PHANGS–JWST First Results: Dust-embedded Star Clusters in NGC 7496 Selected via 3.3 μm PAH Emission

Star clusters are formed in the densest parts of giant molecular clouds (Lada & Lada 2003; Elmegreen 2008; Kruijssen 2012). During their formation and early evolution, they remain surrounded by dust and molecular gas, causing them to be heavily obscured at optical wavelengths and often only detectable at infrared and longer wavelengths. Until now, such young objects were mostly identified using observations of the radio continuum (Johnson et al. 2003) or dust continuum in combination with molecular gas lines (Leroy et al. 2018). Infrared studies of the star clusters inside such nurseries have so far mainly occurred in the Milky Way, Magellanic Clouds, and Local Group (e.g., Großschedl et al. 2019).

Since most stars form in clustered environments (Lada & Lada 2003), embedded clusters represent a very early, critical stage in the star formation process. Therefore, the study of these systems is essential for understanding the mechanisms that drive star formation overall and the properties and evolution of star cluster populations in galaxies.

Previous studies suggest that star clusters remain in this embedded phase within their natal cloud for only a few megayears (∼2–5 Myr; e.g., Whitmore et al. 2014b; Corbelli et al. 2017; Grasha et al. 2018, 2019; Messa et al. 2021; Kim et al. 2021, 2023; Whitmore et al. 2023, this volume). However, it has not been possible to obtain complete samples of embedded clusters beyond the Local Group to constrain these durations more directly, as the angular resolution of previous infrared facilities could not resolve the parsec scales required to identify clusters at larger distances.

The powerful new infrared capabilities of JWST enable embedded young star clusters throughout nearby galaxies to be identified and studied statistically. Complete samples of embedded clusters from PHANGS–JWST, combined with the PHANGS Hubble Space Telescope (HST), Atacama Large Millimeter/submillimeter Array (ALMA), and MUSE data sets, will allow basic open issues in star and cluster formation to be addressed, such as the fraction of star formation that occurs in clusters (cluster formation efficiency), timescales for the clearing of the natal dust and gas, the conditions that lead to cluster formation, and the dependence of these parameters on galactic environment.

In this pilot study, we identify candidate embedded clusters in NGC 7496 using PHANGS–JWST NIRCam and PHANGS-HST UV-optical imaging. We find that selection based on an excess in the F300M − F335M color, indicating polycyclic aromatic hydrocarbon (PAH) 3.3 μm emission, effectively identifies the youngest dusty star clusters. We take a first look at their properties and compare with the census of optically selected clusters in the PHANGS-HST catalog. The rest of this paper is organized as follows. In Section 2, we summarize the observations of NGC 7496 taken as part of the PHANGS–JWST Treasury program. In Section 3, we identify a small set of prototype embedded clusters with strong PAH 3.3 μm emission and use their measured properties to define selection criteria to identify a larger sample of candidate young, embedded clusters. In Section 4, we estimate the masses of the newly identified clusters and take a first look at their luminosity functions (LFs) at 2 and 3 μm. We compare this new embedded cluster sample with the optically selected PHANGS-HST catalog in Section 5 and also examine other properties of the sources, including concentration indices (CIs) and UV-IR spectral energy distributions (SEDs), which also incorporate photometry from PHANGS–JWST MIRI imaging. We summarize our main conclusions and outline future work in Section 6.

The PHANGS–JWST Cycle 1 Treasury program observed its first target, the dusty barred spiral galaxy NGC 7496 located at 18.7 Mpc (Shaya et al. 2017; Kourkchi et al. 2020; Anand et al. 2021), on 2022 July 6. The data were released (simultaneously to the team and public) on 2022 July 14. All 19 galaxies in the PHANGS–JWST sample will eventually be imaged with NIRCam and MIRI from 2.0 to 21 μm using eight filters (NIRCam: F200W, F300M, F335M, and F360M; MIRI: F770W, F1000W, F1130W, and F2100W). Full details of the observing strategy and data reduction are given in the PHANGS–JWST survey description paper in this volume (Lee et al. 2023).

