Optimization by Smoothed Bandpass Calibration in Radio Spectroscopy

Abstract

Abstract We have developed the Smoothed Bandpass Calibration (SBC) method and the best suitable scan pattern to optimize radio spectroscopic observations. Adequate spectral smoothing is applied to the spectrum toward OFF-source blank sky adjacent to a target source for the purpose of bandpass correction. Because the smoothing process reduces noise, we can reduce the integration time for OFF-source scans while keeping the signal-to-noise ratio. Since smoothing is not applied to ON-source scans, the spectral resolution for line features is maintained. An optimal smoothing window is determined by the bandpass flatness evaluated by the spectral Allan variance (SAV). An efficient scan pattern is designed for OFF-source scans within the bandpass-stability time scale estimated by the time-based Allan variance (TAV). We tested the SBC using a digital spectrometer, VESPA, at the VERA Iriki station. For a targeted noise level of 5 $ \times$10$ ^{-4}$ as its ratio to the system noise, the optimal smoothing window was 32–60 channels (ch) over the whole bandwidth of 1024 ch, and the optimal scan pattern was designed as a sequence of 70 s ON $ +$ 10 s OFF scan pairs. The noise level by the SBC has been improved by a factor of 1.74, compared with that by the conventional method. The total telescope-time to reach our target with the SBC was 400 s, which is 1$ /$ 3 of 1200 s required in the conventional way. Improvements in the telescope-time efficiency with the SBC were calculated to be 3$ \times$ (for single-beam scans), 2$ \times$ (dual-beam), and 1.3$ \times$ (on-the-fly). The SBC is useful for optimizing scan patterns for observations from now on, and also for improving the signal-to-noise ratios of archival data if ON- and OFF-source spectra are separately recorded, though the efficiency depends on the spectral stability of the receiving system.