🔭 astrophysics

High-Redshift Galactic Outflows: Orientation Effects, Kinematics, and Metallicity in TNG50 and SERRA

This study utilizes TNG50 and SERRA simulations to analyze high-redshift galactic outflows, revealing that while simulated outflow masses align reasonably well with JWST observations, simulated velocities are significantly lower and detection rates are strongly influenced by galaxy orientation, particularly for massive disc systems.

Original authors: Ivan Kostyuk, Stefano Carniani, Mahsa Kohandel, Andrea Pallottini

Published 2026-05-20

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Ivan Kostyuk, Stefano Carniani, Mahsa Kohandel, Andrea Pallottini

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the early Universe as a bustling construction site where new galaxies are being built. Just like a construction crew, these young galaxies have powerful engines (supernovae and black holes) that blast out clouds of gas. These blasts, called galactic outflows, are crucial because they clear away the raw materials needed to build stars, effectively deciding whether a galaxy will keep growing or stop early.

Recently, the James Webb Space Telescope (JWST) started taking pictures of these ancient construction sites. However, astronomers noticed a puzzle: the telescope is only "seeing" outflows in about 25–40% of the galaxies it looks at. But computer models suggest that every active galaxy should be blasting gas out. Why is the telescope missing so many?

To solve this mystery, the authors of this paper acted like cosmic detectives. They used two different, highly detailed computer simulations of the Universe—TNG50 and SERRA—to create their own "virtual telescopes" and compare them to the real JWST data.

Here is what they found, explained simply:

1. The "Heavy" vs. The "Fast" (Mass and Speed)

The Mass Match: When the scientists looked at how much gas was being blown out, their simulations matched the JWST observations pretty well. The simulations predicted a bit more gas than the telescope saw, which makes sense. Think of it like looking at a campfire: the telescope only sees the bright, hot orange flames (ionized gas), while the simulation sees the flames plus the invisible, hot smoke and embers (multiphase gas). The telescope is just seeing a fraction of the total mess.
The Speed Gap: This is where things got weird. The gas in the simulations was moving 10 times slower than the gas astronomers see in the real Universe. It's as if the simulations are showing a gentle breeze, while the real Universe is screaming with a hurricane. The authors suggest their computer models might be underestimating how hard the "engines" (supernovae and black holes) push the gas.

2. The "Angle of the Camera" (Orientation)

This is the paper's biggest discovery regarding the "missing" outflows.

The Analogy: Imagine a sprinkler system on a lawn. If you stand directly in front of the sprinkler (looking at the "face" of the galaxy), you see a massive, wide spray of water. But if you stand to the side, looking at the sprinkler from the edge (the "edge-on" view), the water stream looks thin and might even look like it's just part of the grass.
The Finding: The simulations showed that if a galaxy is a flat disk (like a pizza), it is much easier to spot its outflows if you are looking at it from the top (face-on). If you look at it from the side (edge-on), the outflows get hidden behind the galaxy's own gas and rotation.
- In the TNG50 simulation, face-on galaxies were about 15% more likely to be detected than edge-on ones. For bigger, flatter galaxies, this difference jumped to 40%.
- This suggests that JWST might be missing many outflows simply because those galaxies are turned sideways relative to us, hiding their "spray" from view.

3. The "Merger Chaos" (Why SERRA was different)

The authors used two different simulations because they work differently.

TNG50 is like a calm, organized construction site where galaxies form nice, flat disks. Here, the "angle" rule worked perfectly.
SERRA is like a chaotic construction site where galaxies are constantly crashing into each other (merging). In this simulation, the outflows were often caused by these crashes (tidal tails) rather than just a steady wind. Because the gas was flying everywhere due to the crash, it didn't matter which way you looked; the outflows were messy and hard to hide. Interestingly, this chaos meant SERRA didn't show the same "angle" effect as TNG50, and it predicted even fewer detectable outflows than what JWST actually sees.

4. The "Dirty Gas" (Metallicity)

The team also checked the "dirtiness" of the gas (how many heavy elements, or "metals," it contains).

They found that the gas being blasted out is actually cleaner (has fewer metals) than the gas sitting inside the galaxy.
Why it matters: Astronomers usually estimate how much gas is being blown out by measuring the brightness of specific chemical lines (like oxygen). If they assume the gas is as "dirty" as the galaxy itself, they will underestimate the total amount of gas. It's like trying to guess the weight of a bag of sand by weighing a bag of gold; if the sand is actually lighter, your math is wrong. The authors suggest we might be underestimating the total mass of these outflows by a factor of two to eight.

The Bottom Line

The paper concludes that the reason we don't see outflows in every galaxy isn't necessarily because they aren't there. It's a combination of:

Hiding: We are looking at some galaxies from the "wrong" angle, hiding the outflows behind the galaxy's disk.
Blindness: Our telescopes might only be seeing the "bright flames" and missing the "smoke," leading us to underestimate the total amount of gas.
Model Limits: Our computer models might be too gentle, not pushing the gas fast enough to match the wild speeds we see in reality.

By understanding these "optical illusions" and physical limitations, astronomers can better interpret what JWST is seeing and get a clearer picture of how galaxies grow and die in the early Universe.

