next previous
Up: High matter density peaks abundances


Subsections

4 Coincidences of high matter density peaks

In this section, we describe the observed pairs of quasars and list the absorption systems with N(H I $) \raisebox{-5pt}{$\;\stackrel{\textstyle >}{\sim}\;$ }2\times
10^{17}$ cm-2 found in the 7 spectra. For each of the systems, we search the adjacent line of sight for the presence of any absorption at the same redshift. When a LLS or a C IV system with rest equivalent width $W_{\rm r} > 0.5$ Å is seen along the second LOS within $\sim $1000 km s-1 from the former LLS, we call this a coincidence. The observed number densities of LLS and C IV absorption systems with $W_{\rm r} > 0.5$ Å are similar at the same redshift (e.g. Steidel et al. 1988; Steidel 1990). We can therefore assume that they trace the same kind of overdensity.

The numbers associated with the coincidences correspond to those in Table 4 and Fig. 15.

4.1 The QSO pair UM680 and UM681

These two QSOs (also called Q0307-195A,B) are separated by 56 arcseconds on the plane of the sky, corresponding to $\sim $ $830{-}940\ h^{-1}$ kpc in the considered redshift interval. Spectra at low and intermediate resolution of this pair have been used in the past to study the correlation of C IV and Lyman-$\alpha $ forest lines (Shaver & Robertson 1983; D'Odorico et al. 1998).

1) $z \simeq 1.788$ - there is no metal system along the LOS of UM680 corresponding to the sub-DLAS at $z_{\rm a}\simeq 1.788$observed in the spectrum of UM681 (see Sect. 3.1). A weak Lyman-$\alpha $ absorption, $\log
N($H I $) \simeq 13.81 \pm 0.05$, is observed at $z_{\rm a} \simeq 1.7876$ (see Fig. 1). From the observed number density of H I Lyman-$\alpha $ absorption lines with column density in the range $13.1 \le \log N($H I$) \le 14$(Kim et al. 2001), the probability for such an absorption to fall in a velocity bin of 200 km s-1 at this redshift is ${\cal P} \sim 0.3{-}0.4$.

2) $z \simeq 2.03$ - the coincident systems observed at $z_{\rm a} \simeq
2.03520$ and $z_{\rm a} \simeq 2.03215$ in the spectra of UM680 and UM681 respectively, are two candidate LLSs which show absorption lines due to the same ionic transitions with a shift of $\simeq $300 km s-1 (see Fig. 5). We detect low ionisation absorption lines due to Al II, Si II and Fe II, together with Al III, Si III and Fe III, the latter only in UM680. The corresponding Si IV and C IV absorption doublets are outside our wavelength coverage, but their presence is discussed in Shaver & Robertson (1983).

It is not possible to constrain the value of the H I column density of both systems due to the complexity of the profile. The Lyman-$\beta$ lines are in a region of the spectrum with low signal-to-noise ratio and probably blended. From the equivalent width ratio of Si II and Fe II to C IV (as measured by Shaver & Robertson 1983) we derive that the systems are likely in a low excitation state and have N(H I ) > 1018 cm-2 (see Bergeron & Stasinska 1986).

  \begin{figure}
\par\includegraphics[width=8.8cm,height=8cm,clip]{MS2352f5.eps}\end{figure} Figure 5: $z \sim 2.03$ - coincident absorption systems in the spectra of UM680 (top panel) and UM681 (bottom panel). The H I Lyman-$\alpha $ transitions are shown with superposed the corresponding Al II $\lambda~1670$ one. The transverse spatial separation between the two LOSs at this redshift is $\sim $ $924\ h^{-1}$ kpc. The dotted lines mark the position of the metal lines at $z_{\rm a} = 2.03215$ and $z_{\rm a} = 2.03520$ (origin of the velocity axes).


  \begin{figure}
\par\includegraphics[width=8.5cm,height=9.6cm,clip]{MS2352f6a.eps...
...{5mm}
\includegraphics[width=8.5cm,height=9.6cm,clip]{MS2352f6b.eps}\end{figure} Figure 6: Left: ionic transitions observed at $z_{\rm a} = 2.12209$ (origin of the velocity axes) in the spectrum of UM681. This redshift is obtained from the fitting of the low ionisation lines; the center of the velocity profile of the high ionisation lines, determined by the N V $\lambda~1238$ Å transition, is shifted by $\simeq $-8.5 km s-1. Right: ionic transitions observed at $z_{\rm a} = 2.12312$ ( $v \simeq 97$ km s-1) in the spectrum of UM680. The origin of the velocity axes is kept at $z_{\rm a} = 2.12209$, redshift of the low ionisation transitions observed in the coincident system in the spectrum of UM681. The transverse separation between the two LOSs at this redshift is $\sim $ $940\ h^{-1}$ kpc.

