MIGHTEE: The dark matter haloes, duty cycle and mechanical feedback from radio-AGN up to $z \sim 2.5$

Using MIGHTEE survey data across three fields, this study reveals that radio-AGN reside in increasingly massive dark matter haloes at lower redshifts with a duty cycle of 5–9%, suggesting that their enhanced clustering relative to non-active galaxies stems from either AGN feedback suppressing stellar growth or a preference for earlier-forming haloes.

The Gist

Here is an explanation of the paper, translated from complex astrophysics into everyday language with some creative analogies.

The Big Picture: The "Party" of the Universe

Imagine the Universe as a giant, sprawling party. Most galaxies are just regular guests, chatting and having a good time. But some galaxies are hosting Active Galactic Nuclei (AGN). Think of these AGN as the "VIPs" of the party. They have a supermassive black hole in their center that is actively eating matter and shooting out massive, high-energy jets of particles (like a firehose of energy). These are the Radio-AGN.

Scientists have long known that these VIPs tend to hang out in the most crowded, expensive neighborhoods (dense clusters of galaxies). But a big question remained: Are they only in these rich neighborhoods because they are rich themselves (massive galaxies), or do they actually prefer the neighborhood regardless of how rich they are?

This paper, using data from the MIGHTEE survey (a massive radio telescope project in South Africa), tries to answer that question by looking at how these VIPs cluster together compared to regular guests who look exactly like them on paper.

---

The Investigation: Finding the VIPs

1. The Search Party (The Data)
The researchers used the MeerKAT radio telescope to scan three huge patches of the sky. They found about 2,000 of these "Radio-AGN" VIPs.

• The Challenge: Radio telescopes see the "firehose" (the jets), but they don't always see the "house" (the galaxy) clearly. To understand the VIPs, they had to cross-reference the radio signals with optical and infrared images to find the actual host galaxies.
• The Control Group: To know if the VIPs are special, you need a control group. The researchers created a "Matched Sample" of regular galaxies. These weren't random; they were carefully selected to have the exact same mass and same age (redshift) as the VIPs.
  • Analogy: Imagine you are studying if people who drive Ferraris live in bigger houses. You wouldn't compare them to people who drive tractors. You would compare them to people who drive BMWs of the same model year and price. That's what they did here.

2. The Clustering Test (The "Who's Who" Game)
The team measured how often these galaxies were found near each other.

• The Result: When they looked at the VIPs (Radio-AGN) and the regular BMW-drivers (Matched Galaxies) separately, they looked very similar.
• The Twist: But when they looked at how the VIPs clustered against the entire universe of galaxies, the VIPs were slightly more "clumpy" than the regulars, especially at lower distances.
• The Metaphor: It's like finding that while VIPs and regular guests might have the same number of friends, the VIPs are slightly more likely to be standing in the center of the room, surrounded by a denser crowd of people, even if they look the same on the outside.

---

The Deep Dive: What's in the "Halo"?

Galaxies don't float in empty space; they sit inside invisible bubbles of dark matter called Halo. Think of a halo as the "foundation" or the "real estate" a galaxy sits on.

1. The Size of the Foundation
The researchers used a statistical model (Halo Occupation Distribution) to guess the size of these dark matter foundations.

• Finding: The VIPs (Radio-AGN) are sitting on foundations that are 1.5 to 1.8 times larger than the foundations of the regular galaxies with the same mass.
• Why? This suggests that to turn on the "firehose" (the AGN), you might need a slightly bigger, more massive neighborhood to provide the fuel (cold gas).
• The Time Travel Aspect: They looked at different eras of the Universe.
  • In the past (High Redshift): The foundations were smaller.
  • In the present (Low Redshift): The foundations are bigger.
  • Analogy: Imagine a plant. In the past, the soil was richer and wetter (more cold gas), so the plant could grow big even in a small pot. Now, the soil is drier, so the plant needs a much bigger pot to get the same amount of water.

2. The "Duty Cycle" (How often do they party?)
The researchers calculated how often these black holes are actually "on."

• The Stat: They are active only about 5% to 9% of the time.
• The Metaphor: Think of a lighthouse. It doesn't shine 24/7; it flashes on and off. A Radio-AGN is like a lighthouse that is only "on" for a few million years, then goes dark for a billion years, then turns back on. Over the history of the Universe, these black holes have likely turned on and off many times.

3. The Energy Bill (Heating the Neighborhood)
When the black hole turns on, it shoots out jets that heat up the surrounding gas.

