The ultraviolet absorption spectrum of the z=2.72 QSO HS 1700+6416. I. Results on heavy-element absorption systems.
Reimers, D.; Vogel, S.
Germany
Abstract
We present the analysis of ultraviolet and optical spectra of the bright, high-redshift quasar HS 1700+6416. Ultraviolet observations in the range from 1150 to 3280A were obtained with the Faint Object Spectrograph (FOS) onboard the Hubble Space Telescope (HST) at a resolution of R=1300. The identification of the numerous absorption lines provides evidence for 15 heavy-element absorption systems, among them 7 Lyman Limit systems (LLS) clearly visible from their Lyman edges in the HST data. The entire spectrum is a superposition of absorption lines by hydrogen, helium and heavy-elements in 7 LLS, at least 8 heavy-element absorption systems and the interstellar medium as well as Lyman lines arising in the more numerous Lyα clouds. Prominent absorption by several ionization stages of C, N and O is detected in almost all heavy-element absorption systems when the corresponding resonance lines fall in the observed wavelength range. The blending problem as a result of the high absorption line density and the low spectral resolution severely affects the quantitative analysis of the data. Defining a local continuum is nearly impossible, instead the global continuum shape was calculated considering the Lyman continuum absorption of the 7 LLS. In order to derive column densities a spectrum synthesis program has been developed. We emphasize that there are uncertainties in deriving the column densities, e.g. unrecognized blending with Lyα or metal lines as well as hidden velocity components. With these caveats in mind column densities measurements have been used to constrain photoionization models in order to derive physical parameters of the absorber systems. Applying the best fitting ionizing background obtained at z=2 (power law with α=-0.6) and a flux at the Lyman limit of logJ(ν_LL_)=-22 in erg/s/Hz/cm2/sr provided the following results: For four redshift systems (z=1.8465, 2.1678, 2.308, 2.433) there is strong evidence for an enhanced oxygence abundance [O/C]>=0.6 relative to carbon (-2.3<=[C/H]<=-1.4) and similarly for an enhanced [N/C]. Unrecognized line blending would have the effect of underestimating the relative oxygen abundance O/C since typically several oxygen ions with several lines (OIII-OVI) contribute with comparable fractions so that blends can be recognized and eliminated while carbon is typically represented by only one line. The N/C ratio is sensitive to line blending and our results need to be confirmed by data with higher resolution. Abundance ratios could be seriously in error if either the dominant carbon and oxygen ions are formed in different subclouds, not resolved with the FOS, or if the global b-values determined from spectral synthesis of the Lyman series lines with neutral hydrogen column densities from the Lyman edges are not representative for subclouds and saturation effects could lead to erroneous heavy ion column densities. Furthermore, the abundance ratios might be incorrect if the assumption of photoionization breaks down. On the basis of our analysis, all heavy-element absorption systems (with the exception of the system at z=1.1572) show enhanced oxygen and nitrogen abundances relative to carbon. No correlation of any of the physical parameters with redshift is obvious, but the relative oxygen overabundance [O/C] seem to increase with decreasing carbon abundance [C/H]. Applying the best fitting ionizing background obtained at z=2 to systems at low redshift is doubtful, but our data are insufficient to draw conclusions about the ionizing background at lower redshifts. In addition, the dominant ion species of oxygen and nitrogen are not observable for absorber clouds at low redshifts. More reliable results for heavy-element abundances in low redshift absorber clouds might be obtained by HS 1700 spectra of much better signal-to-noise as well as higher resolution in the wavelength range from 1150 A to 1600 A observable by the HST. Normalized and calculated spectra along with the line identifications are shown.