[garrethmartin]

103-series tutorial on reading and writing images

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[jrob93]

Looks great to me! Approved, my only suggestion for you to consider is that it may be worth (briefly) explaining that when you have a pure FITs WCS for a visit image it will be suitable for displaying the image, but the most accurate WCS is located with the LSST object and associated methods. Although I think in this particular case deep_coadd WCS is fully represented in FITs, its just that the methods here are described as being applicable to visit/difference images as well. Is this DP1 link still the most valid source for how to deal with WCS & FITS...? https://dp1.lsst.io/overview/known_issues_and_subtleties.html#wcs-fits-approximations-and-misleading-interfaces

[garrethmartin]

Looks great to me! Approved, my only suggestion for you to consider is that it may be worth (briefly) explaining that when you have a pure FITs WCS for a visit image it will be suitable for displaying the image, but the most accurate WCS is located with the LSST object and associated methods. Although I think in this particular case deep_coadd WCS is fully represented in FITs, its just that the methods here are described as being applicable to visit/difference images as well. Is this DP1 link still the most valid source for how to deal with WCS & FITS...? https://dp1.lsst.io/overview/known_issues_and_subtleties.html#wcs-fits-approximations-and-misleading-interfaces

it looks like the DP1 known-issues link is still the right (and only) source... i checked DP2’s own known-issues page and it has no equivalent WCS/FITS section.

i've added something to section 4.2.:

For deep_coadd, the WCS in this header is exact. For visit_image and difference_image, the FITS WCS is only an approximation, which is good enough for display or object finding, but not for precision astrometry. So use the LSST Science Pipelines object's own sky_projection methods instead (via read_archive in Section 4.1), as described in the DP1 known issues documentation.