Deep optical imaging of the field of PC1643+4631A&B, I: Spatial distributions and the counts of faint galaxies

• Published in: arXiv.org
• Main Authors: Haynes, Toby, Cotter, Garret, Baker, Joanne C, Eales, Steve, Jones, Michael E, Rawlings, Steve, Saunders, Richard
• Format: Paper
• Language: English
• Published: Ithaca Cornell University Library, arXiv.org 18.11.1998
• Subjects: Algorithms; Apertures; Big Bang theory; Cosmic microwave background; Error correction; Faint objects; Galaxies; Image detection; Object recognition; Quasars; Red shift; Space telescopes; Stars & galaxies; Systematic errors
• ISSN: 2331-8422
• DOI: 10.48550/arxiv.9811277

We present deep optical images of the PC1643+4631 field obtained at the WHT. This field contains two quasars at redshifts z=3.79 & 3.83 and a cosmic microwave background (CMB) decrement detected with the Ryle Telescope. The images are in U,G,V,R and I filters, and are complete to 25th magnitude in R and G and to 25.5 in U. The isophotal galaxy counts are consistent with the results of Metcalde et al. (1996), Hogg et al. (1997), and others. We find an excess of robust high-redshift Ly-break galaxy candidates with 25.0<R<25.5 compared with the mean number found in the fields studied by Steidel et al. -we expect 7 but find 16 - but we do not find that the galaxies are concentrated in the direction of the CMB decrement. However, we are still not sure of the distance to the system causing the CMB decrement. We have also used our images to compare the commonly used object-finding algorithms of FOCAS and SExtractor: we find FOCAS the more efficient at detecting faint objects and the better at dealing with composite objects, whereas SExtractor's morphological classification is more reliable, especially for faint objects near the resolution limit. More generally, we have also compared the flux lost using isophotal apertures on a real image with that on a noise-only image: recovery of artificial galaxies from the noise-only image significantly overestimates the flux lost from the galaxies, and we find that the corrections made using this technique suffer a systematic error of some 0.4 magnitudes.