[Previous] | [Session 29] | [Next]
J. Scott, J. Bechtold, M. Steinmetz (Steward Observatory, University of Arizona), A. Dobrzycki (Harvard-Smithsonian Center for Astrophysics), V. P. Kulkarni (University of South Carolina), M. Morita (Steward Observatory, University of Arizona)
The proximity effect is the deficit of Lyman \alpha absorption lines in a QSO spectrum near the QSO Lyman \alpha emission line. This is thought to be caused by enhanced photoionization of the Lyman \alpha forest absorbers in the immediate vicinity of the QSO due to the flux of ultraviolet photons from the QSO itself. At large distances from the QSO, the ionization state of the absorbers is mitigated only by the ambient extragalactic ultraviolet radiation field. Thus, the proximity effect has been used as a tool for quantifying the intensity of this UV background at the Lyman limit, J(\nu0).
In Paper I of our series ``A Uniform Analysis of the Ly\alpha forest at z=0 - 5'', we published 39 new moderate-resolution QSO spectra from the Multiple Mirror Telescope of objects at 1.9 < z < 3.6. We supplemented these with 60 QSOs from the literature and analyzed the statistics of the Lyman \alpha forest. In Paper II, we presented measurements of J(\nu0) from the proximity effect analysis based on those data.
In Paper III of this series, we present a set of 270 quasar spectra from the archives of the Faint Object Spectrograph on the Hubble Space Telescope. A total of 151 of these spectra, yielding 906 lines, are suitable for using the proximity effect signature to measure J(\nu0), the mean intensity of the hydrogen-ionizing background radiation field at low redshift. In Paper IV of the series, we will investigate clustering of the low redshift absorbers, and in Paper V, we will present the measurement of the low redshift UV background from these HST/FOS data.
We find evidence for evolution in the background from z~2 to z~0. Using a maximum likelihood technique and the best estimates of each QSO's Lyman limit flux and systemic redshift, we measure J(\nu0)= 7.0+3.4-4.4 \times 10-22 ergs/s/cm2/Hz/sr at z~2 from the ground-based spectra, and J(\nu0)= 1.0+3.8-0.2 \times 10-22 ergs/s/cm2/Hz/sr at 1 < z < 1.67 and J(\nu0)= 6.5+38.-1.6 \times 10-23 ergs/s/cm2/Hz/sr at 0.03 < z < 1 from the HST/FOS spectra. Relaxing the assumption that the spectral shapes of the sample spectra and the background are identical, the best fit HI photoionization rates at these redshifts are found to be 1.9+1.2-1.0 \times 10-12 s-1, 1.3+5.9-0.9 \times 10-12 s-1, and 1.9+8.9-1.3 \times 10-13 s-1.
This large sample of 250 QSO spectra and over 3000 Lyman alpha absorption lines from z~0 to z~4, has allowed us to execute the most systematic analysis of the evolution of the UV background to date. Within the uncertainties, our measurements are consistent with the background predicted by models incorporating the known population of QSOs only. We discuss a theoretical model for the proximity effect that can be used to understand the effect of various systematic uncertainties not treated in the standard proximity effect analysis procedure.
The author(s) of this abstract have provided an email address for comments about the abstract: jscott@as.arizona.edu