🔭 astrophysics

A tight relation between the distribution of globular clusters and dark matter in AS1063

Using deep JWST observations of the galaxy cluster AS1063, the study reveals a tight correlation between the distribution of globular clusters and dark matter, demonstrating that smoothed globular cluster counts can serve as an effective proxy for mapping lensing mass in clusters where direct gravitational lensing constraints are limited.

Original authors: J. M. Diego, C. Goolsby, C. J. Conselice, J. M. Palencia

Published 2026-05-27

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: J. M. Diego, C. Goolsby, C. J. Conselice, J. M. Palencia

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 a giant, invisible cosmic web made of "dark matter" holding a massive galaxy cluster together. We can't see this dark matter directly, but we know it's there because its gravity bends light like a giant lens. Usually, to map this invisible web, astronomers need to find specific background galaxies that are distorted by the lens. But what if there aren't enough distorted galaxies to make a clear picture?

This paper presents a clever new way to map that invisible dark matter using something we can see: Globular Clusters.

Think of a galaxy cluster like a busy city.

The Dark Matter is the invisible foundation and the heavy steel beams holding the skyscrapers up. You can't see them, but they determine the shape of the city.
The Globular Clusters are like thousands of tiny, incredibly tough, self-contained "houses" (dense balls of stars) that orbit the city. Because they are so compact and sturdy, they survive even when the bigger "buildings" (satellite galaxies) around them get torn apart by the city's strong gravitational winds.

The Discovery

The authors used the James Webb Space Telescope (JWST) to take an incredibly deep, high-resolution photo of a galaxy cluster called AS1063. It's like using a super-powerful microscope to look at a city from space.

They found tens of thousands of these tiny "houses" (globular clusters) orbiting in the center of the cluster. They also found thousands more in the outer regions.

The "Magic" Connection

The big question was: Do these tiny "houses" follow the same map as the invisible "steel beams" (dark matter)?

The answer is a resounding yes.
The authors found a "tight relation" between where the globular clusters are and where the dark matter is. It's as if the tiny houses are perfectly arranged to trace the outline of the invisible foundation.

However, there was a small twist: The globular clusters were packed a bit more tightly in the very center than the dark matter was. It's like the houses are huddled closer to the city center than the steel beams are.

The New Tool: The "Smoothing Kernel"

Here is the most practical part of the paper. The authors realized that because these globular clusters are so numerous and follow the dark matter so closely, you can use them as a proxy (a stand-in) to map the dark matter.

They developed a simple mathematical "recipe" (called a smoothing kernel). Imagine you have a picture made of thousands of individual dots (the globular clusters). If you run this picture through a special "blur" filter, the dots blend together to create a smooth, continuous map.

The result? This smooth map of dots looks almost identical to the map of the dark matter derived from complex gravitational lensing models.

Why This Matters

Usually, to map dark matter, you need strong gravitational lensing (where background galaxies are stretched into giant arcs). But this only happens in specific, high-redshift clusters.

Low-redshift clusters (closer to us) are great for seeing globular clusters clearly, but they are too "weak" to create those giant lensing arcs.
The Problem: We couldn't easily map the dark matter in these closer clusters.
The Solution: Now, we can just count the globular clusters, apply the "blur filter," and get a reliable map of the dark matter distribution, even without the lensing arcs.

Summary of Findings

Count: They identified over 28,000 globular clusters in the center and thousands more in the outskirts of AS1063.
Correlation: The distribution of these clusters matches the distribution of dark matter very closely.
Method: By smoothing the discrete points of the clusters, they created a continuous map that matches the mass map derived from lensing.
Application: This method allows astronomers to map dark matter in clusters where traditional lensing methods fail (like nearby, low-redshift clusters), provided they can detect the globular clusters.

In short, the paper shows that if you want to see the invisible skeleton of a galaxy cluster, just look at where the tiny, tough "houses" (globular clusters) are hanging out—they are tracing the invisible dark matter for you.

Technical Summary: A Tight Relation between the Distribution of Globular Clusters and Dark Matter in AS1063

Problem and Motivation
Gravitational lensing is the primary method for mapping the distribution of dark matter (DM) in galaxy clusters, particularly in the inner regions where strong lensing occurs. However, this technique is limited to high-redshift clusters ( $z \gtrsim 0.2$ ) where surface mass densities exceed critical values. Low-redshift clusters, where globular clusters (GCs) are easier to resolve, typically remain subcritical and lack strong lensing constraints. While simulations suggest a tight correlation between the distribution of GCs and DM, observational verification has been hindered by the sensitivity limits of previous telescopes (e.g., HST) in the infrared, where evolved GCs are brightest. This paper addresses the feasibility of using resolved GCs as tracers for the underlying DM distribution, specifically testing whether the discrete distribution of GCs can serve as a proxy for the continuous lensing mass in the massive cluster AS1063 ( $z=0.348$ ).

