Webb Office Hours Session 37: September 11, 2025
Q&A's:
Q1: Where are the Webb Office Hours procedures and guidelines?
A1: Webb Office Hours. Type your question into the WebEx chat. We will asynchronously copy questions from the chat to this main page and work through them as a group. If you have images to share please give WebEx permission to share your screen (you may need to log out and log back in again to enable this feature.)
Q2: The user's group recently got time with NIRSpec and he wants to make sure he's doing things correctly. They're looking at a bright cluster galaxy using mosaic of NRS pointings and he has a few questions. He showed the positioning of the mosaic – different pointings begin to shrink and expand. That doesn't happen with MIRI MRS. Why? (This is G395H).
A2: This is because NIRSpec has two separate detectors and uneven wavelength coverage. Because this is H-grating data, there's light from the entire wavelength range falling on both detectors, but the slices of the IFU slicer are oriented at a diagonal. So, as you go across the detector at some wavelength, spectra begin to fall off the detector. This happens at different wavelengths for different slices. Because of the wavelength gap between the detectors, there is a bit of wavelength range that won't be covered for a given slice before it's picked up again. As you wavelengths between the two detectors, you have fewer slices that are illuminated. So, the mosaic is changing because the spatial coverage is not even.
Q3.1: The user showed the spectra extracted over the entire field of view. For the low SNR region, it seems the spectra really drops to very negative values. Why?
A3.1: Did you do any 1/f cleaning? This would be done in calwebb_detector1 or calwebb_spec2. There's a step called clean_flicker_noise in detector1 which is turned off by default. There's also a similar step called nsclean in spec2. They're the same under the hood, but clean_flicker_noise cleans group by group and NSClean works on the average rate images). In the past, running one or both of these steps has improved the negative values from 1/f noise. The other thing that may be causing it is that sometimes there's a whole pedestal level across the entire detector (not the sky background, but the detector itself) which could be slightly positive or negative in terms of the rates. You can also clean that out with the 1/f noise corrections, and it might also fix some of the wiggles and variations in the spectra that are due to 1/f noise.
Q3.2: Should I do that in calwebb_spec2?
A3.2: You can try one or the other: try turning clean_flicker_nosie on in detector1 and start to compare rate images with and without the cleaning. You can go here for advice for NRS observations: https://jwst-docs.stsci.edu/known-issues-with-jwst-data/1-f-noise#gsc.tab=0
This page also shows an example for MOS data of what you might look for in the images with and without 1/f cleaning. Take a look at the rate images to see how the clean_flicker_noise correction is performing. The page gives you some parameters to start with (specifically for clean_flicker_noise, but it also has notes about NSClean in calwebb_spec2, as well). There are a couple important things to point out: nsigma should be around 1.5 instead of the default of 2, and you should set the bkg_method to None (which is not the default for clean_flicker_noise). NSClean in spec2 has almost the same parameters and runs the same way, but it does it at the rate stage instead of on groups. So, you can also try the options for that step, as well, to see if it works better.
Q4: I have NIRCam data with 10 filters as part of a big program (the VENUS program). When I plot the I2D images from NIRCam with the WCS from the header (using astropy), the images aren't by default aligned with North at the top and east at the right of the plots, do you know why that is?
A4: It looks like you are using the "SIP" header values when reading in the WCS from the header. These headers are approximate and are only made so that the files can be read directly into applications like DS9. It's possible that slight inaccuracies in these header keywords could be causing the default plotting to not have the correct orientation. If you want to access the "gWCS" transforms that are stored in the asdf extensions of the files, these contain the exact transforms directly and could be used to plot the data. More information about the WCS for JWST data is here: https://jwst-pipeline.readthedocs.io/en/latest/jwst/assign_wcs/main.html
Q5: I'm also interested in making multi-color RGB images with several filters, but most of the tools I know of use only three filters. Are there good tools for creating multi-color images with more than 3 filters?
A5: Yes, you can use the JWST pipeline to resample the images onto the same pixel grid to create color images using multiple filters. We show how to do this in the documentation: https://jwst-docs.stsci.edu/known-issues-with-jwst-data/nircam-known-issues/nircam-imaging-known-issues#NIRCamImagingKnownIssues-Resamplingimagesontothesamepixelgrid
Q6:
A6:
Q7:
A7:
