Creating a new .axy

[Paulina Lira]

Hello, I need help creating a new xy list from a Sextractor catalog. I have a series of images with pretty good WCS solutions but I need to adjust the astrometry a bit in order to do image subtraction. Using the index files provided by astrometry.net does not work as only a few stars are found in the ~12' FOV. 

So what I want to do is to use the WCS coordinates that Sextractor determines for one image to adjust the solution of the other until both images have a good match. Just giving the Sextractor catalog the extension .axy does not seem to work since solve-filed still wants to use the index files installed with astrometry.net.

Reading here and there it seems that I need to add ("augment") the stars coming from the index files with those I get from Sextractor. However I don't understand how this works since the WCS solutions of the two catalogs might not coincide. Also, I do not know what command line I need to use to tell solve-field to do this. 

Cheers,

Paulina

[Dustin Lang]

Hi,

Astrometry.net should work fine for images 12 arcmin across.  Which index files do you have?  I would recommend the 5200-series (Gaia) ones, probably scales 02 and 03.

See

http://data.astrometry.net/

and follow the link for "5200-series, LIGHT verison"

you want index-5202-* and index-5203-*.

eg, to download them all,

for ((i=0; i<48; i++)); do

    I=$(printf %02i $i)
    wget https://portal.nersc.gov/project/cosmo/temp/dstn/index-5200/LITE/index-5202-$I.fits
    wget https://portal.nersc.gov/project/cosmo/temp/dstn/index-5200/LITE/index-5203-$I.fits
done

Or you can look up which healpixel your part of sky is in, with this map:

https://github.com/dstndstn/astrometry.net/blob/main/util/hp2.png

If you really want to take your SourceExtractor catalog and build a custom index file from it, you should follow these instructions:

http://astrometry.net/doc/build-index.html

cheers,

--dustin

cheers,

--dustin

--
You received this message because you are subscribed to the Google Groups "astrometry" group.
To unsubscribe from this group and stop receiving emails from it, send an email to astrometry+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/1dd21489-a886-497c-9659-ef97e0547fbcn%40googlegroups.com.

[Paulina Lira]

Hi Dustin, thanks for checking on this.

So, I think I was using the wrong index files (5000 series).

But  I followed the instructions to construct a new index file with Sextractor output and that worked really well. BUT, still the solutions seem not good enough and the subtracted images show lots of dipoles at the position of stars and some galaxies. I include the run below:

solve-field J2211_4.fits --backend-config ../astrometry_config/astrometry.cfg --no-plots --no-remove-lines --uniformize 0 --crpix-center --depth 30,60,90,120,150,180,210,240,270,300,330,360,390,420,450,480 --cpulimit 360 --tweak-order 3 --scale-low 0.34299999999999997 --scale-high 0.357 --scale-units app --ra 332.84792 --dec -1.04028 --radius 0.15 --new-fits none --overwrite --out J2211_4 --dir .

Found an existing WCS header, will try to verify it.
Extracting sources...
simplexy: found 1164 sources.
Solving...
Reading file "./J2211_4.axy"...
Only searching for solutions within 0.15 degrees of RA,Dec (332.848,-1.04028)
Verifying WCS using indices with quads of size [1.16036, 11.6036] arcmin
Verifying WCS with index 1 of 1 (/home/plira/lilith/projects/sibling/analysis/eridanus/speculoos/astrometry_config/J2211_index_file.fits)
Verifying WCS of field 1.
  log-odds ratio 176.135 (3.12202e+76), 114 match, 0 conflict, 1030 distractors, 123 index.
  RA,Dec = (332.85,-1.03618), pixel scale 0.348107 arcsec/pix.
  Hit/miss:   Hit/miss: ---------------------+------------+-+++++++++++++---+-+-+-+--++-----++-+-+---+--+-+---+---+---+-++--
 --> log-odds 176.135
Got 1 solutions.
Field 1 solved: writing to file ./J2211_4.solved to indicate this.
Field: J2211_4.fits
Field center: (RA,Dec) = (332.849916, -1.036473) deg.
Field center: (RA H:M:S, Dec D:M:S) = (22:11:23.980, -01:02:11.304).
Field size: 11.5826 x 11.5781 arcminutes
Field rotation angle: up is 179.753 degrees E of N
Field parity: neg

The huge number of detections is because there is a satellite trail in the image. Still, with 114 matches the solution should be pretty good, right? But it is not  :(

In fact, when running swarp on the images I get the warning: > WARNING: Astrometric approximation too inaccurate for this re-projection.

I will try with the GAIA index files now.

Cheers!

Paulina.

[Dustin Lang]

Are you sure that Swarp is correctly using the SIP distortion solution?  At some point in the past it did not understand SIP.

