Photometric redshift methods Sample Clauses

Photometric redshift methods. There is a growing industry based on the estimation of galaxy redshifts from broad-band photometry. The basic photometric redshift technique consists of using the colours of a galaxy in a selection of medium- or broad-band filters as a crude approximation of the galaxy’s spectral energy distribution or SED, in order to derive its redshift and spectral type. The technique is very efficient compared to spectroscopy since the signal-to-noise in broad-band filters is much greater than the signal-to-noise in a dispersed spectrum and, furthermore, a whole field of galaxies can be imaged at once while spectroscopy is limited to individual galaxies or those that can be positioned on slits or fibres. However, photometric redshifts are only approximate at best and are sometimes subject to complete misidentifications. For many applications though, large sample sizes are more important than precise redshifts and photometric redshifts may be used to good effect. Photometric redshifts have been used extensively in recent years on the ultra-deep and well-calibrated Hubble Deep Field observations. The most commonly used approach is the template-fitting technique, implemented, for example in the ‘Hyper-z’ package. This involves compiling a library of template spectra – either theoretical SEDs from population synthesis models (see Section 5.2.2) or empirical. Then, the expected flux through each survey filter is calculated for each template SED on a grid of redshifts, with corrections for interstellar, intergalactic and Galactic extinction where necessary. A redshift and spectral type are then estimated for each observed galaxy by minimizing χ2 with respect to redshift, z, and spectral type SED. Given the growing use of photometric redshifts in giant future surveys such as those to be undertaken by VISTA, it is important to refine existing methods and, if possible, implement better, faster ones. Lahav and collaborators have developed XXX-z, a photo-z method that utilizes Artificial Neural Networks. Unlike the standard template-fitting photometric redshift technique, a large spectroscopically-identified training set is required but, where one is available, XXX-z produces photometric redshift accuracies at least as good as, and often better than the template-fitting method. Furthermore, inputs other than galaxy colours, such as morphology, angular size and surface brightness, may be easily incorporated. The method has been applied successfully to SDSS data as well as to semi-...