GEMINI (July2006) Gemini IRAF package v1.9 GEMINI (July2006) Release date: July 28, 2006 This document describes the revisions relative to the Gemini IRAF package v1.8. This version of the Gemini IRAF package is compatible with, and requires, the IRAF patch release V2.12.2a (http://iraf.net). Also, the August 2005 version of FITSUTIL is mandatory. Note that the external packages have not yet been moved to IRAF.net. They can be obtained from the former IRAF website (http://iraf.noao.edu). PRIMARY REVISIONS: 1) New package for the reduction of NIFS data (gemini.nifs) 2) Support for Michelle polarimetry imaging data reduction 3) New mid-IR spectroscopy routines 4) More new tasks 5) More new functionalities and parameter changes 6) Bug fixes DETAILS OF THE REVISIONS: 1) New package for NIFS data (gemini.nifs). The current package offers support for the processing of NIFS IFU data. The gemini.nifs packages uses some of the GNIRS spectroscopy tasks: the NIFS package automatically loads the GNIRS package to obtain access to those tasks. The NIFS package itself contains the NIFS specific tasks. The NIFSINFO task offers an overview of the tasks found in the package. Data reduction examples for the baseline calibration, the telluric standards, and the science observations are available by invoking NIFSEXAMPLES. (i) New tasks in NIFS: - nifs.nfacquire -- Collapse flip-mirror frame into an image of the field for acquisition. - nifs.nfdispc -- Display a series of channel maps from a data cube. - nifs.nfextract -- Extract a 1-D spectrum from a transformed frame. - nifs.nffitcoords -- Compute 2-D dispersion and distortion maps. - nifs.nffixbad -- Fix bad pixels in NIFS frames. - nifs.nfimage -- Collapse unprocessed disperse spectral frame into a 2-D image. - nifs.nfmap -- Extract a channel map from a NIFS data cube. - nifs.nfprepare -- Prepare NIFS data for reduction. - nifs.nfsdist -- Establish S-distortion calibration. - nifs.nftelluric -- Correct telluric absorption in transformed frame. - nifs.nftransform -- Rectify 2-D NIFS spectra. - nifs.nifcube -- Create data cubes from NIFS format files. Documentation: - nifs.nifsinfo -- Overview of the NIFS package. - nifs.nifsexamples -- Print data reduction examples to the screen. Utility tasks: - nifs.nfcheck (hidden task) - nifs.nfckcal (hidden task) - nifs.nfpad (hidden task) - nifs.nfwcs (hidden task) 2) Support for the new Michelle imaging polarimetry mode. The Michelle imaging polarimetry mode has recently been commissioned at Gemini North. This release includes new tasks for the data reduction of such data. The MIDIRINFO task offers a section with information specific to imaging polarimetry with Michelle. A data reduction example for imaging polarimetry is available by invoking MIDIREXAMPLES. (i) New polarimetry tasks in MIDIR - midir.mipsplit -- Write polarimetry waveplate images to series of simple FITS file. - midir.mipsstk -- Combine individual Stokes images from MIPSTOKES into a single set of images. - midir.mipstack -- Straight average of nod images. - midir.mipstokes -- Calculate Stokes parameters for each nod pair. - midir.miptrans -- Calculate Stokes parameters form stacked polarimetry file. (ii) New functionalities/parameters in MIDIR for polarimetry support midir.mireduce: midir.mprepare: - New functionalities: * Recognize polarimetry data based on header content * Process polarimetry data 3) New mid-IR spectroscopy routines for Michelle and T-ReCS. Several new tasks have been added to the MIDIR package for the reduction of Michelle and T-ReCS spectroscopy data. Those tasks address such issues as flat field correction, de-fringing, and telluric correction. A 'wrapper' spectroscopy reduction task is now available in MSREDUCE. (i) New spectroscopy tasks in MIDIR - midir.msabsflux -- Telluric correction and absolute calibration - midir.msdefringe -- Defringing of T-ReCS or Michelle 1-D spectra - midir.msflatcor -- Bias and flat corrections for Michelle spectra - midir.msreduce -- Wrapper routine for reduction of T-ReCS or Michelle spectra (ii) New functionalities/parameters in MIDIR for spectroscopy support A large suite of spectral energy distribution files for spectrum calibration has been added in midir$data/. Please refer to the README file in midir$data/. midir.mistack: - New functionalities: * The "frametype" option will work with Michelle NOD mode spectroscopic observations. If frametype is either "src" or "ref" one gets the stacked sky frames