3) $z \simeq 2.122$ - the QSO UM681 presents a metal system at its emission redshift ( $z_{\rm a} \simeq 2.12209$) with lines due to C IV, N V, O VI and S IV and also weak low ionisation lines (see Fig. 6). This system, although characterized by highly ionised transitions, has a velocity spread of less than $\sim $250 km s-1 and does not show any evidence of partial coverage. Furthermore, the presence of singly ionised absorption lines and the symmetric velocity profile favour an absorber with a dense core. Therefore, although the system is located in the vicinity of the quasar it is probably not associated with it.

In addition, there is a very similar absorption system along the LOS of UM680, at $z_{\rm a} \simeq 2.12312$, corresponding to a velocity shift of $\sim $100 km s-1 (see Fig. 6). The transverse spatial separation between the two LOSs at this redshift is $\sim $ $940\ h^{-1}$ kpc. The latter system is located at $\sim $2000 km s-1 from the emission redshift of UM680; the same arguments as before are valid to reject the hypothesis that this is due to gas associated with either of the two quasars. The observed H I Lyman-$\alpha $ and Lyman-$\beta$ absorption lines for this system are consistently fitted with a main component of column density $\log
N($H I $) \raisebox{-5pt}{$\;\stackrel{\textstyle >}{\sim}\;$ }
17.3$.

Figure 7 shows the H I Lyman-$\alpha $ emission region in the two QSO spectra. The coinciding Lyman-$\alpha $ absorptions at $z_{\rm a} \sim
2.122$ are shown, together with the associated N V $\lambda~1238$ lines (the N V $\lambda~1242$transitions fall outside the observed wavelength range). Another pair of Lyman-$\alpha $ absorptions is observed at $z_{\rm
a} \sim 2.099$, which shows an associated N V doublet in the spectrum of UM680, while does not have any detected associated metal line in UM681.

Shaver & Robertson (1983) suggest the existence of a uniform, 1 Mpc diameter, gaseous disk associated with UM681 to explain the coincidence at $z_{\rm a} \sim
2.122$. The presence of a further coincidence at $\sim $2000 km s-1 from this one, favours the thesis that the absorptions are due to a coherent gaseous structure embedding both quasars and possibly small galactic objects. Deep imaging of the field could possibly shed light on the nature of the absorbers and of the ionising processes at work in the gas.

4.2 The QSO pair Q2344+1228 and Q2343+1232

The first spectra of this QSO pair were presented by Sargent et al. (1988), the two objects are separated by 5 arcmin on the plane of the sky, corresponding to a transverse spatial separation of $\simeq $$5\ h^{-1}$ Mpc in the considered redshift range. The remarkable feature is the presence of a DLAS in each of the LOS (see Sect. 3).

  \begin{figure}
\par\includegraphics[width=8.5cm,height=7.7cm,clip]{MS2352f7.eps} %
\end{figure} Figure 7: Lyman-$\alpha $ emission region in the spectra of the QSOs UM680 (top panel) and UM681 (bottom panel). Marked are the two coinciding H I Lyman-$\alpha $ lines at $z_{\rm a} \simeq 2.099$ and 2.122 and other interesting metal absorption lines (see text). The two LOSs are separated by $\sim $ 940 h-1 kpc at z = 2.122.


  \begin{figure}
\par\includegraphics[width=8.5cm,height=7.7cm,clip]{MS2352f8.eps} %
\end{figure} Figure 8: H I Lyman-$\alpha $ emission regions in the spectra of Q2344+1228 (top panel) and Q2343+1232 (bottom panel). The transverse spatial separation between the two LOSs in this region is $\sim $$5\ h^{-1}$ Mpc. The dashed vertical lines mark: in the top panel, the position of the H I Lyman-$\alpha $ emission at $z_{\rm e}=2.773$; in the bottom panel, the labeled emission lines at $z_{\rm e}=2.549$, with the shifted rest wavelengths by Tytler & Fan (1992).

