# #+++++++++++++++++++++++++++++++++++++++++++++++++++++++ #.COPYRIGHT (C) 1993 European Southern Observatory #.IDENT irspec_extended.hlp #.AUTHOR Cristian Levin (ESO/La Silla) #.KEYWORDS Infrared spectroscopy, Irspec #.PURPOSE This is the help file for the extended help # feature of XIrspec. # #.VERSION 1.0 Package Creation 17-MAR-1994 #------------------------------------------------------- # ~HELP_CONTEXT This is the context Irspec. ~HELP_HELP You can get an extended help for each button of the interface by pressing the right mouse button when the cursor is over the interface button. ~HELP_TUTORIAL Not available for the moment. ~HELP_VERSION This is the version 1.0 of the Graphical User Interface for the context Irspec. comments about the interface can be sent to : clevin@eso.org (Cristian Levin) comments about the commands can be sent to: midas@eso.org (MIDAS support) ~MAIN_BADPIX To create a table containing the position of bad pixels. The IRSPEC array has a fixed pattern of bad pixels which, which can be cleaned by by substituting the values of the bad pixels with the average of the neighboring good pixels. A default table of bad pixels (irsbadpix.tbl) is automatically created when starting the IRSPEC context. If the default table is going to be used, cleaning of frames is done implicitly when reducing the standard star or object, and BADPIX is not explicitly required. ~MAIN_FLAT To create a normalized flat from an halogen frame. The flat will be used in the reduction of the standard star and the object. The normalization procedure consists of dividing the original frame, cleaned of the bad pixels, by the average value of the central 6 rows. Hence, the normalized output image contains also the detector response at low spatial frequencies. Use the explicit MIDAS commands FIT/FLAT if you need a more normalized flat. The vignetted rows are set to large values in order to make them vanishing after division by the flat. For applications up to 2.5 microns the input, flat image is usually a measurement of the halogen lamp with counts level as close as possible to those in the astronomical frames. ~MAIN_FLUX First step for flux calibration of IRSPEC data. The user must provide an ASCII file with his/her favorite guesses of the star fluxes at a number of wavelengths, e.g. the values of F(lambda) at the effective wavelengths of photometric filters in case of photometric standard stars. The file must contain, in columns 1 and 2, the wavelength and corresponding flux, respectively. Columns 3 and higher may contain comments. Example 1.25 1.04e-3 J=6.12, flux in 1e-11 erg cm-2 s-1 mu-1 2.20 196 K=5.73, same units 1.65 4.34e-2 H=5.99, same units The command will then create a MIDAS table containing a properly sampled F(lambda) vs. lambda interpolation within the wavelength limits given in the input ascii file. ~MAIN_STANDARD Open the window for standard star reduction. To create a 1-D response frame which contains the conversion from counts/sec to flux units for the measurements taken at a given central wavelength. ~MAIN_OBJECT Open the window for object reduction. ~MAIN_EXTRACT Open the window for spectrum extraction. ~MAIN_MERGE Open the window for the merge of overlapping 1D spectra into a table. The command also allows optimizing the connection on the overlapping parts and excluding a given number of pixels at both edges of the spectrum. ~BADPIX_DEFINE ~BADPIX_APPLY ~FLUX_APPLY Interpolation of the flux values provided in the ASCII table 'Broad band flux table', to create a MIDAS table 'Output flux table'. The 'Broad band flux table' must contain, in columns 1 and 2, the wavelength and corresponding flux, respectively, of the standard star used to calibrate the spectra. The 'Output flux table' contains wavelengths (column :wl) and fluxes (column :flux) interpolated over the input values according to the Interpolation method. The star flux is sampled at wavelength intervals specified in 'Wavelength step', default = 0.01. 'Fit degree' is the degree of the polynomial/spline, default = 2. For the transformation from magnitudes to fluxes you may use e.g. Wilson et al. 1972, ApJ 177, 533 Bessel 1979, PASP 91, 589 ~STANDARD_REDUCE To create a 1-D response frame which contains the conversion from counts/sec to flux units for the measurements taken at a given central wavelength. 'Flux table' is a previously created MIDAS table with properly sampled flux values of the standard star. 'Response frame' is the output 1-D frame containing the instrumental response (counts per second per unit flux). 'Y-positions' are the positions of the lower and upper standard star continua in the sky-subtracted frame. Wavelength calibration ---------------------- The wavelength dispersion on the IRSPEC array is linear within a small fraction of the pixel size. Hence, wavelength calibration simply means to modify the x-start and x-step descriptor values of the image. Standard: A quite precise - usually within one pixel - estimate of the wavelength scale is available directly from the IRSPEC descriptors. The standard wavelength calibration works automatically. High acc: A more accurate wavelength calibration can be reached if the frame contains emission lines with known wavelengths. Up to 2.3 microns the OH lines in the sky frames are a very convenient calibrator (Oliva & Origlia 1992, A&A 254, 466). If High accuracy is selected, specify a 'Reference frame' which will contain the precise wavelength parameters. The 'High accuracy' method works interactively. Sky subtraction --------------- In most of the infrared the