P h o t o m e t r y   O b s e r v a t i o n   G u i d e l i n e s 


     (An observation = 1 frame in each of V, R, and I)


STANDARDS

     Aim for 20+ observations of standards for a night; these must have good
     counts (say 25K or more) since they set the exinction and transformation 
     coefficients that apply to all unknowns.

     Bright standards may be done as soon as it is too dark for flats and
     before it is dark enough for parallax regions.

     5-6 observations of 1 or 2 of the very red ones (W359, VB8, GL866) are
     needed in order to get second-order transformation terms. A good way is 
     to do one twice, a set immediately followed by another set, then again 
     an hour or so later if possible.

     The red/blue extinction stars (B-V > 1.2) must be followed from the 
     meridian to an airmass around X = 2 (or a few tenths greater than the
     largest program airmass).


UNKNOWNS

     Should be done on the meridian if possible, but certainly at airmass
     smaller than 1.5 (unless there are some special, desperate reasons for
     going to higher X - since the extinction coefficient is multiplied by
     the air mass in the reductions, the smaller the airmass, the smaller
     the influence of the exinction coefficient and its error).

     On a night of photometry, when doing a parallax region always do a
     frame or two of the other color - then that too could be reduced for
     magnitude and colors at little cost in observing time.

     Unknowns should be observed on three separate nights to check for
     reliability in the photometry and possible variability - single
     observation on one night should never be considered adequate for
     publication.


PRINCIPLES AND PRACTICE

     Differential photometry is relative photometry, and you can do it
     in small fields like ccd fields or over small disstances of arc -
     people who work on eclipsing binaries, e.g., use a standard very
     close by, and just get delta m's between the variable and std. If
     the extinction coeficient is, say, 0.1 magnitudes, in some color,
     and if you observe a star at an airmass say only a tenth of an
     airmass different from your standard star, then the effect of the
     extinction is only 0.1 x 0.1 or 0.01 magnitudes. Thus the
     magnitude of your star is just the delta m between it and your
     standard added with the mag of the standard, or more succinctly,
     mag(unknwn) = mag(std) + delta mag + 0.01. And if the color of
     the two stars is very similar you don't have to worry about
     getting onto the standard system (but if there were a large
     color difference between standard and unknown there might also be
     a small correction tht is a function of color owing to different
     color senstivity of detector + filter + telescope + site).

     So the reduction is just some simple arithmetic if you know the
     extinction coefficient and any color correction, and if the
     differences in airmass and color between standard and unknown are 
     small, you can't make much of an error, even if the extinction is
     not well determined. 

     But real men do all sky photometry, and that is usually what is
     meant when you say photometry, and if you do not mean that you (I)
     ((a few others at least)) say "differential photometry". That
     means standards all over the sky, at different sometimes large
     airmass differences (as large as 1. say), and then you need
     enough standards to reduce your data to some level of accuracy
     while getting or assuming extinction coefficients, and getting or
     assuming the polynomial coefficients needed to correct the
     instrumental magnitudes to "the" standard system. At the 1-meter
     we really need to have several measurements of one or more
     extremely red stars to make it all work properly. 






A p e r t u r e   P h o t o m e t r y   R e d u c t i o n s       


BIAS SUBTRACTION

     CCD #6 has no overscan region, so bias subtraction must be done manually.  
     Each set of frames must be preceded and followed by a separate bias frame.  
     Use MISTAT to find the mean counts for each bias frame, then for each bias 
     "pair", find the average of the two means and use ICSUB to subtract that 
     value from each of the frames in the set bracketed by the pair.


OTHER PREPROCESSING

     The remainder of the preprocessing proceeds as for astrometry, up to and 
     including RASTAT.  However, during the CPOS phase, be sure to include a
     few extra well-exposed stars wherever possible, especially when the object
     of interest is faint.  This will help later in judging the "curve of growth"
     with increasing apertures, and in the case of faint objects, is essential to 
     making aperture corrections.


DOFOTO

     At this point, each .DST frame should have a corresponding .CEN file.  Use
     the routine DOFOTO to obtain a set of corresponding .MAGS files.  DOFOTO
     is a front-end to the Figaro routine FOTO.  FOTO handles only one frame at
     a time and allows a maximum of 7 aperture sizes.  DOFOTO handles multiple
     frames, allows 2 or 3 passes through FOTO (i.e., 14 or 21 aperture sizes),
     and calls the C program APDIFFS, which lists the differences in magnitudes    
     between successive aperture sizes (type "dofoto" by itself for the syntax
     and usage details).  At this point, the option for a third pass through FOTO
     (apertures 15-21) should NOT be used (more on this later).  Obtain landscape
     hardcopies of all the .MAGS files with the routine MPRINT (mprint *.mags).


SELECTING APERTURE MAGNITUDES

     Here's where things can become somewhat subjective.  Go through each .MAGS
     listing and observe the curve of growth by examining the "Aperture Deltas".
     Here is a sample (truncated--it actually extends out to 14-13):

          *** Aperture Deltas ***
      
     Object    2-1    3-2    4-3    5-4    6-5    7-6    8-7    9-8   10-9  ...
      
          1 -0.978 -0.304 -0.125 -0.057 -0.029 -0.019 -0.011 -0.006 -0.005  ...
          2 -0.914 -0.295 -0.118 -0.055 -0.028 -0.017 -0.011 -0.007 -0.005  ...
          3 -0.892 -0.306 -0.121 -0.058 -0.031 -0.018 -0.011 -0.007 -0.005  ...
          4 -0.914 -0.297 -0.119 -0.054 -0.029 -0.017 -0.011 -0.008 -0.005  ...
          5 -0.894 -0.307 -0.121 -0.055 -0.029 -0.019 -0.012 -0.009 -0.004  ...

