Surveys with the xNTD and CLAR

We compare strawman designs for two Square Kilometre Array demonstrators, the Extended New Technology Demonstrator (xNTD) and the Canadian Large Adaptive Reflector (CLAR), in the context of survey capability and suitability for addressing key survey science programmes. We first show that, all else being equal, the survey speed for detecting point sources is proportional to the product of field-ofview and the square of the effective collecting area for both interferometer and single-antenna designs. For extended emission, the survey speed is proportional to field-of-view alone for single-antennas, and field of view times the square of the filling factor of the array for interferometers. CLAR offers a very large collecting area, but a small field of view and only modest resolution. xNTD offers the converse, with only moderate collecting area but a large field of view and high resolution. Based on strawman specifications, the two instruments are comparable in speed for point source surveys, with filling factor obviously impacting xNTD for extended emission. However, practical considerations such as spectral baseline removal and continuum confusion likely mean that although the two instruments would be closely comparable for H I galaxy redshift work, xNTD would be far superior for continuum point-source work because of the lower confusion limit in its smaller (synthesised) beam. The large, filled aperture of CLAR would, however, make it the instrument of choice for detecting the cosmic web, and for pulsar work. From a science point of view the two instruments are thus largely complementary.