CALIBRATION STEPS The calibration steps are listed by keyword below. The differences between poa_calfos and calfos are emphasized: CNT_CORR (no changes w.r.t. 'calfos') Convert from raw counts to count rates by dividing each data point by the exposure time and correcting for disabled diodes. If 'DEFDDTBL' is true, the disabled diodes are taken from the Unique Data Log (UDL), otherwise, disabled diodes are found in the file 'DDTHFILE'. OFF_CORR (major changes w.r.t. 'calfos') Data are corrected for zero-point offsets in the dispersion direction. These include: (a) the geomagnetically-induced image motion problem (GIMP), (b) a compensation for errors made in the on-board GIMP correction, (c) a compensation for X-shifts introduced by commanded Y-shifts (YBASE updates), (d) drifts as a result of variable temperature in the optical bench and detector head, and (e) drifts as a result of long term variation of the optical bench geometry and the magnetic environment. The data are shifted in the FOS X-direction (the dispersion direction); currently this shift is calculated with sub-pixel resolution, but applied in integer pixels in order to avoid resampling the data. The offsets are based on model calculations of the geo-magnetic field strength at the spacecraft's position. These calculations require the 'norad' table of the daily HST ephemeris parameters. This table is within the release. We would like to stress that ALL data throughout the operational life of FOS REQUIRE this OFF_CORR correction (whether or not the on-board GIMP correction was already performed). Therefore, 'poa_calfos' requires that the 'OFF_CORR' header keyword is set to 'PERFORM'. The pre-processor 'poa_preproc_fos' performs a header modification to the input data, in order to set this keyword to 'PERFORM'. In the unusual case where you do not want the OFF_CORR correction to be performed on the data; you can reset the OFF_CORR keyword to a value of 'OMIT' using tasks like 'hedit' or 'eheader' (after you have run the pre-processor 'poa_preproc_fos'). In this case, when you run 'poa_calfos' the OFF_CORR correction will still be calculated, but not applied. You can view the calculations (which would have been applied if OFF_CORR=PERFORM) in the new '.poa' output file which contains a sort of "history" of the POA calculations and results. When you run 'poa_calfos' the STDOUT POA related messages show the resultant GIMP correction (ie rootname y1010105t): *** X-OFFSET corrections calculated from POA model *** POA ygimp: det,id: BLUE 2 POA ygimp: mode,grp/ystp,xoff: SPECTROSCOPY 1 -2.664 POA x-offset for MODE = SPEC is -2.664 1/4 diodes = -3 array pix *** End of POA calculations *** Each GROUP or IMAGE spectrum has an offset calculation; each offset will be printed to STDOUT. The POA offset correction is listed twice: REAL value calculation and INTEGER round-off (pixels). The actual applied correction uses the integer offset value. The POA "keywords" in the output '.poa' file, can also be used to calculate this offset value. The "keywords" OFF_DIOD and OFF_PIX store the calculated POA offset value in diodes and pixels, respectively. The data are shifted in the x direction by OFF_PIX. In "words", this calculation equation looks like this: "offset" = - "aperture/grating zero point" - "GIMP related offsets" - "YBASE related offset" - "temperature related offset" Using the POA information in the '.poa' output file, one can translate it to: XOFFSET(diodes) = - YAPGRTX0 - YOFFXP*0.15 - YYBSXSCL*(YBASE-YYBASE0)*0.018898118 - YTMPXSCL*YMEANTMP XOFFSET(array pixels) = NINT(XOFFSET(diodes)*NXSTEPS/4) The output spectrum is shifted by XOFFSET(array pixels). This is equivalent to the value of OFF_PIX in the '.poa' file. PPC_CORR (no changes w.r.t. 'calfos') Correct the raw count rates for saturation in the detector electronics. Requires table 'ccg2' containing the paired-pulse correction table. BAC_CORR (no changes w.r.t. 'calfos') Subtract the background from sky and object spectra. If no background was taken, a default reference background, 'BACHFILE', is used. The observed background is smoothed with a median, followed by a mean filter before subtraction. The filter widths are contained in table 'ccs3'. No smoothing is done to the reference file background if used. GMF_CORR (no changes w.r.t. 'calfos') If a reference background is used in the 'BAC_CORR' calibration step, it will be scaled according to the geomagnetic position of the spacecraft at the time of observation. The background will be scaled to a mean count rate appropriate for the geomagnetic position. Requires table 'ccs8' containing predicted background count rates. SCT_CORR (no changes w.r.t. 'calfos') Subtract scattered light from object and sky spectra. Can only be performed for detector/grating combinations that contain a region within the recorded spectrum that is not sensitive to dispersed light from which the level of scattered light is measured. Requires table 'ccs9' containing the beginning and ending diode positions for the region of no sensitivity for each detector/grating combination. FLT_CORR (major changes w.r.t. 