Clustering analysis of medium-band selected high-redshift galaxies

This paper presents a clustering analysis of high-redshift galaxies selected via medium-band photometry using IBIS and DESI data, revealing that the samples consist of overlapping Ly$\alpha$ emitters and Lyman break galaxies with correlation lengths of $3-4\,h^{-1}$Mpc and linear biases of$1.8-2.5$, thereby validating their utility for future large-scale structure cosmological surveys.

The Gist

Here is an explanation of the paper, translated from "astrophysicist" to "everyday human," using some creative analogies.

The Big Picture: Mapping the Cosmic Web

Imagine the universe as a giant, invisible spiderweb made of dark matter. Galaxies are the dew drops sitting on this web. For decades, astronomers have been trying to map this web to understand how the universe expands and how gravity works.

Usually, they map the web by looking at "dew drops" that are easy to see and are relatively close to us (like neighbors in a city). But to understand the history of the universe, they need to look at the "dew drops" from the distant past—when the universe was much younger. This is the "High Redshift" zone (roughly 2 to 3.5 billion years after the Big Bang).

The problem? The galaxies from that era are faint, hard to find, and look very different from the ones we see nearby. This paper is about a new, clever way to find them and measure how they cluster together.

The Challenge: Finding Needles in a Haystack

Think of the universe as a giant, dark ocean.

• Traditional Tracers: Usually, astronomers use "Luminous Red Galaxies" (LRGs) or "Quasars" (bright beacons). But at high redshifts (far away), these beacons are too dim or too rare. It's like trying to find a specific type of fish in the deep ocean, but your net only catches the big, colorful ones that live near the surface.
• The New Tracers: The authors are looking for two specific types of "fish":
  1. Lyman-alpha Emitters (LAEs): These are young, star-forming galaxies that glow brightly in a specific color of ultraviolet light (Lyman-alpha). They are like fireflies flashing in the dark.
  2. Lyman Break Galaxies (LBGs): These are galaxies where the light gets "cut off" at a specific wavelength because of hydrogen gas. They are like objects that suddenly disappear when you look through a specific colored filter.

The Tool: The "Medium-Band" Camera

To find these faint, distant galaxies, the team used a special camera technique called Medium-Band Imaging.

The Analogy:
Imagine you are trying to find a specific song in a radio station that plays a mix of genres.

• Broad-band filters are like listening to the whole station at once. You hear everything, but you can't isolate the specific song you want.
• Narrow-band filters are like tuning in to a single, tiny frequency. You hear the song perfectly, but you might miss the context or the volume is too low.
• Medium-band filters (what this paper uses) are like tuning into a specific "genre" channel. It's wide enough to catch the signal clearly but narrow enough to isolate the specific "color" of light coming from these distant galaxies.

The team used the IBIS survey (a camera on a telescope in Chile) to take pictures of a patch of sky using these medium-band filters. This allowed them to spot galaxies that were glowing with that specific "Lyman-alpha" light, effectively creating a list of "targets" to study.

The Investigation: The DESI Telescope

Once they had their list of targets from the camera, they needed to confirm exactly how far away they were. For this, they used DESI (Dark Energy Spectroscopic Instrument).

The Analogy:
Think of the camera as a security guard taking a photo of a crowd and spotting people wearing red hats.

• DESI is like a team of detectives who run up to those people with red hats and ask, "Who are you? How old are you?"
• However, there's a catch: The detectives only have 5,000 microphones (fibers) to talk to 5,000 people at once. If two people wearing red hats are standing too close together, the microphones can't reach both. This is called "Fiber Collision."

Because of this limitation, the team couldn't get a perfect 3D map of every single galaxy. Instead, they used a clever workaround: Angular Clustering.

The Workaround:
Instead of trying to measure the exact distance between every pair of galaxies (which the fiber collision messed up), they looked at how the galaxies were grouped on the sky (2D).

• Imagine looking at a crowd from a balcony. You can't measure the exact distance between two people in the crowd, but you can see if they are standing in tight little groups or scattered randomly.
• By measuring how "clumpy" the galaxies are on the sky, and knowing how deep the "slice" of the universe they are looking at is, they could mathematically reconstruct how the galaxies are clustered in 3D space.

The Findings: What Did They Learn?