for the nods, while if the value is "dif" one gets the difference between NOD A and NOD B, which shows the source spectrum in positive and negative forms as is usual for near-IR observations dithered along the slit. 4) More new tasks. (i) GEMTOOLS - gemtools.gemcube -- Build data cube. Used by nifs.nifcube. - gemtools.gemdate -- Get FITS format date string (hidden task). - gemtools.gemisnumber -- Check if a string is a number, and if so, of which type (hidden task). (ii) GNIRS - gnirs.nstelluric -- Calculate and apply telluric correction (this is a re-release, as the task was pulled out of v1.8) (iii) MIDIR - midir.mcheckheader -- Make Michelle headers GSA-compatible. (GSA=Gemini Science Archive) (iv) NIRI - niri.nisupersky -- Make deeper object masks for NISKY. 5) More new functionalities and parameter changes. (i) GEMTOOLS gemtools.gemarith: gemtools.gemexpr: - New default value: * logfile = "" gemtools.gemcombine: - New parameters * masktype * maskvalue (ii) GMOS gmos.gdisplay: - New functionality: * Flagging of saturated pixels and/or pixels above some threshold. - New parameters: * fl_sat * signal gmos.gfquick: - New functionality: * Outputs P,Q coordinates intead of X,Y. - New default values: * xpos1 = "default" (was 1210) * nslit = "default" (was 2) * yoffu = "default" (was 12) gmos.gbias: gmos.gifringe: - New functionality: * Expose two new GEMCOMBINE parameters (masktype, maskvalue) instead of using hardcoded values. - New parameters: * masktype * maskvalue gmos.gmosaic: - New functionalities: * Support for variance and data quality frames. * Option to convert a 'short' image to 'real' before mosaicing. - New parameter: * fl_vardq * var_ext * dq_ext * fl_real gmos.gprepare: - New functionality: * Set the keyword DISPAXIS (instead of relying on GSAPPWAVE). gmos.gscut: - New MDF column: * Calculate and add column SPECCEN (y-center of spectrum, type 'real') to MDF. This is for future development. gmos.gsflat: - New functionality: * Expose two new GEMCOMBINE parameters (masktype, maskvalue) instead of using hardcoded values. - New parameters: * masktype * maskvalue - Modified header keyword: * Changed the slit correction time stamp to GSLITCOR as GSSLITFUNC conflicts with the GSSLITFUNCTION (not released) time stamp. gmos.gstransform: - New parameters: * gratingdb * filterdb * key_dispaxis * dispaxis (iii) GNIRS gnirs.nsflat: - New functionality: * Expose two new GEMCOMBINE parameters (masktype, maskvalue) instead of using hardcoded values. - New parameters: * masktype * maskvalue - New default value: * process : for cross-dispersed data 'auto' defaults to 'fit' (used to default to 'trace') * statsec = "default" * fitsec = "default" For statsec and fitsec, "default" means "[*,*]" for NIRI, and "MDF" for everything else. This is a cosmetic change mostly. gnirs.nsheaders: - New parameter: * key_instrument gnirs.nsprepare: - New default value: * bpm = "gnirs$data/gnirs_2005sep24_bpm.fits" (this a new BPM for gnirs) gnirs.nsreduce: gnirs.nsslitfunction: gnirs.nssky: gnirs.nssdist: gnirs.nscombine: gnirs.nsstack: gnirs.nschelper: - New functionality: * Expose two new GEMCOMBINE parameters (masktype, maskvalue) instead of using hardcoded values. - New parameters: * masktype * maskvalue gnirs.nswavelength: - New parameter: * fl_median (default is 'yes' for GNIRS data, 'no' otherwise) - New default values: * coordlist = "gnirs$data/lowresargon.dat" * threshold = 100. (was 0.) gnirs.nvnoise: - New functionality: * Bad pixel fixing when calculating statistics. - New parameters: * statsec * fl_boundary gnirs$data/nsappwave.fits: - New functionalities: * Support for NIFS * New lens designations for Short cameras gnirs$data/config.fits: - New functionality: * New lens designations for Short cameras (iv) MIDIR midir.miflat: - New functionality: * MIFLAT now outputs a MEF file rather than a simple FITS file. midir.mireduce: - New functionalities: * Ability to "rescue" incomplete raw data files. * Output a variance frame. * Ability to define the region to use for registration. * Requires flat (produced by MIFLAT) to be a MEF file. - New parameters * fl_rescue * fl_variance * region midir.miregister: - New functionalities: * Write World Coordinate System header keywords to the science extension. (Instead of just the Primary Header Unit.) * Output a variance frame. * Ability to define the region to use for registration. - New parameters: * fl_variance * regions midir.mistack: - New functionalities: * Write World Coordinate System header keywords to the science extension. (Instead of just the Primary Header Unit.) * Output a variance frame. - New parameter: * fl_variance midir.mprepare: - New functionality: * Ability to "rescue" incomplete raw data files. * Support for polarimetry data. - New parameter: * fl_rescue gnirs.nswavelength for Michelle via gnirs$data/michelle.dat: gnirs.nswavelength for T-ReCS via gnirs$data/trecs.dat: - New default value: * nswavelength.coordlist = "gnirs$data/sky.dat" (iv) NIRI niri.nisky: - New functionality: * Always write NISKYMSK keyword, even if fl_keepmasks=no. - New parameter: * fl_nifastsky gnirs.nswavelength for NIRI via gnirs$data/niri.dat - New default value: * nswavelength.coordlist = "lowresargon.dat" 6) Bug fixes. This list is not exhaustive. gmos.gireduce: * Fixed behaviour when fl_over+ and fl_trim-. The code was resetting DATASEC and marking the image as TRIMMED even though that was not the case. gmos.gsappwave: * Fixed a typo in the WCS of the variance (VAR) plane (e.g. C1_1 -> CD1_1). gmos.gscalibrate: * Fixed variance and data quality plane propagation. gmos.gscut: * GSCUT expects the database to be named 'database', yet APFIND was using whatever specred.database was set to. This dependency on the specred.database user setting has been removed. In the future, we plan to remove the hardcoding the database name. gmos.gsextract: * Fixed extraction of the variance (VAR) plane from APALL output. gmos.gsflat: * Fixed a major bug that was introduced in Gemini 1.7 (and patched in 1.7.2 and 1.8.1). When several flats were used as input, GSFLAT was using only the first input flat in the list. This occured only when fl_detec=no. gmos.gsstandard: * The parameter 'caldir' could not easily be reset because of hardcoded checks. It offers more flexibility now. gmos.gstransform: * Correctly handle variance and data quality planes. * Correctly set NEXTEND keyword value when fl_vardq = yes. This miscount might have affected some users' personal software depending on NEXTEND. gnirs$nsflat: * Correctly propagate the input's header keywords. * Fixed and improved the handling of NIRI darks to address specific problems reported by some users. Please refer to the new NIRI spectroscopy example for the correct reduction procedure (niri.niriexamples) * The 'fitsec' and 'statsec' features were not working correctly when set to something else than the data's default (MDF or [*,*]). This has been fixed for 'process=fit' (now recommended for all data). gnirs.nsreduce: * Correctly propagate the DQ plane header keywords. * Checks whether the outputs already exists at the *beginning* of the task rather than go through a lengthy process and crash if they do. gnirs.nsappwave: * Large (~300%) improvement in efficiency, and more robust. * Actually propagate the MDF. The previous version was losing the MDF along the way. gnirs$data/gnirs-xd-short-32-mdf.fits: * This MDF was incorrect for the higher orders. This was not noticed earlier because before Jul/Aug 2005, the higher orders could not be seen properly because of the poor XD filter transmission in the blue. midir.miregister: * The tmp files were not deleted properly resulting in all frames in for all images a list of files, up to the current file, being stacked. * Number of files that can be processed in one call set to 100. * Set EXTNAME and EXTVER for variance plane. midir.mistack: * The tmp files were not deleted properly resulting in all frames in for all images a list of files, up to the current file, being stacked. * Number of files that can be processed in one call set to 100. * Remove system dependence on the command "sed". This was problem for some Unix users. * Set EXTNAME and EXTVER for variance plane. midir.mprepare: * Remove system dependence on the command "sed". This was problem for some Unix users. * Make EXTVER (the extension version) increment correctly for the science extensions. * Actually handle already prepared images instead fo just flagging them. midir.tprepare: * FITS extensions of the output file now have EXTNAME and EXTVER set. 7) Updates to reduction examples (i) gnirs.gnirsexamples - Improved cross-dispersed (XD) and longslit (LS) examples. (ii) midir.midirexamples - Now included are examples for michelle spectroscopy and polarimetry. (iii) niri.niriexamples - imaging * Added an 'unlearn' warning. - spectroscopy * Updated to reflect the correct way to handle the darks. * Added an 'unlearn' warning.