Photometric selection of emission-line galaxies, clustering analysis and a search for the integrated Sachs-Wolfe effect

We investigate the use of simple colour cuts applied to the Sloan Digital Sky Survey (SDSS) optical imaging to perform photometric selections of emission-line galaxies (ELGs) out toz < 1. Our selection is aimed at discerning three separate redshift ranges: 0.2 less than or similar to z less than or similar to 0.4, 0.4 less than or similar to z less than or similar to 0.6 and 0.6 less than or similar to z less than or similar to 1.0, which we calibrate using data taken by the COMBO-17 survey in a single field (S11). We thus perform colour cuts using the SDSS g, r and i bands and obtain mean photometric redshifts of <(z)over bar>(low) = 0.32 +/- 0.08, (z) over bar (mid) = 0.44 +/- 0.12 and (z) over bar (hi) = 0.65 +/- 0.21. We further calibrate our high-redshift selection using spectroscopic observations with the AAOmega spectrograph on the 4-m Anglo-Australian Telescope, observing approximate to 50-200 galaxy candidates in four separate fields. With just 1 h of integration time and seeing of approximate to 1.6 arcsec, we successfully determined redshifts for approximate to 65 per cent of the targeted candidates. We compare our spectroscopic redshifts to the photometric redshifts from the COMBO-17 survey and find reasonable agreement between the two. We calculate the angular correlation functions of these samples and find correlation lengths of r(0) = 2.78 +/- 0.08, 3.71 +/- 0.11 and 5.50 +/- 0.13 h(-1) Mpc for the low-, mid-and high-redshift samples, respectively. Comparing these results with predicted dark matter clustering, we estimate the bias parameter for each sample to be b = 0.72 +/- 0.02, b = 0.93 +/- 0.03 and b = 1.43 +/- 0.03. We calculate the two-point redshift-space autocorrelation function at z approximate to 0.6 and find a clustering amplitude of s(o) = 6.4 +/- 0.8 h(-1) Mpc. Finally, we use our photometric sample to search for the integrated Sachs-Wolfe signal in the Wilkinson Microwave Anisotropy Probe (WMAP) 5-yr data. We cross-correlate our three redshift samples with the WMAP W, V, Q and K bands and find an overall trend for a positive signal similar to that expected from models. However, the signal in each is relatively weak, with the results in the WMAP W band being w(Tg)(< 100 arcmin) = 0.25 +/- 0.27, 0.17 +/- 0.20 and 0.17 +/- 0.16 mu K for the low-, mid-and high-redshift samples, respectively. Combining all three galaxy samples, we find a signal of w(Tg)(< 100 arcmin) = 0.20 +/- 0.12 mu K in the WMAP W band, a significance of 1.7 sigma. However, in testing for systematics where the WMAP data are rotated with respect to the ELG sample, we found similar results at several different rotation angles, implying the apparent signal may be produced by systematic effects.