Astron. Astrophys. 333, 841-863 (1998)

3. Comments on individual objects

Here, we briefly summarize spectroscopic data previously obtained on our targets and indicate the main results provided by the new spectra.

Table 3. Absorption lines detected in the spectrum of PKS 0454+039

3.1. EX 0302-223

Metal-rich absorption systems have been mentioned for this object at (Mg II: Bergeron, unpublished) and (Fe II and Mg II: Petitjean & Bergeron 1990). The FOS spectrum (Fig. 1), clearly shows that the Ly line from the latter is damped and reveals several strong features from C II, C IV, N I, O I, Si II, Si III and Si IV. No less than four new metal systems are detected. In agreement with the prediction made in Paper I, relatively strong lines from Fe II and Mg II are seen at a redshift , similar to that of a bright spiral galaxy located  kpc from the QSO line of sight. Second, a strong C IV doublet together with Ly is present at . In addition to Ly , Ly and Ly the FOS spectrum shows weak features from C II, C III, Si II and Si III at ; this system induces a Lyman edge near 2120 Å which is clearly apparent in the IUE spectrum (Lanzetta et al. 1995). On the other hand, no convincing line is seen from the Mg II system (the line at 2371 Å could be Al II 1670 at , but this feature seems to be too strong given that we do not detect Fe II 1608). Finally, a narrow high ionization system with strong O VI lines is detected at . The HST data do not confirm the and 0.9874 candidate damped Ly lines proposed by Lanzetta et al. (1995) (the former turns out to be Ly at 1.3284).

Fig. 1. FOS spectrum of EX 0302-223 (G270H). Long tick marks correspond to absorption lines from the DLAS

3.2. PKS 0454+039

Before this study, two absorption systems were known in this object, at and 1.1537 (Burbidge et al. 1977, Caulet 1989; Steidel & Sargent 1992). Steidel et al. (1995) have shown that the first one is a low metallicity (about 1/10 solar) DLAS, as indeed suggested by the strength of Mg II and Fe II lines, while in the second system, the Ly line is not damped. More than twenty metal lines are seen at together with a Lyman edge near 1700 Å (Fig. 2). The Si II 1304 line is definitely blended with another (Ly -only) line since i) it appears too strong with respect to e.g. Si II 1260 and ii) the match in redshift is not satisfactory. In addition to several metal lines from Si II, Si III, Si IV, C II and C III, the system displays a beautiful set of 9 lines from the Lyman series ending with a partial Lyman edge; the presence of N V and O VI lines in this system remains uncertain. One additional motivation for observing this QSO is the presence of two intervening galaxies detected by Steidel et al. (1993). The closest one is a dwarf star-forming galaxy at from which we do detect Mg II absorption. The second galaxy is a more luminous one at and  kpc which could induce C IV absorption; no lines are seen from this galaxy but we detect a new C IV doublet at .

Fig. 2. Same as Fig. 1 for PKS 0454+039 (G190H and G270H spectra)

3.3. 3C 196

The 21 cm absorption system at is known to display very strong associated Fe II and Mg II lines as well as Mn II and Ca II (Foltz et al. 1988; Boissé & Boulade 1990). An additional system is present at . The G160L HST data presented by Cohen et al. (1996) show that, unfortunately, the latter produces a Lyman edge nearly coincident with the Ly line at , which renders the determination of very difficult. An additional source of uncertainty in the analysis is related to the possible contribution of scattered light which makes the zero of the intensity scale ill defined. Our new (post-COSTAR) spectrum, although of lower integration time, is of interest in this regard. In order to perform a quantitative comparison, we retrieved from the HST archive the spectrum obtained in 1992 and analyzed by Cohen et al. (1996). Using the few narrow lines detected at either , 0.437 or 0.871 (Fig. 3), we measure shifts of 14 and 6 Å in the wavelength scales of the late and new spectrum respectively (these values are not accurate but the relative shift of 8 Å is well constrained by the data). Once corrected, the two spectra appear in quite good agreement but a systematic difference is seen in the blue wing of the Ly emission line (Fig. 3). If not instrumental, such an effect could be due to variable Ly absorption from gas ejected by the QSO; the time elapsed between the two observations is 2.05 yr and during that interval the amount or ionization degree of the gas may have changed (see e.g. Schartel et al. 1997 for another example of variable absorption). An alternative possibility is intrinsic emission line variability. Variable BAL-type absorption is supported by two facts: i) the gas responsible for the narrow absorption (to which the higher velocity gas is probably associated) is known to cover only partially the broad line region (Cohen et al. 1996) and ii) only a small variation of of about  cm^-2 is needed to account for the strength of the effect (assuming optical thinness).

