This page archives Slack comments from the splinter session on MRS background of the Improving JWST Data Products Workshop (IJDPW).
Macarena Garcia Marin
Mellissa Shahbandeh, at t this point you are the only person signed in for this exercise; I will join you and likely will rerun the pipeline on the same dataset (wanted to do that anyway). The idea is to evaluate how well your code manages to subtract the background on this dataset, without using the associated background observation in the program. And then compare with the pipeline output.
Ideally, one of the outcomes of the exercise should be how well your code runs on other datasets.
Mellissa Shahbandeh
Thanks very much Maca! should I use level 3 data or run it from stage 0? or doesn't matter?
Macarena Garcia Marin
I think you should follow your standard process, same as you follow for your science data.
Macarena Garcia Marin
Some logistics for tomorrow. The session will run from 9:30 to 12:30 with a floating break (meaning coffee will be available in the cafeteria starting from TBC time and we can break whenever we want/need.
Mike Engesser
Hi Maca - I will join for this exercise and as well, though I will need to excuse myself at some point for a meeting on darks
Macarena Garcia Marin
oh great!
David Law
Should also be there, for a chink of time at least
Macarena Garcia Marin
Thanks David. IF we have enough critical mass could also have a discussion on potential pipeline areas of improvement
Mellissa Shahbandeh
M_Shanhandeh_wr140_vs_bgs_median.pdf
Macarena Garcia Marin
Thanks to everyone who joined today's discussion, and specially to @Melissa Shahbandeh and @Mike Engesser for sharing their code and results. For context, both Melissa's (IFU) and Mike's (IFU and Imaging) code are optimized for faint transients science (or more generally faint point sources). Both of their codes run well on the very bright target proposed as an example (WR140), but Mike identified improvements he plans to implement on his.
Macarena Garcia Marin
Main discussion points:
- There is agreement that the proposed codes are ideally suited to be tools external to the pipeline, and not to be an integral part of it
- Both cover the need to remove backgrounds from very faint point sources. Even in cases where a science program had a dedicated background, they were often too bright and additional process was needed. Also work on brighter targets (some improvement underway)
- Melissa’s software is already available on GitHub, we discussed the possibility of adding it to the External Community Tools page
- Mike will further work on his code, that offers imaging capabilities, with the idea of making it broadly available to the community