🔭 astrophysics

Halo mass of ULIRGs at z~2

This paper presents a clustering analysis of approximately 3,000 ultraluminous infrared galaxies at redshift $z\sim 2$ , revealing that they reside in dark matter halos with characteristic masses around $10^{12.5-12.7}\ h^{-1}\rm M_{\odot}$ that are expected to evolve into massive groups or clusters by $z=0$ , with a notable difference in satellite galaxy contributions between the COSMOS and BOOTES fields.

Original authors: Jun Pan, Ming Li, Cheng Cheng, Jasheng Huang

Published 2026-08-18

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Jun Pan, Ming Li, Cheng Cheng, Jasheng Huang

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

In the vast history of the universe, there was a time when galaxies were growing faster and more furiously than they ever have since. Astronomers call this era "cosmic noon," a period roughly 10 billion years ago when the universe was about a third of its current age. During this time, the rate at which stars were being born reached its highest peak, and the galaxies that would eventually become the massive, elliptical giants we see today were being assembled. To understand how these cosmic structures formed, scientists look at the invisible scaffolding that holds them together: dark matter. This mysterious substance does not emit light, but its gravity creates massive bubbles, or halos, that trap ordinary gas and dust, allowing stars to ignite. By studying how galaxies cluster together in space, researchers can weigh these invisible halos and learn how the most energetic galaxies of the early universe were built.

A team of astronomers has now turned its attention to a specific group of these ancient powerhouses: ultraluminous infrared galaxies. These are not ordinary galaxies; they are so bright in infrared light that they shine with the energy of trillions of suns, fueled by intense bursts of star formation and sometimes by hungry black holes at their centers. Because they are shrouded in thick dust, they are difficult to spot in visible light, but they glow brightly in the infrared. The researchers wanted to know exactly how massive the dark matter halos hosting these galaxies were at cosmic noon. To find out, they gathered a massive sample of nearly 3,000 of these galaxies from two different patches of sky, known as the COSMOS and BOOTES fields. By measuring how closely these galaxies were grouped together, the team could calculate the weight of the dark matter bubbles they live in.

The study focused on galaxies located at a distance corresponding to a time when the universe was about a third of its current age. The astronomers used data from the Spitzer Space Telescope, which captured light at specific wavelengths to identify galaxies that were actively forming stars while filtering out those dominated by active black holes. They selected galaxies based on their brightness at a wavelength of 24 micrometers and their colors in four different infrared bands. This careful selection process resulted in a catalog of 722 galaxies in the COSMOS field and 2,268 galaxies in the larger BOOTES field. With these numbers, the team could map out the positions of these galaxies and measure how often they appeared near one another compared to a random distribution.

Using a statistical method that compares the observed clustering to theoretical models, the researchers determined the mass of the dark matter halos surrounding these galaxies. They found that these ultraluminous galaxies live in halos with a mass of roughly 5 trillion times the mass of our sun. This result is significant because it is lower than some previous estimates suggested, indicating that these extreme star-forming galaxies do not necessarily require the most massive dark matter bubbles to exist. The team also looked at whether these galaxies were alone in their halos or if they had smaller companion galaxies orbiting them. In the COSMOS field, the data suggested that these galaxies were almost entirely alone, acting as the single dominant member of their dark matter halo. However, in the BOOTES field, the clustering pattern indicated that about 15 percent of these galaxies might be satellite companions orbiting within the same massive halo as a central galaxy.

The researchers noted that this difference between the two fields might be due to the natural variation in the universe, where different regions have slightly different densities of matter. Despite this variation, the overall picture is clear: the dark matter halos hosting these intense star-forming galaxies at cosmic noon are substantial, but not the most massive structures in the universe. By tracing the evolution of these halos forward in time, the team calculated that they would eventually grow into halos with masses between 10 and 100 trillion solar masses by the time the universe reaches its current age. This suggests that the ultraluminous galaxies of the past are the ancestors of the most massive elliptical galaxies we see today. The study provides a clearer view of how the most energetic galaxies of the early universe were connected to the invisible cosmic web, offering a precise measurement of the environments in which they thrived.

