The multiple coherence scales of C IV at cosmic noon

By analyzing high-resolution spectra of quasar pairs at cosmic noon, this study reveals that C IV absorption systems exhibit multiple coherence scales, with a small-scale coherence of approximately 4.7 kpc corresponding to individual clouds and a large-scale coherence of roughly 654 kpc reflecting the extent of C IV-enriched regions and galaxy clustering.

The Gist

Imagine the universe not as empty space, but as a vast, invisible ocean. In this ocean, galaxies are like islands. But what's interesting is that these islands aren't just sitting there; they are surrounded by a foggy, gaseous "halo" called the Circumgalactic Medium (CGM). This fog is made of gas that is constantly flowing in and out of the galaxies, like a cosmic breathing cycle.

The problem? This fog is invisible. You can't see it with a telescope. It's too faint and too spread out.

So, how do astronomers study this invisible fog? They use quasars. Think of quasars as incredibly bright, distant lighthouses shining through the cosmic ocean. When the light from a quasar travels to Earth, it passes through this fog. The fog absorbs some of the light at specific colors, leaving dark "fingerprints" (absorption lines) in the spectrum. By studying these fingerprints, astronomers can tell what the fog is made of.

In this specific paper, the researchers are looking for a specific type of fingerprint: C IV (Carbon IV). This is a signpost for gas that is "warm" and "cool" (relatively speaking, in cosmic terms) and is found around galaxies when the universe was about half its current age (a time astronomers call "cosmic noon").

The Big Question: How Big is the Fog?

For a long time, we knew the fog existed, but we didn't know its shape or size. Is it a giant, continuous cloud stretching for millions of miles? Or is it made of tiny, isolated puddles?

To figure this out, the team used a clever trick: Quasar Pairs.

Imagine you are standing in a foggy forest. If you look at a single tree (a single quasar), you can see the fog in front of it, but you don't know if the fog is a thin mist or a thick wall. But, if you have two trees standing very close together, and you look at them both, you can compare what you see.

• If the fog looks exactly the same in both directions, the fog is likely a big, continuous sheet.
• If the fog looks totally different in one direction than the other, the fog is likely made of small, scattered clumps.

The researchers gathered data on 12 pairs of quasars (some naturally close together, some magnified by gravity like a lens). They looked at the "fog" (C IV gas) between these pairs, measuring the distance between the two lines of sight from a tiny fraction of a kilometer up to millions of kilometers.

The Discovery: A Three-Layered Structure

The team found that the fog isn't just one thing. It has three distinct scales, like a Russian nesting doll:

1. The Giant Envelope (The "Neighborhood"):
   At large distances (up to about 650,000 light-years or 200 kiloparsecs), the gas behaves like a giant, connected cloud surrounding a galaxy. This is the "enriched region" where the galaxy has polluted the space around it with heavy elements (like carbon). It's like a neighborhood where everyone shares the same air.

2. The Flat Middle (The "Suburbs"):
   Between about 5,000 and 500,000 light-years, the gas correlation flattens out. It's like walking through the suburbs of a city; you see houses (galaxies), but the specific "fog" of one house doesn't strongly overlap with the next. The gas is there, but it's not tightly connected in a specific way at this scale.

3. The Tiny Clumps (The "Clouds"):
   Here is the most exciting part. When they looked at very small scales (less than 5,000 light-years, or about 4.7 kpc), they found a sharp spike in correlation. This means the gas isn't a smooth, continuous sheet. Instead, it's made of individual, distinct clouds.

   • Analogy: Imagine looking at a cloud in the sky. From far away, it looks like a single white blob. But if you zoom in, you realize it's actually made of millions of tiny, separate water droplets. The researchers found that the C IV gas is made of these "droplets" or "clouds" that are only a few thousand light-years across.

Why Does This Matter?

This discovery changes how we think about the "breathing" of galaxies.

• It's not a smooth wind: The gas flowing in and out of galaxies isn't a smooth, continuous stream. It's more like a spray of mist or a collection of rain clouds.
• Galaxy Clustering: The size of these giant "enriched regions" (650,000 light-years) matches how galaxies themselves cluster together. This suggests that the gas and the galaxies are deeply linked; the gas traces the structure of the galaxies themselves.
• The "Cloud" Size: The fact that the gas breaks down into small clouds (about 5,000 light-years wide) tells us that the physics of the gas is complex. It's turbulent and clumpy, not calm and smooth.

The Bottom Line

Think of the universe around a galaxy like a giant, multi-layered soup.

• The big pot is the galaxy cluster.
• The broth is the giant cloud of gas surrounding the galaxy (650,000 light-years wide).
• But if you look closely at the broth, you see it's actually full of tiny, distinct dumplings (the individual gas clouds, about 5,000 light-years wide).

This paper is the first time we've been able to clearly see both the "pot" and the "dumplings" at the same time. It tells us that the universe is much more structured and "clumpy" than we previously thought, even in the vast spaces between galaxies.

Technical Summary

Here is a detailed technical summary of the paper "The multiple coherence scales of C iv at cosmic noon" by Cortés-Muñoz et al. (2026).

1. Problem Statement

The spatial and kinematic structure of the Circumgalactic Medium (CGM)—the diffuse gas surrounding galaxies—remains poorly constrained observationally. While single lines of sight (toward quasars or galaxies) provide information on gas abundance and kinematics, they cannot unambiguously determine the spatial clustering or physical extent of these gas structures. Previous studies have focused largely on large-scale clustering (megaparsec scales), leaving the small-scale structure (sub-kiloparsec to kiloparsec scales) of metal-enriched gas largely unexplored. Specifically, the coherence lengths of individual C IV-bearing clouds and the size of the metal-enriched regions around galaxies at "cosmic noon" ($z \approx 2$) are unknown.

