🔭 astrophysics

Challenges of standard halo models in constraining galaxy properties from CIB anisotropies

This study demonstrates that while current halo models can fit cosmic infrared background anisotropy data well, they fail to reliably recover key physical parameters like peak star formation efficiency mass due to fundamental limitations in their treatment of halo bias and matter clustering, necessitating improved cosmological ingredients for robust parameter constraints.

Original authors: Athanasia Gkogkou, Guilaine Lagache, Matthieu Béthermin, Abhishek Maniyar

Published 2026-02-05

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Athanasia Gkogkou, Guilaine Lagache, Matthieu Béthermin, Abhishek Maniyar

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 universe as a giant, dark ocean. We can't see the water (dark matter) directly, but we can see the bioluminescent fish (galaxies) that live inside it. These fish glow with infrared light because they are busy giving birth to new stars.

Astronomers try to map this ocean by looking at the "flicker" or "static" of this light, known as the Cosmic Infrared Background (CIB). To make sense of this static, they use a mathematical recipe called the Halo Model. Think of this model as a set of instructions that says: "If you have a dark matter 'bubble' of this size, it should contain this many glowing fish, and they should shine this brightly."

This paper is like a quality control check. The authors asked: "If we feed our recipe the exact data it was designed to create, can it correctly figure out the original instructions?"

Here is the breakdown of their experiment and what they found:

1. The Experiment: The "Fake Universe"

The authors built a simulated universe (called SIDES-Uchuu). This isn't a real observation; it's a computer program where they know the "true" answers. They know exactly how big the dark matter bubbles are, how many fish are in them, and how efficiently those fish are making stars.

They then took this simulated data and tried to fit their standard recipe (the M21 Halo Model) to it, pretending they didn't know the answers beforehand.

2. The Good News: The Picture Looks Right

When the authors compared the model's output to the simulated data, the pictures matched perfectly.

The Analogy: If you take a photo of a forest and then use a computer program to guess how many trees are in it, the program might guess the total number of trees and the overall shape of the forest correctly.
The Result: The model successfully predicted the overall brightness of the universe and the general rate at which stars are being born.

3. The Bad News: The "Why" is Wrong

Here is where the paper gets interesting. Even though the picture looked right, the internal details were wrong.

The Analogy: Imagine you are trying to guess the recipe for a cake just by tasting the final product. You might guess the cake is sweet and fluffy (correct), but you might guess it was baked at 400°F when it was actually baked at 350°F. You got the result right, but the process was wrong.
The Result: The model failed to accurately identify which size of dark matter bubble is the most efficient at making stars. It consistently guessed that the "best" bubbles were much larger than they actually were in the simulation.

4. The "Perfect Match" Test

To be sure the problem wasn't just a difference in how the simulation and the model spoke to each other, the authors created a "Simplified Universe" (SSU).

They forced the simulation to speak the exact same language as the model. They removed all the complex, messy real-world variables and made the simulation follow the model's rules perfectly.
The Shock: Even with this perfect match, the model still failed to recover the correct answers.

5. Why Did It Fail?

The authors dug deeper and found the culprit. The problem wasn't the "fish" (the galaxies) or how they shine; the problem was the map of the ocean (the dark matter structure).

The Culprit: The model uses a simplified, straight-line approximation to describe how dark matter clumps together. In reality, dark matter clumps in a more complex, curved way.
The Metaphor: It's like trying to measure the distance between two cities using a ruler on a flat map, when the Earth is actually round. The ruler (the model) gives you a number, but because it ignores the curve of the Earth (the complex physics of the universe), your calculation of the "true distance" is off.

Summary

The paper concludes that while our current "recipe" for understanding the universe's background light is good at predicting what the light looks like, it is currently bad at telling us the true physical rules behind it.

To fix this, astronomers need to upgrade the "map" part of their recipe to include more complex physics about how dark matter clumps, rather than just focusing on the galaxies themselves. Until then, we can trust the pictures, but we should be careful about the specific numbers we derive from them.