• The Calculation: They calculated the total energy these jets have dumped into the universe.
• The Result: It's a massive amount of energy (enough to heat up a galaxy group significantly).
• Why it matters: Without this heating, the gas in these neighborhoods would cool down too fast and collapse to form too many stars. The AGN acts like a thermostat, blowing hot air to stop the neighborhood from freezing over and forming too many new stars.

---

The Conclusion: Why are they different?

So, why do these Radio-AGN sit in bigger, more crowded neighborhoods than their look-alikes? The paper suggests two main theories:

1. The "Feedback" Theory: The AGN might be causing the difference. By blowing away gas and stopping star formation, the AGN keeps the galaxy's mass low (so it looks like a "regular" galaxy) while the dark matter halo underneath remains massive. It's like a person who stops eating (stopping star growth) but still lives in a mansion (the halo).
2. The "Real Estate" Theory: Maybe the black holes only turn on in neighborhoods that formed a long time ago. These older neighborhoods are naturally more crowded and clustered, giving the black hole more fuel to start its fire.

Summary in One Sentence

This study found that super-energetic black holes (Radio-AGN) live in slightly more massive and crowded dark matter neighborhoods than regular galaxies of the same size, likely because they need that extra "real estate" to fuel their jets, which in turn act as cosmic thermostats to regulate how stars are born in the universe.

Technical Summary

Here is a detailed technical summary of the paper "MIGHTEE: The dark matter haloes, duty cycle and mechanical feedback from radio-AGN up to $z \sim 2.5$."

1. Problem Statement

Active Galactic Nuclei (AGN), particularly radio-loud AGN, are known to reside in massive dark matter haloes and are more strongly clustered than non-active galaxies. However, a fundamental question remains: Is this enhanced clustering solely a consequence of radio-AGN preferentially hosting massive galaxies, or do they occupy distinct environments beyond this mass dependence?

Previous studies have been limited by either small survey areas (high cosmic variance) or shallow flux limits (missing lower-luminosity AGN). Furthermore, it is unclear if the clustering strength depends on radio luminosity or if the typical host halo mass evolves significantly with redshift. This paper aims to resolve these uncertainties by comparing the clustering of radio-AGN against a control sample of galaxies matched in both stellar mass and redshift, utilizing the deep, wide-area MIGHTEE survey.

2. Methodology

Data Sources:

• Radio Data: Utilized Data Release 1 (DR1) from the MIGHTEE (MeerKAT International GHz Tiered Extragalactic Exploration) survey, covering three fields: COSMOS, XMM-LSS, and CDFS. The survey offers high sensitivity ($\sim 1.2-3.6 \, \mu\text{Jy beam}^{-1}$) and resolution ($\sim 5$ arcsec).
• Optical/NIR Data: Cross-matched with deep photometric data from VISTA (VIDEO, UltraVISTA) and HSC-SSP to obtain stellar masses and photometric redshifts ($z$-PDFs).
• Sample Construction:
  • AGN Sample: $\sim 2,000$ radio sources selected by a redshift-dependent luminosity threshold ($L_{\text{AGN}}(z)$) ensuring 95% purity (excluding star-forming galaxies).
  • Control Sample: A "matched galaxy sample" constructed by randomly selecting optical/NIR galaxies that match the AGN sample's joint distribution of stellar mass and redshift (accounting for $z$-PDFs).
  • Redshift Bins: The analysis was divided into three bins: $0 < z < 1$,$1 < z < 1.5$, and$1.5 < z < 2.5$.

Statistical Analysis:

• Clustering Measurement: Measured the Angular Two-Point Correlation Function (TPCF), $\omega(\theta)$, using the Landy-Szalay estimator. Both auto-correlations (AGN vs. AGN, Matched vs. Matched) and cross-correlations (AGN/Matched vs. full optical/NIR population) were calculated.
• Error Estimation: Uncertainties were derived using the jackknife resampling method (100 patches) to account for cosmic variance and bin correlations.
• Halo Occupation Distribution (HOD) Modeling:
  • Used the CCL (Core Cosmology Library) to fit HOD models to the correlation functions.
  • The model assumes a standard 5-parameter framework: $M_{\text{min}}$ (mass for central galaxy), $M_0$ (mass for satellites), $M_1$ (mass for one satellite), $\sigma_{\ln M}$ (transition smoothness), and $\alpha_s$ (satellite scaling).
  • Joint Fitting: To constrain AGN parameters robustly, the study performed a joint fit to the optical/NIR auto-correlation, the AGN cross-correlation, and the matched galaxy cross-correlation simultaneously. This leverages the large statistics of the optical/NIR sample to constrain the bias of the background population.