Technical Summary: Halo Mass of ULIRGs at Cosmic Noon

Problem and Motivation
The epoch at redshift $z \sim 2$ , known as "cosmic noon," represents a critical phase in galaxy formation where the cosmic star formation rate density peaked. Understanding the host dark matter halos of strongly star-forming galaxies during this period is essential for tracing the assembly of massive galaxies. However, spectroscopic surveys in the optical redshift desert ( $1.4 < z < 2.5$ ) remain observationally challenging, often yielding samples of only a few hundred galaxies. Consequently, clustering analyses in this regime frequently rely on photometric samples, which can suffer from contamination by Active Galactic Nuclei (AGNs) and uncertainties in redshift distributions. While previous studies have investigated the clustering of submillimeter galaxies (SMGs) and ultraluminous infrared galaxies (ULIRGs), there is a need for robust constraints on the host halo masses of ULIRGs selected via infrared flux and colors to mitigate AGN confusion and improve statistical significance.

Methodology
The authors present a clustering analysis of approximately 3,000 ULIRGs at $z \sim 2$ uniformly selected from the COSMOS and BOOTES fields.

Sample Selection: The sample consists of 722 objects in COSMOS and 2,268 in BOOTES, selected using MIPS 24 $\mu$ m flux ( $F_{24\mu m} > 350\mu$ Jy) and specific IRAC color criteria ( $[3.6]-[4.5] > 0$ and $[5.8]-[8.0] < 0$ ). These criteria target the redshifted 7.7 $\mu$ m PAH emission and the 1.6 $\mu$ m stellar bump while screening out power-law AGNs.
Angular Correlation Functions (ACFs): The authors measured the angular correlation functions for both fields using the Landy–Szalay estimator. They employed a jackknife resampling technique ( $N_{jk}=100$ for COSMOS, $N_{jk}=81$ for BOOTES) to estimate the covariance matrix and applied iterative corrections for the integral constraint.
Modeling: Two theoretical frameworks were applied to the measured ACFs:
1. Linear Bias Model: Assuming linear theory and a scale-independent bias, the authors fitted the ACFs to derive the linear bias parameter ( $b_g$ ) and the corresponding characteristic halo mass ( $M_h$ ).
2. Halo Occupation Distribution (HOD): Using the Zheng et al. (2005, 2007) model, the authors fitted the ACFs to constrain the occupation of central and satellite galaxies within dark matter halos. This model includes five parameters ( $\log M_{min}, \log M_1, \log M_0, \sigma_{\log M}, \alpha$ ) and incorporates a number density constraint to break degeneracies between bias and abundance.

Key Results

Halo Masses: Both modeling approaches yield consistent characteristic halo masses.
- Linear Model: Yields $\log M_h/(h^{-1}M_\odot) = 12.64 \pm 0.50$ (COSMOS) and $12.72 \pm 0.13$ (BOOTES).
- HOD Model: Yields occupation-weighted effective halo masses of $\log M_{eff}/(h^{-1}M_\odot) = 12.50^{+0.26}_{-0.28}$ (COSMOS) and $12.89^{+0.11}_{-0.12}$ (BOOTES).
Satellite Fraction: A notable difference exists between the two fields regarding satellite occupation. The HOD fit for the BOOTES field suggests a non-negligible satellite fraction ( $f_{sat} \sim 0.15$ ), whereas the COSMOS field is consistent with a nearly zero satellite fraction ( $f_{sat} \sim 0$ ), despite COSMOS having a higher surface density of galaxies.
Evolution: Using the halo mass evolution model of Behroozi et al. (2013), the authors project that these $z \sim 2$ ULIRGs reside in halos that will evolve into systems with masses of $\sim 10^{13.34} - 10^{13.86} h^{-1}M_\odot$ by $z=0$ .
Comparison with Previous Work: The derived halo masses are approximately one order of magnitude lower than those reported by Farrah et al. (2006) but are in broad agreement with results for SMGs at similar redshifts and consistent with Starikova et al. (2012).

Significance and Claims
The paper claims to present the largest ULIRG sample at $z \sim 2$ used for clustering analysis to date. By directly fitting ACFs with linear bias and HOD models rather than relying solely on correlation lengths, the study provides robust constraints on the typical host halo mass of ULIRGs at cosmic noon. The authors emphasize that their results suggest these galaxies reside in halos of $\sim 10^{12.7} h^{-1}M_\odot$ , which is lower than some previous estimates. The study also highlights the potential role of cosmic variance, noting the field-to-field variation in satellite fractions between COSMOS and BOOTES. The authors conclude that while their current data supports the use of the standard Zheng HOD model, future larger-area surveys (e.g., with Euclid or Roman) will be necessary to more precisely constrain HOD parameters and determine if the observed environmental dependence of ULIRG multiplicity is a universal feature of galaxy evolution.

More like this