HOW-TO VODIA (OLD)¶
Introduction¶
VODIA (VST OmegaCAM Difference Image Analysis) is a package optimized to detect small photometric variations in crowded fields. The software is based on the DIA package (Difference Image Analysis, written by Woźniak 2000, (Acta Astron. 50, 421; see also Woźniak et al. 2002, Acta Astron. 52, 129) that makes use of the “optimal PSF matching algorithm” with a space-varying kernel (Alard & Lupton 1998, ApJ 503, 325; Alard 2000, A&AS 144, 363). The software includes 6 (+1) independent C programs (corresponding to 6 different steps) and does not use any external library. The inspect method makes use of a fortran program for preliminary analysis of the results. The different steps are:
- wcs2pixThe astrometric solution is used to aligne (and cut accordingly) the scientific images.
- MstackTo obtain the reference frame from the stacking of the best seeing images.
- GetpsfExtract PSF from Reference to extract PSF photometry from the difference images.
- AgaA convolution kernel is found and used to create the difference images.
- GetvarDetects variable candidates.
- PhotProduces light curves of variable candidates.
- 6a.
- PhotRefProduces light curves of all objects detected in the reference image (the user can choose between Phot and PhotRef depending on the scientific goal).
Astro-WISE implementation¶
VODIA has been incorporated into Astro-WISE by A. Volpicelli and F. Getman. This is done by providing a Python wrapper around the C programs which stores the input and output to the Astro-WISE database/data storage. One FORTRAN program is used only by the inspect method for preliminary analysis of the results.
Compiling and installing the C code¶
The VODIA code can be found in Astro-WISE CVS at:
opipe/Experimental/VODIA/VODIA.tar.gz
Extract this archive to whatever directory you like. To compile run ‘make’. To install: ‘make install’. Installation will copy binaries to user bin directory or you can copy them to system directory manually. Up to now, the code is tested for 32-bit machines only.
Running VODIA¶
[import all python scripts]
awe> from astro.main.VariabilityFrame import VariabilityFrame
awe> from astro.external import Dia
awe> from astro.main.DiaConfig import DiaMstackConfig,DiaGetpsfConfig,
DiaAgaConfig,DiaGetvarConfig,DiaPhotConfig
[select science frames]
awe> q = ( RegriddedFrame.OBJECT.like('projectname')) &\
( RegriddedFrame.filter.name == '#842' ) &\
( RegriddedFrame.chip.name == 'ccd50')
awe> regs = [reg for reg in q]
[creates variability object]
awe> v = VariabilityFrame()
[put our list of images as input]
awe> v.regridded_frames = regs
[create the list of images to produce the reference (optional)]
awe> q = ( RegriddedFrame.OBJECT.like('projectname')) & \
( RegriddedFrame.filter.name == '#842' ) &\
( RegriddedFrame.chip.name == 'ccd50') &\
( RegriddedFrame.psf_radius < 0.6 )
awe> v.regridded_frames_for_reference=list(q)
[run VODIA (all programs automatically)]
awe> v.make()
[to interactively check the results]
awe> v.inspect()
[Example of changing one parameter (C_MIN=detection threshold) and run again
getvar and phot]
awe> print v.getvar_config.C_MIN
awe> v.getvar_config.C_MIN = 0.9
awe> v.make_getvar()
awe> v.make_phot()
[check again new results]
awe> v.inspect()
[storing the file with the light curves]
awe> v.store()
[commit the results to database]
awe> v.commit()
The output consists of one ASCII file with N lightcurves (N being the number of sources analyzed).