1. It's a Mixed Bag: The galaxies they found aren't just one type. They are a mix of the "fireflies" (LAEs) and the "disappearing acts" (LBGs). About half of the galaxies in their sample are actually LBGs that happen to be glowing brightly in Lyman-alpha.
2. The Clumping Factor: They measured how strongly these galaxies clump together. They found that these high-redshift galaxies live in "halos" (clouds of dark matter) that are about 3 to 4 times the size of our local galaxy group.
3. The Bias: In astronomy, "bias" means how much more likely a galaxy is to be found in a crowded area compared to the average. These galaxies are "biased" tracers—they prefer to hang out in the busiest neighborhoods of the universe. Their bias is about 1.8 to 2.5, which is a sweet spot for future studies.

Why Does This Matter? (The Future)

This paper is essentially a "test drive" for the next generation of space surveys.

• The Simulation Problem: To understand these results, astronomers need to run supercomputer simulations. The authors found that their current simulations were barely powerful enough to model these galaxies. Future simulations will need to be much more detailed (like upgrading from a 2D map to a high-definition 3D model) to get it right.
• The Future Promise: If we can select these galaxies better (using the medium-band technique), we will be able to map the universe's expansion history with incredible precision. This will help us answer big questions: Is dark energy changing? How did the first galaxies form?

Summary

The authors took a new type of camera filter to find faint, ancient galaxies. They used a powerful telescope to confirm their identities, overcame technical limitations by looking at how they group together on the sky, and discovered that these galaxies are a mix of two types living in massive dark matter clouds. This work proves that the "medium-band" technique is a winning strategy for mapping the early universe in the future.

Technical Summary

Here is a detailed technical summary of the paper "Clustering analysis of medium-band selected high-redshift galaxies" by Ebina et al.

1. Problem Statement

Next-generation cosmological surveys (Stage IV and V) aim to probe the large-scale structure (LSS) of the universe at high redshifts ($2.3 < z < 3.5$) to constrain dark energy and fundamental physics. However, traditional tracers like Luminous Red Galaxies (LRGs) and Emission Line Galaxies (ELGs) become sparse at$z \gtrsim 2$.

• The Challenge: While Lyman-alpha emitters (LAEs) and Lyman break galaxies (LBGs) are promising tracers at these redshifts, they are difficult to select efficiently over large volumes. Narrow-band imaging is effective but slow for wide fields. Broadband "dropout" techniques are efficient but suffer from contamination and lack precise redshift resolution.
• The Gap: There is a need to characterize the clustering properties, host halo masses, and interloper fractions of galaxies selected via medium-band photometry, a technique that offers a balance between efficiency and redshift precision, but for which the clustering statistics and bias parameters at $z \sim 3$ are not yet well established.

2. Methodology

The authors utilize a hybrid dataset combining photometric imaging from the Intermediate Band Imaging Survey (IBIS) and spectroscopic follow-up from the Dark Energy Spectroscopic Instrument (DESI).

Data Sources

• Imaging (IBIS): Uses the Dark Energy Camera (DECam) on the Blanco telescope. The survey employs 5 adjacent medium bands (M411–M517) spanning $4000\text{--}5300$Å. The study focuses on the XMM-LSS field ($6.11 \text{ deg}^2$).
• Spectroscopy (DESI): Ancillary program observations provided redshifts for a subset of the IBIS targets. This data is crucial for determining the redshift distribution ($dN/dz$) and the interloper fraction ($f_{int}$).
• Target Selection: Three novel selection criteria were applied using the medium bands:
  1. Emission Line Peaks: Using colors between three adjacent bands to identify Ly$\alpha$ emission peaks.
  2. Synthetic Broadband (LAE style): Constructing a synthetic $g$-band from medium bands to apply traditional LAE color cuts.
  3. Synthetic Broadband (LBG style): Using a single medium band against an average of the others to detect the Lyman break.
     The sample is split into "Total" (fainter limits, $m_{fib} \approx 25.3\text{--}25.5$) and "Wide" (brighter limits, $m_{fib} \approx 25.0$) subsets, covering redshifts $2.26 < z < 3.41$.

