Lyman-alpha absorption in the spectrum of the Z = 4.5 QSO BR 1033-0327
We present the results of echelle spectroscopy of the Lyɑ forest region in the spectrum of the bright (m_R = 18.5) optically selected z = 4.5 QSO BR 1033-0327. These results, which extend Lyɑ forest absorption line studies up to a redshift of z= 4.5 at a resolution of 12 km s^(-1), represent the first high-resolution study of Lyɑ forest systems at such high redshifts. We compare the properties of the Lyɑ forest systems in the redshift range 3.7 < than z < than 4.3 with those inferred from similar observations at 1.86 < than z < than 3.27 and find some clear indications of redshift dependence in these. First, the ionizing flux from the quasar appears to affect the Lyɑ line density further from the QSO than at lower redshifts, implying that the Lyman limit background flux at redshift z ~ 4.2 is J_v ~ 1 to 3 x 10^(-22) ergs cm^(-2) s^(-1) sr^(-1). This is about a factor of 10 lower than estimates at z ~ 2.5. Second, the Doppler parameter distribution for systems where the effect of the QSO ionizing flux is small (3.7 < than z < than 4.3) has an excess at values ~ 20 km s^(-1), compared with lower redshifts; we suggest this may be due to the lower ionizing flux at z ~ 4.2, which would reduce the heating of the gas. Finally, while there is no evidence that the shape of the H I column density distribution changes significantly, the redshift dependence of the total number of systems is stronger than that found at lower redshifts, with the number of systems per unit redshift ~ (1+ z)^(4.6). Also, over the entire redshift range there is no significant correlation between Doppler parameter and H I column density and no signal in the two-point correlation function down to velocity separations of Δv ~ 100 km s^(-1). There is no evidence for continuously distributed HI absorption (the Gunn-Peterson effect) at these redshifts.
© 1994 American Astronomical Society. Received 1993 August 9; accepted 1993 December 16. We are grateful to the CTIO staff for assistance at the telescope, John Webb for development of the profile-fitting software, Michael Rauch for the computer program to examine the two-point correlation function, Gary Ferland for his photoionization code, Avery Meiksin, Jordi Miralda-Escude, Martin Rees, and Bob Williams for useful conversations, and a referee for very helpful detailed comments on the manuscript.
Published - 1994ApJ___428__574W.pdf