In this analysis, we examine the JWST data together with UV-optical imaging in five filters from the PHANGS-HST survey (Lee et al. 2023; F275W, F336W, F438W, F555W, and F814W). We compare our dust-embedded clusters with the sample from the PHANGS-HST cluster catalog (Thilker et al. 2022; Turner et al. 2021; Deger et al. 2022; Lee et al. 2023), which provides age and stellar mass estimates from SED fitting in NGC 7496. Here we consider the 263 clusters in the HST catalog that have passed human visual inspection and are classified as class 1 (symmetric compact cluster) and class 2 (asymmetric compact cluster; Whitmore et al. 2021; Deger et al. 2022). This PHANGS-HST sample spans a wide range of properties, with ages between 1 Myr and 13.7 Gyr and masses between 500 and 4.6 × 10⁶ M_⊙ with a median of 1.5 × 10⁴ M_⊙. Of these, 131 are 10 Myr and younger, with masses between 600 and 4.3 × 10⁵ M_⊙ with a median of 8000 M_⊙.

3.1. Identification of Embedded Cluster Prototypes

To begin our studies of dust-embedded young star clusters with JWST, we visually inspect the images to identify a small sample of sources that are obvious candidates for dusty clusters and can serve as prototypes to guide our analysis. We look for bright compact sources in the NIRCam F335M filter that also appear in at least one other PHANGS–JWST NIRCam image (F200W, F300M, or F360M) but are either faint or not visible by eye in the PHANGS-HST UV-optical images (A_V ≳ 10). We choose compact objects that are slightly more extended than point sources (i.e., FWHM ≳ 1.8 pixels ∼10 pc), following a process similar to that for optical star clusters in PHANGS-HST imaging (Thilker et al. 2022; Whitmore et al. 2021; Deger et al. 2022; Lee et al. 2023). We choose to start our inspection with the NIRCam F335M image because it is the highest-resolution image obtained by PHANGS–JWST that captures PAH emission and hot dust. ⁴¹ We identify an initial set of 12 prototypical clusters. The bottom panels of Figure 1 show examples of these prototypes marked in yellow. Aperture photometry is performed to examine the properties of these objects and to develop selection criteria to select embedded cluster candidates, as discussed further in the next section.

Zoom In Zoom Out Reset image size

3.2. Source Detection and Photometry

To obtain positions and photometry for our set of 12 prototype embedded clusters, as well as search for embedded clusters in a more systematic way, we use routines in the Photutils (Bradley et al. 2022) Astropy package for photometry. First, we use the peak-finding algorithm find_peaks to detect sources in the F335M image. The level of the background varies significantly across the image. So, in order to estimate local background levels, we use the function SExtractorBackground to build a 2D model of the background. Then, we search for local maxima in square regions that are 11 pixels on a side.

To derive accurate source positions, we use the function centroid_quadratic, which determines the centroid by fitting 2D quadratic polynomials. We extract photometry from a circular aperture with a radius of 0093 (3 NIRCam short-wavelength pixels) that encloses more than 50% of the energy for a point source (0073) in the F335W filter. We subtract the background computed from an annulus between 0217 and 031 away from the source center. The units of the NIRCam imaging are megajanskys per steradian, so we multiply by the pixel area in steradians, given by the parameter PIXAR_SR in the header of the image, to obtain flux densities in megajanskys. We then convert the measured flux densities into the AB magnitudes using the usual expression:

$$\begin{eqnarray*}&&{m}_{\mathrm{AB}}=-2.5{\mathrm{log}}_{10}\left(\tfrac{{f}_{\nu }}{3\,631\times {10}^{-6}\ \mathrm{MJy}}\right).\end{eqnarray*}$$

For this initial analysis, we forgo aperture corrections to the photometry to calculate total source fluxes. The measurement of accurate total fluxes for individual clusters across the PHANGS–JWST filter set will be the subject of dedicated future work to address the challenges presented by the 10-fold decrease in resolution from 2 to 21 μm. Nevertheless, our sample selection strategy, which is primarily based on the F300M – F335M color (described in Section 3), should not be greatly affected, since the resolution in these adjacent bands is quite similar, with a difference in the point-spread function (PSF) FWHM of only 0011.

