Spatially Resolved AGN Ionization and Star Formation at Cosmic Noon with JWST/JEMS

Using JWST/JEMS observations of approximately 200 galaxies at Cosmic Noon ($z\approx 2.5-2.9$), this study demonstrates that AGN hosts exhibit significantly more extended ionized gas than control galaxies, with a strong correlation between gas extent and AGN luminosity suggesting that AGN activity often dominates gas ionization on galaxy-wide scales during this epoch.

The Gist

Imagine the universe when it was about 2 to 3 billion years old. Astronomers call this era "Cosmic Noon." It was the galaxy's "golden age," a time when stars were being born at a frantic pace and supermassive black holes at the centers of galaxies were feasting on gas, glowing with intense energy.

For a long time, scientists have debated a cosmic tug-of-war: Who is the real boss of the gas in these galaxies? Is it the stars (the "star formation" team) or the Active Galactic Nuclei (AGN) (the black hole team)? Both produce intense radiation that ionizes (charges up) the gas around them, but they do it in different ways.

This paper is like a high-resolution detective story using the James Webb Space Telescope (JWST) to settle the argument. Here is the breakdown in simple terms:

1. The Tools: A Cosmic "Flashlight" and "X-Ray Vision"

The team used JWST's powerful camera (NIRCam) to look at about 200 galaxies. They used a special trick called medium-band imaging.

• The Analogy: Imagine trying to see a specific color of light (like a neon sign) in a room full of bright white lightbulbs. Usually, the white light washes out the neon. JWST's filters act like special sunglasses that block out the white lightbulbs, letting the team see only the neon signs.
• The Neon Signs: They looked for two specific types of "neon":
  1. [O III] + Hβ: A bright glow that can be caused by either hot stars or black holes. It's like a generic "energy" light.
  2. Paβ (Paschen Beta): A specific glow from hydrogen that is very hard to hide with dust. It's a better tracer for where the gas is actually being ionized, especially in dusty, messy environments.

2. The Investigation: Measuring the "Glow"

The scientists didn't just count the galaxies; they measured how far out the glow extended.

• The Analogy: Think of a campfire.
  • Control Galaxies (No Black Hole): The light comes from the fire itself (stars). The glow is usually compact, like a cozy campfire.
  • AGN Galaxies (With Black Hole): The black hole is like a massive, blinding spotlight. The team wanted to see if this spotlight lit up a much larger area of the forest (the galaxy) than the campfire did.

3. The Findings: The Black Hole Wins (But the Stars Help)

Here is what they discovered:

• The Black Hole's Reach is Bigger: In galaxies with active black holes, the ionized gas glowed significantly further out (about 30% larger) than in galaxies without black holes.
• The "Cone" Shape: Many of the black hole galaxies showed a "cone" or "bicone" shape in the gas.
  • The Analogy: Imagine a lighthouse beam cutting through fog. The black hole shoots out radiation in a cone shape, lighting up the gas in a specific direction, whereas stars light up the gas more evenly around them.
• The Size vs. Power Rule: They found a clear rule: The brighter the black hole, the bigger the cone of gas it lights up. This relationship was surprisingly similar to what we see in nearby, older galaxies, suggesting the physics hasn't changed much over billions of years.
• The Dusty Twist: In some cases, the "hydrogen glow" (Paβ) was actually larger than the "oxygen glow" ([O III]).
  • The Analogy: Imagine a dusty room. The bright white light (oxygen) gets blocked by the dust, but the red light (hydrogen) passes through. This told the scientists that in some galaxies, the gas is so dusty that the black hole's light is being hidden, but the stars are still doing their work.

4. The Big Picture: A Team Effort

The most important conclusion is that at Cosmic Noon, both teams are playing.

• The galaxies with the biggest, most extended gas clouds were the ones with both active black holes and intense star formation.
• The black holes seem to dominate the "big picture" (lighting up huge cones of gas), but the stars are still contributing significantly to the local neighborhoods.

