The Arizona CDFS Environment Survey


Here, we further describe the parameters of the survey, including details regarding the sampling rate, redshift success rate, and comparison to other spectroscopic and photometric redshift measurements. More current and complete information regarding ACES is included in Cooper et al. 2012.

Work is underway to examine conflicts internal to the catalog (e.g., where a source has two high-quality redshift measurements that significantly disagree). In addition, we are looking at comparisons with other data sets to test the robustness of our redshifts. As a simple sanity check, we compare the spectroscopic redshifts from ACES with the photometric redshifts from COMBO-17 (courtesy of Wolf et al. 2004). In general, there is excellent agreement....

The deficiency of redshifts around z ~ 0.9 results from our inability to resolve the [O II] doublet. That is, the ACES spectrum of an emission-line galaxy at z ~ 0.9 would yield an unresolved [O II] emission doublet at λ ~ 7100Å, while Hβ and [O III] would be redward of our spectral window (λ > 9200Å). Now, we are unable to easily distinguish this single emission line from Hα (i.e., a galaxy at z ~ 0.08), as at that redshift Hβ and [O III] would be blueward of our spectral window (λ < 5650Å). As a result, many objects at z ~ 0.9 are classified as Q=2. Using broad-band color info, we hope to recover these objects in the future.

Unlike many previous spectroscopic campaigns in the Chandra Deep Field South, ACES extends well beyond the GOODS-S footprint to cover the full 30' x 30' ECDFS. The following plot shows the distribution of ACES redshifts (green and red points) on the sky relative to the COMBO-17 R-band photometric catalog (smaller black points). The distribution of high-quality public redshifts in the field is shown in comparison (cyan diamonds). The ACES redshift catalog more than doubles the number of high-quality spectroscopic redshifts in the ECDFS, while also significantly extending the spatial coverage.

The black points show the distribution of sources in the COMBO-17 optical catalog with R < 24.5. The green points denote those sources with a secure redshift from ACES, while the red points indicate failures in the current catalog --- some of these will likely be converted to green points with improved data reduction techniques. Finally, the cyan diamonds show the location of high-quality redshifts from the following sources: Le Févre et al. 2004, Vanzella et al. 2005, Vanzella et al. 2006, Mignoli et al. 2005, Ravikumar et al. 2007, Szokoly et al. 2004, Balestra et al. 2010, and finally some yet-to-be-published DEEP2/DEIMOS redshifts.

The completeness of the ACES sample is quite high at bright magnitudes (R <  23). The left-hand plot below shows the magnitude distribution for the COMBO-17 parent R-band catalog (solid black line) alongside the magnitude distribution for the complete ACES sample of spectroscopic targets (green dashed line) and for the sample of sources with an accurate (Q = -1,3,4) ACES redshift (red dotted line). The right-band plot below then shows the ratio of these distributions (i.e., the sampling fraction as a function of magnitude) for the two classes of objects: (i) those sources targeted by ACES and (ii) those sources for which an accurate redshift is measured. At magnitudes brighter than R = 23, the sampling rate of the survey is roughly 80%, with accurate redshifts measured for more than half of all sources at R < 23. When folding in available public redshifts in the field, the effective sampling rate for the survey is higher yet (see dashed lines below).

Given the large field-of-view of IMACS, every ACES slitmask covers the central portion of the ECDFS. In spite of this, the sampling rate is relatively uniform across the field at R < 23 (see plot of target sampling rate here), though there is some decrease towards the corners of the field. The following plots show the redshift completeness as a function of spatial position in the CDFS field for sources with R < 23, as contained in the current (2011jun01) ACES redshift catalog (left plot) and in the ACES redshift catalog plus the set of public redshifts listed above (right plot). The public data sets primarily cover the central GOODS-S region, bringing the redshift success rate in the central portion of the field to roughly 80%. In each plot, the red %s to the right of the color bar gives the portion of the CDFS area with a target completeness greater than the corresponding level in the color bar (e.g., ACES currently achieves a redshift completeness of greater than 60% across 51% of the 30’ x 30’ extended CDFS region).

The redshift completeness for sources at RAB < 23 is computed in a sliding box with height and width of ~35”, as illustrated in the upper right-hand corner of each plot. The black dashed line in each plot shows the extent of the COMBO-17 parent R-band catalog (roughly 30' x 30'), while the magenta box designates the location of upcoming WFC3-IR observations as part of CANDELS. The redshift completeness for the ACES sample is relatively high and also uniform across the field, allowing environment statistics to be robustly measured across nearly the entire ECDFS region.