3.3. Selection Criteria for Embedded Cluster Candidates

In Figure 2, we show color–magnitude diagrams (CMDs) based on F335M and F300M photometry; the former captures flux from the PAH 3.3 μm feature, while the latter primarily probes the stellar and dust continuum (Lee et al. 2023; Sandstrom et al. 2023a, 2023b; Chastenet et al. 2023a, 2023b, this volume). Panel (a) in Figure 2 shows the 12 visually identified prototype embedded clusters (yellow stars) together with the sample of F335M sources detected using Photutils find_peaks (black dots). This latter sample is comprised of a mixture of sources including stellar clusters of different ages; main sequence, red giant branch, and asymptotic giant branch stars; and background galaxies.

Zoom In Zoom Out Reset image size

These CMDs are analogous to those commonly used to select emission-line sources in extragalactic narrowband imaging surveys (e.g., Ly et al. 2011; Lee et al. 2012), and similar strategies can be adopted to identify and characterize sources with PAH emission. Sources that show a significant F300M − F335M color excess (i.e., a continuum-subtracted F335M flux) are identified as PAH emitters. The locus of points around F300M − F335M = 0 represents continuum sources with little to no PAH emission. These objects are typically associated with star clusters older than 10 Myr. as shown in panel (c) of Figure 2. Not surprisingly, this distribution becomes broader at fainter F335M magnitudes due to increasing noise in the F300M − F335M measurement (compare with Figure 3 in Lee et al. 2012). Sources with PAH emission show significant F300M − F335M color excess and are found to the right of the locus, and we find that all 12 prototypes are in this region.

Based on this clear separation in the CMD, we select a larger sample of bright PAH emitters for examination, i.e., F300M − F335M > 0.3, F335M < 24, and F300M < 25. This selection provides a sample of 67 objects, shown with cyan symbols in panel (a) of Figure 2. In this figure, the lines indicate the limits F300M − F335M > 0.3 and F335M < 24. We note there are additional sources (beyond the 67 cyan candidates) that fall inside this box; however, these sources are fainter than F300M = 25 mag and not included in our sample.

We visually inspect each of these 67 cluster candidates and divide them into three categories as follows.

• 1.  
  Embedded: optical emission very faint or undetected, similar to the 12 prototype objects (N = 28).
• 2.  
  Intermediate/ambiguous: optical emission detected, but it is not clear whether the optical source is the counterpart to the F335M source due to positional differences, and lower resolution at 3 μm (N = 16).
• 3.  
  Visible/exposed: clear optical counterpart (N = 23).

Examples of objects in each category are shown in Figure 1. We will compare our new, dust-embedded cluster sample with the optically selected PHANGS-HST catalog in Section 5.

Luminosity and mass functions for the youngest cluster populations in galaxies give direct constraints on the initial cluster mass function, analogous to the initial stellar mass function. One of the major weaknesses in current studies of star formation is that the fraction of the most recently formed dust-obscured clusters has been unknown and unaccounted for. These initial results provide a preview of the work with JWST that will measure these fundamental distributions.

4.1. Luminosity Function

For clusters with similar ages and reddening, the LF of clusters can serve as a proxy for their mass function. The LF can be approximated, at least to first order, by a power-law shape of the form ${dN}/{dL}\propto {L}^{\alpha }$ (Whitmore et al. 1999; Fall 2006; Whitmore et al. 2014a). Values of the power-law index α ∼ −2 are typical for young star clusters (e.g., Larsen 2009; Whitmore et al. 2014a).

