The 1- m discontinuity in the extragalactic background light spectrum: an artefact of foreground subtraction

Abstract

Several recent papers claim the detection of a near-infrared (NIR) extragalactic background light (EBL) intensity at 1.25-4 μm that exceeds the integrated light of galaxies by factors of >3. When combined with a claimed optical detection of the EBL at 0.80 μm the EBL excess emission spectrum has a discontinuity at ∼1 μm. This discontinuity has given rise to an interpretation in terms of ultraviolet radiation emanating from the first generation of massive stars at redshifts of 7-20 (so-called Population III stars). The interpretation of the NIR excess emission as being of extragalactic origin depends crucially on the model used in the subtraction of the zodiacal light (ZL), the dominant foreground contaminant. We estimate the ZL at 0.80 μm using on the one hand the measurement by Bernstein et al., with corrections for some omitted effects of atmospheric scattering and calibration, and on the other hand the model of Kelsall et al. There is in neither case any evidence for a step in the EBL at ∼1 μm. We emphasize that in order to avoid systematic effects it is essential to use the same ZL model for both the NIR (1.25-4 μm) and optical (0.80 μm) data. We emphasize, however, that our analysis does not allow a statement on the overall level of the NIR EBL. The contribution of the diffuse galactic light to the 'EBL excess' emission is estimated. It is found to be significant at 3-4 μm and should be carefully evaluated in future measurements which aim at detecting an EBL signal at the level of ∼10 nW m -2 sr -1 , that is, at the level of the integrated light of (known) galaxies.