The emission redshift of Q2343+1232 reported by Lu et al. (1998), $z_{\rm e} \simeq 2.549$, is consistent with the position of the emission lines observed in the Sargent et al. (1988) spectrum (Si IV+O IV] and C IV) and with the O I emission in our spectrum (marked in Fig. 8), when the shifted rest wavelengths computed by Tytler & Fan (1992) are used. Likely, the peak observed at $\lambda \sim 4375$ Å is partly due to the N V emission, while the maximum of the Lyman-$\alpha $ emission is strongly absorbed. We identify two absorption systems at $z_{\rm a} > z_{\rm
e}$: a N V doublet and the corresponding Lyman-$\alpha $ absorption at $z_{\rm a} \simeq 2.5698$ ( $\Delta v \simeq
1750$ km s-1), together with another possible Lyman-$\alpha $ line at $z_{\rm a}
\simeq 2.579$ ( $\Delta v \simeq 2500$ km s-1). They do not show any signature of partial coverage and they could be explained by the presence of a cluster of galaxies of which the QSO itself is a member (e.g. Weymann et al. 1979).

  \begin{figure}
\par\includegraphics[width=8.8cm,height=8cm,clip]{MS2352f9.eps} %
\end{figure} Figure 9: $z \sim 2.171$ - coincidence between the LLS at $z_{\rm a} \simeq 2.17115$ (origin of the velocity axes) in the spectrum of Q2343+1232 (bottom panel) and a H I Lyman-$\alpha $ absorption without associated metals in the spectrum of Q2344+1228 (top panel). Overplotted on the Lyman-$\alpha $ absorptions are the corresponding C IV $\lambda~1548$ and $\lambda~1550$ spectral regions showing a detectable absorption only in Q2344+1232. The two LOSs are separated by $5\ h^{-1}$ Mpc.

4) $z \simeq 2.171$ - in the spectrum of Q2343+1232, we identify a metal system at $z_{\rm a} \simeq 2.171$ which could be a LLS on the ground of the column density ratios of the observed transitions (Bergeron & Stasinska 1986). In particular, O I/C IV $\sim
0.6$, Mg II/C IV $\sim 0.4$ and Si II/Si IV $\sim 3.7$. No metal system is detected along the companion LOS within $\sim $3000 km s-1. However, a complex H I Lyman-$\alpha $ absorption is present (see Fig. 9) whose velocity profile appears to match that of the Lyman-$\alpha $ in Q2343+1232 when shifted red-ward by $\sim $240 km s-1.
  \begin{figure}
\par\includegraphics[width=8.8cm,height=8cm,clip]{MS2352f10.eps}\end{figure} Figure 10: $z \sim 2.43$ - coincident absorption systems in the spectra of Q2344+1228 (top panel) and Q2343+1232 (bottom panel). The two LOS are separated by $\sim $ $5.3\ h^{-1}$ Mpc. The H I Lyman-$\alpha $ transitions are shown with superposed: the corresponding Si IV doublet at $z_{\rm a} =2.4271$ (top) and the Si II $\lambda~1260$ plus the Si IV doublet shifted upward by 0.2 for clearness (bottom). The dotted vertical lines mark from left to right: the position of the weak metal complex in Q2343+1232 at $z_{\rm a} = 2.42536$, the main component of the metal absorption in Q2344+1228 and the main component of the low ionisation metal lines in Q2343+1232 at $z_{\rm a} = 2.43125$ (origin of the velocity axes).

5) $z \simeq 2.43$ - in the spectrum of Q2343+1232, the DLAS at $z_{\rm a}
\simeq 2.43$ (see Sect. 4.2) coincides with a metal system at $z_{\rm a} \simeq 2.4271$ (redshift of the Si IV main component) in the companion LOS, showing only high ionisation lines (C IV was detected by Sargent et al. 1988) and a strongly saturated Lyman-$\alpha $ ( $W_{\rm r}
\simeq 2$ Å) (see Fig. 10). The H I absorption is likely not a LLS since singly ionised lines are not detected (like Mg II and Fe II). An acceptable fit of the profile is obtained with two main components at N(H I $) \sim
10^{15}$ and $\sim $ $6.3\times10^{15}$ cm-2, which however should be considered as lower limits.

6) $z \simeq 2.54$ - the DLAS at $z_{\rm a} \simeq 2.53788$ in the spectrum of Q2344+1228 (see Sect. 4.1) does not have a corresponding metal system on the LOS to Q2343+1232, but it is indeed at $\sim $940 km s-1 from the H I Lyman-$\alpha $ emission at the redshift of this quasar (see Fig. 8).

