🔭 astrophysics

Measurement of Substructure from the Kinematics of the GD-1 Stellar Stream

By measuring the intrinsic radial velocity dispersion of the GD-1 stellar stream using 160 member stars, this study finds that the observed kinematic heating is best explained by impacts from numerous low-mass or a single compact dark matter subhalo, suggesting a potential deviation from standard Cold Dark Matter predictions regarding subhalo compactness and abundance.

Original authors: Jacob Nibauer, Ana Bonaca, Adrian M. Price-Whelan, David N. Spergel, Jenny E. Greene

Published 2026-06-05

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Jacob Nibauer, Ana Bonaca, Adrian M. Price-Whelan, David N. Spergel, Jenny E. Greene

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 Milky Way galaxy as a giant, invisible ocean of dark matter. According to our best theories, this ocean isn't smooth; it's filled with thousands of tiny, invisible whirlpools called subhalos. These whirlpools are clumps of dark matter that are too small to hold stars, so we can't see them directly.

To find these invisible whirlpools, astronomers look at stellar streams. Think of a stellar stream like a long, thin ribbon of stars left behind by a globular cluster (a ball of stars) that got torn apart by the galaxy's gravity. If the ocean were perfectly smooth, this ribbon would flow in a straight, calm line. But if the ribbon bumps into an invisible whirlpool, it gets a "kick," creating ripples, gaps, or speeding up in certain spots.

This paper focuses on a specific ribbon called GD-1. The authors, led by Jacob Nibauer, wanted to see if this ribbon was being kicked around by these invisible dark matter whirlpools.

The Investigation: Measuring the "Jitter"

The team gathered data on 160 stars in the GD-1 stream from four different telescopes. They measured how fast these stars were moving toward or away from us.

The Analogy: Imagine a group of runners on a track. If they are all running at exactly the same speed, the group is "cold" and orderly. If they are jostling each other, some running faster and some slower, the group has "velocity dispersion" or "jitter."
The Finding: The GD-1 stream is much "jittery" than expected. In the middle section of the stream, the stars are moving with a speed variation of about 4.8 km/s.
The Surprise: If the Milky Way were smooth and empty of these dark matter whirlpools, the stream should be very calm, with a jitter of only about 1 km/s. The fact that GD-1 is so "hot" (jittery) suggests it has been hit by something.

The Culprit: Dark Matter Subhalos

The authors used computer models to figure out what could cause this jitter. They tested two main scenarios:

The "Many Small Hits" Scenario: The stream was hit by many small, invisible dark matter clumps.
The "One Big Hit" Scenario: The stream was hit once by a single, very dense, massive dark matter clump.

The Verdict: Both scenarios could explain the data, but with a twist. The dark matter clumps involved had to be much more compact (denser and tighter) than standard theories predicted.

The Analogy: Standard theory says dark matter clumps are like fluffy cotton balls. The data suggests they are more like hard, dense marbles. A fluffy cotton ball wouldn't kick the ribbon hard enough to cause the observed jitter; a dense marble would.

What This Means for Dark Matter

The study puts a limit on how much of the Milky Way is made of these subhalos. They found that subhalos make up about 5% of the galaxy's total mass (with a possible range of 2% to 13%). This fits with the standard "Cold Dark Matter" theory.

However, the shape of these clumps is the real story. The data suggests that at low masses, dark matter clumps are denser than the standard "Cold Dark Matter" model predicts. This hints that our understanding of the dark matter particle itself might need a tweak. It could be that dark matter interacts with itself in ways we didn't expect, making the clumps collapse into tighter, denser shapes.

Summary

The Stream: GD-1 is a ribbon of stars that is moving too erratically to be explained by a smooth galaxy.
The Cause: It has been bumped by invisible dark matter clumps.
The Twist: These clumps aren't the "fluffy" ones we expected; they are "dense marbles."
The Conclusion: While the amount of dark matter fits current theories, the structure of the smallest clumps suggests the standard theory might be slightly off, pointing toward new physics about what dark matter actually is.

The authors note that this is just the beginning. With more data from future telescopes, they hope to map these invisible clumps across the entire sky to finally solve the mystery of dark matter.

Technical Summary: Measurement of Substructure from the Kinematics of the GD-1 Stellar Stream

Problem Statement
In the Cold Dark Matter (CDM) paradigm, galaxies are predicted to host a vast population of dark matter subhalos, ranging from galactic scales down to Earth-mass scales. However, subhalos below the galaxy formation threshold ( $\approx 10^8 M_\odot$ ) are expected to be devoid of baryonic matter, rendering them invisible to direct observation. Consequently, testing the CDM paradigm on small scales requires indirect gravitational tracers. Stellar streams, which are kinematically cold structures formed by the tidal disruption of globular clusters, are sensitive to perturbations from low-mass subhalos. While previous studies have utilized the morphological gaps and spurs in streams like GD-1 to infer subhalo interactions, this work addresses the need to constrain the intrinsic velocity dispersion of the stream. A measured velocity dispersion significantly higher than that predicted by smooth Milky Way potentials would indicate heating driven by dark matter substructure.

