PHOTOSOL

STScI • ACDSD • MAST • CASB • Documentation

Products
GSC
DSS
GSPC
Science

Publications
Data Access
Related Science 
Missions
HST
GEMINI
VLT
NGST
Virtual Observatory
XMM
Facilities
Plate Scanning
COMPASS ooDB 
Staff Pages

Last Updated Jan 2001

Copyright © 2001 The Association of Universities for Research in Astronomy, Inc. All Rights Reserved.

Purpose :

Compute a Photometric Plate Solution (Modified 5/99).

Procedure :

This program uses a photometric catalog (either GSCP-I, GSPC-II or TYCHO) and for these stars reads the integrated density (Gaussian volume) on the plate, pixel coordinates and reference magnitude from the MCO file. It then computes the Chebyshev coefficients from a least-squares method of the log of the integrated density (Gaussian volume) against the reference magnitude. More detailed information can be found in
  • photosol .
    • There are a number scenarios where the program will abort gracefully. If this happens an abort file region_name_photosol.abort will be created and a message as to why the program aborted is written to the file. The existence of this abort file is useful when running a script. The script can check for the existence of the abort file before proceeding with the next task. A few reasons why photosol will abort are: the MCO, GSH or color tables are missing; the MCO file does not contain the desired magnitudes (at least B and V or V and R), there are not enough reference to perform the reduction, or important data could not be read from the GSH file.
    Major Changes with version 3.2(5/99)
  • Since the general function was in practive difficult to use we added the option to artificially add 2 points at the plate limit. These objects are assign the a minimum integrated density for the quoted plate limit from literature. The fainter of the two has an integrated density based on the minimum for deblended objects with a magnitude of the plate limit - delta error in magnitude. The bright of the two has an integrated density from intentory with a magnitude of the plate limit + delta error in magnitude.
  • A number of addition vaules are written to the GSH header in order to support the determination of the photometric error by APPLY_CALIB. One of these values is the locations of the sequences used in the calibration. There will probably be cases where the location can not be found - this is when the GSPC-II sequence is not found in the sequnece location table. If this happens it culd be due to an old naming style (rerunning findcat for GP2 should fix it) or it might be a case where the sequence is not in the table and it should be. Please contact Jane Morrison when this happens and she will update the table.
  • A few bugs where removed

    • Major Changes with version 3.1 (2/99)
  • The option to reject stars interactively is possible by using the qualifier /interactive.
  • A general function exists for several plate types (later all plate types) which was created by Jorge Garcia. The general function is used to reject objects fall from the function and also defines photometric the function in the faint extrapolated region of the photometric reference stars
  • Color plotting was added to the rejected objects. When plotting to the screen color is the only option. However, if color is desired on the postscript files then the qualifier /COLORPS must be used.
  • By default only the first three orders of the photometric solution are plotting to calibration plot. However, the program calculates up to the 6th order. If a solution of an order higher than 3 is desired to be plotted on the graphs, then use the qualifier /Porder = up to order 6.
  • A plot was added to show which stars are rejected because their colors where not on the color tables. These are usually stars of extreme color. On magnitude in the bandpass can be determined for these stars, so the Johnson V magnitude from the reference catalog vs photometric parameter is plotted

    •
    Major Changes with version 2.4 (9/98)
  • The following qualifiers have been deleted: /apply, /color_terms, /bvuse, vruse, /bv_color_table, /vr_color_table, /print.
  • The following qualifiers have been added: /color_table =(list of tables to be used) /Cweight and /ratio.
  • /rank_color_table=name of table.
  • The qualifiers defining the iteration were split up to be /iterate (logical), /delta_sigma=0.2, /max_iter = 20 and /sigma = 2.5.
  • Two plots were added: Location of the rejected stars on the plate and the residuals vs. magnitude (reference magnitude in the bandpass of the plate). In order to keep similar plots together a few of the plots were reorganized and a 6th plot was added.

