Dust Curves with a Twist: JWST Shows One Size Doesn’t Fit All

Title: Unveiling the trends between dust attenuation and galaxy

properties at z∼2−12 with JWST

Authors: V. Markov, S. Gallerani, A. Pallottini,  M. Brada, S. Carniani, R. Tripodi,, G. Noirot,, F. Di Mascia, E.Parlanti, N. Martis

First Author’s Institute:  Faculty of Mathematics and Physics, University of Ljubljana, Slovenia/ Scuola Normale Superiore, Pisa, Italy

Status: Preprint on arXiv.

When light from a distant galaxy reaches our telescopes, it’s already been through a lot. It’s traveled billions of years and, along the way, passed through a haze of cosmic dust—tiny particles within galaxies that absorb and scatter light, reshaping how a galaxy’s spectrum appears to us. For years, astronomers have relied on a few classic templates—like the Calzetti law or the SMC curve—to correct for the light lost to dust. These attenuation curves describe how starlight dims at different wavelengths, and they’re crucial for estimating basic properties like a galaxy’s star-formation rate or stellar mass. But most of these curves were calibrated on local galaxies. What about the first galaxies, forming in the early universe? Would their dust behave the same way? Thanks to near-infrared capabilities of JWST’s NIRSPec and NIRCam instruments, today’s authors set out to find out. In their new study, they used deep spectroscopy and photometry to explore how dust attenuation curves actually behave across 173 galaxies spanning redshifts 2 to 12—a time range that covers nearly 90% of cosmic history. The results are a wake-up call: there is no single, universal dust law. Galaxies don’t just differ in how much dust they have—they differ in how that dust shapes light. The attenuation curve appears to vary from galaxy to galaxy, and those variations appear to track changes in age, dust content, star formation, and metallicity.

Measuring Dust by Letting It Speak

Instead of assuming a fixed shape for each galaxy’s dust curve, the team used the SED-fitting code BAGPIPES to let two key features float freely:

  • Slope (S): how quickly the curve rises into the ultraviolet (higher = steeper).
  • UV bump strength (B): the prominence of the 2175 Å bump, famously strong in the Milky Way and weak or absent elsewhere.

With this flexible approach, the team didn’t just fit each galaxy’s star light—they captured how its dust behaves, too. And once they had dust curves for all galaxies, they looked for patterns. What kinds of galaxies have steep slopes? Who shows a UV bump? How does that all change with time?

Dust Curves Are Not One-Size-Fits-All

To investigate, the authors grouped galaxies by a range of physical properties—attenuation, star formation rate, stellar mass, age, metallicity, and more. The results, shown in Figure 1, reveal trends: dust attenuation curves vary systematically depending on a galaxy’s internal conditions.

 A grid of three panels, each showing average dust attenuation curves for galaxies grouped by a different property: AV, star formation rate, stellar mass. Each panel shows how the curve shape—its steepness and bump—varies between groups.
Figure 1: Dust curve shapes for galaxies grouped by three different properties. Each panel shows how the median attenuation curve varies with one physical parameter. Steeper curves tend to appear in older, more massive, or less star-forming galaxies, while flatter curves dominate in dustier or burstier ones. (Image Credit: A subset of Figure 3 from today’s paper)

Galaxies with higher V-band attenuation (AV) tend to have shallower curves overall (less steep rise in the UV) and weaker UV bumps—consistent with radiative transfer effects that mute the light from deeply embedded stars. In contrast, galaxies with older stellar populations or higher stellar mass display steeper curves, likely due to the buildup of small, UV-sensitive grains over time. Meanwhile, young, bursty galaxies show flatter slopes, pointing to the destructive power of their intense radiation fields.

To quantify these relationships, the authors computed Pearson correlation coefficients, as shown in Figure 2. For example the slope anti-correlates with AV (r = –0.35) and specific SFR (r = –0.38), and positively correlates with stellar age (r = 0.47). UV bump strength also drops with increasing AV. While many of these parameters are themselves interconnected, the correlations help show which factors are most strongly tied to the dust curve’s shape.