In Figure 3, we present the LFs of our embedded cluster candidates in the F200W and F300M bands, which sample flux from the stellar continuum, with some contribution from the dust continuum (red squares). These distributions increase starting from the brightest magnitudes (to the right), eventually flatten, and then drop off again (F200W filter). The flattening toward fainter magnitudes is almost certainly due to incompleteness.

Zoom In Zoom Out Reset image size

For comparison, we also show the LF of optically selected clusters from the PHANGS-HST survey (Thilker et al. 2022; Turner et al. 2021; Deger et al. 2022; Lee et al. 2023), with no restriction on age (shown in orange) and clusters with best-fit ages younger than 10 Myr (shown in blue). We also plot the LF of the combined cluster sample, i.e., PHANGS-HST clusters plus the 67 embedded candidates (shown with green triangles). The LF for the full HST+JWST cluster sample (green) has a best-fit power-law index of −2.01 in F200W and −1.77 for F300M (shown as the green lines in each panel). The fit over only the PHANGS-HST clusters (orange) gives very similar index values, −1.91 in F200W and −1.73 for F300M (orange lines). On the other hand, while there are only a few data points before our embedded cluster distribution (red points) flattens at magnitudes fainter than M_F200W ≥ −8, the bright points suggest a behavior similar to the optically selected clusters. The LF in the F300M filter, however, appears to be steeper than the PHANGS-HST clusters. This difference could be due to the imposed selection criteria that involve cuts using the F300M filter (Section 3.3).

Based on our preliminary results, we suspect that the LF for embedded star clusters presents a similar shape to that for optical clusters. However, our sample is not large enough to derive accurate power-law indices. In future work, we will extend this work to the full sample of 19 PHANGS–JWST galaxies, which will be based on total fluxes computed with appropriate aperture corrections. A larger sample will better constrain the shape of the LF, as well as the embedded cluster mass function, which again, is effectively the initial cluster mass function. Comparison with what we have already learned for visible clusters will provide important insights into the survival of clusters in transition to becoming exposed/visible.

4.2. Estimate of Cluster Masses

We compare our sample of 67 dust-embedded cluster candidates with the PHANGS-HST human-inspected class 1 + 2 cluster catalog (N = 263) and find that our analysis of the PHANGS–JWST imaging has identified 59 new embedded cluster candidates in NGC 7496. These objects are of particular importance because they are likely to be among the youngest clusters in this galaxy and are needed to construct a complete census. In this context, it is notable that the size of our new sample is comparable to the number of young HST clusters (59 younger than 2 Myr; 94 younger than 5 Myr) and may double the census of the population younger than ∼2 Myr.

Of the eight clusters in common between the two samples, we classify seven as visible/exposed and one as partially embedded/intermediate. All are found to be younger than 11 Myr (six having ages less than 1 Myr), with masses of ∼1–7 × 10⁴ M_⊙. (As mentioned earlier, the full PHANGS-HST cluster sample, meanwhile, contains clusters as old as 13 Gyr.)

We can gain additional insight by examining the infrared properties of clusters of all ages from the optically selected PHANGS-HST sample. Panels (c) and (d) in Figure 2 show the location of clusters in different age intervals in the F300M − F335M color versus F335M magnitude diagram. In panel (c), we can see that clusters older than 100 Myr (magenta triangles) have F300M − F335M = 0, indicating that these older sources do not have PAH emission. This panel also shows that nearly all clusters with ages between 10 and 100 Myr (green plus signs) and 5–10 Myr (yellow diamonds) have F300M − F335M = 0. On the other hand, panel (d) shows that all HST clusters with an F300M − F335M color excess (i.e., strong PAH emission) have ages ≤2 Myr (cyan crosses). These very young, optically selected clusters have a similar range of F300M − F335M values as our 67 embedded candidates, although most of them have a significantly fainter F335M magnitude. Panel (d) of Figure 2 also shows clusters with estimated ages between 2 and 5 Myr (orange squares). While a small fraction of these 2–5 Myr clusters have slightly higher F300M − F335M values than clusters older than 5 Myr, a surprising number of them show little to no PAH emission. Assuming that the age dating from PHANGS-HST is correct for these clusters, this result indicates that measurable PAH emission may drop very rapidly (in 2 Myr) in clusters of moderate mass.