2. Methodology

The authors utilized a unique dataset of quasar pairs (both projected and gravitationally lensed) to probe the CGM across a wide range of transverse separations.

• Sample Construction:
  • Data Source: High-resolution spectra ($R \approx 45,000$) from VLT/UVES and Keck/HIRES.
  • Sample Size: 12 quasar pairs total: 8 projected pairs and 4 lensed pairs.
  • Redshift Range: $1.6 \lesssim z \lesssim 3.3$(emission redshifts$2.08 \leq z \leq 3.22$).
  • Transverse Separations ($\Delta r$): Ranging from sub-kiloparsec ($\sim 20$ pc) to a few megaparsec ($\sim 2.4$ Mpc).
• Absorption Analysis:
  • Identified C IV $\lambda\lambda 1548, 1550$ doublet absorption systems.
  • Fitted Voigt profiles to 141 C IV systems, extracting 620 individual velocity components (327 on line-of-sight A, 293 on line-of-sight B).
  • Ensured 90% completeness for column densities $\log(N/\text{cm}^{-2}) > 12.5$.
• Statistical Analysis:
  • Computed the two-point correlation function, $\xi(\Delta v, \Delta r)$, using the generalized Landy & Szalay (1993) estimator.
  • Calculated the Projected Transverse Correlation Function (PTCF), $\Xi(\Delta r)$, by integrating $\xi$ over velocity space (assuming Hubble flow).
  • Employed bootstrapping ($N_{BS}=100$) to estimate uncertainties, addressing potential underestimation of variance in Poisson-limited samples.
  • Tested three models for $\Xi(\Delta r)$: a single power law, a broken power law (one break), and a double broken power law (two breaks).

3. Key Contributions

• Largest Archival Sample: This study presents the largest archival sample of high-resolution echelle spectra for closely separated quasar pairs used to measure C IV clustering.
• Multi-Scale Analysis: It bridges the gap between large-scale (galaxy clustering) and small-scale (cloud internal structure) studies, covering scales from 20 pc to 2.4 Mpc.
• Novel Modeling: The authors introduce a double broken power-law model to describe the C IV correlation function, identifying two distinct physical regimes (enriched regions vs. individual clouds) rather than a single scaling relation.

4. Key Results

The analysis of $\xi(\Delta r)$ and $\Xi(\Delta r)$ revealed a complex, multi-regime structure:

1. Large Scales ($\Delta r \gtrsim 500$ kpc):

   • The correlation amplitude decreases with separation, consistent with the clustering of galaxies.
   • The data aligns with the galaxy-galaxy autocorrelation function (Durkalec et al. 2015), suggesting C IV traces the distribution of galaxies at these scales.

2. Intermediate Scales ($5 \text{ kpc} \lesssim \Delta r \lesssim 500$ kpc):

   • The correlation function flattens (becomes constant).
   • This plateau indicates that the survey is probing distances smaller than the typical size of the metal-enriched regions surrounding galaxies. Once $\Delta r$ is smaller than the region size, the probability of finding a pair does not increase significantly.

3. Small Scales ($\Delta r \lesssim 5$ kpc):

   • The correlation amplitude increases sharply as separation decreases, peaking at the smallest scales ($\sim 0.1$ kpc).
   • This rise is attributed to pairs of velocity components belonging to the same individual C IV absorption system (clouds).

Derived Coherence Lengths:
By fitting the double broken power-law model (Eq. 13), the authors inferred two distinct coherence scales:

• $r_1 = 654^{+100}_{-87}$ kpc: Represents the characteristic size of the C IV-enriched regions (the "halo" of metal-enriched gas) around galaxies at $z \approx 2$.
• $r_2 = 4.70^{+1.60}_{-1.19}$ kpc: Represents the characteristic size of individual C IV-bearing clouds within the CGM.

Velocity Clustering:

• The velocity correlation function $\xi(\Delta v)$ at small separations ($\Delta r < 10$ kpc) was found to be consistent with previous measurements from single lines of sight and independent measurements using gravitational arcs (Lopez et al. 2024), validating the consistency of point-source and extended-source measurements.

5. Significance

• Tracer of Galaxy Clustering: The results confirm that C IV is an excellent tracer of galaxy clustering at cosmic noon, as the large-scale correlation matches galaxy autocorrelation.
• CGM Structure: The discovery of a sharp break at $\sim 5$ kpc provides the first robust observational constraint on the size of individual cool-warm CGM clouds, suggesting they are significantly smaller than the overall metal-enriched halos.
• Methodological Validation: The study demonstrates that combining projected and lensed quasar pairs allows for a comprehensive mapping of the CGM from sub-kiloparsec to megaparsec scales, resolving ambiguities inherent in single-line-of-sight studies.
• Future Directions: The authors note that while the current data constrains scales down to $\sim 10$ pc, further high-resolution data on lensed systems is required to precisely measure the coherence length at scales $< 1$ kpc.

In summary, this paper establishes that the CGM at cosmic noon is structured hierarchically: galaxies are surrounded by large ($\sim 650$ kpc) metal-enriched regions, which are themselves composed of smaller ($\sim 5$ kpc) discrete clouds of C IV gas.