     Notice that as the aperture sizes increase, the differences in magnitudes
     between successive apertures decreases.  Choose the aperture for which
     the delta is less than or equal to -0.010.  In this case, that would be
     aperture 9.  In this example, the only object of interest is object 1.
     However, by selecting additional objects as suggested above, it's possible
     to see that the curve of growth is fairly consistent for all the stars       
     chosen.  On the hardcopy, circle in pen the magnitude listed for the object
     of interest under the appropriate aperture heading.


APERTURE CORRECTIONS

     Here's another (truncated) sample:

     *** Aperture Deltas ***
      
     Object    2-1    3-2    4-3    5-4    6-5    7-6    8-7    9-8   10-9  ...

          1 -0.901 -0.301 -0.105 -0.044 -0.024 -0.015 -0.011 -0.008 -0.010  ...
          2 -0.916 -0.283 -0.099 -0.044 -0.022 -0.014 -0.008 -0.006 -0.005  ...
          3 -1.125 -0.304 -0.111 -0.047 -0.024 -0.014 -0.009 -0.007 -0.005  ...

     In this case, things aren't so consistent--the curve of growth for object 1
     is not keeping pace with the other two.  It would be tempting to choose
     aperture 9 and move on, however, the "majority rule" would suggest aperture
     8 instead.  By inspecting the actual magnitudes (or by foreknowledge of the
     situation), the discrepancy can be attributed to the relative faintness of
     object 1.  In this case, a magnitude correction is probably called for. The
     steps for making the correction are as follows:

          1) Find the "majority rule" aperture (this may often be dictated by
             a single well-exposed star, chosen expressly for this purpose).
             Call this apX.

          2) Find the aperture size that best represents the FWHM radius of the 
             image.  Call this apY.  For example, with a pixel size of 0.57 arcsec, 
             if the recorded seeing value is 2.0 arcsec, that makes a FWHM diameter 
             of 3.5 pixels, or a FWHM radius of ~1.8 pixels, so choose apY = 2.

          3) For all the well-exposed objects (i.e., those used in making the
             "majority rule" determination), calculate mag(apX) - mag(apY), and
             then the average of these deltas.

          4) Add this average delta value to the mag(apY) of the object of interest.
             The result is the corrected magnitude.

          5) Record the result and the calculations involved in pen on the .MAGS
             hardcopy for that frame.


THE "3" OPTION

     In the event of especially poor seeing, the aperture deltas may not "bottom out"
     to -0.010, even with an aperture radius of 14 pixels.  In that case, it's possible
     to extend the aperture range by rerunning DOFOTO on the frame(s) involved using
     the "3" option.  This a judgment call.  If extending the range by only one or two
     more apertures helps, it may be ok to use the result, but data with such poor seeing
     should probably be called into question.  The "3" option may perhaps be useful only
     to gauge the severity of the situation.
    

THE CCDPHOT INPUT FILE

     Make a copy of the obslog file from the observing run involved, and edit it so
     that it contains only the information for the photometry frames.  Sets of frames
     (i.e., V, R, I frames at the "same" airmass) should be separated by blank lines.  
     For example,

     240     4.0 10:56:17  +07:01:59  10:52:22.5  1.2739 W359     PBL 240 I 4
     242    20.0 10:56:17  +07:01:59  11:00:25.3  1.2739 W359     PBL 242 R 20
     243    80.0 10:56:17  +07:01:59  11:04:46.6  1.2746 W359     PBL 243 V 80
      
     262    11.0 10:56:17  +07:02:02  13:02:42.4  1.5141 W359     PBL 262 I 11
     263    55.0 10:56:17  +07:02:03  13:06:41.0  1.5321 W359     PBL 263 R 55
     264   240.0 10:56:17  +07:02:03  13:09:44.5  1.5466 W359     PBL 264 V 240
      
     220    15.0 09:23:25  -45:24:06  09:27:12.1  1.0310 E4-A     PBL 220 I 15
     222    12.0 09:23:25  -45:24:06  09:31:44.8  1.0313 E4-A     PBL 222 R 12
     223    12.0 09:23:25  -45:24:06  09:34:16.7  1.0316 E4-A     PBL 223 V 12

     Use the routine PREPHOT to create a template for the CCDPHOT input file.  Let's
     say the edited obslog copy is called "obscopy" and we want to call the CCDPHOT
     input file "pbl2.dat".  The template is created with the command

          prephot obscopy pbl2.dat 

     The resulting template will look like this:

     nn, 1.2741, vv.vvv,  80, rr.rrr,  20, ii.iii,   4, bb.bbb,   0, 'W359'
     nn, 1.5309, vv.vvv, 240, rr.rrr,  55, ii.iii,  11, bb.bbb,   0, 'W359'
     nn, 1.0313, vv.vvv,  12, rr.rrr,  12, ii.iii,  15, bb.bbb,   0, 'E4-A'

     Edit this file as follows:

          1)  segregate the standards from the unknowns, with the standards
              at the top.

          2)  replace the fields "nn", "vv.vvv", "rr.rrr", etc., with each object's
              index, V magnitude, R magnitude, etc., respectively.

          3)  insert at the very top a line of the form

                   'RUN#:DDMONYY', NSTDS

              where "NSTDS" is the number of standard observations, for example

                   'PBL2:03MAR96', 016

          4)  insert above the start of the unknown observations a line containing the
              number of unknowns, for example

                   08


CCDPHOT 

     In the directory containing the CCDPHOT input file, make a copy of the file
     ~/dat/foto/stds.dat.  You are now ready to run CCDPHOT.  Type

          ccdphot

     and reply to the prompts.