Analysis Techniques

• Angular Clustering: Due to fiber assignment complications in the DESI ancillary program (which prevent a clean 3D analysis), the authors perform a pseudo-3D clustering analysis. They compute the angular auto-correlation function, $w(\theta)$, within narrow redshift slices ($\Delta z \sim 0.2$) and project these into a transverse correlation function, $w_p(R)$.
• HOD Modeling: The clustering measurements are interpreted using Halo Occupation Distribution (HOD) models. The authors use the "standard" 5-parameter HOD model (AbacusSummit suite) to assign galaxies to dark matter halos. They scan parameter space to find best-fit models for correlation length ($r_0$) and bias.
• Perturbation Theory (PT): Mock catalogs generated from N-body simulations (AbacusSummit) are analyzed using one-loop Eulerian Perturbation Theory (via velocileptors) to measure scale-dependent bias, stochastic terms, and counterterms.
• Extended Models: The authors also test the High Mass Quenched (HMQ) HOD model, which suppresses central galaxies in massive halos, to see if it better fits the data given the active star-forming nature of the tracers.

3. Key Contributions

1. First Clustering Measurement: This is the first measurement of clustering statistics for high-redshift galaxies selected specifically via medium-band photometry combined with DESI spectroscopy.
2. Sample Characterization: The study quantifies the composition of the sample, revealing it is a mixture of LAEs and LBGs. Cross-matching with traditional LBG catalogs (H09) shows that $\sim 46\%$ of the high-$z$ ($2.9 < z < 3.41$) sample are LBGs, confirming that medium-band selections capture a continuum of Ly$\alpha$ equivalent widths.
3. Interloper Mitigation: The authors demonstrate that refining the sample with tighter broadband color cuts significantly reduces the interloper fraction ($f_{int}$) from $\lesssim 0.4$ to $\lesssim 0.15$, which is critical for accurate angular clustering measurements.
4. Simulation Requirements: The paper establishes specific resolution requirements for future N-body simulations needed to model these tracers, noting that standard resolutions may be insufficient for the low-mass halos hosting these galaxies.

4. Key Results

• Clustering Amplitude: The study detects clustering in the $z \sim 3$ sample and a marginal detection in the $z \sim 2.5$ sample.
  • Correlation Length: $r_0 \simeq 3\text{--}4 \, h^{-1}\text{Mpc}$.
  • Linear Bias: $b \sim 1.8\text{--}2.5$.
  • These values are consistent with previous LAE studies and faint LBG samples, despite the different selection techniques.
• HOD Parameters:
  • Best-fit models indicate the galaxies reside in halos with masses $\sim 10^{10}\text{--}10^{12} \, h^{-1}M_\odot$.
  • The "Wide" sample (brighter) shows slightly stronger clustering than the "Total" sample, consistent with higher luminosity tracers residing in more massive halos.
  • The satellite fraction ($f_{sat}$) is low ($\sim 3\text{--}12\%$).
• Scale-Dependent Bias:
  • At $z=3$, the scale-dependent bias is more pronounced than at $z=2.5$, driven by the decorrelation between the galaxy and matter fields.
  • The linear bias $b_1$ at $z=3$ falls in the range $1.8\text{--}2.6$, which is lower than typical broadband-selected LBGs (which often have$b_1 > 3$) but higher than some LAE simulations.
• HMQ vs. Standard HOD: The HMQ model provides a slightly better fit ($\chi^2$) than the standard model, suggesting that central galaxies in very massive halos may be suppressed (consistent with radiative transfer effects or quenching), but the difference is within current observational uncertainties.
• Fiber Assignment Impact: The study highlights that the specific dithering fiber assignment scheme used for this ancillary program introduces complex, density-dependent selection effects that invalidate standard 3D clustering corrections, necessitating the angular clustering approach.

5. Significance and Future Outlook

• Cosmological Potential: The medium-band selected samples are excellent tracers for future high-$z$ cosmology. They exhibit lower bias and smaller "Fingers of God" effects compared to traditional dropout LBGs, making them less susceptible to non-linear modeling uncertainties.
• Forecast: The authors forecast that a full IBIS survey covering $3000 \text{ deg}^2$, combined with CMB lensing cross-correlations, could constrain the growth of structure parameter$\sigma_8$at$z \sim 3$ to a precision of 5.1%. This would be the tightest constraint on clustering at such high redshifts.
• Simulation Needs: The results imply that future simulations for high-$z$ surveys must resolve halos down to $\sim 10^{10} \, h^{-1}M_\odot$ (particle masses $< 10^9 \, h^{-1}M_\odot$) in boxes larger than $2 , h^{-1}\text{Gpc}$ to accurately model the host halos of these tracers.
• Strategic Value: This work validates medium-band photometry as a viable, efficient strategy for selecting high-redshift tracers for Stage V spectroscopic surveys (like DESI-II), bridging the gap between narrow-band depth and broadband efficiency.