Project

General

Profile

Mask-G1

Prerequisites

Single-epoch catalog with reasonable astrometric solution (<1") and images with associated footprint mask produced, guiding star catalog, star-mask (currently, these star-masks are not produced by Mangle, but a standalone-code being written by Roberl Gruendl will probably replace it).

Test a)

General checking of bright star and diffraction spike masking in single-epoch images. Motivation: we want to avoid objects originating from these artifacts going into the coadd catalogs. Therefore we need a mask informing where the exclusion areas are (redshift-independent).

This test can be updated incrementally as the SV proceeds.

Procedure

1 - Randomly select ~20 stars with magnitude < 14 (saturated?) in the field, from the guiding star catalog for instance.
2 - Use cutout tool, ds9, to look for RA,DEC positions of these stars in the images and the mask.
3 - With a Mangle star-mask or other, we should be able to combine it with the final (SE or coadd) catalogs, to check if the exclusion radii are enough to eliminate the artifacts arising from these objects (would be straightforward with Mangle).

Verdict

Pass if all stars and diffraction spikes have been masked out.

Consequences

In case of failure, evaluate with larger sample the amount of stars missed. Verify if it's a single fluke or a more systematic failure of identification of bright stars in the masking pipeline.

Test b)

This test checks that the amount of wrongly masked areas is small (i.e., masked areas which do not contain a saturated star).

This test can be updated incrementally as the SV proceeds.

Procedure

1 - Create catalog I of bright stars (i<14, TBC) from guiding star catalog. (What is the magnitude at which stars will begin to saturate, per band, for DECam?)
2 - Match catalog I with the star-mask positions (the latter in Mangle format, TBC)

Tools: positional matching tool in python
Input: catalog I, star-mask
Output: number of matches

Verdict

Pass if number of matches ~= number of star-mask positions.

Consequences

If the number of matches is greater than the number of star-mask positions, there are bright stars from the catalog which have not been masked. Check visually if those correspond to actual saturated stars. If so, star-masking has to be reviewed. If not, the input catalog has to be revised (lower the magnitude limit). This leads to an iterative process that could be lengthy to converge, we have to understand when it is enough for our purposes.

If the number of matches is smaller than the number of star-mask positions, there are more masked-out areas than bright stars. Check visually if these areas contain actual saturated stars. If so, the input catalog has to be revised (increase the magnitude limit). If not, star-masking has to be reviewed.

Test c)

In this test we check that few catalog objects are inside the star-mask.

Procedure

1 - Create catalog I of objects passing quality cuts (and with moderate <~21 brightness to avoid junk as much as possible).
2 - Produce an object catalog O with objects within the star-polygons.

Tools: polyid (Mangle software tool)
Input: catalog I, star-mask
Output: catalog O with objects within the star-polygons

Verdict

Pass if size(catalog_O) < size(0.01*catalog_I).

Consequences