next previous
Up: The absorption spectrum

8. Conclusions

We have presented high-resolution spectra of the quasar PKS 2126-158.

The analysis of the Lytex2html_wrap_inline1969 forest shows that:

  1. The peak value of the Doppler parameter distribution is tex2html_wrap_inline2779 km s-1. This value is large enough to be in agreement with a standard ionization model, but not too high (tex2html_wrap_inline2783 km s-1) to make the Lytex2html_wrap_inline1969 clouds contribution to baryonic matter exceed the standard nucleosynthesis value (Hu et al. 1995).
  2. The plot of the statistical sample of Lytex2html_wrap_inline1969 lines in the tex2html_wrap_inline2265 plane shows no significant correlation between the two parameters.
  3. The column density distribution shows a cutoff, due to incompleteness and blending, for tex2html_wrap_inline2387, for higher values it is well described by a power-law tex2html_wrap_inline2795 with tex2html_wrap_inline2383 (Hu et al. 1995; Giallongo et al. 1996).
  4. The number density of lines per unit redshift is in agreement with a standard power-law evolution of the type tex2html_wrap_inline2799 with tex2html_wrap_inline2801 (Giallongo et al. 1996).

Clustering properties of the Lytex2html_wrap_inline1969 lines have been investigated in another paper (Cristiani et al. 1997). Lytex2html_wrap_inline1969 of higher column density (tex2html_wrap_inline2807) are found to cluster significantly on scales around 100 km s-1.

From the analysis of metal absorptions the following results have been obtained:

  1. Two new C IV absorption systems have been detected at z=3.2165 and z=2.5537.
  2. A mean metallicity of tex2html_wrap_inline2815 dex solar has been found using the metal systems at z=2.8195, z=2.9072 and z=2.9675. In order to make the column densities of the intervening systems compatible with realistic assumptions about the cloud sizes and the [Si/C] ratios, it is necessary to assume an increase of a factor 10 in the jump at 4 Ryd of the standard spectrum of the ionizing UV background (Haardt & Madau 1997).
  3. Merging the present data with those obtained at comparable resolution for the two quasars Q0055-26 and Q0000-26, has provided a relative large sample (71 doublets) of C IV absorptions, complete down to a column density tex2html_wrap_inline2827. The C IV column density distribution is well fitted by a single power-law, with tex2html_wrap_inline1981.
  4. The mode of the Doppler parameter distribution is tex2html_wrap_inline1983 km -1.
  5. The clustering properties of the individual components of the C IV features have been investigated making use of a TPCF in the velocity space. A significant signal is obtained for scales smaller than 200-300 km s-1, tex2html_wrap_inline2839. The result is consistent with previous findings (Sargent et al. 1988; Heisler et al. 1989) when the increased resolution and s/n ratio of the present data is taken into account.
  6. Correlations between total equivalent widths and number of components and between total equivalent widths and velocity spread of the individual C IV systems are observed. The two-point correlation functions for lower column density C IV absorption systems, recently estimated by Womble et al. (1997) and Songaila & Cowie (1996), show a weaker signal than the present data. These three pieces of evidence suggest a trend of decreasing clustering amplitude with decreasing column density. An analogous behaviour has been observed for Lytex2html_wrap_inline1969 lines by Cristiani et al. (1997).

Acknowledgements

We thank P. Madau, C. Porciani and S. Savaglio for enlightening discussions. SC acknowledges the support of the ASI contract 95-RS-38 and of the TMR-network of the European Community "Galaxy Formation and Evolution''.


next previous
Up: The absorption spectrum

Copyright by the European Southern Observatory (ESO)
web@ed-phys.fr