MIRI CR showers failure near bad pixels

[stscijgbot-jp]

Issue JP-3536 was created on JIRA by David Law:

In some final testing of the MIRI CR showers code for MRS data I ran across a failure case in which the code erroneously flags pixels around large blobs masked with the DO_NOT_USE bit for bright extended sources.

Test case is jw01523001001_03101_00001_mirifulong_rate, which is a 5-group, 10-int observation of a bright planetary nebula.

Figure showers.png attached shows the ordinary reduction on the left, zoomed up on a large clump of known bad pixels identified in the bad pixel mask.  As intended, these are NaN-valued in the rate file and thus show up as green.  On the right is the reduction with shower flagging enabled, which shows a bright white halo around the masked pixels.  Typical count rates in this halo are given as about 3000 DN/s, compared to about 1500 DN/s in the surrounding region.

There also seems to be some excessive flagging at the very reddest end of Ch4C, and atop ultra bright emission lines in the same image.

[stscijgbot-jp]

Comment by David Law on JIRA:

Looks like this may be a rediscovery of https://jira.stsci.edu/browse/JP-3140

[stscijgbot-jp]

Comment by David Law on JIRA:

Looks like this was successfully fixed by spacetelescope/stcal#248 closing as complete.  (It didn't get called out by that PR, but this issue was one of the ones motivating those changes)