Technical Summary: High-Redshift Galactic Outflows: Orientation Effects, Kinematics, and Metallicity in TNG50 and SERRA

Problem Statement
Galactic outflows, driven by supernovae and active galactic nuclei (AGN), are critical for regulating star formation and galaxy evolution. While cosmological simulations predict that fast outflowing gas should be ubiquitous in star-forming galaxies, recent James Webb Space Telescope (JWST)/NIRSpec observations of low-mass ( $M_\star \sim 10^7 M_\odot$ ) galaxies at high redshifts ( $z > 3$ ) reveal a lower occurrence rate of warm ionized outflows (25–40%) than expected. This discrepancy raises questions regarding the intrinsic incidence of outflows, the influence of viewing angle (projection effects), and the physical conditions of the outflowing gas. The authors aim to resolve these tensions by characterizing outflows in high-redshift galaxies using two distinct simulation frameworks and comparing them directly to JWST/JADES survey data.

Methodology
The study utilizes two cosmological hydrodynamic simulations with fundamentally different numerical approaches:

TNG50: A high-resolution cosmological simulation using the AREPO code (moving mesh, MHD) with subgrid models for star formation and feedback (kinetic wind particles).
SERRA: A suite of zoom-in simulations using the RAMSES code (adaptive mesh refinement) with on-the-fly radiative transfer and a non-equilibrium chemical network.

Sample Selection:

TNG50: $\sim 6 \times 10^4$ galaxies with stellar masses $M_\star > 10^{7.5} M_\odot$ in the redshift range $z = 3–6$ .
SERRA: $\sim 3 \times 10^3$ galaxies with halo masses $M_h \ge 10^9 M_\odot$ (corresponding to $M_\star \gtrsim 10^{7.5} M_\odot$ ) in the range $z = 4–6$ .

Outflow Identification Algorithm:
To distinguish outflowing gas from the galactic disk and inflows within the immediate vicinity of galaxies (mimicking observational apertures), the authors developed a novel separation method:

Gaussian Mixture Model (GMM): Gas particles are mapped into a 5-dimensional space (3 spatial coordinates, radial velocity, and Star Formation Rate). A multimodal Gaussian profile ( $n_{modes}=3$ ) is fitted to separate particles into: (1) galactic gas (random motion near the center), and (2–3) outflow components.
Additional Criteria: Particles with substantial velocity components perpendicular to the galactic plane ( $|v_z| > 2\sigma$ ) are also classified as outflows to capture tangentially launched gas.
Observational Mimicry: Outflow properties (mass, velocity) are calculated within a physical radius corresponding to a $0.6''$ aperture ( $\sim 3.5–4.7$ kpc) to match JWST/NIRSpec MOS observations.

Key Results

Outflow Masses: Both simulations reproduce the observed outflow masses within $\sim 0.5$ dex, though TNG50 predicts slightly higher values. The authors suggest this offset implies that optical emission lines (H $\alpha$ , [O III]) capture only a fraction ( $\sim 35\%$ ) of the total multiphase outflow. SERRA results align more closely with JADES observations, potentially due to different feedback implementations (preventative feedback vs. kinetic winds).
Outflow Velocities: A significant discrepancy exists in kinematics. Simulated outflow velocities are typically an order of magnitude lower than those inferred from JWST observations. In TNG50, velocities are prescribed to scale with halo velocity dispersion, which may underpredict wind speeds compared to reality.
Metallicity: Outflows exhibit significantly lower metallicities than their host galaxies ( $\langle Z_{out}/Z_{gal} \rangle \approx 0.52$ in TNG50 and $0.16$ in SERRA). This is attributed to outflows sweeping up unenriched inflowing gas or originating from younger, less enriched regions. Consequently, mass estimates derived from metal-line luminosity may be systematically underestimated by a factor of 2–8.
Orientation Dependence (TNG50): Detectability is strongly dependent on viewing angle. Face-on galaxies are $\sim 15\%$ more likely to show detectable outflows than edge-on systems. For massive, disc-shaped galaxies ( $M_\star > 10^{8.5} M_\odot$ ), this difference increases to nearly $40\%$ , as edge-on views suffer from overlapping velocity distributions with the galactic disk.
Orientation Dependence (SERRA): Unlike TNG50, SERRA shows minimal orientation dependence. This is attributed to the high frequency of mergers in the SERRA sample ( $\sim 60\%$ ), where tidal features and merger remnants dominate the kinematic signatures, masking the underlying galactic morphology and outflow direction.
Temporal Evolution: In intermediate-mass galaxies, outflow activity correlates with specific star formation rate (sSFR) and declines toward lower redshifts. In massive galaxies, outflow activity resurges at later times ( $z < 3$ ) driven by central AGN feedback, even as sSFR declines.

Significance and Claims
The paper claims to provide a robust framework for comparing simulated and observed outflows by developing a GMM-based identification method that operates at the same spatial scales as current telescopes. By utilizing two independent simulations, the authors disentangle intrinsic physical trends from model-specific artifacts.

The study concludes that the lower-than-expected incidence rate of outflows in JWST observations can be partially explained by orientation effects (particularly in disc-dominated systems) and sensitivity limitations regarding the multiphase nature of the gas. However, the persistent velocity discrepancy suggests that current simulation models may systematically underpredict wind speeds or that observational velocity inference methods differ significantly from simulation definitions. The authors emphasize that metallicity biases in mass estimation and the complex kinematics of merging systems (as seen in SERRA) are critical factors often overlooked in simple comparisons.

1. The "Heavy" vs. The "Fast" (Mass and Speed)

2. The "Angle of the Camera" (Orientation)

3. The "Merger Chaos" (Why SERRA was different)

4. The "Dirty Gas" (Metallicity)

The Bottom Line

More like this