The optical extragalactic background light: Revisions and further comments

We revise the measurements in our previous work of foreground zodiacal light (ZL) and diffuse Galactic light (DGL) that were used to measure the extragalactic background light ( EBL). These changes result in a decrease of 8 and an increase of 0: 3 in units of 10(-9) ergs s(-1) cm(-2) sr(-1) angstrom(-1) ("cgs'' units) in the ZL and DGL flux, respectively. We therefore obtain revised values for the EBL of 6 +/- 4, 10 +/- 5, and 7 +/- 4 cgs in the HST WFPC2 U ( F300W), V ( F555W), and I (F814W) bands, respectively, fromsources fainter than m(V) similar to 23 AB mag. The revisions are dominated by the details of the tropospheric scattering models used tomeasure the ZL. We discuss these results in the context of faint number counts and diffuse EBL measurements at other wavelengths. In particular, we note that unless the slope of the galaxy counts increases beyond m(V) similar to 30 AB mag, unresolved sources will contribute < 0.2 cgs, which is far below the uncertainties achievable for any diffuse EBL measurement in the foreseeable future. Therefore, the best constraints on faint sources come from the resolved sources themselves. As in our earlier work, models are still required to derive the bolometric EBL (0.1 - 1000 mu m) due to uncertainties in the mid-infrared; consequently, our previous discussions of the bolometric EBL are not affected by the revisions presented here. Finally, we discuss the nature of the extended point-spread function (PSF) of ground-based telescopes and its impact on surface brightness measurements. In particular, we show that the slope and amplitude of extended PSFs vary considerably between telescopes and with time. We find no conclusive, single cause of extended PSFs, although atmospheric scattering is ruled out.