[Paulina Lira]

It is all wrapped around within zogy (https://github.com/pmvreeswijk/ZOGY), so I hope it is properly done!

If you have an example of a swarp run, that would be great. I will look myself too.

Paulina.

[Dustin Lang]

I don't use SWarp and have never heard of ZOGY.

For diagnosing whether the match is good, I would look at the "red-green" plot (the *-indx.png) plot -- if you have lots of red circles perfectly encircling green circles, then it's a good solution.  Based on the "hit-miss" meter, it looks like the brightest stars in your image are not well localized?  This can happen when there is saturation.  Sometimes --downsample 2 helps with that.  Or maybe that's the satellite trail.  Any way you can mask that out?

--dstn

[Paulina Lira]

Great tip, thanks!!! 

ZOGY = Zackay, Ofek & Gal-Yam, 2016, ApJ, 830, 27  (https://iopscience.iop.org/article/10.3847/0004-637X/830/1/27/pdf)

:)

Paulina

[Paulina Lira]

I downloaded the GAIA index files and it is working like a charm....

Happy 2022!!!

Paulina.

[Paulina Lira]

Hi Dustin,

I'm using astronmetry in a new set of images now and I've encountered the following problem:

solve-field J0249_2.fits --backend-config /home/plira/lilith/projects/sibling/analysis/eridanus/speculoos/astrometry_config/astrometry.cfg --no-plots --no-remove-l
ines --uniformize 0 --crpix-center --depth 30,60,90,120,150,180,210,240,270,300,330,360,390,420,450,480 --cpulimit 360 --tweak-order 3 --scale-low 0.8089799999999999 --scale-high 0.85902 --scale-units app --ra 02 49 13.00 --dec -08 15 25.6 --radius 0.5 --new-fits no
ne --overwrite --out J0249_2 --dir .
Reading input file 1 of 5: "J0249_2.fits"...

Found an existing WCS header, will try to verify it.
Extracting sources...

simplexy: found 2230 sources.
Solving...
Reading file "./J0249_2.axy"...
Only searching for solutions within 0.5 degrees of RA,Dec (2,-8)
Verifying WCS using indices with quads of size [5.56946, 55.6946] arcmin
Verifying WCS with index 1 of 1 (/home/plira/lilith/projects/sibling/analysis/eridanus/speculoos/astrometry_config/index-5202-16.fits)

Verifying WCS of field 1.

--> log-odds 0
Got 0 solutions.
Field 1 did not solve (index index-5202-16.fits, field objects 1-30).
Field 1 did not solve (index index-5202-16.fits, field objects 31-60).
Field 1 did not solve (index index-5202-16.fits, field objects 61-90).
Field 1 did not solve (index index-5202-16.fits, field objects 91-120).
Field 1 did not solve (index index-5202-16.fits, field objects 121-150).
Field 1 did not solve (index index-5202-16.fits, field objects 151-180).
Field 1 did not solve (index index-5202-16.fits, field objects 181-210).
Field 1 did not solve (index index-5202-16.fits, field objects 211-240).
Field 1 did not solve (index index-5202-16.fits, field objects 241-270).
Field 1 did not solve (index index-5202-16.fits, field objects 271-300).
Field 1 did not solve (index index-5202-16.fits, field objects 301-330).
Field 1 did not solve (index index-5202-16.fits, field objects 331-360).
Field 1 did not solve (index index-5202-16.fits, field objects 361-390).
Field 1 did not solve (index index-5202-16.fits, field objects 391-420).
Field 1 did not solve (index index-5202-16.fits, field objects 421-450).
Field 1 did not solve (index index-5202-16.fits, field objects 451-480).
Field: J0249_2.fits
Did not solve (or no WCS file was written).

Reading input file 2 of 5: "49"...
ERROR: Image type not recognized: Unknown image type "["cannot open `49' (No such file or directory)"]"
augment-xylist.c:590:backtick Failed to run command: /usr/local/astrometry/bin/image2pnm --infile 49 --uncompressed-outfile /tmp/tmp.uncompressed.fvud8s --outfile /tmp/tmp.ppm.nCLhaq --ppm --mydir /usr/local/astrometry/bin/solve-field
ioutils.c:568:run_command_get_outputs Command failed: return value 255

What I fail to understand is why it is looking for a solution around RA,Dec (2,-8) when the pointing of the image is --ra 02 49 13.00 --dec -08 15 25.6.

Any help is most appreciated!!

Paulina.

[Dustin Lang]

Hi,

If you want to give HH:MM:SS format for --ra and --dec, they MUST have the colons, not be space-separated.

ie,

--ra 02:49:13.00 --dec -08:15:25.6

cheers,

--dustin