3. Key Contributions

1. Multi-Field Clustering Analysis: Unlike previous single-field deep studies, this work combines three widely separated fields (COSMOS, XMM-LSS, CDFS), significantly reducing cosmic variance and allowing clustering measurements out to $z \sim 2.5$.
2. Rigorous Control Sample: The construction of a control sample matched in both stellar mass and redshift (using full $z$-PDFs) isolates the environmental effects of AGN activity from simple mass dependence.
3. HOD Modeling of Cross-Correlations: The paper implements a sophisticated joint fitting approach to extract HOD parameters for AGN from cross-correlation data, overcoming the statistical limitations of small AGN sample sizes.
4. Energy Budget Calculation: The study quantifies the total mechanical energy injected by radio jets into the intergalactic medium over cosmic time, comparing it to the heating required to explain observed gas entropy.

4. Key Results

Clustering and Bias:

• Auto-correlation: No statistically significant difference was found between the auto-correlation of AGN and the matched galaxy sample in any redshift bin ($p$-values $> 0.8$).
• Cross-correlation: Significant differences emerged when cross-correlating with the full galaxy population. In the lowest redshift bin ($0 < z < 1$), AGN showed stronger clustering ($\sim 1.5\times$) than matched galaxies on small scales.
• Bias Parameters ($b$):
  • $z \approx 0.76$: $b = 1.94 \pm 0.07$
  • $z \approx 1.25$: $b = 2.50^{+0.11}_{-0.18}$
  • $z \approx 1.75$: $b = 3.38^{+0.27}_{-0.38}$
  • The bias increases with redshift, indicating AGN reside in increasingly rare, high-density peaks at earlier times.

Dark Matter Halo Masses:

• The typical host halo mass decreases with increasing redshift:
  • $z \approx 0.76$: $\log_{10}(\langle M_h \rangle/M_\odot) = 13.44^{+0.08}_{-0.08}$
  • $z \approx 1.25$: $\log_{10}(\langle M_h \rangle/M_\odot) = 13.17^{+0.07}_{-0.06}$
  • $z \approx 1.75$: $\log_{10}(\langle M_h \rangle/M_\odot) = 13.03^{+0.09}_{-0.10}$
• Comparison with Control: Radio-AGN reside in haloes significantly more massive than their stellar-mass-matched counterparts:
  • At $z < 1$: AGN haloes are $1.54^{+0.47}_{-0.33}$ times more massive.
  • At $1.5 < z < 2.5$: AGN haloes are **$1.82^{+1.04}_{-0.57}$** times more massive.
  • At $1 < z < 1.5$, the difference is less significant ($1.11^{+0.25}_{-0.20}$).

Duty Cycle and Feedback:

• Duty Cycle: Estimated at $\sim 5-9\%$ across the redshift range.
• Active Lifetime: The cumulative active time is $\sim 880$ Myr over $0 < z < 2.5$, implying radio-AGN undergo multiple episodes of activity (restarting jets) rather than a single continuous phase.
• Mechanical Feedback: The total kinetic energy deposited by radio jets per halo over $0 < z < 2.5$is estimated at **$\sim 10^{53}$ J**. This is sufficient to account for the observed excess heating of intra-group/cluster gas (0.5–1 keV per particle) beyond gravitational collapse.

5. Significance and Conclusion

This study confirms that radio-AGN inhabit distinct, more massive dark matter haloes than non-active galaxies of the same stellar mass. The findings suggest two primary drivers for this environmental preference:

1. Assembly Bias: AGN may preferentially reside in haloes that formed earlier and are more strongly clustered.
2. Feedback Effects: Jet-mode feedback may suppress star formation in AGN hosts, reducing their stellar mass while leaving the halo mass unchanged. This would cause AGN hosts to appear in lower stellar mass bins than non-active galaxies of the same halo mass.

The observed decrease in typical halo mass with redshift is attributed to the higher abundance of cold gas at earlier cosmic epochs, which fuels AGN activity even in lower-mass haloes. The results provide strong evidence that radio-AGN feedback is a dominant mechanism for heating the intergalactic medium in galaxy groups and clusters, playing a crucial role in regulating galaxy evolution.

Limitations: The primary limitation is the relatively small number of AGN ($\sim 2000$), leading to large statistical uncertainties. Future surveys (SKAO, LSST, Euclid) with larger source counts and better redshift information will allow for tighter constraints and finer redshift binning.