The Survey

2dFLenS is a southern-hemisphere spectroscopic redshift survey using the 2dF-AAOmega multi-fibre spectroscopic system at the Australian Astronomical Observatory. Its purpose was to extend the coverage of spectroscopic-redshift observations that overlap with deep optical imaging surveys performed to measure weak gravitational lensing.

Sky coverage

2dFLenS targeted a series of regions in the Southern Galactic Cap (SGC) and Northern Galactic Cap (NGC), shown as the red-shaded areas in the figure on the left. The principal focus of the observations was the area imaged by the Kilo-Degree Survey which lacked existing coverage by the Baryon Oscillation Spectroscopic Survey. 2dFLenS also conducted observations in southern sky regions of two other lensing surveys, CFHTLenS and RCSLenS, which lacked previous deep spectroscopic follow-up. Our observations also overlap with the footprint of the Dark Energy Survey.

We tiled the 2dFLenS observation regions with 2-degree diameter circular pointings of the 2dF+AAOmega spectroscopic system. The 2dFLenS project was allocated a total of 53 AAT nights spread across the 14B, 15A and 15B semesters. Observations for the project were carried out between 17 September 2014 and 5 January 2016.

At the end of the survey we had succeeded in observing 275 2dF field centres, including almost all the fields overlapping with KiDS. The figure on the left illustrates the final status of the survey coverage, which amounted to a total of 731 square degrees.

Target selection

2dFLenS obtained spectra for two main target samples.

First, we targeted Luminous Red Galaxies (LRGs), which can be used to measure the gravitational lensing signal imprinted in the apparent shapes of background sources, in order to test gravitational physics. Bright LRGs in the redshift range z < 1 can be readily selected using colour and magnitude cuts, enabling photometric-redshift calibration by cross-correlation over this range.

Secondly, we observed a nearly-complete subset of galaxies in the magnitude range 17 < r < 19.5, in order to facilitate photometric-redshift determination and source classification by direct calibration.

Since KiDS imaging observations were still ongoing when 2dFLenS commenced, we instead used VST-ATLAS for target selection, which is sufficiently deep for the selection of the 2dFLenS samples.

The LRG target selection for 2dFLenS employed similar colour and magnitude cuts to those utilized by the Sloan surveys, creating "LOWZ-like", "CMASS-like" and "eBOSS-like" samples. The redshift distributions of these selections, together with the magnitude-limited sample, are illustrated in the left-hand figure.

AAT observations

2dFLenS configurations were observed in 45-minute integrations, matched to the fibre re-configuration time of the 2dF instrument. We used the 580V and 385R AAOmega gratings, producing a total wavelength range of each observation from 370 to 880 nm.

2dFLenS data was reduced using the 2DFDR software developed at the AAO. The left-hand figure shows spectra of 2dFLenS LRGs stacked by rest-frame wavelength in dz=0.1 redshift slices. Prominent spectral features are indicated by the vertical dotted lines. Redshifts were obtained by a combination of manual and automatic methods making use of the runz, autoz and marz codes.

Chris Blake,