All Figures

$\begin{figure} \par\setlength{\unitlength}{1mm} \begin{picture} (120,95) \put(0... ...02]{h4441f1l.eps} } \end{picture}\hspace*{\fill} \parbox[b]{55mm}{ }\end{figure}$ Figure 1: Top row: example PMAS data of HE 0435-1223, for different wavebands. The first three images are quasi-monochromatic (3.3 Å bandwidth), the right-hand image is averaged over a wide passband. Each image measures $8''\times 8''$ m and each spatial pixel is $0\hbox{$.\!\!^{\prime\prime}$ }5 \times 0\hbox{$.\!\!^{\prime\prime}$ }5$ . The orientation is standard (North is to the top, East is to the left). The four QSO components are clearly resolved. Notice the strong shift of image centroids as a function of wavelength (cf. Fig. 2), caused by differential atmospheric refraction. Middle row: corresponding model images consisting of four superposed Gaussians (for details see text). Bottom row: residuals after subtracting the four point sources.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=82mm,clip]{h4441f2.eps}\end{figure}$ Figure 2: Centroid displacement as a function of wavelength, mainly due to differential atmospheric refraction. Each point corresponds to one quasi-monochromatic spectral layer. In the later fitting passes, the measured centroids are replaced by the polynomial approximation plotted by the solid lines.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.15cm,clip]{h4441f3.eps}\end{figure}$ Figure 3: Same as Fig. 2, but for the full width at half maximum (average of major and minor axis) of the Gaussian PSF.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics*[width=178mm,clip]{h4441f4.eps}\end{figure}$ Figure 4: Extracted and coadded spectra of the four QSO components. The thin lines near zero represent the 1 $\sigma$ error arrays. The adopted local continuum levels, used to derive emission line properties, are indicated by the dashed line segments.
Open with DEXTER

In the text

$\begin{figure} \includegraphics[width=11.9cm,clip]{h4441f5.eps}\end{figure}$ Figure 5: Pairwise comparison of normalised emission line profiles (upper row: C IV; lower row: C III]). The light thick lines represent component A, the dark thin lines denote components B, C, and D, respectively. Below each panel we show also the difference between the profiles.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[angle=90,width=42mm,clip]{h4441f6a.ps}\hspace*{4mm} \includegraphics[angle=90,width=42mm,clip]{h4441f6b.ps}\end{figure}$ Figure 6: Left panel: $4\times 4$ supersampled image of HE 0435-1223, coadded over a simulated 5500-7200 Å passband (see text for details). Right panel: residual image after PSF subtraction, supersampled and coadded in the same way. Despite the significant PSF residuals, the lensing galaxy is clearly detected. The positions of the four QSO components and the lensing galaxy (from Paper I) are marked by crosses.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.8cm,clip]{h4441f7.eps}\end{figure}$ Figure 7: Redshift estimation for the lensing galaxy. Left panel: $\chi ^2$ vs. redshift for model fit of 10 Gyr SSP to the binned residual spectrum at position of the lensing galaxy. Right panel: residual spectrum with best-fit SSP model spectrum superimposed. For details see text.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=82mm,clip]{h4441f2.eps}\end{figure}$	Figure 2: Centroid displacement as a function of wavelength, mainly due to differential atmospheric refraction. Each point corresponds to one quasi-monochromatic spectral layer. In the later fitting passes, the measured centroids are replaced by the polynomial approximation plotted by the solid lines.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=8.15cm,clip]{h4441f3.eps}\end{figure}$	Figure 3: Same as Fig. 2, but for the full width at half maximum (average of major and minor axis) of the Gaussian PSF.
Open with DEXTER

$\begin{figure} \par\includegraphics*[width=178mm,clip]{h4441f4.eps}\end{figure}$	Figure 4: Extracted and coadded spectra of the four QSO components. The thin lines near zero represent the 1 $\sigma$ error arrays. The adopted local continuum levels, used to derive emission line properties, are indicated by the dashed line segments.
Open with DEXTER

$\begin{figure} \includegraphics[width=11.9cm,clip]{h4441f5.eps}\end{figure}$	Figure 5: Pairwise comparison of normalised emission line profiles (upper row: C IV; lower row: C III]). The light thick lines represent component A, the dark thin lines denote components B, C, and D, respectively. Below each panel we show also the difference between the profiles.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=8.8cm,clip]{h4441f7.eps}\end{figure}$	Figure 7: Redshift estimation for the lensing galaxy. Left panel: $\chi ^2$ vs. redshift for model fit of 10 Gyr SSP to the binned residual spectrum at position of the lensing galaxy. Right panel: residual spectrum with best-fit SSP model spectrum superimposed. For details see text.
Open with DEXTER