4.3 The QSO triplet Q2138-4427, Q2139-4433 and Q2139-4434

The quasar Q2139-4434 ( $z_{\rm e} = 3.23$) was observed at intermediate resolution together with its companion Q2138-4427 ( $z_{\rm e} = 3.17$) by Francis & Hewett (1993). They are separated by 8 arcmin on the plane of the sky. Francis and Hewett observed common strong Lyman-$\alpha $ absorptions at $z \sim 2.38$ and $z \sim 2.85$ and further imaging of the field revealed the presence of a cluster of galaxies at $z \sim 2.38$ (Francis et al. 1996, 1997, 2001a). Wolfe et al. (1995) confirmed the damped nature of the system at $z \sim 2.85$ in the spectrum of Q2138-4427. We obtained high resolution spectra of Q2138-4427, Q2139-4434 and of Q2139-4433 ( $z_{\rm e} =
3.220$, R = 19.97; Hawkins & Véron 1996). The latter two QSOs are separated by 1 arcmin on the plane of the sky.

  \begin{figure}
\par\includegraphics[width=8.8cm,height=6cm,clip]{MS2352f11.eps}\end{figure} Figure 11: Q2138-4427: H I Lyman-$\alpha $ absorption line at $z_{\rm a} = 2.38279$.


  \begin{figure}
\par\includegraphics[width=8.8cm,height=9cm,clip]{MS2352f12.eps}\end{figure} Figure 12: $z \sim 2.38$ - coincident absorption systems in the spectra of Q2139-4434 (mid panels) and Q2138-4427 (bottom panels), with a transverse separation of $\sim $$9\ h^{-1}$ Mpc. The left panels show the Si II $\lambda~1304$ transition and the right panels show the Fe II $\lambda~2382$ one. In the top row, the corresponding regions in the spectrum of Q2139-4433 are plotted. At this redshift, the separation between the LOSs to Q2139-4433 and Q2139-4434 is $\sim $$1\ h^{-1}$ Mpc and between Q2139-4433 and Q2138-4427 is $\sim $ $7.7\ h^{-1}$ Mpc. The origin of the velocity axes is set at $z_{\rm a} = 2.38279$. The other vertical dotted line marks the position of the system in Q2139-4434 at $z_{\rm a} = 2.37977$.

7) $z\simeq 2.38$ - the strong Lyman-$\alpha $ absorption at $z_{\rm a} \simeq
2.38$ in the spectrum of Q2138-4427 has at least one visible damped wing in the velocity profile (see Fig. 11) implying a column density N(H I $)
\raisebox{-5pt}{$\;\stackrel{\textstyle >}{\sim}\;$ }10^{19}$ cm-2. Unfortunately, the spectra of Q2139-4433 and Q2139-4434 do not cover the wavelength region where the corresponding H I Lyman-$\alpha $ lines should fall, while the spectrum of Q2138-4427 does not cover that of the C IV doublet at this redshift. In the low resolution spectrum of Q2139-4434 by Francis & Hewett (1993), an absorption line with equivalent width $\sim $20 Å is present at this redshift, which would correspond to a Lyman-$\alpha $ line with N(H I $) \sim
7\times 10^{19}$ cm-2. We do not detect C IV absorption at this redshift in the spectra of Q2139-4433 and Q2139-4434 but we identify neutral and singly ionised transition lines (C II, O I, Si II and Fe II) with a simple two-component velocity profile in Q2139-4434. Figure 12 shows two coincident transitions in Q2138-4427 and Q2139-4434, they have a minimal velocity separation of around 150 km s-1, while the two LOSs are at a transverse separation of $\sim $$9\ h^{-1}$ Mpc.

  \begin{figure}
\par\includegraphics[width=8.8cm,height=9cm,clip]{MS2352f13.eps}\end{figure} Figure 13: $z \sim 2.73$ - LLS at $z_{\rm a} \simeq 2.73557$ (origin of the velocity axes) in the spectrum of Q2139-4434 (middle panel). The H I Lyman-$\alpha $ absorption line is shown with superposed the lines of the C IV doublet. The other panels plot the corresponding H I Lyman-$\alpha $ line in the spectrum of Q2139-4433 (top) and of Q2138-4427 (bottom). No metal lines are associated to the latter two hydrogen absorptions. At this redshift, the transverse spatial separation between the LOSs to Q2139-4434 and Q2139-4433 is $\sim $$1\ h^{-1}$ Mpc, between Q2139-4434 and Q2138-4427 is $\sim $$9\ h^{-1}$ Mpc and between Q2139-4433 and Q2138-4427 is $\sim $$8\ h^{-1}$ Mpc.

