On the atmospheric limitations of ground-based submillimetre astronomy using array receivers

The calibration of ground-based submillimetre observations has always been a difficult process. We discuss how to overcome the limitations imposed by the submillimetre atmosphere. Novel ways to improve line-of-sight opacity estimates are presented, resulting in tight relations between opacities at different wavelengths. The submillimetre camera SCUBA, mounted on the James Clerk Maxwell Telescope (JCMT), is the first large-scale submillimetre array, and as such is ideal for combating the effects of the atmosphere. For example, we find that the off-source pixels are crucial for removing sky noise. Benefiting from several years of SCUBA operation, a data base of deep SCUBA observations has been constructed to help us understand better the nature of sky noise and the effects of the atmosphere on instrument sensitivity. This has revealed several results. First, there is evidence for positive correlations between sky noise and seeing and sky noise and sky opacity. Furthermore, 850-mum and 450-mum sky noise are clearly correlated, suggesting that 450-mum data may be used to correct 850-mum observations for sky noise. Perhaps most important of all: if off-source bolometers are used for sky noise removal, there is no correlation between instrument sensitivity and chop throw, for chop throws out to 180 arcsec. Understanding the effects of submillimetre seeing is also important, and we find that the JCMT beam is not significantly broadened by seeing, nor is there an obvious correlation between seeing and pointing excursions.