    •
    General Function
    The general function is used to reject "bad" objects which fall far from the general function and also to defined the photometric calibration to be used in the faint extrapolated region of the photometric reference star calibration. The qualifiers that define the general function are:
    /GFREJECT (true by default) reject objects which are /GFDELTA_MAG from the general function solution.
    /GFDELTA_MAG is actually read in from the general function table, but the option exist to supply the value by the CLD.
    /GENFUN (by default true) use the general function and find the shifted function valid in the faint extrapolated region.
    /GFSHIFTPT is used in shifting the general function to fall as close as possible to the photometrically determined calibration line. Location where to attach the general function so it is continuous with the photometrically determined solutions = minimum of reference star photometric parameter (log integrated density) + /GFSHIFTPT.
    The general function can be used to reject start "far" from the function, with the idea being that these objects are probably junk of some sort or mismatches. In order to reject objects a first preliminary first-order least squares fit is performed. Using this solution the general function is shifted so it is continuous at a designated region (the minimum of the photometric parameter value + /GFSHIFTPT value). After this rough zero point shift objects are rejected if they fall above/below a certain delta magnitude from the 1st order polynomial. The delta magnitude is read in from the general function file or can be changed by the user by the qualifier /GFdelta_mag. Also the general function can be used to define the photometric calibration in the faint extrapolated region of the photometric reference stars. If this function is desired, /GENFUN is set true (which it is by default). After the photometric calibration is determined, the general function is again shifted so it is continuous with the photometric solution (determined by the reference stars) at the point defined by the minimum of the photometric + /GFSHIFTPT. The shifted general function Chebyshev polynomial is then written to the GSH file.
    How the iteration works
    There are 3 command line qualifiers that control the iteration SIGMA, DELTA_SIGMA, MAX_ITER
    MAX_ITER: is the maximum number of times to iterate.
    SIGMA : is the number of standard deviations the object can be away from the sigma of fit and not be rejected.
    DELTA_SIGMA : if (last sigma from fit - this sigma from fit) < DELTA_SIGMA stop iterating.
    The default values were as follows (check current CLD for values):
    DELTA_SIGMA: 0.2
    MAX_ITER : 20
    SIGMA : 2.5
    Example on how it works in practice
    A fit is preformed and the sigma_from_fit is found (from least squares). For each object a delta magitude is determined (magnitude in bandpass - mag from fit)
    If delta mag > SIGMA * sigma_from_fit reject object.
    If (sigma_from_last_iteration - sigma_from_this_iteration) < DELTA_SIGMA stop iterating.
    If number of iterations > MAX_ITER - stop iterating.
    The iteration qualifiers correspond to how plate_solution iterated and reject stars. The program iterates until the iteration number is less than max_iter and the (previous formal sigma - this formal sigma) < deltasig. A star is rejected when its residual is > /sigma* formal sigma. 
    Notes on Color Tables
    The qualifier /color_table= (list of color tables) is used to specify which color tables are used. If it is not used then the default values are used. Caution if you use this qualifier then the default values are not used anymore and only those tables given are used. Therefore this qualifier has also taken over the job of the /bvuse and /vruse qualifiers. For example, if the user does not want to use BV colors then they should use /color_table = VR_color.table and only the VR color table is used. /rank_color_table = rank color table is a table that is used when a star has more than one color available (it has more than 2 magnitudes). This table gives a rank for each possible color for each bandpass. The idea is to give the highest weight to the color closest to the bandpass of the plate. Only the color closest to the bandpass of plate (ie the one with the highest weight) is used in calibration, the other ones are given a weight of 0 and essentially ignored. This table is an ASCII table. Read the first few lines at the top if the top it you want to create your own table. If so it is probably best to copy the present table and in your own personal copy edit the rankings or to suit your case.
    Check the list of qualifiers given below, if your still confused then contact Jane Morrison.

