit \textit{and} preferred models as opposed to the $\chi^2_{red}$ and Bayesian evidence, especially in cases involving nonlinear models.
J. Hibbard, D. Rapetti, J. Burns, et. al.
Fri, 21 Apr 23
37/60
Comments: 25 pages, 12 figures, submitted to ApJ