Go to: ​http://svn.pan-starrs.ifa.hawaii.edu/trac/ipp/wiki/Zoology

This is a wiki describing the false detection zoology tools/methods. This is NOT the page cataloging what we find.

Everything at the moment lives at /data/ipp009.0/heather. We can move this around or add to it.

Files in the directory (useful for creating region files, or further investigating interesting candidates)

• heather.blog.txt -- some of what I've done. Not complete.

• sas_skycells.sql
• sas_skycells.txt --these are list of sas footprints (filter, skycell, stack_id). Used by my perl scripts

• footprint-?-not-matched-psphot.fits -- these are the lists of false detections
• g.txt, r.txt, i.txt. z.txt, y.txt -- these are the lists of false detections in txt format (limited number of columns)

• make_region.pl -- perl script to make the region files. You may tweak this to get better (different) region files (example- changing the color or information displayed)
• grab_stacks.pl -- this just grabs the stacks and puts them in stack_id directories. Should only need to be run once.

• stack_id directories -- these contain the stacks (everything). Likely you will only need to look at the unconv.fits.
• *reg files -- these are the region files for ds9.

how to load up ds9 with regions

go to /data/ipp009.0/heather directory, then run something like this (choose the appropriate stack_id):

ds9 336341/RINGS.V3.skycell.1316.079.stk.336341.unconv.fits -region 336341.reg 

Note: to generate the commands interesting a filter:

/data/ipp009.0/schastel/sas_commands.sh <filter>

e.g.:

/data/ipp009.0/schastel/sas_commands.sh g

This probably loads too much information. You can turn off the text by going to Region -> Region Parameters -> Show Region Text

By default, I've just loaded the region files with the idet column (the object's ssn), to make it easier to investigate.

You can do screen grabs with gimp, or by zooming in and going to File -> Save Image.

Database containing detections features (see Action A-002)

It's on neverland.

CREATE DATABASE Zoology;
USE Zoology;

CREATE TABLE DetectionFeatures (
  id            INT NOT NULL PRIMARY KEY AUTO_INCREMENT
  , stack_id    BIGINT(20) NOT NULL
  , INDEX(stack_id)
  , filter      CHAR(10) NOT NULL
  , INDEX(filter)
  , skycell_id  CHAR(46) NOT NULL
  , INDEX(skycell_id)
  , idet        CHAR(18)
  , x_psf FLOAT
  , y_psf FLOAT
  , x_psf_sig FLOAT
  , y_psf_sig FLOAT
  , posangle FLOAT
  , pltscale FLOAT
  , psf_inst_mag FLOAT
  , psf_inst_mag_sig FLOAT
  , psf_inst_flux FLOAT
  , psf_inst_flux_sig FLOAT
  , ap_mag FLOAT
  , ap_mag_raw FLOAT
  , ap_mag_radius FLOAT
  , peak_flux_as_mag FLOAT
  , cal_psf_mag FLOAT
  , cal_psf_mag_sig FLOAT
  , ra_psf DOUBLE
  , dec_psf DOUBLE
  , sky FLOAT
  , sky_sigma FLOAT
  , psf_chisq FLOAT
  , cr_nsigma FLOAT
  , ext_nsigma FLOAT
  , psf_major FLOAT
  , psf_minor FLOAT
  , psf_theta FLOAT
  , psf_qf FLOAT
  , psf_qf_perfect FLOAT
  , psf_ndof INTEGER
  , psf_npix INTEGER
  , moments_xx FLOAT
  , moments_xy FLOAT
  , moments_yy FLOAT
  , moments_m3c FLOAT
  , moments_m3s FLOAT
  , moments_m4c FLOAT
  , moments_m4s FLOAT
  , moments_r1 FLOAT
  , moments_rh FLOAT
  , kron_flux FLOAT
  , kron_flux_err FLOAT
  , kron_flux_inner FLOAT
  , kron_flux_outer FLOAT
  , flags BIGINT(64)
  , flags2 BIGINT(64)
  , n_frames INTEGER
  , padding INTEGER
  , ps_zp DOUBLE
  , exptime DOUBLE
)

Files used to populate the database (fitsExtract.py is located in ~ipp/sch/zoology)

python fitsExtract.py /data/ipp009.0/heather/footprint-r-not-matched-psphot.fits > data/populate.r.sql
python fitsExtract.py /data/ipp009.0/heather/footprint-g-not-matched-psphot.fits > data/populate.g.sql
python fitsExtract.py /data/ipp009.0/heather/footprint-i-not-matched-psphot.fits > data/populate.i.sql
python fitsExtract.py /data/ipp009.0/heather/footprint-z-not-matched-psphot.fits > data/populate.z.sql
python fitsExtract.py /data/ipp009.0/heather/footprint-y-not-matched-psphot.fits > data/populate.y.sql

Division of labor

we each should take a filter. Ken's priority is g and y, as those are the messiest. the reasons for false detections are likely filter dependent.

• heather : g
• serge :
• roy :

After those three, we can arm wrestle for what's left ;)