We also examine the CIs of infrared and optically selected clusters. The CI provides a measure of the compactness of the source and is useful to distinguish between clusters and stars (e.g., Chandar et al. 2010; Whitmore et al. 2014b; Deger et al. 2022). In particular, Whitmore et al. (2023, this volume) found that a CI estimated from the JWST F200W filter significantly improves the separation of stars and clusters. The CI values were estimated following the procedure in Whitmore et al. (2023, this volume), as the magnitude difference measured between 1 and 4 pixel apertures (0031 and 0124) in the F200W filter. In Figure 4, we present the results. The vertical dashed line indicates CI =1.4, a limit to separate clusters from stars (Whitmore et al. 2023, this volume). The 67 dust-embedded clusters from this work are nearly all to the right of the line, meaning they are extended relative to a point source, although there are a few fainter objects that fall to the left of the line. The dispersion in the size measurements clearly increases toward fainter magnitudes.

Zoom In Zoom Out Reset image size

Next, we examine the overall spatial distribution of the sample, as well as the location of our dust-embedded cluster candidates relative to other structures and components of the interstellar medium (ISM). In Figure 1, we see a strong concentration of embedded clusters (shown in red) along a dust lane in the northern spiral arm, which forms a ridge just next to young (age < 10 Myr) optically selected clusters. Jumping to the left panel of Figure 5, we see that the strongest PAH emitters (shown as green dots without an accompanying magenta circle) are mostly located where we see the most prominent dust features in the optical HST images, along the spiral arms. Sources with F300M − F335M > 0.3 but faint F335M magnitudes (see magenta box in Figure 2(a)) are enclosed with a magenta circle. Recall that we decided to apply cuts in magnitude to study only the brightest sources (F335M < 24 and F300M < 25) to simplify this initial analysis. However, by imposing these limits, weaker PAH emitters are excluded from inspection of the image; these appear to coincide with the more subtle dust lanes and potentially may also be dust-embedded clusters. In future work, we will test selection criteria that employ a color excess curve to identify fainter embedded clusters.

Zoom In Zoom Out Reset image size

The middle and right panels of Figure 5 show a CO(2–1) ALMA image taken as part of the PHANGS-ALMA survey (Leroy et al. 2021) and the Hα VLT-MUSE image of NGC 7496 from the PHANGS-MUSE survey (Emsellem et al. 2022). Our 67 PAH-emitting sources are closely associated with molecular gas and current star formation. Only 9% of our candidates appear unassociated with significant peaks of CO emission, Hα, or both. These 9% all have no obvious optical counterpart in the HST images, and half have CI values similar to stars, indicating that they are very compact and possibly not embedded clusters.

Finally, we examine the SEDs of the embedded clusters from optical wavelengths with HST to the IR with NIRCam and MIRI, which provide further insights into their properties. We show preliminary SEDs of the 12 embedded cluster prototypes in Figure 6. These show a clear deficit in optical flux relative to the infrared. In fact, we were able to identify possibly associated faint optical emission for only three of these objects. For the nine embedded clusters, without UV-optical detections, we show upper limits in the HST bands. For this first look, the photometric measurements in all bands have been performed with the same 0155 aperture, which leads to a relative underestimate of the measured fluxes in the MIRI bands, as no aperture corrections have been applied as discussed before. To produce proper SEDs for these clusters in future work, we will need to correctly distinguish between crowded point sources and their corresponding diffuse PAH emission with more careful source extraction that takes the different PSFs into account. These are some of the challenges that need to be addressed in order to perform proper SED fitting for embedded clusters and more accurately constrain their physical properties (e.g., stellar masses, dust masses, ages, and reddenings). In addition to the distinctive PAH feature at 3.3 μm, we see a dip at 10 μm indicating silicate absorption, which is caused by protostellar envelopes (Crapsi et al. 2008) and has been observed in young stellar objects (e.g., Großschedl et al. 2019). This spectral feature supports our conclusion that these candidates are very young embedded clusters.