'calfos') Remove diode-to-diode sensitivity variations and fine structure by multiplying by the flat field response. Requires the flat field response file 'FL1HFILE'. A second flat field file, 'FL2HFILE', is required for paired-aperture or spectropolarimetry observations. Please note that the flat fields have been reprocessed using the POA correction. These new flat fields have the same name as the previously used flat fields, except that they located within the STPOA package. The locations of the appropriate POA flat fields (FL1HFILE, FL2HFILE) are added to the FOS datafile using the POA pre-processor 'poa_preproc_fos'. The pre-processor will check to make sure that there is an equivalent redone POA flat field, and update the .d0h header with it's location. If a new flat field with the POA corrections does not exits, then the original flat field will be used. The pre-processor writes a message to STDOUT telling the user whether new flats will be used. Please see the help file for 'poa_preproc_fos' for more information on the POA pre-processor. SKY_CORR (no changes w.r.t. 'calfos') Subtract the sky from the object spectra. The routine smooths the sky using a median and mean filter, scales the sky by the aperture size, and shifts the sky before the subtraction. The routine requires table 'ccs3' containing the filter widths, the aperture size table 'ccs0', the emission line position table 'ccs2', and the sky shift table 'ccs5'. WAV_CORR (major changes w.r.t. 'calfos') Compute a vacuum wavelength scale for each object or sky spectrum. Wavelengths are computed using coefficients stored in table 'ccs6'. The 'ccs6' wavelength calibration file has been completely changed from it's original 'calfos' format. This file is now located within the STPOA package. The POA pre-processor 'poa_preproc_fos' updates the CCS6 header keyword with the new POA file name and location; a message is printed to STDOUT, telling the user that the wavelength calibration file has been changed in the header. The new 'ccs6' file contains 10 coefficients to calculate the wavelength dispersion. Some modes, like PRISM, are still being done using the original 3+ coefficients, but most other modes have been upgraded to use 10 coefficients for the solution. All coefficients used for processing, are printed to STDOUT. For FOS data modes where new coefficients have not yet been calculated, the old coeffients are used. FLX_CORR (no changes w.r.t. 'calfos') Convert the object spectra to absolute flux units by multiplying by the inverse sensitivity curve. Requires the inverse sensitivity file 'IV1HFILE'. A second inverse sensitivity file, 'IV2HFILE', is required for paired aperture or spectropolarimetry observations. This method of flux calibration for pre-COSTAR observations is superceded in POA_CALFOS v2.0 by the APR_CORR, AIS_CORR, and TIM_CORR steps (see below). APR_CORR (no changes w.r.t. 'calfos') Correct for changes in aperture throughput due to changes in OTA focus position. Also correct the observed countrates to match that of the AIS reference aperture. Requires OTA focus history table 'ccsa', relative aperture throughput table 'ccsb', and aperture throughput vs focus table 'ccsc'. AIS_CORR (no changes w.r.t. 'calfos') Convert the object spectra to absolute flux units by multiplying by the average inverse sensitivity (AIS) curve for the reference aperture. Requires the AIS file 'AISHFILE'. This method of flux calibration for pre-COSTAR data supercedes the FLX_CORR step in POA_CALFOS versions 2.0 and later and requires that the APR_CORR step be performed first. TIM_CORR (no changes w.r.t. 'calfos') Correct the flux calibrated object spectra for changes in instrument sensitivity over time. Requires the table 'ccsd' containing correction factors as a function of wavelength and time for each detector/grating combination. ERR_CORR (no changes w.r.t. 'calfos') Compute the propagated error at each point in the spectrum. MOD_CORR (no changes w.r.t. 'calfos') Perform ground software mode dependent reductions for time-resolved, spectropolarimetry, and rapid-readout observations. The spectropolarimetry reductions require the the Wollaston and Waveplate parameter table 'ccs4', and the retardation reference file, 'RETHFILE'. If the spectropolarimetry data were acquired during the post-COSTAR epoch as indicated by the Science header keyword/value pair, KYDEPLOY=T, and a post-COSTAR calibration is available, the post-COSTAR polarimetry correction reference file, 'PCPHFILE', is also required for a proper calibration. However, it should be noted that if no post-COSTAR polarimetry calibration file for the 'PCPHFILE' calibration file keyword is provided, 'poa_calfos' will not abort. Instead, warning messages will be issued and the data in the output 'c3h' file will still be calibrated, but the additional post-COSTAR correction will not have been applied.