Sky noise removal for submillimeter spectroscopy based on statistical methods

Removing sky emission is essential to extract astronomical signals for submillimeter spectroscopy with ground-based single-dish telescopes, however, conventional switching methods not only cause baseline instability but result in low observing efficiency of on-source. Here we present two statistical approaches to efficient sky removal. For a heterodyne receiver, we develop an off-point-less observing method by a frequency-modulating local oscillator (FMLO; Taniguchi et al., PASJ, in press), which is three times more efficient than the conventional method. For an ultra-wideband spectrometer (DESHIMA; Endo et al. 2019a, 2019b), we develop a sky removal method applicable to continuum observations by using an atmospheric model.