2 Observations and data reduction

In September 2000, we obtained high resolution spectra of three QSO pairs with the UV and Visual Echelle Spectrograph (UVES, Dekker et al. 2000) mounted on the Kueyen telescope of the ESO VLT (Cerro Paranal, Chile). The journal of observations is reported in Table 1.

 

 

Table 1: Journal of observations September 2000.

Object	Mag	$z_{\rm e}$	Wvl. range	$t_{\rm exp}$
			(nm)	(sec)
UM680	18.6	2.144	305-387 / 477-680	7950
UM681	19.1	2.122	305-387 / 477-680	11 600
Q2344+1228	17.5	2.773	376-498 / 670-103	3600
Q2343+1232	17.00	2.549	376-498 / 670-103	3600
Q2139-4433	20.18	3.220	413-530 / 559-939	9000
Q2139-4434	17.72	3.23	413-530 / 559-939	7200

Spectra were taken in dichroic mode with a slit of 1.2'' and binning of $2\times 2$ pixels. A binning of $3\times 2$ pixels was adopted for one of the spectra of the faintest object Q2139-4433. The resolution is $\sim$37 000 and $\sim$35 000 in the blue and in the red portion of the spectra respectively. Wavelength ranges in the blue arm were chosen in order to cover most of the Lyman-$\alpha$ forest of each object. Another paper will be devoted to the detailed discussion of the lines in this region (D'Odorico et al. in preparation).

Data reduction was carried on by using the specific UVES pipeline (see Ballester et al. 2000) in the framework of the 99NOV version of the ESO reduction package, MIDAS. The continuum was determined by manually selecting regions not affected by evident absorption and by interpolating them with a spline function of 4th degree.

Metal absorption systems were detected, in general, by first identifying C IV or Mg II doublets and then looking for other ionic transitions at the same redshift. Atomic parameters for the lines were taken from Verner et al. (1994). New oscillator strengths were adopted for most of the Fe II transitions (Bergeson et al. 1994, 1996; Raassen & Uylings 1998). Lines were fitted with Voigt profiles in the LYMAN context of the MIDAS reduction package. The reported errors on column densities are the $1~\sigma$ errors of the fit computed in MIDAS. They possibly underestimate the real error on the measure since they do not take into account the uncertainty on the continuum level determination. Furthermore, they are the result of a single fitting model which is not univocal in certain column density regimes and for heavily blended systems (Fontana & Ballester 1995). In the cases in which the column density was weakly constrained and the MIDAS procedure could not converge to a unique solution (e.g. for saturated H I Lyman-$\alpha$ lines), indicative values of the column density were obtained by use of the interactive fitting program XVoigt (Mar & Bailey 1995).

We could analyse also the UVES spectrum of Q2138-4427 (B = 18.9, $z_{\rm e} \simeq 3.17$), with similar resolution and wavelength coverage. The detailed description of its reduction and of the two DLASs present in it, will be given elsewhere (Ledoux et al. in preparation).