Summary

Think of a galaxy at Cosmic Noon as a busy city.

• Stars are the streetlights: They light up the immediate blocks nicely.
• Black Holes are the massive stadium floodlights: They can light up the entire city and the surrounding countryside, but they often shine in specific beams (cones).

This paper used the sharpest "eyes" in history (JWST) to prove that while the stars are doing a lot of work, the black holes are the ones turning on the stadium lights, shaping the gas on a massive scale. However, because the city is so dusty and chaotic, you often need to look at the "red light" (hydrogen) to see the full picture of what's happening.

Technical Summary

1. Problem Statement and Scientific Context

The period known as "Cosmic Noon" ($z \approx 2-3$) represents the peak epoch of both cosmic star formation and black hole accretion. During this era, Active Galactic Nuclei (AGN) and intense star formation frequently coexist within the same galaxies, making it difficult to disentangle their respective contributions to the ionization of interstellar gas.

• The Challenge: While the rest-frame optical [O III]$\lambda$5007 line is a standard tracer of ionized gas, it is produced by both massive stars and AGN. Previous high-redshift studies were limited by small sample sizes, observational biases (focusing on luminous quasars), and insufficient spatial resolution/sensitivity to detect extended, low-surface-brightness emission.
• The Gap: It remains unclear whether AGN radiation or stellar processes dominate gas ionization at $z \sim 3$, how the spatial extent of ionized gas scales with AGN luminosity at these redshifts, and how dust obscuration affects these measurements.

2. Methodology

The authors utilized data from the JWST Extragalactic Medium-band Survey (JEMS) in the GOODS-S field to conduct a spatially resolved analysis of $\sim 200$ galaxies at $2.5 < z < 2.9$.

• Data & Tracers:
  • JWST/NIRCam Medium-band Imaging: Used filters F182M, F210M, F430M, F460M, and F480M.
  • Tracer 1 ([O III]+H$\beta$): Mapped using F182M (line) minus F210M (continuum). This traces ionized gas in both AGN Narrow Line Regions (NLR) and star-forming H II regions.
  • Tracer 2 (Pa$\beta$): Mapped using F460M/F480M (line) minus F430M (continuum). As a near-IR hydrogen recombination line, Pa$\beta$ is less sensitive to dust extinction than optical lines, providing a complementary view of ionized gas and star formation.
• Sample Selection:
  • Initial sample of 307 objects with $SNR > 5$ in relevant filters.
  • Final sample of 208 galaxies after visual inspection, removal of contaminants, and redshift verification.
  • Sub-samples:
    • AGN Hosts: 33 galaxies (16%) identified via multiwavelength diagnostics (X-ray, MIR SED, radio, color-color, and SED modeling).
    • Pa$\beta$-detected: 32 galaxies (15%) showing clear Pa$\beta$ emission.
    • Control: 175 non-AGN galaxies.
• Analysis Techniques:
  • Emission Line Mapping: Continuum-subtracted maps constructed with scaling factors ($f$) derived from AGNfitter-rx (for [O III]+H$\beta$) and CIGALE (for Pa$\beta$) to account for continuum slopes.
  • Size Measurements: Spatial extents measured using a uniform 4$\sigma$ surface brightness threshold. Metrics included average radial size ($R_{avg}$), maximum radial size ($R_{max}$) of the largest feature, and maximum extent ($D_{max}$) from the galaxy center.
  • SED Modeling: Used AGNfitter-rx (Bayesian MCMC) to derive AGN luminosity ($L_{AGN}$), host galaxy luminosity, Star Formation Rate (SFR), and stellar mass ($M_*$).

3. Key Contributions

• Statistically Robust Sample: This is the first study to apply spatially resolved ionization diagnostics to a statistically significant sample ($N \approx 200$) at Cosmic Noon, moving beyond the small samples of luminous quasars typical of previous work.
• Dual-Tracer Approach: The simultaneous mapping of [O III]+H$\beta$ and Pa$\beta$ allows for a differential analysis of ionization sources, leveraging the dust-insensitivity of Pa$\beta$ to probe obscured regions.
• Refined Size-Luminosity Relation: The study establishes a new, tighter constraint on the [O III] extent vs. AGN luminosity relation at $z \sim 3$, correcting for previous biases related to surface brightness limits and sample size.