Zoom In Zoom Out Reset image size

In this work, we have used new NIRCam imaging of the nearby spiral galaxy NGC 7496, the first galaxy to be observed for the PHANGS–JWST Treasury survey, to identify very young clusters embedded in natal dust and gas. Our selection method relies on detection of PAH 3.3 μm emission, which is closely associated with the ISM of very recently formed clusters.

We begin by identifying 12 prototype dust-obscured clusters based on visual inspection of both PHANGS–JWST NIRCam and PHANGS-HST optical images. We find that all 12 show strong PAH emission at 3.3 μm and fall in a region of the F335M versus F330M − F335M CMD that clearly separates from the vast majority of detections. To first order, the PAH emission at 3.3 μm appears to correlate with other dust emission (Sandstrom et al. 2023a, 2023b; Dale et al. 2023; Chastenet et al. 2023a, 2023b, this volume) and so should represent a sensitive, high-resolution tracer of dust-reprocessed UV light from young stars. Based on this, we develop criteria to select a larger sample of candidates. This selection is primarily based on the F330M – F335M color excess and yields a sample of 67 candidate embedded clusters in NGC 7496. By comparing the F300M – F335M colors of this sample to optically selected clusters in the PHANGS-HST catalog that have ages determined through SED fitting of five-band UV-optical photometry, we estimate that the embedded clusters are likely to be younger than ∼2 Myr.

We find that our new sample may significantly increase the census of young clusters from the PHANGS-HST catalog; the number of clusters younger than 2 Myr could be increased by a factor of 2. A more careful analysis of the selection functions and completeness of both the HST and JWST cluster samples will be required to provide an accurate assessment of the fraction of young clusters that are embedded and missing from optical catalogs. The eight clusters in common between the two catalogs have very young SED ages (six are ∼1 Myr). These results underscore the importance of our new PHANGS–JWST imaging for studying the earliest stages of star and cluster formation.

Preliminary optical/infrared SEDs for the visually selected prototype candidates show PAH emission and possible silicate absorption. The CI measurements confirm that the majority of our 67 sources are more extended than stars/point sources. These characteristics provide additional evidence that most of our selected candidates are star clusters. We also take a first look at the 2 and 3 μm LFs for our sample and estimate stellar masses, finding values of 10⁴–10⁵ M_⊙.

The strongest and brightest PAH emitters are mainly located in prominent dust lanes within the main spiral arms of the galaxy, next to young optically detected clusters from PHANGS-HST. The spatial distribution of our candidates shows that they are also correlated with Hα and CO(2–1) emission.

In future work, we will expand this analysis to all 19 galaxies in the PHANGS–JWST sample and examine the completeness of samples based on the F300M – F335M color excess for selecting young embedded star clusters. Significant effort will be required to develop techniques to obtain robust total fluxes across the 13 PHANGS-HST and JWST filters from 2750 Å to 2 μm due to the factor of 10 decrease in resolution. Such techniques are needed to produce SEDs appropriate for constraining the physical properties of these clusters through the fitting of stellar population and dust models. Ultimately, a complete census of star clusters from PHANGS-HST and JWST will enable basic unsolved problems in star formation and cluster evolution to be addressed through new constraints on the initial cluster mass function, star cluster formation efficiencies, and feedback timescales associated with the clearing of the natal gas and dust.

