λ 6-cm observations of fluctuations in the 3 K cosmic microwave background

Abstract

We describe the design and execution of the first in a series of experiments to measure fluctuations in the cosmic microwave background (CMB) at λ6 cm on angular scales from 10 to 60 arcmin. At this wavelength the atmosphere has negligible effect on the observations and a high degree of repeatability in the data is demonstrated. Particular attention is paid to identifying and then eliminating systematic effects which have limited previous experiments. These effects repeat in hour angle (HA) and are due to the telescope environment. A careful discussion is given of the statistical procedures used, in order that our results can be critically compared with previous work. This includes a description of (a) the removal of systematic effects arising from the HA baseline and known sources in the field, (b) the hypothesis tests required in obtaining an estimate of the CMB fluctuations, and (c) the derivation of fluctuations on scales larger than the beam. Appendices give details of some important analytical procedures. Observations were made of regions surrounding the north celestial pole (NCP) and the galaxy clusters A 576 and 2218. The most sensitive results were obtained for the NCP region where a deep λ6-cm survey for weak sources already existed. The 95 per cent confidence upper limit for fluctuations on a scale of 10 arcmin around the observing track in the NCP region was $$T_b\lt0.86$$ mK, corresponding to $$\Delta T/T\lt3.0\times10^{-4}$$. This upper limit was 4.6 × 10⁻⁴ for 30-arcmin scales and 2.3 × 10⁻³ for 60-arcmin scales. Such limits appear to rule out existing theories of galaxy formation by adiabatic perturbations and a comparison with the predictions of isothermal theories is made. By using the mean levels at three observing centres in the NCP region, a tentative indication of anisotropy at a level of $$9\times10^{-5}$$ was found. The searches for decrements in A 576 and 2218 revealed the presence of significant emission at the position of both clusters. Any discussion of decrements in these clusters must take account of this central emission as well as source confusion in the immediate surroundings.