4. Key Results

A. Morphologies and Spatial Extents

• Morphologies: AGN hosts exhibit more extended and asymmetric [O III]+H$\beta$ morphologies (conical/biconical, knotted, extended) compared to the control sample, which is dominated by compact or knotted structures. Approximately 25% of AGN show conical morphologies.
• Extent Differences:
  • AGN vs. Control: AGN hosts have systematically larger ionized gas extents than control galaxies.
    • Median $R_{max}$ for [O III]+H$\beta$: 2.2 kpc (AGN) vs. 1.1 kpc (Control).
    • Median $R_{max}$ for Pa$\beta$: 2.3 kpc (AGN) vs. 1.1 kpc (Control).
  • Tracer Comparison: [O III]+H$\beta$ is modestly more extended than Pa$\beta$ by $\sim 0.1$ dex ($\sim 0.2-0.5$ kpc). However, in some dust-obscured systems, Pa$\beta$ appears more extended, suggesting dust attenuation suppresses the optical [O III]+H$\beta$ signal.

B. The [O III] Extent–AGN Luminosity Relation

• Slope Measurement: The authors derived the slope ($m$) of the relation $\log(R) \propto m \log(L_{AGN})$.
  • For $R_{max}$: $m \approx 0.21$.
  • For $R_{avg}$: $m \approx 0.19$.
• Comparison to Low-$z$: These slopes are consistent with the shallow end of relations measured at low redshift ($z < 0.5$), which typically range from $m \sim 0.2$ to $0.5$.
• Interpretation: The shallow slope suggests a "matter-bounded" regime, where the spatial extent of the NLR is limited by the availability and distribution of gas in the host galaxy rather than solely by the intensity of the AGN radiation field (which would produce a steeper, "ionization-bounded" slope).

C. Pa$\beta$ and Stellar Contributions

• Pa$\beta$ Detection: 52% of AGN and only 9% of control galaxies were detected in Pa$\beta$.
• Correlations:
  • Pa$\beta$ extent shows a weak negative correlation with AGN luminosity but a steep positive correlation ($m \sim 0.9$) with host galaxy luminosity and SFR.
  • This indicates that Pa$\beta$ emission is strongly driven by star formation, even in AGN hosts.
• Overlap: Pa$\beta$-detected galaxies occupy similar regions of parameter space (color, SFR, stellar mass) as AGN hosts, suggesting that high-mass, dusty, star-forming galaxies are the primary sites for both AGN activity and detectable Pa$\beta$ emission.

5. Significance and Implications

• Dominance of Ionization: The results suggest that at Cosmic Noon, AGN activity likely dominates the ionization of gas on large scales (kpc scales) in galaxies with mixed activity, as evidenced by the larger extents in AGN hosts. However, stellar processes remain significant contributors, particularly in compact, high-surface-brightness regions.
• Gas Availability: The shallow slope of the extent-luminosity relation implies that the gas reservoir of the host galaxy is the limiting factor for the size of the ionized region, rather than the AGN power itself.
• Dust and Obscuration: The study highlights the importance of dust. In some systems, dust obscuration suppresses optical [O III] emission, making near-IR tracers like Pa$\beta$ essential for a complete census of ionized gas.
• Methodological Advance: The work demonstrates the power of JWST medium-band imaging to perform spatially resolved studies of ionization physics in the early universe, overcoming the limitations of ground-based IFU observations.

In conclusion, this paper provides a comprehensive view of how AGN and stars shape the gaseous reservoirs of galaxies at Cosmic Noon, revealing that while AGN drive large-scale ionization, the physical extent of this ionization is heavily regulated by the host galaxy's gas content and dust properties.