CFHT Information Bulletin, number 38, First Semester 1998

OSIS observation of the Dark(?) Lens Cluster AXJ2019+112

Kneib1,, Soucail1, G., Jaunsen2, A., Hattori3, M., Hjorth4, J., Yamada3, T.

1) OMP, 14 Av. E. Belin, 31400 Toulouse, France, 2) SHS, Norway, 3) Institute for Astronomy, TohoKu University, Sendai, Japan, 4) NORDITA, Denmark.

Gravitational Lensing efficiently probes the mass content of our Universe. Despite multi-wavelength observations of multiple-quasar systems, a large number of them are not fully understood. In particular, wide separation double quasars, and systems which require a unusually high mass-to-light ratio -- the so-called `Dark-Lenses' -- retain some mystery of the nature of their gravitational deflector(s).

Recently, a first step in the understanding of these Dark-Lens systems has been made with the ASCA/ROSAT X-ray satellites. Hattori et al. (1997) have detected a cluster-like emission AXJ2019+112, centered on the MG2016+112 lens system (Figure 15). Thanks to the detection of the Iron line on the ASCA spectrum, a redshift of zFe~1 was measured for the cluster X-ray emission, in good agreement with the redshift of the D galaxy (Schneider et al. 1985). Yet, the detection of a cluster has been of some surprise because no galaxy concentration had been previously detected in the direction of the lens system, leading Hattori et al. to name this cluster a "Dark Cluster". However, the lens system lies at low galactic latitude which makes any deep optical search difficult. Recently, Jaunsen and Hjorth have acquired at the Nordic Optical Telescope deep images around MG2016+112 (Figure 16) in good seeing conditions. With the photometry of this field (in R and I), a sample of faint and red galaxies was selected. This galaxy sample can then be studied spectroscopically to test the existence of galaxies in the Dark Cluster: this constituted the aim of our proposal.

Thanks to is tip-tilt mirror, its small field of view and a thinned CCD, the OSIS instrument is particularly well suited for such a programme. Two nights were allocated on August 3-4, 1997 for a multi-object spectroscopy of the field. Profiting of excellent weather conditions, we designed two masks of 19 and 18 slits. The R150 grism was used which gives a final resolution of 14Å. Both masks were exposed a total of 6 hours (6 × 3600s).

A total of 45 spectra were extracted (some slits contained multiple objects). IAB magnitude of targeted objects range from 22.3 to 23.2 (4 objects brighter were included to fill in the masks). Thanks to a careful pre-selection of the candidates, only 10 objects have been identified as stars, and 3 objects could not be identified. 32 objects have a measured redshift but 5 of them are tentative. We found 7 galaxies at the X-ray cluster redshift: 4 red-ellipticals (one of them being the D galaxy) and 3 star-forming galaxies. 22 galaxies have a redshift from 0.328 to 0.98 and 2 have a tentative redshift larger than 3. Considering the difficulty of detecting z ~ 1 elliptical galaxies (the 4000Å break is shifted to the I band where sky emission lines are pre-eminent) and their faintness, we have been successful in getting them, thanks to the long exposure time. This positive result demonstrates the existence of a structure around z ~ 1. The nature of this Dark Lens is therefore unveiled and can probably be simply explained by the presence of a "normal" galaxy cluster.

We also had a closer look at the MG2016+112 lens system. The NICMOS-2 image shown in Figure 15, clearly reveals the nature of the main deflector D: a giant elliptical galaxy (z = 1.004 from our spectrum, Figure 17) which coincides with the X-ray centroid. The A/B/C multiple images are typical of a "fold" system (cf. Nair & Garrett 1997) where C is the "merging" of 2 images: indeed, our spectrum of C confirms the detection of a component at the quasar redshift at position C (Schneider et al. 1985). Furthermore, the line intensity ratio between B and C are changing with wavelength (C/B decreasing from red to blue) in a similar way as has been found in broad band images (Nair & Garret 1997 for a summary) suggesting some absorption in the line of sight, most probably due to an intervening disk galaxy at the cluster redshift or higher. Moreover, B1 redshift is measured at z = 3.269 which confirms the interpretation of the narrow-filter (centered on Lyman-) detection of Schneider et al. 1986. Both, these new spectroscopic evidence and the NICMOS-2 data, will shed new light on the understanding of this gravitational system.

Considering the difficulty of the search, these OSIS observations were very successful, thanks to excellent weather conditions and the great efficiency of the OSIS instrument and CFHT telescope. However, here we reached the limit of 4m-class telescope capabilities: an integration time of 6 hours was necessary to achieve the spectroscopic identification of most objects (only 4 out of 45 could not be identified) with a median IAB magnitude of 22.7. Follow-up of these complex distant lensing systems will be better pursued with 8m-class telescopes; nevertheless, we have demonstrated that a well focussed observation on CFHT can still have a high scientific return.


Hattori, M., Ikebe, Y., Asaoka, I., Takeshima, T., Böhringer, H., Mihara, T., Neumann, D.M., Schindler, S., Tsuru, T., Tamura, T., 1997, Nature, 388, 146

Nair S., & Garrett, M.A., 1997, MNRAS, 284, 58.

Schneider, D.P., Lawrence, C.R., Schmidt, M., Gunn, J.E., Turner, E.L., Burke, B.F., Dhawan, V., 1985, ApJ, 294, 66

Schneider, D.P., Gunn, J.E., Turner, E.L., Lawrence, C.R., Hewitt, J.N., Schmidt, M., Burke, B.F., 1986, AJ, 91, 991

Editor: Dr. T. M. C. Abbott,
Copyright © 1998, CFHT Corporation. All rights reserved.