Astron. Astrophys. 318, 729-740 (1997)

6. The associations between other samples of galaxies

The study of any possible association between faint blue and galaxies has been extended to a consideration of other samples of galaxies. A colour index test was applied to the brighter galaxies in order to select the candidates. It is of interest to investigate whether a relaxation of the selection criteria to include brighter galaxies of all colours affects the conclusions. Equally, any possible clustering between the brighter () galaxies and all faint () galaxies, the faint blue with faint red galaxies, and between the faint blue galaxies and themselves have been investigated.

The selection criteria for the samples of galaxies are summarised in Table 4. The brighter images were subjected to a star/galaxy classification as described in Sect. 5.1 for the candidates. This reduced the number of images from 38 to 21. As in the case of the faint blue galaxies, no classification was attempted for the faint and faint red samples. Unlike the case of the faint blue samples, however, the samples will include modest numbers of main sequence stars of the Galactic halo and thick disc. Following the method of Sect. 5.1, star number densities were predicted for each galaxy sample of each of the four fields and image number densities were taken from Metcalfe et al. (1995). The resulting fractional star contaminations are presented in Table 5. The effect of stars is in general small, although the fourth field, lying at intermediate galactic latitude, does experience some significant contamination to the red galaxies sample.

Galaxy-galaxy separations were calculated between the members of different samples, as presented in Table 6. The table lists the total number of galaxy-galaxy separations and the number of separations in the range 10 to 60 arcsec.

Table 6. Galaxy-galaxy separation results for additional samples of galaxies. The numbers of galaxy-galaxy separations between the various samples are presented for the observational data for both the entire range of separations and for the restricted range to . Error estimates are taken from Sect. 7. For comparison the results of the Monte Carlo simulations for randomly distributed points are presented in columns 5 and 6 for the same separation ranges. For the faint blue-faint red study, simulated case (a) refers to random blue and the observed red galaxies; case (b) refers to the observed blue and simulated red galaxies. In the faint blue-faint blue study, case (a) represents the statistics for one observed and one random blue sample, while case (b) refers to the intercorrelations of a single random sample

Detailed Monte-Carlo simulations were performed for each pair of samples to predict the number of separations in the 10 to 60 arcsec range, using large numbers of random points to reduce numerical noise, normalising the results to the observed total number of separations in each frame. The bright-faint blue galaxies simulation used the observed bright galaxy data, but simulated random faint blue images for each frame. Two simulations were performed for the faint blue-faint red samples: one (case (a) in Table 6) used random blue galaxies and the observed red on each frame; the other (case (b)) used the observed blue and random red galaxies on each frame. The bright-faint galaxy study again used the observed bright galaxy data, with simulated random faint images on each frame. The simulation of the association of faint blue galaxies with themselves was firstly (case (a)) performed using the observed blue galaxies and random points on each data frame, then again (case (b)) with a single random sample of 2000 galaxies per frame computing the separations internal to the random sample. All simulation results were normalised to the observed total number of separations.

© European Southern Observatory (ESO) 1997

Online publication: July 3, 1998
helpdesk.link@springer.de