Frontier Fields: Combining HST, VLT, and Spitzer data to explore the z ~ 8 Universe behind the lensing cluster MACSJ0416.1−2403

¹ Instituto de Astrofísica, Facultad de Física, Pontificia Universidad Católica de Chile, 306 Santiago 22, Chile
e-mail: nlaporte@astro.puc.cl
² Instituto de Astrofísica de Canarias (IAC), 38200, La Laguna, Tenerife, Spain
³ Departamento de Astrofísica, Universidad de La Laguna (ULL), 38205, La Laguna, Tenerife, Spain
⁴ IRAP, CNRS – 14 Avenue Edouard Belin – 31400 Toulouse, France
⁵ Instituto de Astronomía, Universidad Nacional Autónoma de México, Mexico
⁶ Millennium Institute of Astrophysics, Santiago, Chile
⁷ Space Science Institute, 4750 Walnut Street, Suite 205, Colorado, 80301 Boulder, USA
⁸ Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218, USA
⁹ Centro de Astro-Ingeniería, Pontificia Universidad Católica de Chile, Vicuna Mackenna 4860, Santiago, Chile

Received: 22 September 2014
Accepted: 2 December 2014

Abstract

Context. The Hubble Space Telescope (HST) Frontier Fields (HFFs) project started at the end of 2013 with the aim of providing extremely deep images of six massive galaxy clusters. One of the main goals of this program is to push several telescopes to their limits to provide the best current view of the earliest stages of the Universe. The analysis of the initial data has already demonstrated the huge capabilities of the program.

Aims. We present a detailed analysis of z ~ 8 objects behind the HFFs lensing cluster, MACSJ0416.1-2403, combining 0.3−1.6 μm imaging from HST, ground-based K_s imaging from VLT HAWK-I, and 3.6 μm and 4.5 μm Spitzer Space Telescope. The images probe to 5σ depths of ≈29 AB for HST, 25.6 AB for HAWK-I, and ≈0.310 and 0.391 μJy at 3.6 and 4.5 μm, respectively. With these datasets, we assess the photometric properties of z ~ 8 galaxies in this field, as well as their distribution in luminosity, to unprecedented sensitivity.

Methods. We applied the classical Lyman break (LB) technique, which combines non detection criteria in bands blueward of the Lyman break at z ~ 8 and color-selection in bands redward of the break. To avoid contamination by mid-z interlopers, we required a strong break between optical and near-infrared data. We determined the photometric properties of z ~ 8 selected candidates using spectral energy distribution (SED)-fitting with standard library templates. The luminosity function at z ~ 8 is computed using a Monte-Carlo method taking advantage of the SED-fitting results. A piece of cautionary information is gleaned from new deep optical photometry of a previously identified z ~ 8 galaxy in this cluster, which is now firmly detected as a mid-z interloper with a strong ≈1.5 mag Balmer break (between F606W and F125W). Using the SED of this interloper, we estimated the contamination rate of our MACSJ0416.1−2403 sample, and that of previous samples in Abell 2744 that were based on HFF data, we highlight the dangers of pushing the LB technique too close to the photometry limits.

Results. Our selection reliably recovers four objects with m_F160W ranging from 26.0 to 27.9 AB that are located in modest-amplification regions (μ < 2.4). Two of the objects display a secondary break between the IRAC 3.6 μm and 4.5 μm bands, which could be associated to the Balmer break or emission lines at z ~ 8. The SED-fitting analysis suggests that all of these objects favor high-z solutions with no reliable secondary solutions. The candidates generally have star formation rates around ~10 M_⊙/yr and sizes ranging from 0.2 to 0.5 kpc, which agrees well with previous observations and expectations for objects in the early Universe. The sample size and luminosity distribution are consistent with previous findings.