Using GMOS pre-imaging

Information about the co-added images supplied by Gemini

For MOS programs scheduled during the current semester, the PIs will receive separate instructions by e-mail on how to retrieve the reduced and co-added images of their targets from the Gemini Science Archive. Due to the short time available to design and prepare the masks, it is strongly recommended to use these images as the basis for the mask designs, rather than to start from the raw imaging data.

The reduced and co-added images have been produced using the Gemini IRAF package which includes reduction software for GMOS. The software correctly takes into account the gaps between the CCDs and the misalignment of the CCDs relative to each other.

The images have been co-added with the task imcoadd, which is also part of the Gemini IRAF package. The co-added images are average images, thus they are scaled to the exposure time of the individual images used to create the co-added image (see the keyword EXPTIME in the header of the primary header unit). Further, the co-added images are in e-/pixel.

For images taken under photometric conditions, an approximate standard calibration is supplied.

How to prepare an Object Table

The Object Table is a FITS table that contains, as a minimum, the following columns:

ID	Unique object id (integer)
RA	RA in hours (real)
DEC	Dec in degrees (real)
x_ccd	X coordinate of object position in pixels (real)
y_ccd	Y coordinate of object position in pixels (real)
MAG	(Relative) magnitude of object (real)

The order of these columns is important. The column names are case sensitive.

Other columns may be present in the table and may help the user selecting objects for the mask. The user may find the following columns useful:

slitsize_x	Slit width in arcsec (real)
slitsize_y	Slit length in arcsec (real)
slittilt	Slit position angle in degrees (real)
slitpos_y	Slit position in the Y-direction (spatial) relative to the object position in arcsec (real)

The default is slittilt=0. slittilt is measured counter-clockwise relative to the default position. The column slitpos_y is useful for designing slits where the objects are not all in the centers of the slits. Tweaking the slit positions in the Y-direction may allow the user to fit more slits in a given mask. The example script objtexample2.cl shows the use of the columns slitsize_x, slitsize_y, slittilt, and slitpos_y in more detail.

NOTE: If the image is binned, the Object Table MUST contain a header keyword "PIXSCALE" containing the pixel scale in arcsec/pixel.

Two IRAF tasks app2objt and stsdas2objt are supplied to help turn the output from common photometry programs into valid Object Tables.

      app2objt - Convert a table from daofind/apphot/daophot into an Object Table
   stsdas2objt - Convert FITS or STSDAS tables into an Object Table

These tasks are part of the released version of the Gemini IRAF package and can be found in the package gemini.gmos.mostools (ie. the user needs to load the packages "gemini", "gmos" and "mostools").

The task app2objt may be used to convert the output from the IRAF tasks daofind, phot, allstar or nstar into valid Object Tables.

The task stsdas2objt may be used to convert any FITS or STSDAS table into a valid Object Table. The task is specifically designed to handle FITS output tables from SExtractor.

The user may also prepare the Object Table with any other software capable of writing FITS tables. In this case, it is recommended to process the FITS table with the task stsdas2objt to ensure that the column names and the format is correct.

To make the preparation of the Object Table go a little easier, the user may find the following two example IRAF scripts useful.

   objtexample1.cl - How to prepare an Object Table from output from apphot
   objtexample2.cl - How to prepare an Object Table from output from SExtractor

Making the best use of imaging from other telescopes

If the user has prior imaging from other telescopes, it may be possible to use GMOS images for the mask design obtained with quite short exposure times and/or in worse seeing conditions than needed for the spectroscopy. If the GMOS images contain a sufficient number of objects (> 50) with good signal-to-noise then these objects may be used to "boot-strap" the pixel positions of the science targets.

It is recommended that the user proceeds as follows.

• Based on the bright (but unsaturated) objects in the GMOS image, determine the transformation between GMOS pixel coordinates (x_ccd,y_ccd) and the (RA,DEC) from the previous observations.
• Transform all (RA,DEC) from the previous observations to GMOS pixel coordinates (x_ccd,y_ccd).
• Make an STSDAS table or a FITS table that has the following columns

  ID	Unique object id (integer)
  RA	RA a in hours (real)
  DEC	Dec in degrees (real)
  x_ccd	X coordinate of object position in pixels (real)
  y_ccd	Y coordinate of object position in pixels (real)
  MAG	(Relative) magnitude of object (real)
  plus any other columns the user would like to include

  The table may be made from ASCII input using the task tcreate in the IRAF package tables.ttools which is part of the STSDAS software distributed by STScI.
• Process the table with the task stsdas2objt to ensure that a valid Object Table is produced. It is recommended to recalculate (RA,DEC) using the World Coordinate System in the image headers. To do so, set the flag fl_wcs to "yes" in the task stsdas2objt. (RA,DEC) are used for plotting only, while (x_ccd,y_ccd) is used to define the positions of the slit.

Convert the Object Table into an Object Definition File (ODF)

Use the Gemini MOS Mask Preparation Software (GMMPS) to create an Object Definition File (ODF)

Last Update September 19, 2007; Kathy Roth, Ilona Soechting
Previous update September 7, 2007; Rodrigo Carrasco