SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 360, 853-860 (2000)

Previous Section Next Section Title Page Table of Contents

2. Observations-reductions

The data were taken with SOFI, the near-IR (1 to 2.5 µm) imaging spectrograph on the ESO NTT. Two grisms were used to cover the 1 to 2.5 µm wavelength range: a "blue grism", which covers the region from 0.95 to 1.64 µm and a "red grism" which covers the region 1.53 to 2.52 µm. With a 1" slit, the spectral resolution is around 600. The observations with the blue grism were taken on the night of 1998 June 13, for a total integration time of 4560 seconds and the observations with the red grism were taken on the night of 1999 January 6, with a total integration time of 2400 seconds. Although the seeing for both nights was good, 0.6" - 0.8", neither night was photometric.

The slit was aligned with the two images of the quasar. As is standard practice in the infrared, the object was observed at two positions along the slit. The strong and highly variable night sky features were effectively removed by subtracting the resulting spectra from each other. The 2-D sky-subtracted spectra were then flat-fielded, registered, and added.

The two dimensional combined frames were spatially deconvolved in order to extract the spectrum of the lensing galaxy. For this purpose, we used the method outlined by Courbin et al. (1999, 2000). The algorithm is a spectroscopic extension of the so-called "MCS image deconvolution algorithm" (Magain et al. 1998). It spatially deconvolves 2-D spectra of blended objects, using the spectrum of a reference point source. It also improves their spatial sampling and decomposes them into the individual spectra of point sources (the two quasar images) and extended sources (the lensing galaxy). One also obtains a two-dimensional residual map, i.e., the difference between the data and the deconvolved spectrum (reconvolved by the spectrum of the PSF), in units of the photon noise. The quality of the deconvolution is checked using the residual map, which should be flat with a mean value of 1. The different products of the deconvolution are shown in Fig. 1 for the spectrum taken with the red grism.

[FIGURE] Fig. 1. Two dimensional spectra of HE 1104-1805. From top to bottom, (i) the 2-D near-IR combined (1.5-2.5 µm) spectrum (seeing [FORMULA] 0.6 ", pixel size [FORMULA] 0.14"), (ii) its deconvolved version (resolution of 0.14 ", pixel size [FORMULA] 0.07"), (iii) the deconvolved spectrum of the lens alone, and (iv) the residual map (see text).

As the data were obtained before we developed our spectra deconvolution algorithm, we did not observe in an optimal way, in the sense that no reference spectrum was obtained (a spectrum of a star in the field of view). We aligned the slit along the two quasar components, as is usually done for such observations. However, the seeing of the data taken with the red grism was good enough to derive the PSF spectrum from the brighter quasar itself. This was not possible with the data taken with the blue grism.

For the observations taken with the red grism, the spectrum of the lensing galaxy, was extracted from the 2-D deconvolved spectrum (extended component only) with standard aperture extraction techniques. The quasar spectra are a product of the deconvolution process and therefore do not show any contamination by the lensing galaxy. For the observations taken with the blue grism, we used wide apertures to extract the quasar spectra from the original data. The lens is therefore contaminating the spectra of the quasar, but by virtue of its very red color, the contamination is negligible.

All extracted 1-D spectra were then divided by that of a bright star and multiplied by a blackbody curve that has a temperature that is appropriate for the spectral type of the star. Before the division, spectral features that were visible in the spectra of the bright star, such as the Pachen and Bracket lines of hydrogen, were removed by interpolation.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: August 23, 2000
helpdesk.link@springer.de