    Calling Sequence :

    PHOTOSOL MCO filename (*.MCO)
    [/CATALOG = 3 letter designation for reference catalog]
    Use either GPC, GP2, or TOC {def = GPC}
    [/CATLIST=list of 3 letter reference catalogs] GP2, TOC
    This allows you to read in multiple catalogs from an MCO file

    Input and Output files :

    [/[no]NEW ]
    Create a new DID file
    [/[no]ANNOTATE]
    Write to DID file
    [/[no]UPDATE_GSH]
    Update GSH calibration keywords
    [/[no]DUMPFILE=name]
    {def=region_name.PHOTO_OUT_catalog_names}
    [/[no]IDFILE=name]
    {def=region_name_ID.table}
    [/USERMCO=filename]
    [/COLOR_TABLE = List of color tables]
    {default= use the BV and VR color tables in task_pars. These are defined by the the qualifiers /BV_COLOR_TABLE, /VR_COLOR_TABLE and /TYCHO_COLOR_TABLE}
    If /COLOR_TABLE is used then only the color tables given will be used no defaults are used. This allows the user to eliminate certain colors from being used by not specifying the color table.
    [/RANK_COLOR_TABLE = {default = taskpars:rank_colors.table}
    Table contains how to rank the colors in each bandpass
    [/GFUNCTION = {default =taskpars:general_function_2_99.table]
    General function table
    [/PLATE_LIMIT_TABLE= {default =taskpars:plate_limit_4_99.table]
    Plate limit table
    [/SEQUENCE_TABLE= {default =taskpars:gspcII_sequence_info_4_99_hand.table]
    GSPC II sequence table
    [/VTABLE]
    Not used yet (it will be the vignetting table).

    Faint end of extrapolation :

    [/ARTIFICIAL]
    Force the solution to go use the weighted artificial points

    General function variables :

    [/[no]GENFUN {def= GENFUN}]
    Write the general function coefficients in the extrapolate region
    [/GFSHIFTPT {def = 0.1}]
    Amount to offset from last valid point to determine the region where the zero point shift is determined.
    [/GFDELTA_MAG= n]
    /GFDELTA_MAG= n, where n is a tolerance in magnitude (+/-) above which an object is rejected.
    [/[no]GFREJECT {def = /GFREJECT}]
    Reject stars using the general function

    Controls iteration :

    [/[no]ITERATE=
    Iterate - rejecting stars (yes/no)
    [/DELTA_SIGMA= {def = 0.2}
    Continue iteration until last sigma - this sigma le Delta_sigma
    [/MAX_ITER= {def = 20}
    Maximum number of iterations allowed
    [/SIGMA= {def = 2.5}

    Type of Calibration :

    [/ORDER = valid orders 1,2,3]
    {default = 2} Order of Chebyshev polynomials
    [/EXTRAPOLATION
    Extrapolation mode of Chebyshev Polynomials outside fitting range
    (Valid Extrapolation modes- AV,TN,No, {default = AV})
    [/PHOTO_TYPE= density or Gaussian_volume ]
    {default=density}
    Which type of photometric parameter to use. Only options are density (integrated density or Gaussian_volume. At this time (2/99) the general function can only be used with integrated density.
    Resulting photometric coefficient name: PMD (density)
    Gaussian volume is an experimental variable. If you are having problems with using Gaussian volume as your photometric calibrator check to make sure the MCO file has Gaussian volume. Also obviously objects will have a Gaussian volume only where the centroider converged.
    [/CWEIGHT {default = true}]
    Cweight stands for catalog weight. If more than one reference catalog is used then stars are weighted according to which catalog the are from. The weight assigned is based on the range of magnitude of the reference stars/number of reference stars from each catalog. In general, this weighting was chosen to give the GSPC stars a higher weight over the TYCHO stars.}
    [/[no]VIGNETTING]
    {default Vignetting] Apply (or not not apply) the vignetting correction.

    Properties of valid objects :

    [/REJECT=(number_reject_1, number_reject_2, ....)]
    Specify stars to be rejected by the index number from TABLE 1 produced by photosol.
    [/XRANGE=(lx,ux)]
    Limit stars used to this specific region ( x limit)
    [/YRANGE=(ly,uy)]
    Limit stars used to this specific region ( y limit)
    [/MAGRANGE=(lm,um)]
    Limit stars used to this specific region (catalog magn)
    [/PAR_RANGE=(ly,uy)]
    Limit stars used to this specific region (photometric parameter range)
    [/[no]ONLY_STARS]
    {default only stars}] Reject objects based on classification
    [/[no]BLEND
    {default noblend}] Reject objects which were blended
    [/[no]GVOL
    {default gvol}] Reject objects with invalid Gaussian volumes
    [/[no]RATIO]
    {true by default}
    Rejected objects base on the ratio test of their area_above_threshold values.

