|
Description
This paper describes the methods used to obtain the spectroscopic data and construct redshift catalogs for the Canada-France Deep Redshift Survey (CFRS). The full data set consists of more than 1000 spectra, of objects with 17.5=<I_AB_=<22.5, obtained from deep multislit data with the MARLIN and MOS-SIS spectrographs at the Canada-France-Hawaii Telescope (CFHT). The final spectroscopic catalog contains 200 stars, 591 galaxies with secure redshifts in the range 0=<z=<1.3, six QSOs, and 146 objects with very uncertain or unknown redshifts, leading to an overall success rate of identification of 85%. In addition, 67 objects affected by observational problems have been placed in a supplemental list. We describe here the instrumental setup and the observing procedures used to gather this large data set efficiently. New optimal ways of packing spectra on the detector to increase significantly the multiplexing gain offered by multislit spectroscopy are described. Dedicated data reduction procedures have been developed under the IRAF environment to allow for fast and accurate processing. Very strict procedures have been followed to establish a reliable list of final spectroscopic measurements. Fully independent processing of the data has been carried out by three members of the team for each data set associated with a multislit mask, and final redshifts were assigned only after the careful comparison of the three independent measurements. A confidence class scheme was established. We strongly emphasize the benefits of such procedures. Finally, we present the spectroscopic data obtained for 303 objects in the 0000-00 and 1000+25 fields. The success rate in spectroscopic identification is 83% for the 0000-00 field and 84% for the 1000+25 field.
|
