[Paper Review] Photometric H alpha and [O II] Luminosity Function of SDF and SXDF Galaxies: Implications for Future Baryon Oscillation Surveys

Masanao Sumiyoshi, Tomonori Totani+10 more|ArXiv.org|Feb 12, 2009

Galaxies: Formation, Evolution, Phenomena3 citations

TL;DR

This study estimates the Hα and [O II] emission line luminosity functions of galaxies at z = 0.5–1.7 using photometric redshifts and star-formation rate estimates derived from multi-wavelength imaging in the Subaru Deep Field and Subaru XMM-Deep Field. It demonstrates that color-color selections using Bi′z′ and Bz′K filters can efficiently target emission line galaxies for future baryon acoustic oscillation (BAO) surveys with 50–60% success rates, requiring depths of B ≈ 24–25 for optimal redshift coverage.

ABSTRACT

Efficient selection of emission line galaxies at z > 1 by photometric information in wide field surveys is one of the keys for future spectroscopic surveys to constrain dark energy using the baryon acoustic oscillation (BAO) signature. Here we estimate the H alpha and [O II] line luminosity functions of galaxies at z = 0.5-1.7 using a novel approach where multi-wavelength imaging data is used to jointly estimate both photometric redshifts and star-formation rates. These photometric estimates of line luminosities at high-redshift use the large data sets of the Subaru Deep Field and Subaru XMM-Newton Deep Field (covering \sim 1 deg^2) and are calibrated with the spectroscopic data of the local Sloan Digital Sky Survey galaxies. The derived luminosity functions (especially H alpha) are in reasonable agreement with the past estimates based on spectroscopic or narrow-band-filter surveys. This dataset is useful for examining the photometric selection of target galaxies for BAO surveys because of the large cosmological volume covered and the large number of galaxies with detailed photometric information. We use the sample to derive the photometric and physical properties of emission line galaxies to assist planning for future spectroscopic BAO surveys. We also show some examples of photometric selection procedures which can efficiently select these emission line galaxies.

Motivation & Objective

• To estimate the Hα and [O II] emission line luminosity functions of high-redshift galaxies (z = 0.5–1.7) using photometric data.
• To assess the feasibility of photometric selection methods for identifying emission line galaxies as targets in future spectroscopic BAO surveys.
• To provide quantitative guidance on required survey depth and selection efficiency for 8–10 m telescopes with wide-field spectrographs.
• To calibrate photometric line luminosity estimates using low-redshift SDSS spectroscopic data.
• To derive physical properties (stellar mass, extinction, size, bias) of emission line galaxies for cosmological survey planning.

Proposed method

• Uses multi-wavelength photometry from the Subaru Deep Field (SDF) and Subaru XMM-Deep Field (SXDF) covering ~0.85 deg².
• Employs a joint photometric redshift and star-formation rate estimation technique based on SED fitting to derive line luminosities.
• Calibrates photometric line luminosities against spectroscopic measurements of local SDSS galaxies to ensure accuracy.
• Applies color-color selection criteria (Bi′z′ and Bz′K) to identify high-redshift emission line galaxies in photometric surveys.
• Derives luminosity functions in three redshift bins (0.5–1.0, 1.0–1.4, 1.4–1.7) using the calibrated sample.
• Estimates stellar mass, extinction, FWHM size, and linear bias from photometric and redshift estimates using structure formation theory.

Experimental results

Research questions

• RQ1 Can photometric redshift and SED-fitting techniques accurately estimate Hα and [O II] line luminosities at z > 0.5?
• RQ2 What is the number density and distribution of emission line galaxies suitable for BAO surveys in the redshift range z = 0.5–1.7?
• RQ3 How efficient are simple two-color photometric selections (Bi′z′ and Bz′K) in isolating high-redshift emission line galaxies?
• RQ4 What imaging depth (in B-band) is required to achieve sufficient target density for BAO surveys with exposure times <1 hour?
• RQ5 How do the physical properties (mass, size, extinction, bias) of emission line galaxies vary with redshift and luminosity?

Key findings

• The derived Hα and [O II] luminosity functions are in reasonable agreement with prior spectroscopic and narrow-band survey results, validating the photometric method.
• Sufficient numbers of emission line galaxies exist for BAO surveys in the redshift ranges 0.5 < z < 1.0 and 1.0 < z < 1.4, detectable with <1 hour exposures on 8–10 m telescopes.
• For 1.0 < z < 1.4, a B-band depth of ≈24 is sufficient to select target galaxies with 50–60% success rate using Bi′z′ or Bz′K color cuts.
• For 1.4 < z < 1.7, a deeper B-band depth of ≈25 is required to achieve comparable selection efficiency.
• Both Bi′z′ and Bz′K color selections perform comparably in terms of depth requirement and success rate, with no clear advantage for either.
• Two-color selection alone is insufficient for efficient targeting in the high-redshift bin 1.4 < z < 1.7, suggesting need for multi-color or N-dimensional photometric methods.