Level 3 Error for Pure Parallel NIRCam WFSS with NGROUPS=1

[stscijgbot-jp]

Issue JP-3801 was created on JIRA by Tyler Pauly:

During operations processing of new data in jw05398, a pure parallel WFSS program, errors were returned during Level 3 processing of o025, a direct image component of the program observed with the setting NGROUPS=1.

The NIRCam team looked into it and reported that data taken with NGROUPS=1 can be subject to poor calibration; the data in question appear to have large bias residuals remaining during level 3 image processing, resulting in the outlier_detection step flagging all pixels as NaN/DO_NOT_USE.

[stscijgbot-jp]

Comment by Tyler Pauly on JIRA:

Link to relevant slack discussion: https://stsci.slack.com/archives/C9KF4MVEV/p1731071916799629

[stscijgbot-jp]

Comment by Alicia Canipe on JIRA:

Just an update: we're trying to figure out why NGROUPS=1 didn't work for the PID 5398 observations when it did seem to work for other programs that have used NGROUPS=1: 

• PID 1058 obs 105, 106, 297
• PID 1409 obs 132
• PID 2560 obs 17
• PID 6078 obs 1

[stscijgbot-jp]

Comment by Alicia Canipe on JIRA:

Tyler Pauly was obs 25 of 5398 the only one that had issues? Obs 27, 28, 25, 26 also all have NGROUPS=1 and the strange pattern, but it sounds like they made it though stage 3 okay? 

[stscijgbot-jp]

Comment by Tyler Pauly on JIRA:

Obs 25 was the reported error from Ops, but I pulled o028 from MAST and tried image3, which failed for both LW and SW data. I'm guessing that 26 and 27 were not attempted because the direct image level 3 products were not generated.

[stscijgbot-jp]

Comment by Alicia Canipe on JIRA:

Okay thanks! That helps