[Paper Review] Galaxy and Mass Assembly (GAMA): maximum likelihood determination of the luminosity function and its evolution

J. Loveday, P. Norberg+7 more|arXiv (Cornell University)|May 5, 2015

Galaxies: Formation, Evolution, Phenomena1 references4 citations

TL;DR

This paper presents a modified joint stepwise maximum likelihood (JSWML) method to simultaneously measure the r-band luminosity function (LF) and its evolution in the GAMA-II survey, correcting for radial density variations and luminosity/density evolution. It finds that blue galaxies exhibit stronger luminosity evolution (fading by ~0.5 mag) and higher luminosity density evolution than red galaxies, with significant degeneracy between luminosity and density evolution parameters.

ABSTRACT

We describe modifications to the joint stepwise maximum likelihood method of Cole (2011) in order to simultaneously fit the GAMA-II galaxy luminosity function (LF), corrected for radial density variations, and its evolution with redshift. The whole sample is reasonably well-fit with luminosity (Qe) and density (Pe) evolution parameters Qe, Pe = 1.0, 1.0 but with significant degeneracies characterized by Qe = 1.4 - 0.4Pe. Blue galaxies exhibit larger luminosity density evolution than red galaxies, as expected. We present the evolution-corrected r-band LF for the whole sample and for blue and red sub-samples, using both Petrosian and Sersic magnitudes. Petrosian magnitudes miss a substantial fraction of the flux of de Vaucouleurs profile galaxies: the Sersic LF is substantially higher than the Petrosian LF at the bright end.

Motivation & Objective

• To develop a robust method for measuring the galaxy luminosity function (LF) in the presence of radial density variations and cosmic evolution.
• To correct for luminosity and number density evolution in the GAMA-II survey over a redshift baseline of z ≈ 0.35.
• To compare the evolution of blue and red galaxy populations in terms of luminosity and density changes.
• To assess the reliability of the method using simulated data and comparison with local SDSS data.
• To provide evolution-corrected LFs using both Petrosian and Sérsic magnitudes, highlighting flux loss in standard photometry.

Proposed method

• Adaptation of Cole's (2011) joint stepwise maximum likelihood (JSWML) method to include radial density corrections and evolution modeling.
• Use of density-corrected Vmax values to account for radial overdensities in the survey volume.
• Simultaneous fitting of non-parametric LF estimates and evolution parameters (Qe for luminosity, Pe for density) across redshift bins.
• Application of the method to GAMA-II data with spectroscopic redshifts (nQ ≥ 3) and Petrosian/Sérsic magnitudes.
• Use of simulated data to test method performance and validate recovery of the true LF and evolution parameters.
• Comparison of results with local SDSS data to assess accuracy and consistency.

Experimental results

Research questions

• RQ1 How can the galaxy luminosity function be accurately measured in a flux-limited survey with significant radial density variations and cosmic evolution?
• RQ2 What are the relative contributions of luminosity and number density evolution to the observed LF in the GAMA-II survey?
• RQ3 How do the evolution parameters differ between blue and red galaxy populations?
• RQ4 To what extent do Petrosian magnitudes underestimate the luminosity function compared to Sérsic magnitudes?
• RQ5 Can the method reliably recover the true LF and evolution parameters despite degeneracies between Qe and Pe?

Key findings

• The overall galaxy population shows a luminosity fade of approximately 0.5 mag in the r-band and a comoving number density decrease by a factor of ~1.6 since z ≈ 0.5.
• Blue galaxies exhibit significantly stronger luminosity evolution than red galaxies, with luminosity density evolution differing at the ~5σ level.
• The luminosity function derived from Sérsic magnitudes is substantially higher at the bright end than that from Petrosian magnitudes, due to flux missed in the latter.
• The luminosity function is poorly described by a Schechter function, especially for red galaxies, due to excess counts at both faint and bright magnitudes.
• A degeneracy exists between luminosity evolution (Qe) and density evolution (Pe), with Qe ≈ 1.4 - 0.4Pe, limiting precise individual parameter constraints.
• The method successfully recovers the evolution-corrected LF to high accuracy, as validated by simulations and comparison with SDSS data.