Technical Summary: Challenges of Standard Halo Models in Constraining Galaxy Properties from CIB Anisotropies

Problem Statement
The halo model, combined with Halo Occupation Distribution (HOD) prescriptions, is the standard framework for interpreting Cosmic Infrared Background (CIB) anisotropies to extract physical properties of star-forming galaxies and their connection to large-scale structure. While recent implementations, such as the Maniyar et al. (2021) (M21) model, utilize more physical parameterizations linking star formation rate (SFR) to baryonic accretion rates (BAR), the reliability of these models in recovering intrinsic astrophysical parameters remains uncertain. Specifically, it is unclear whether discrepancies in recovered parameters (e.g., the halo mass at peak star formation efficiency, $M_{max}$ , and the peak efficiency, $\eta_{max}$ ) arise from the specific HOD assumptions (galaxy emission) or from fundamental limitations in the halo model framework itself (e.g., halo bias and matter clustering approximations).

Methodology
The authors employ a rigorous validation approach by fitting the M21 HOD model to mock CIB data generated from the SIDES-Uchuu simulation, a hybrid of the SIDES infrared dust simulation and the Uchuu N-body cosmological simulation. The analysis proceeds in three stages:

Initial Fitting: The M21 model is fitted to mock CIB power spectra (auto- and cross-power spectra at Planck frequencies: 217, 353, 545, and 857 GHz) and Star Formation Rate Density (SFRD) data derived from the standard SIDES-Uchuu simulation. The fitting uses a Markov Chain Monte Carlo (MCMC) approach with a full covariance matrix to account for frequency correlations.
Simplified Simulation (SSU): To isolate the impact of the HOD parameterization, the authors create a "Simplified SIDES-Uchuu" (SSU) simulation. This version explicitly aligns with the M21 model's assumptions by:
- Using the analytic SFR parameterization (Eq. 3) dependent solely on halo mass and redshift, rather than the redshift-evolved main sequence.
- Utilizing effective Spectral Energy Distribution (SED) templates dependent only on redshift, rather than templates varying with the mean radiation field ( $\langle U \rangle$ ).
- Disabling gravitational lensing ( $\mu=1$ ).
- Optionally disabling the quenching of satellite galaxies to match the model's treatment.
Component Analysis: The authors systematically compare the simulation outputs with the halo model's constituent terms (one-halo, two-halo, and shot noise) to trace the origin of any remaining discrepancies. They also test the impact of introducing scatter in the SFR–halo mass relation and SED templates.

Key Results

Fit Quality vs. Parameter Recovery: While the M21 HOD model provides a statistically good fit to the mock CIB power spectra (low $\chi^2$ ), it fails to accurately recover the intrinsic astrophysical parameters of the simulation. Specifically, the halo mass at which star formation efficiency peaks ( $M_{max}$ ) is consistently overestimated by the model, particularly at $z < 2$ .
Persistence of Discrepancies: Even when fitting data generated from the SSU simulation (which strictly adheres to the M21 model's HOD assumptions), significant offsets persist. The model recovers emission-related components (SFRD, emissivity) within 5%, but exhibits a scale- and redshift-dependent offset exceeding 20% in the two-halo term of the CIB power spectrum.
Source of Discrepancy: The authors attribute the mismatch in the two-halo term to limitations in the halo model's cosmological ingredients, specifically the use of linear matter power spectra and scale-independent halo bias approximations.
Impact of Scatter: Incorporating scatter in the SFR–halo mass relation and SED templates significantly affects the shot noise component (altering it by $\sim 50\%$ ) but has a modest impact (less than 10%) on the clustered component of the power spectrum.
Shot Noise Recovery: The model struggles to recover the exact intrinsic shot noise values, with best-fit values falling outside the $1\sigma$ statistical uncertainty of the intrinsic values, likely due to degeneracy between shot noise and the one-halo term.

Significance and Conclusions
The paper concludes that the inability of current standard halo models to reliably recover physical parameters from CIB clustering is not solely due to the complexity of galaxy emission physics (the HOD ingredients). Instead, a significant portion of the discrepancy stems from the cosmological framework itself. The authors argue that recovering accurate physical parameters from CIB anisotropies requires improvements to the halo model's cosmological ingredients, specifically the adoption of scale-dependent halo bias and nonlinear matter power spectra, in addition to careful modeling of emission physics. This validation highlights the necessity of refining the theoretical framework before these models can be confidently used to forecast results for future line intensity mapping (LIM) experiments or to derive precise constraints on galaxy evolution from CIB data.