The authors would like to thank the anonymous referee for constructive comments that helped improve the quality of this work. This work is based on observations made with the NASA/ESA/CSA JWST and NASA/ESA Hubble Space Telescopes. The data were obtained from the Mikulski Archive for Space Telescopes at the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS 5-03127 for JWST and NASA contract NAS 5-26555 for HST. The JWST observations are associated with program 2107 and those from HST with program 15454. The specific JWST observations analyzed can be accessed via 10.17909/9bdf-jn24. We have also used higher-level data products developed and released by PHANGS-HST 10.17909/t9-r08f-dq31 (image products) and 10.17909/jray-9798 (catalog products). We are grateful to Scott Fleming at STScI and his team for their excellent support in helping us provide these products to the community through https://archive.stsci.edu/hlsp/phangs-hst and https://archive.stsci.edu/hlsp/phangs-cat. Based on observations collected at the European Southern Observatory under ESO program 1100.B-0651 (PHANGS-MUSE; PI: Schinnerer). This paper makes use of the following ALMA data: ADS/JAO.ALMA#2017.1.00886.L, ALMA is a partnership of ESO (representing its member states), NSF (USA) and NINS (Japan), together with NRC (Canada), MOST and ASIAA (Taiwan), and KASI (Republic of Korea), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO and NAOJ. J.M.D.K. gratefully acknowledges funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program via the ERC Starting Grant MUSTANG (grant agreement No. 714907). COOL Research DAO is a Decentralized Autonomous Organization supporting research in astrophysics aimed at uncovering our cosmic origins. E.J.W. acknowledges the funding provided by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—Project-ID 138713538—SFB 881 ("The Milky Way System", subproject P1). M.B. acknowledges support from FONDECYT regular grant 1211000 and by the ANID BASAL project FB210003. T.G.W. acknowledges funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant agreement No. 694343). J.K. gratefully acknowledges funding from the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) through the DFG Sachbeihilfe (grant No. KR4801/2-1). M.C. gratefully acknowledges funding from the DFG through the Emmy Noether Research Group (grant No. CH2137/1-1). F.B. would like to acknowledge funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant agreement No. 726384/Empire). R.S.K. acknowledges financial support from the European Research Council via the ERC Synergy Grant "ECOGAL" (project ID 855130), the Deutsche Forschungsgemeinschaft (DFG) via the Collaborative Research Center "The Milky Way System" (SFB 881—funding ID 138713538—subprojects A1, B1, B2, and B8), and the Heidelberg Cluster of Excellence (EXC 2181-390900948) "STRUCTURES", funded by the German Excellence Strategy. R.S.K. also thanks the German Ministry for Economic Affairs and Climate Action for funding in the project "MAINN" (funding ID 50OO2206). E.R. acknowledges the support of the Natural Sciences and Engineering Research Council of Canada (NSERC), funding reference No. RGPIN-2022-03499. K.G. is supported by the Australian Research Council through Discovery Early Career Researcher Award (DECRA) Fellowship DE220100766 funded by the Australian Government. K.G. is supported by the Australian Research Council Centre of Excellence for All Sky Astrophysics in 3 Dimensions (ASTRO 3D) through project No. CE170100013. S.D. is supported by funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant agreement No. 101018897 CosmicExplorer). K.K. gratefully acknowledges funding from the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) in the form of the Emmy Noether Research Group (grant No. KR4598/2-1; PI: Kreckel). H.H. acknowledges the support of the Natural Sciences and Engineering Research Council of Canada (NSERC), funding reference No. RGPIN-2022-03499.

J.P. acknowledges support by DAOISM grant ANR-21-CE31-0010 and the Programme National "Physique et Chimie du Milieu Interstellaire" (PCMI) of CNRS/INSU with INC/INP, cofunded by CEA and CNES. G.A.B. acknowledges the support from ANID Basal project FB210003.

Facilities: JWST - James Webb Space Telescope, HST - , ALMA - , MUSE - .

Software: Astropy (Astropy Collaboration et al. 2013,2018, 2022). This research made use of Photutils, an Astropy package for detection and photometry of astronomical sources (Bradley et al. 2022) AstroPy (astropy.org)