    Plotting :

    [/[no]PLOT ]
    {default= false : do not plot to the screen}. If /plot is used then the plots are printed to the screen}
    [/PORDER ]
    {def=3} Order of polynomials to plot
    [/XPMIN = lx]
    {def=4} X min for plotting
    [/XPMAX = rx]
    {def=8} X max for plotting
    [/YPMIN = ly]
    {def=7} Y min for plotting
    [/YPMAX = uy]
    {def=24} Y max for plotting
    [/TEST_POINT = x]
    {def=5.25} test value for comparing results

    Misc :

    [/COMMENT="String"]
    write to DID file
    [/VERBOSE]
    Print out extra information to working window and plots.
    [/DEBUG]

    Availability :

    • GSSS AXP - yes
    • GSSS VAX - yes
    • PIXELS - NO
    REJECTION CODES.
    There are a number of reasons a star can be rejected and not used in the calibration. If a star is rejected then a reject code is assigned to the star and printed in table 2 (under the column DROPPED). For the most updated version look in photosol_local.inc. Description of codes is as follows:
    -1 = Object has an extreme color and is off the color table.
    -2 = The object does not have the required magnitudes to form the desired color
    -3,-4,-5 = held for other color problems
    -6 = Rejection based hand edited flag in the MCO file
    -7 = Rejection based on classification. The qualifier /ONLY_STARS is present by default, so only objects classified as stars are used in the calibration. To turn this qualifier off type: /NOONLY_STARS.
    -8 = Rejected because the photometric parameter (integrated density or Gaussian volume) was not valid.
    -9 = Rejected by hand. (Rejected using the qualifier /reject = )
    -10 = Rejected because it was out of range. Either not in the x,y range or not in the magnitude range specified by /XRANGE, /YRANGE, and /MAGRANGE, respectively.
    -11 = Rejected because the delete flag was set.
    -12 = Rejected because object failed the ratio above threshold test.
    -13 = Rejected because object fell out of range of photometric parameter .
    -14 = Rejected because object was a blend.
    -15 = Rejected because Gaussian volume was invalid.
    -16 = Rejected based on general function (fell far from general function).
    -17 = Rejected interactively.
    > 0 = Rejected in least squares iteration. Number corresponds to iteration rejected.

    History/Author :

    • Original Implementation-Helmet Jenkner for GSC I processing in 1989
    • 3/97 adaption by Jane Morrison adaption to GSC II processing
    • 12/97- Present Improvements by Jane Morrison see description of modifications above.
    • Added CATLIST to program. This allows the user to select more than one reference catalog.
    • Added USEBV and USEVR. These qualifiers allow the user to eliminate reference stars of a certain color.
    • rank_Color added. If a star has more than one color then this qualifier allows the user to weight a star according to it color. Weight_color refers to the color closest to the bandpass of the plate. The other color is given a weight of (1-weight_color).
    • Took out the postscript version of table 1 and 2 and made them separate ASCII files.
    • Added a check if a line written to the did file is too long (usually this is the command line). If it is then it is broken up into lines of n length.
    • Added the abort file.
    • 4/98 re-organization of photosol to support NT port of VMS photosol.
    • 9/98 M. Jackson added a number of new features to photosol to make it more general and support the NT porting these include: weighting of reference catalogs, how the color tables are used (added the /color_table qualifier), and weight certain colors from a single star higher than other based on the weight_color.table.
    • 9/98 J. Morrison changed the way photosol iterates and reject stars, added a new plot (residuals vs. magnitude) and made a few other changes to make the porting of VMS to NT easier. If the photometric parameter is Gaussian_volume, then the plots and tables produced have GV in their extension name. From J. Garcia's suggestion a qualifier was added to reject stars based on a ratio of area_above_threshold values.
    • 2/99 and 5/99 updates done by J. Morrison (look at the top pf the documenation page to see what was done.

    Items to be added in the future :

  • Add a vignetting table

    •