A matrix of colored squares showing the strength and direction of correlations between dust curve parameters (slope and bump strength) and galaxy properties like age, SFR, mass, and AV. Color and number inside each square indicate the Pearson correlation coefficient.
Figure 2: A matrix of Pearson correlation coefficients quantifying how the attenuation slope (S) and UV bump strength (B) relate to different galaxy properties. Strongest trends – those cells with a very high or low r value – include flatter slopes with increasing AV and sSFR, and steeper slopes with older stellar ages. The bump strength also weakens with more dust. (Image Credit: Figure 4 from today’s paper)

This observational picture gains another dimension in Figure 3. The right panel shows how galaxies are distributed in redshift depending on the shape of their dust curves—specifically the slope (S) and UV bump strength (B). Galaxies with flatter curves and weaker UV bumps (Calzetti-like) tend to dominate at higher redshift, while those with steeper slopes and/or stronger bumps (SMC- or MW-like) are more common at lower redshifts. This trend suggests that attenuation curves themselves may evolve over cosmic time, reflecting changes in how dust is produced, processed, and enriched in galaxies.

Two panels: the left shows galaxies plotted by dust slope and bump strength, color-coded by redshift, with four labeled quadrants. The right shows redshift distributions for each quadrant. High-redshift galaxies cluster in the quadrant with flat slopes and weak bumps.
Figure 3: Where galaxies fall in dust curve space. The left panel shows each galaxy’s attenuation slope (S) and UV bump strength (B), divided into four quadrants based on median values. The right panel shows how the redshift distribution varies for galaxies in each quadrant. (Image Credit: Figure 5 from today’s paper)

What Shapes the Curve?

Having identified these trends, the natural question becomes: what physical processes actually shape the attenuation curve? The authors outline several key mechanisms that may drive the diversity we observe.

One is radiative transfer: In dusty galaxies, UV light is heavily absorbed before it escapes, especially from young, deeply embedded stars. This makes the observed curve appear flatter—not because the grains are different, but because we only see the light that survives.

Geometry matters too. Supernovae and stellar winds can clear out cavities, letting stars shine from relatively dust-free regions. This uneven star–dust mixing can flatten the curve even in dusty galaxies.

Then there’s the grain-level physics. Small dust grains are particularly vulnerable to UV radiation. In high-sSFR galaxies, those intense radiation fields can destroy or modify the smallest grains—the very ones that steepen the UV slope. That’s why galaxies with high specific star-formation rates tend to show the flattest attenuation curves. Older galaxies, on the other hand, tend to accumulate steeper curves over time. This likely reflects the survival and growth of small, UV-sensitive grains as radiation fields weaken.

Finally, chemical evolution shapes the dust landscape. Early galaxies likely produced dust primarily through supernovae, which tend to yield larger grains. As galaxies mature, processes like grain shattering and growth begin to populate the ISM with smaller particles and carbon-rich species like PAHs. These small grains both steepen the slope and contribute to the UV bump—especially in metal-rich environments. This connection appears in Figure 4, where galaxies with higher measured oxygen abundances tend to show stronger UV bumps.

A scatter plot of UV bump strength versus gas-phase metallicity (oxygen abundance). The plot shows a tentative evidence: galaxies with higher metallicity generally show stronger UV bumps.
Figure 4:  UV bump strength versus oxygen abundance (a proxy for metallicity) in galaxies with emission-line measurements. There’s a tentative trend: galaxies with higher metallicity tend to show stronger UV bumps, consistent with the idea that carbonaceous bump-producing grains form more easily in enriched environments.(Image Credit: Figure 8 from today’s paper).

Together, these processes describe a kind of dust lifecycle. Galaxies start out with relatively flat, featureless attenuation curves. As they build up metals and process grains through time, the curves steepen and develop structure. But intense star formation can interrupt this process, flattening the curve again—at least temporarily.

A Dust Curve Is a History Book

These results show that attenuation curves are not static assumptions—they’re fingerprints of a galaxy’s physical state. By tracking how the curve shape varies with dust content, age, SFR, and metallicity, we get a window into how dust evolves over cosmic time.

That has big implications. For observers, it means we can’t safely apply a single dust correction to every galaxy, especially at high redshift. For theorists and simulators, it’s a call to reproduce not just how much dust galaxies contain, but how that dust behaves.

Dust doesn’t just obscure the light. It encodes a story—of growth, destruction, enrichment, and time. And with JWST, we’re finally beginning to read it

Astrobite edited by: Ryan White

Featured Image Credit: Niloofar Sharei (Using Canva)

Author

  • Niloofar Sharei

    I’m an Astronomy PhD candidate at UC Riverside studying how galaxies grow through star-forming clumps. I track how these clumps emerge, evolve, and sometimes survive long enough to reshape their galaxies. When I’m not thinking about cosmic blobs, I’m reading, hiking, or listening to Bach.

    View all posts

Submit a Comment

Your email address will not be published. Required fields are marked *