Methodology
The authors analyze the GD-1 stellar stream using a combination of observational data and perturbation theory modeling:

Data Collection and Velocity Dispersion Measurement:
- The study utilizes 160 unique member stars identified from the data-driven GD-1 catalog (Starkman et al. 2025) with membership probability $>75\%$ .
- Radial velocities ( $v_r$ ) are compiled from four spectroscopic surveys: DESI (early data release), SDSS DR9, LAMOST DR8, and MMT Hectochelle.
- To constrain binary star contamination, the authors use repeat observations of 34 stars across different instruments.
- A hierarchical Bayesian model is employed to separate observational errors, binary-induced velocity jitter ( $\sigma_{binary}$ ), and the intrinsic stream velocity dispersion ( $\sigma_{\Delta v_r}$ ). The stream is binned into three segments along its length ( $\phi_1 \in [-60, -40], [-40, -20], [-20, 0]$ deg) to measure local dispersion.
Dynamical Modeling:
- The authors employ the streamsculptor framework, which uses perturbation theory to model the cumulative effect of hundreds of subhalo impacts on the stream.
- Unperturbed stream models are generated using a particle-spray approach for progenitor masses of $2 \times 10^4 M_\odot$ and $10^5 M_\odot$ and dynamical ages of 3.5, 5, and 8 Gyr, within a static Milky Way potential (MilkyWayPotential2022).
- Subhalos are modeled as Hernquist profiles. The study explores two scenarios to break degeneracies between subhalo number density and internal structure:
  - Model I: Assumes a Warm Dark Matter (WDM) scenario with a suppression in the subhalo mass function (SHMF) below a half-mode mass ( $M_{hm}$ ) and a constant mass-size relation ( $r_s/r_{s,cdm}$ ).
  - Model II: Assumes a standard CDM SHMF ( $M_{hm}=0$ ) but allows the mass-size relation to vary as a broken power-law with mass, testing for mass-dependent concentration.
Inference:
- The authors use Approximate Bayesian Computation (ABC) to compare simulated stream realizations against the observed velocity dispersion.
- Simulations are generated for various subhalo population parameters (normalization, $M_{hm}$ , concentration). Samples are accepted if the Euclidean distance between the simulated and observed velocity dispersion (and stream width constraints) falls within a tolerance $\epsilon$ .

Key Contributions and Results

Intrinsic Velocity Dispersion: The study measures the intrinsic radial velocity dispersion of GD-1 to be approximately $2\text{--}5$ km s $^{-1}$ across its length. The region with the highest dispersion ( $\phi_1 \in [-40, -20]$ deg) shows a value of $4.8^{+1.2}_{-0.7}$ km s $^{-1}$ . This represents a $4\sigma$ deviation from unperturbed stream models formed in a smooth potential, which predict dispersions of only $\approx 0.7\text{--}1.1$ km s $^{-1}$ .
Subhalo Population Constraints: The observed dispersion can be explained by a population of dark matter subhalos. The inferred fraction of the Milky Way's mass in subhalos is $f_{sub} = 0.05^{+0.08}_{-0.03}$ (68% confidence), consistent with CDM expectations ( $\approx 7.5\%$ ).
Compactness of Subhalos: A primary finding is the preference for subhalos that are more compact than standard CDM predictions.
- In Model I, if the subhalo mass fraction is $10\%$ , the scale radius is constrained to $r_s/r_{s,cdm} = 0.3^{+0.3}_{-0.2}$ (68% confidence).
- In Model II, there is a preference for increasingly compact subhalos at low masses ( $M < 10^7 M_\odot$ ). For subhalos with $M_{200} = 10^6 M_\odot$ , the concentration $c_{-2}$ is constrained to be $>190$ at 68% confidence, significantly higher than the CDM prediction of $\approx 43$ .
Alternative Scenarios: The data can be explained either by numerous impacts with low-mass subhalos or by a single impact with a very compact subhalo ( $M \gtrsim 10^8 M_\odot$ ). The authors note that a single impact with a compact subhalo is sufficient to generate the observed dispersion, consistent with previous interpretations of the GD-1 "spur" feature.

Significance and Claims
The paper claims to provide a new, independent constraint on dark matter substructure using the kinematics of stellar streams, complementing previous constraints derived from stream density morphology. The results suggest a possible deviation from the standard CDM mass-size relation at low subhalo masses, where subhalos are expected to be purely dark matter dominated. Specifically, the preference for higher concentrations (or smaller scale radii) than CDM predictions aligns with alternative dark matter theories such as Self-Interacting Dark Matter (SIDM) or Atomic Dark Matter, which predict enhanced central densities.

The authors emphasize that their constraints are empirical and agnostic to specific particle physics models, serving as a benchmark to test various dark matter theories. They acknowledge that degeneracies exist between the number of subhalo impacts and their concentration (the "rate-concentration degeneracy"), which could be broken in future work by jointly modeling stream density and kinematics. The study concludes that while the subhalo mass fraction is consistent with CDM, the internal structure of these subhalos may require physics beyond the standard CDM paradigm, particularly at low masses.

The Investigation: Measuring the "Jitter"

The Culprit: Dark Matter Subhalos

What This Means for Dark Matter

Summary

More like this