Abstract
Lyman break galaxies (LBGs) at z ~ 3-4 are targeted to measure the fraction
of Lyman continuum (LyC) flux that escapes from high redshift galaxies.
However, z ~ 3-4 LBGs are identified using the Lyman break technique which
preferentially selects galaxies with little or no LyC. We re-examine the
standard LBG selection criteria by performing spectrophotometry on composite
spectra constructed from 794 U_nGR-selected z ~ 3 LBGs from the literature
while adding LyC flux of varying strengths. The modified composite spectra
accurately predict the range of redshifts, properties, and LyC flux of LBGs in
the literature that have spectroscopic LyC measurements while predicting the
existence of a significant fraction of galaxies outside the standard selection
region. These galaxies, termed Lyman continuum galaxies (LCGs), are expected to
have high levels of LyC flux and are estimated to have a number density ~30-50
percent that of the LBG population. We define R_obs(U_n) as the relative
fraction of observed LyC flux, integrated from 912A to the shortest restframe
wavelength probed by the U_n filter, to the observed non-ionising flux (here
measured at 1500A). We use the 794 spectra as a statistical sample for the full
z ~ 3 LBG population, and find R_obs(U_n) = 5.0 +1.0/-0.4 (4.1 +0.5/-0.3)
percent, which corresponds to an intrinsic LyC escape fraction of f_esc = 10.5
+2.0/-0.8 (8.6 +1.0/-0.6) percent (contamination corrected). From the composite
spectral distributions we estimate R_obs(U_n) ~16 +/-3, f_esc ~33 +/-7 percent
for LCGs and R_obs(U_n) ~8 +/-3, f_esc ~16 +/-4 percent for the combined LBG +
LCG z ~ 3 sample. All values are measured in apertures defined by the UV
continuum and do not include extended and/or offset LyC flux. A complete high
redshift galaxy census and total emergent LyC flux is essential to quantify the
contribution of galaxies to the epoch of reionisation.
Users
Please
log in to take part in the discussion (add own reviews or comments).