UR: Something from Nothing: Analyzing JWST Pixel Data

The Undergraduate Research series is where we feature the research that you’re doing. If you are an undergraduate that took part in an REU or similar astro research project and would like to share this on Astrobites, please check out our submission page for more details. We would also love to hear about your more general research experience!

Feras Natsheh is a recent graduate of Whitman College, where he earned his BA in Physics-Astronomy. This post describes part of his senior research at Whitman College under the supervision of Dr. Jessica Sutter. This work was also presented as a poster at the 2024 Whitman Undergraduate Conference. After graduating, Feras joined as a post-baccalaureate research fellow at the New Mexico Consortium, working for Aurorasaurus, an aurora citizen science project sponsored by NASA and NSF. He hopes to pursue graduate studies in Astrophysics or Space Physics after concluding his current role.

Getting the Background Right

The James Webb Space Telescope (JWST) has been providing a plethora of new data about galaxy formation and structure. In particular, the Physics at High Angular resolution in Nearby GalaxieS (PHANGS) team has been using JWST to study the structure of nearby galaxies. My aim in this study is to address challenges specific to the study of these nearby galaxies, such as removing instrumental noise, accurately matching background levels, and calibrating astrometry for extended sources. One of the most exciting aspects of this data is how it can reveal faint emission that was undetectable to previous infrared observatories. But, as with any new instrument, there is still a lot to learn about how to best use JWST! In particular, without a precise knowledge of any remaining noise in the data, it becomes uncertain where we are seeing previously unobservable infrared emission and where we are only detecting the remaining static that gets introduced by the observing and calibration process. It’s like looking at an overexposed photo where bright spots blend into the background. Without properly adjusting the exposure, you can’t tell whether those bright areas are part of the scene or artifacts of overexposure. Similarly, understanding JWST noise levels allows us to know genuine emissions from observational residues. (For more details on the PHANGS data processing, check out this paper.) Near- and mid-infrared wavelengths are crucial for exploring star formation and the interstellar medium (ISM) in galaxies. Stars form deep within dusty molecular clouds, making infrared observations more effective than optical ones for detecting these initial stages of star formation. Additionally, emissions from polycyclic aromatic hydrocarbons (PAHs) occur at these wavelengths, providing essential insights into the PAH’s size and how they evolve within the ISM, which is key to comprehending their overall lifecycle. (For a more rigorous discussion of PAH observations, check out this paper.)

Pixels in Perspective

We primarily considered observations of NGC 1559 (shown in Figure 1), part of a JWST Cycle 2 mid- and near-infrared dataset using MIRI and NIRCam observations, respectively. This dataset selected disk galaxies in the southern hemisphere that were moderately inclined, actively star-forming, and within a distance of 17 Mpc. We searched for some of the emptiest regions in the images of NGC 1559 in different wavelength filters. These regions represent the “zero-level” regions. Even if these regions contain no astrophysical sources, there will still be counts in these pixels due to intrinsic properties of the camera like bias and dark current. We can get better estimates of these properties to accurately remove them from images by calibrating them with the zero-level regions.

A summary of our work is displayed in Figure 2. Here, we show the data for the zero-level regions in two filters from both the NIRCam and MIRI observations. The histograms represent the probability distributions of pixel values in the zero-level regions (which we expect to be centered at or near zero), with fits shown in the orange curves.

Characterizing and analyzing the background zero levels in 55 nearby galaxies obtained through the PHANGS-JWST Cycle 2 Treasury Program data is pivotal for several reasons. It significantly enhances image quality, signal-to-noise ratio, spectral analysis, and the ability to detect diffuse emission (such as PAHs). This aids in the study of PAH features, dust continuum, recombination lines, and the stellar continuum emission across local star-forming galaxies. With this refined understanding of sky background and noise levels, we can better investigate the faintest, smallest structures within galaxies and their role in shaping galaxy evolution over time. Future efforts will expand to include additional galaxies across a broader range of masses and star formation rates, focusing on identifying the optimal resolutions that maximizes signal-to-noise ratios while preserving the fine structures that only JWST can observe.

Edited by: Brandon Pries

Featured image credit: JWST via NASA