EDP Sciences
Free Access
Volume 376, Number 3, September IV 2001
Page(s) 756 - 774
Section Cosmology
DOI https://doi.org/10.1051/0004-6361:20011051
Published online 15 September 2001

A&A 376, 756-774 (2001)
DOI: 10.1051/0004-6361:20011051

The Canada-France deep fields survey

I. 100 000 galaxies, 1 deg $^\mathsf{2}$: A precise measurement of $\mathsf{\omega(\theta)}$ to $I{_{AB}\sim25}$
H. J. McCracken1, O. Le Fèvre1, M. Brodwin2, S. Foucaud1, S. J. Lilly3, D. Crampton3 and Y. Mellier4, 5

1  Laboratoire d'Astrophysique de Marseille, Traverse du Siphon, 13376 Marseille Cedex 12, France
2  University of Toronto, Department of Astronomy, 60 St. George Street, Toronto, Ontario, Canada
3  Herzberg Institute for Astrophysics, 5071 West Saanich Road, Victoria, British Colombia, Canada
4  Institut d'Astrophysique de Paris, 98bis boulevard Arago, 75014 Paris, France
5  Observatoire de Paris, DEMIRM, 61 avenue de l'Observatoire, 75014 Paris, France

(Received 10 April 2001 / Accepted 13 July 2001)

Using the University of Hawaii's 8K mosaic camera (UH8K), we have measured the angular correlation function $\omega(\theta)$ for 100 000 galaxies distributed over four widely separated fields totalling ~$1\deg^2$ and reaching a limiting magnitude of $I_{AB}(3\sigma,3'')\sim 25.5$. This unique combination of areal coverage and depth allows us to investigate the dependence of $\omega(\theta)$ at $1\arcmin$, $A_\omega(1\arcmin)$, on sample median magnitude in the range $19.5< I_{AB-{\rm med}}< 24$. Furthermore, our rigorous control of systematic photometric and astrometric errors means that fainter than $I_{AB-{\rm med}}\sim 22$ we measure $\omega(\theta)$ on scales of several arc-minutes to an accuracy of $30\%$. Our results show that $A_\omega(1\arcmin)$ decreases monotonically to $I{_{AB}\sim25}$. At bright magnitudes, $\omega(\theta)$ is consistent with a power-law of slope $\delta =
-0.8$ for $0.2\arcmin< \theta< 3.0\arcmin$ but at fainter magnitudes we detect a slope flattening with $\delta \sim -0.6$. At the $3\sigma$ level, our observations are still consistent with $\delta =
-0.8$. We also find a clear dependence of $A_\omega(1\arcmin)$ on observed (V-I)AB colour. In the magnitude ranges 18.5< IAB< 24.0 and 18.5< IAB< 23.0 we find galaxies with 2.6< (V-I)AB< 2.9 (the reddest bin we consider) have $A_\omega(1\arcmin)$'s which are ~$10\times$ higher than the full field population. On the basis of their similar colours and clustering properties, we tentatively identify these objects as a superset of the "extremely red objects" found through optical-infrared selection. We demonstrate that our model predictions for the redshift distribution for the faint galaxy population are in good agreement with current spectroscopic observations. Using these predictions, we find that for low-$\Omega$ cosmologies and assuming a local galaxy correlation length r0=4.3 h-1 Mpc, in the range $19.5< I_{AB-\rm med}< 22$, the growth of galaxy clustering (parameterised by $\epsilon$), is $\epsilon\sim0$. However, at $22< I_{AB-\rm med}< 24.0$, our observations are consistent with $\epsilon\gtrsim 1$. Models with $\epsilon\sim0$ cannot simultaneously match both bright and faint measurements of $A_\omega(1\arcmin)$. We show how this result is a natural consequence of the "bias-free" nature of the "$\epsilon$" formalism and is consistent with the field galaxy population in the range 22.0< IAB< 24.0 being dominated by galaxies of low intrinsic luminosity.

Key words: cosmology: large-scale structure of Universe -- observations: galaxies -- general -- astronomical data bases: surveys

Offprint request: H. J. McCracken, henry.joy.mccracken@astrsp-mrs.fr

© ESO 2001

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.