| Abstract: Detecting primordial gravitational waves is a top priority of next generation CMB surveys like CMB-S4 or the Simons Observatory. It was once hoped that the amplitude of the primordial signal would be large enough to make discrimination from galactic foregrounds and instrumental systematics a simple affair. As upper limits from current surveys march down, however, it is becoming increasingly likely that these will be the limiting factors. Recent results from the Planck satellite indicated that galactic dust emission exhibits significantly greater complexity than expected, which would produce a large bias if not accounted for, significantly decrease constraining power per-dollar if marginalized over, and either way have major implications for experimental design. I will present the results of a re-analysis of these data that finds no evidence for such complexity. I will also present a new data-driven method to predict and filter out instrumental contamination that should be sufficient for CMB-S4. |