CHOOSING REF STARS  (EHV 2021.0314)


1. In the star's parallax directory (in /regions/), display first image in
   frame 1, and recent image in frame 2. 
   Blink between the two in ds9 to see if the pointing is consistent.
   If there is a difference, leave up the image that best represents most of
   the images taken.

   MAKE SURE this image is in the pi filter (given in the observing list).
   You can do this in iraf; for example, with file '20210202.09.063.o.fits':
     hselect 20210202.09.063.o.fits '$I, filter2' yes
   Note that for this purpose, ov = nv = V filter.

2. Display the setup page from the telescope:
     gv /nfs/recons2/CTIOPI.setups/starname.point.ps &
   Leave this open, or at least take a screenshot of it.
   You will need to look at it side-by-side with the ds9 image.

3. In ds9, circle all the ref stars suggested in the pointing setup page. 
   These are the targets that will be taken into account by the observer: 
   they should ALWAYS be away from bad columns, not saturated, etc., so they
   are safe starting choices.

4. In ds9, circle additional potential ref stars. 

   * Do NOT choose any stars that are brighter than the "exposure" star in the
     pointing setup page (star with box around it).

   * Choose stars that are brighter than 5,000 counts.

   * If possible, choose stars >100 px from the frame edges, and ideally very
     close to the pi star.

   * If possible, do not choose any stars that overlap with other stars' PSFs
     (too close on the sky). This overlap will make their precise positions
     seeing-dependent (bad).

   * You must have >5 refs --- choose as many as possible here (>20 is ok!). 
     You can delete the extras later.

   * If possible, choose a ref set that is evenly distributed around the pi
     star. The PERFECT ref set would be a circle with pi star in the center.

B. (Bonus) Flip the ds9 image (Zoom -> Invert Y) and display the same field in
   Aladin with DSS2 and/or 2MASS images. Display the Gaia catalog on top of it. 
   Then select each of your potential ref stars in Aladin (one by one) to check
   their distances using the parallaxes. Parallax <3 mas is ok, <2 mas is ideal. 
   ==> If any of your circled refs fail this test, remove them from your set 
       (select their circles and press 'delete' on the keyboard).

5. Step back and assess your ref field. Does it look reasonable? If you have
   >20 refs, delete the worst ones until you have 20 or fewer. Refer to step 4!

6. Add numbers to the stars in the field to indicate the order of tagging. You
   can add numbers to circle regions in ds9 by double-clicking on each in turn,
   *or* you can change your region shape to Text and add text labels next to each
   circle.  Try to label them in a way convenient for tagging, like clockwise or
   counter-clockwise around the field.
   (Remember, everyone who reduces this field for years to come will have to
   tag them in this same order, unless they redo this ref-choosing process)

7. Save an image of the labeled field, either by taking a screenshot or using 
   File -> Print... -> check "File" option and input the path. Place this image
   in that star's directory on /nfs/recons4/CTIOPI/regions
   --> Ultimately, you will need to make a real setup page using 'setfield' in IRAF. 
       But that step cannot happen until you have started the first astrometry 
       reduction for this field, because the star.settings file must exist.