Azimuth/Elevation converter for Astrometry.net

Tips and techniques article, especially for DSLR citizen science data.

Get Astrometry.net ≥ 0.67 or use the astrometry.net cloud service.

python3 -m pip install -e .

The main program used is PlateScale.py. Auxiliary scripts are under scripts/

PlateScale.py

The main script most users would use to register a star field image to Azimuth and Elevation is "PlateScale.py".

The "--args" command line option allows passing through a variety of parameters to solve-field, which underlies this program. Type solve-field -h or man solve-field for a brief description of the nearly 100 options available.

Be sure to enclose the options in quotes. For example, to specify that the image field is at least 20 degrees in extent:

python PlateScale.py ~/data/myimg.jpg --args "--scale-low 20"

Citizen science images often contain extraneous items in the image field of view. These can very easily break solve-field, which is designed for professional science-grade imagery from telescopes and narrow to medium field of view imagers (at least to 50 degree FOV). To mitigate these issues, judicious use of arguments passed to solve-field via --args is probably a good start.

The parameters I find most useful for citizen science images include:

--scale-low <scale>: lower bound of image scale estimate

--scale-high <scale>: upper bound of image scale estimate

--depth <number or range>: number of field objects to look at, or range
          of numbers; 1 is the brightest star, so "-d 10" or "-d 1-10" mean look
          at the top ten brightest stars only.

For extraneous regions of the image, try making a copy of the original image that has the offending regions cropped out. If the original image is in a lossy format such as JPEG, consider saving in a lossless format such as PNG after cropping.

Astrometry.net installed on local computer

python PlateScale.py myimg.fits 61.2 -149.9 2013-04-02T12:03:23Z

gives netCDF .nc with az/el ra/dec and PNG plots of the data. Both files contain the same data, just for your convenience.

61.2 -149.9 is your WGS84 coordinates, 2013-04-02T12:03:23Z is UTC time of the picture.

wcs.fits from the Astrometry.net website

Download from nova.astrometry.net solved image the "new-image.fits" and "wcs.fits" files, then:

python PlateScale.py 61.2 -149.9 2013-04-02T12:03:23Z new-image.fits

Notes

• 2MASS index

• Tycho index

• ways to use astrometry.net

• astrometry.net source code releases

• astrometry.net GitHub

• article on good robustness of Astrometry.net to shaky, streaked images.

Download star index files

python downloadIndex.py

Edit file /etc/astrometry.cfg or similar:

Be sure add_path points to /home/username/astrometry-data, where username is your Linux username. Don't use ~ or $HOME.

Uncomment inparallel to process much faster.

Optionally, set minwidth smaller than the smallest FOV (in degrees) expected. For example, if NOT using a telescope, perhaps minwidth 1 or something.

PlotGeomap.py

Plot an image registered to Latitude and Longitude, assuming the image features all occurred at a single altitude. This technique is used in aeronomy assuming a certain altitude of auroral or airglow emissions. This approximation is based on colors representing particle dynamics at a range of altitudes, approximated by a single altitude. For example, if a short wavelength filter (blue) was applied to the auroral image, one might assume the emissions were at about 100 km altitude.

Related

For source extraction or photometry, see my AstroPy-based examples.