Effectiveness of Ground Subtraction

Of principal concern is whether and on what timescales the ground signal is repeatable; if it is stable over eight hours, the common signal can be removed from the eight fields at only a small cost in signal/noise on the residual CMBR component. Fortunately, repeated observations demonstrate that the ground signal is stable over several days, and this subtraction scheme has proven successful.

A selection of five such fields is shown in Figure 4, after ground subtraction. In the presence of residual near-field contamination of the visibilities (say, by the ground), these maps would be dominated by large-scale noise across the map. On the contrary, real structure in the far-field will appear enveloped by the primary beam, as can plainly be seen in these images. As the final panel demonstrates, the enveloping of the signal is well matched to the measured beam profile, approaching the theoretical noise floor far from the center of the map.

More quantitative analysis of the distribution of ground-subtracted visibilities confirms this qualitative impression, and shows no evidence for residual ground contamination for u-v radii $> 50\lambda$, and little residual signal even for the shortest baselines.

 

Figure: Images of five DASI CMB fields. The two concentric circles represent the -3 dB and -10 dB taper of the beam, respectively. The lower right panel shows the rms pixel values for field 5 (lower middle panel) as a function of radius (black points), the primary beam taper normalized to the first rms pixel value (solid line), and the theoretical rms image noise, determined from the scatter in 8-s visibility data (dashed line). The rms noise tapers with the primary beam sensitivity until it becomes dominated by the instrument noise at the edges of the plots, indicating the origin of the signal is on the sky. The signal level is consistent with that expected from the CMB.

Go back to DASI home page