sky emission contains bright emission lines, OH transitions up to 2.3 microns and molecular bands (CO2, CH4) at longer wavelengths. The intensity of the OH transitions varies on short time scales, so the 'object - sky' difference frames may contain residual sky lines. It also happens that the grating moves by tiny amounts between object and sky frames, which deteriorates the sky subtraction considerably. If such occurs, the sky frame may be multiplied by a factor different from 1.0 and shifted by an amount different from 0. Factors and shifts can be entered or obtained interactively by specifying factor = 0 ~OBJECT_REDUCE Wavelength calibration ---------------------- The wavelength dispersion on the IRSPEC array is linear within a small fraction of the pixel size. Hence, wavelength calibration simply means to modify the x-start and x-step descriptor values of the image. Standard: A quite precise - usually within one pixel - estimate of the wavelength scale is available directly from the IRSPEC descriptors. The standard wavelength calibration works automatically. High acc: A more accurate wavelength calibration can be reached if the frame contains emission lines with known wavelengths. Up to 2.3 microns the OH lines in the sky frames are a very convenient calibrator (Oliva & Origlia 1992, A&A 254, 466). If High accuracy is selected, specify a 'Reference frame' which will contain the precise wavelength parameters. The 'High accuracy' method works interactively. Sky subtraction --------------- In most of the infrared the sky emission contains bright emission lines, OH transitions up to 2.3 microns and molecular bands (CO2, CH4) at longer wavelengths. The intensity of the OH transitions varies on short time scales, so the 'object - sky' difference frames may contain residual sky lines. It also happens that the grating moves by tiny amounts between object and sky frames, which deteriorates the sky subtraction considerably. If such occurs, the sky frame may be multiplied by a factor different from 1.0 and shifted by an amount different from 0. Factors and shifts can be entered or obtained interactively by specifying factor = 0 ~OBJECT_FLUX To flux calibrate a spectrum which could either be a 2-D long-slit image or a 1-D spectrum. A 'Response frame' must have been previously created using the 'Std star...' command. ~MERGE_APPLY Merge overlapping 1-D spectra (images) into a table. The spectra must be ordered in wavelength and must be overlapping. The 1-D spectra to be connected must be named with a prefix and a number, eg spec0001, spec0002, spec0003, etc. Prefix: The files prefix, i.e. "spec" in the above example Interval: Give here the first and last number of the files to be merged. Step: Increment of file-ID numbers, default = 1. ref#: In the process of forcing the connection of the overlapping regions, one of the 1-D images (the reference image) is left untouched. If ref=0 (default) the reference image is the central 1D frame of frames specified. Example ------- prefix: spec interval: 4, 12 step: 3 ref#: 8 will cause the 1-D spectra, contained in files spec0004, spec0008, spec0012, to be merged, with file spec0008 taken as the reference. NOTE ---- The connection between the spectra is optimized by applying a multiplicative factor to the spectra (NB the input spectra are left untouched). This implies that the command works properly only with spectra with strong and well defined continua while it may produce totally unreliable results with spectra with faint continua affected by variable offsets. ~EXTRACT_SKY Asks the user for the sky windows using the cursor over the display window. The coordinates are then displayed in the input form. ~EXTRACT_OBJECT Asks the user for the object limits using the cursor over the display window. The coordinates are then displayed in the input form. ~EXTRACT_FIT Fit a polynomial to the data in two windows along the Y-axis. Two modes can be selected: Same spatial profile -- Here it is assumed that the normalized spatial profile is the same for all columns and only one polynomial is fitted to the mean spatial profile. Independent profile -- In this case an independent polynomial is fitted for every row of the spectrum, filtering also the cosmic rays. PARAMETERS: The following parameters must be set in the input form: a) Sky limits (pixels) -- starting and ending Y coordinates of the sky windows b) Polynomial fit degree -- degree of the fit to be used. c) Radius -- used for cosmics rays rejection. ~EXTRACT_AVERAGE The standard method. The sky is taken from the sky limits values of the input window. The result is summed or averaged depending on the extraction method parameter. ~EXTRACT_WEIGHT Extract spectrum from CCD frame. The rows of the spectrum are added with weights which are chosen for optimal S/N-ratio of the resulting spectrum. Cosmic ray hits are removed by analyzing the profile perpendicular to the dispersion (assumed to be along the X-axis). The method is only suited for spectra of (spatially) unresolved sources. The method is very similar to the one described by Horne (1986, PASP 98, 609). PARAMETERS: The extraction algorithm uses the following parameters: a) Object limits (pixels) -- Approximate Y coordinates of the object (starting and ending). b) Extraction iterations -- Number of iterations in the extraction process. c) Read-out-noise (e-) -- Read out noise of the CCD in electrons. d) Inverse gain factor -- Conversion factor from ADU to electrons. (e-/ADU) e) Threshold for cosmic rays -- Threshold used for the rejection of cosmic ray hits.