Methodology
The authors utilized deep, high-resolution images from the James Webb Space Telescope (JWST) obtained as part of the GLIMPSE program (Program ID 3293). The dataset includes observations from the short-wavelength filters (F090W, F115W, F150W, F200W) in the main module and a second module extending the field of view.

GC Identification: The team applied a two-step high-pass filtering process (using an "À Trous" wavelet approach, specifically $G_{0.5} - G_5$ followed by $G_1 - G_3$ ) to enhance point sources while suppressing diffuse emission and large-scale structures.
Masking: To prevent spurious detections, the authors meticulously masked bright sources, diffraction spikes, extended galaxies, and known lensed arcs. They also masked the central 1" region of the Brightest Cluster Galaxy (BCG) during automated detection, performing visual identification only in this specific area.
Catalog Construction: Point sources were detected in the combined filtered image. The final catalog includes 28,026 detected GCs in the main module. To account for masked regions, the authors employed a Monte Carlo simulation to "inpaint" missing GCs based on the local density, adding approximately 3,500 estimated sources.
Mass Tracing: The discrete distribution of GCs was converted into a continuous 2D surface density map ( $\rho_{gc}$ ) using a convolution kernel $W \propto (SL + d)^{-\alpha}$ . The authors determined that a power-law index of $\alpha = 1.6$ and a softening length of $SL = 0.9"$ provided the best fit to the lensing convergence ( $\kappa$ ) derived from a separate lens model (Paper-I).
Contamination Assessment: Data from the second JWST module (extending to 1.18 Mpc) was analyzed to estimate the level of contamination from background compact galaxies. The surface density at the largest radii was used as an upper limit for this contamination.

Key Results

Correlation: A very tight correlation exists between the surface number density of GCs ( $n_{GC}$ ) and the lensing convergence ( $\kappa$ ). The authors derive a scaling law $\langle \kappa \rangle = 0.4 n_{GC}^{0.7}$ for regions where $n_{GC} > 4 \text{ arcsec}^{-2}$ .
Spatial Distribution: The distribution of GCs closely mirrors the lensing mass distribution. When smoothed with the derived kernel, the GC density map reproduces the contours of the lensing convergence map with remarkable agreement, particularly in regions constrained by strong lensing.
Radial Profiles: The radial profile of GCs remains constant between 3" and 13", then declines. At radii $r > 13"$ , the GC number density falls approximately as $r^{-2}$ , consistent with N-body simulations. This decline is slightly steeper than the lensing mass profile in the outer regions, a discrepancy attributed to the lensing model's bias beyond the constrained radius and the shallower slope of the smoothing kernel used for GCs.
Core Structure: The GC distribution exhibits a large core, resembling the DM distribution more closely than the intracluster light (ICL), which shows a mismatch with the projected mass in the lens model.
Mass Estimate: Within a 60" radius, the total mass in GCs is estimated at $8.47 \times 10^9 M_\odot$ , implying an average GC mass of $\sim 3 \times 10^5 M_\odot$ , consistent with Milky Way GCs and simulations.
Contamination: The estimated contamination from background compact galaxies is approximately 0.015 GC per kpc $^2$ (roughly 15% of the inner density), which was subtracted to refine the radial profiles.

Significance and Claims
The paper demonstrates that GCs behave as collisionless particles similar to DM and can serve as effective tracers for the dark matter distribution in galaxy clusters. The primary contribution is the demonstration that a simple smoothing kernel can transform the discrete GC distribution into a continuous 2D map that matches the lensing convergence.

The authors claim this method offers a viable alternative for inferring mass distributions in clusters where gravitational lensing constraints are scarce or absent, such as low-redshift clusters that do not produce strong lensing. While the study focuses on AS1063, the authors suggest this approach could be applied to other massive clusters where GCs are detectable but lensing data is limited. The work validates the theoretical expectation of a tight GC-DM relationship using actual JWST data, providing a new tool for mass mapping in the era of deep infrared surveys.

The Discovery

The "Magic" Connection

The New Tool: The "Smoothing Kernel"

Why This Matters

Summary of Findings

More like this