Fig. 3. G160L spectra of 3C 196 obtained in 1994 as part of our program (thick line) and in 1992 by Cohen and collaborators (thin line) after correction of the wavelength scale (see text). The main absorption features from the systems at (DLAS) and are indicated (thick and thin tick marks respectively). Note the difference between the two spectra in the blue wing of the Ly emission line

3.4. Q 1209+107

The G160L FOS spectrum (Fig. 4), although of relatively poor S/N ratio, provides significant information on the already known metal systems at , 0.6295 and 1.8434. First, these data confirm that the 0.6295 system is damped (see Sect.  4.1.4); additional narrow absorptions (Ly and N I 1200) as well as a Lyman break are detected at this redshift. Ly at 0.3930 is also present and there are a few other possible features at 1.8434 (O III 702 at 1995 Å and O III 833 at 2369 Å).

Fig. 4. G160L spectrum of Q1209+107

3.5. PKS 1229-021

This object has already been the subject of several detailed optical studies. In particular, the high resolution data published by Lanzetta & Bowen (1992) suggest a high metallicity for the intervening gas since strong Mn II lines are detected. We indeed find a large number of metal absorptions from the system at . In 1991, one of us observed this object with the IUE in order to detect the damped Ly line and determine the H I column density; surprisingly, near the expected wavelength of this feature, a cut-off was seen (this IUE spectrum is shown in the catalog of Lanzetta et al. 1993). Steidel et al. (1994a) detected a Mg II doublet at 0.7570 and proposed that this system is responsible for the observed Lyman edge. Our data indicate that this Mg II doublet belongs to an extensive metal line system with absorptions from Si II, Si III, Si IV, C IV, N V and O VI. We detect an additional C IV system at . Furthermore, a strong Ly line is seen at with several associated lines from the Lyman series; a careful examination of the optical data published by Steidel et al. (1994a) suggested the presence of weak associated Mg II lines near 5120 Å. Measurements performed on the spectrum that C. Steidel kindly communicated to us confirm that shallow Mg II absorption is indeed present (the 2796 Å line is seen at  Å and with  Å). On the basis of our higher resolution and better S/N ratio UV data, we find that the partial cut-off near 1670 Å is in fact due to this system. Fortunately, the damped Ly line at can nevertheless be seen, superimposed onto the attenuated continuum (Fig. 5).

Fig. 5. Same as Fig. 1 for PKS 1229-021 (G190H and G270H spectra)

3.6. 3C 286

We do not detect any additional narrow metal system in the new FOS spectrum (Fig. 6), which makes the line identification much easier for this QSO than for the other ones observed at high resolution. The profiles of the O VI, N V and to a lesser extent, of the Ly and C IV emission lines suggest the presence of broad absorption from highly ionized gas at . In particular, two sharp edges are seen near 1925 and 2310 Å which, when attributed to the O VI and N V doublets, correspond to about the same redshift, , that is an infall velocity of 2700 km s^-1 relative to the QSO. From the DLAS, we detect several new (mostly low-ionization) species in addition to those - Fe II, Mg II, Mg I, Zn II, Cr II and Ca II - already seen by Meyer & York (1992) and Cohen et al. (1994). Given the strength of the Ly line, the C IV doublet is remarkably weak; the second (1551 Å) doublet line can barely be seen. C II 1335 lies just at the red end of the Ly emission line and its equivalent width strongly depends on the adopted shape for the adjacent continuum ( Å in the normalized spectrum presented in Fig. 17). Weak N I 1200 and Si III 1206 absorption can also be seen in the blue wing of the damped Ly line but their significance is difficult to assess quantitatively. Doublets from O VI, N V and Si IV are not detected, with good upper limits.

Fig. 6. Same as Fig. 1 for 3C 286 (G190H and G270H spectra)

© European Southern Observatory (ESO) 1998

Online publication: April 28, 1998

helpdesk.link@springer.de