8) $z \simeq 2.73$ - the system at $z_{\rm a} \simeq 2.73557$ in the spectrum of Q2139-4434 is again a candidate LLS on the ground of the observed ionic transitions. No metal lines are detected within $\sim $3000 km s-1 of this absorption redshift along the LOS of Q2139-4433 and of Q2138-4427. On the other hand, the velocity profile of the observed H I Lyman-$\alpha $ absorptions follows that of the C IV absorption associated to the LLS (see Fig. 13). Unfortunately, we cannot disentagle the velocity structure of the LLS Lyman-$\alpha $ absorption since our spectrum does not extend to the region where the higher lines in the Lyman series are located.

  \begin{figure}
\par\includegraphics[width=8.8cm,height=9cm,clip]{MS2352f14.eps}\end{figure} Figure 14: $z \sim 2.85$ - coincident absorption systems in the spectra of Q2139-4433 (top panel), Q2139-4434 (middle panel) and Q2138-4427 (bottom panel). The H I Lyman-$\alpha $ transitions are shown in the top and mid row and the H I Lyman-$\gamma $ corresponding to the DLAS in the spectrum of Q2138-4427 is in the bottom row. Overplotted are the corresponding C IV doublets, no metal absorption is observed in the spectrum of Q2139-4434. The dotted vertical lines mark the position of the main components in C IV at $z_{\rm a} = 2.85153$ (origin of velocity axes) and at $z_{\rm a} = 2.85262$. The transverse spatial separations between the three LOSs are about the same as those reported in the caption of Fig. 13.

9) $z \simeq 2.85$ - the DLAS at $z_{\rm a} \simeq 2.85$ in the spectrum of Q2138-4427 coincides with a complex H I Lyman-$\alpha $ absorption in the spectrum of Q2139-4434, with no detectable associated metal transitions. On the other hand, we identify a saturated Lyman-$\alpha $ absorption ( $W_{\rm r} \simeq 1$ Å) and a C IV doublet at $z_{\rm a} \simeq 2.85262$ in the spectrum of Q2139-4433 (see Fig. 14), partially superposing in redshift upon the C IV absorption associated to the DLAS. The transverse spatial separation between the two LOSs at this redshift is $\sim $$9\ h^{-1}$ Mpc.

This correlation could be interpreted as due to a gaseous structure perpendicular to the LOSs and extending over several Mpc in the direction defined by the three quasars.

 

 
Table 4: Summary of observed coincidences.
Objects Ident. Redshifta $\Delta s^{b}$ $\Delta v_{\rm min}^{c}$ log N(H I) $W_{\rm r}(\lambda1548)$ log N(Fe II)
      (h-1 Mpc) (km s-1)   Å  
  1 1.7874 0.87   13.8 out <11.8
    1.78865   >3000 19.0 out 14.5
UM680 2 2.0352 0.92 300 >18 0.4d 12.8
UM681   2.03215     >18 0.7d 13.4
  3 2.12312 0.94 100 (Si II) >17.3 0.5 <12.7
    2.12209     >17.3 0.44 <12.6
  4 2.17115 5 >3000 >17.3 0.34 13.1
    2.167       <0.01 <11.8
Q2343+1232 5 2.43125 5.3 110 (Si IV) 20.35 1.1e 14.7
Q2344+1228   2.4271     >15.9 0.7e <12.5
  6 2.549f 5.3        
    2.53788     20.4 out 14.1
  7 out 1     <0.014 <12.3
    2.37977 9 150 (Fe II) 20 <0.008 13.4
    2.38279 8   >19 out e.w. 1.2h
Q2139-4433 8 2.73258     16.5 <0.03 <12.9
Q2139-4434   2.73557   >3000 >17.3 0.6 13
Q2138-4427g   2.7323       <0.005 <11.9
  9 2.85262   0 (C IV)   0.5 <13
    2.85378     14.8 <0.007 <12.6
    2.85153     20.9 0.8 e.w. 0.2h
a The reported redshifts correspond to the main component of the associated metal absorption, if present; or to the strongest H I Lyman-$\alpha $ absorption closer to the redshift of the high density system.
b Transverse spatial separation between the lines of sight; in the case of the triplet it refers to the distance to the following object in the list.
c Minimal velocity separation between metal absorption lines of the same ionic species in the coupled lines of sight.
d C IV $\lambda\ 1548$ rest equivalent width from Shaver & Robertson (1983).
e C IV $\lambda\ 1548$ rest equivalent width from Sargent (1987).
f Emission redshift of the paired QSO.
g Precise column density determination for the metal lines in the spectrum of Q2138-4427 will be reported by Ledoux et al. (in preparation).
h Rest equivalent width in Å.



next previous
Up: High matter density peaks abundances

Copyright ESO 2002