DEIMOS spectroscopy of z = 6 protocluster candidate in COSMOS – a massive protocluster embedded in a large-scale structure?

Abstract

ABSTRACT We present the results of our Keck/DEep Imaging Multi-Object Spectrograph spectroscopic follow-up of candidate galaxies of i-band-dropout protocluster candidate galaxies at z ∼ 6 in the COSMOS field. We securely detect Lyman α emission lines in 14 of the 30 objects targeted, 10 of them being at z = 6 with a signal-to-noise ratio of 5–20; the remaining galaxies are either non-detections or interlopers with redshift too different from z = 6 to be part of the protocluster. The 10 galaxies at z ≈ 6 make the protocluster one of the riches at z > 5. The emission lines exhibit asymmetric profiles with high skewness values ranging from 2.87 to 31.75, with a median of 7.37. This asymmetry is consistent with them being Ly α, resulting in a redshift range of z = 5.85–6.08. Using the spectroscopic redshifts, we recalculate the overdensity map for the COSMOS field and find the galaxies to be in a significant overdensity at the 4σ level, with a peak overdensity of δ = 11.8 (compared to the previous value of δ = 9.2). The protocluster galaxies have stellar masses derived from Bagpipes spectral energy distribution fits of $10^{8.29}\!-\!10^{10.28} \rm \, M_{\rm \odot }$ and star formation rates of $2\!-\!39\, \rm M_{\rm \odot }\rm \, yr^{-1}$, placing them on the main sequence at this epoch. Using a stellar-to-halo-mass relationship, we estimate the dark matter halo mass of the most massive halo in the protocluster to be $\sim 10^{12}\rm M_{\rm \odot }$. By comparison with halo mass evolution tracks from simulations, the protocluster is expected to evolve into a Virgo- or Coma-like cluster in the present day.