🔭 astrophysics

Detectable subhalo impacts in Milky Way streams

This paper presents a statistical framework to forecast detectable subhalo impacts on Milky Way stellar streams using LSST and Via data, identifying five promising streams with significant expected impact rates and demonstrating how such observations can constrain dark matter particle properties across different cosmological models.

Original authors: Junyang Lu, Elias Bernreuther, Tongyan Lin, Vincent S. H. Lee, Ana Bonaca, Ethan O. Nadler

Published 2026-08-21

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Junyang Lu, Elias Bernreuther, Tongyan Lin, Vincent S. H. Lee, Ana Bonaca, Ethan O. Nadler

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

The universe is not empty; it is filled with invisible scaffolding known as dark matter. While we cannot see this substance directly, its gravity holds galaxies together and shapes the cosmos. According to our best theories, this dark matter should exist in vast numbers of small, clumpy chunks called subhalos, ranging from the size of a large galaxy down to something as small as a star cluster. However, most of these tiny clumps are completely dark, containing no stars or gas to reveal their presence. They are ghostly islands floating in the dark, invisible to our most powerful telescopes. Finding them is crucial because their abundance and behavior could tell us whether our current understanding of dark matter is correct or if a different, stranger type of particle is at work.

One of the few ways to catch a glimpse of these invisible ghosts is to watch how they disturb the stars around them. Imagine a long, thin ribbon of stars stretching across the sky, formed when a small cluster of stars is torn apart by the gravity of the Milky Way. These ribbons, known as stellar streams, are incredibly sensitive to any passing object. If a dark subhalo were to fly close to such a stream, its gravity would tug on the stars, creating a ripple, a gap, or a wobble in the ribbon's smooth path. By studying these disturbances, astronomers hope to weigh the invisible subhalos and count how many exist, effectively mapping the dark matter that surrounds our galaxy.

A team of researchers has now built a sophisticated method to predict exactly how many of these dark encounters we should be able to see in the near future. They focused on a catalog of over one hundred known stellar streams in our galaxy, but they knew that not all of them were suitable for this delicate work. Some streams are too short, too faint, or too complex to model accurately. The team narrowed their list down to fourteen promising streams that could be reliably simulated using a computer model that tracks the movement of individual stars. For each of these streams, they calculated how long the stream has existed and how fast it moves, factors that determine how many dark subhalos it might have encountered over its lifetime.

The researchers then simulated the passage of thousands of dark subhalos past these streams to see which encounters would leave a detectable mark. A major challenge in this work is distinguishing a true impact from the natural, smooth curves of the stream itself. A distant subhalo might pull on the entire stream, making it look slightly different, but this effect can be confused with the normal uncertainties in how we model the stream's shape. To solve this, the team developed a statistical technique that fits a smooth curve to the stream's path and then looks for deviations from that curve. This allows them to ignore the broad, gentle shifts caused by distant objects and focus only on the sharp, localized ripples that signal a close encounter with a dark subhalo.

Their analysis reveals that with the data expected from upcoming major surveys, we are on the verge of detecting these dark encounters. The study identifies five specific streams that are the most likely places to find these impacts. The most promising target is a stream called Jet, which is expected to show about five detectable impacts in the coming years. Following Jet are the streams Orphan-Chenab, ATLAS-Aliqa Uma, GD-1, and Palomar 5, which are predicted to show between zero and one and a half impacts each. In contrast, the researchers found that with the data we have today, we would likely see almost no clear signs of these dark encounters, highlighting how much more powerful future telescopes will be for this specific task.

The team also explored how different theories about the nature of dark matter would change these numbers. If dark matter is "warm," meaning the particles move faster and form fewer small clumps, the number of detectable impacts would drop by a factor of four. Similarly, if dark matter is "fuzzy," a theory where the particles behave like waves, the number of impacts would also be suppressed by about four times. Conversely, if dark matter particles interact with each other in a specific way, the number of impacts could increase. However, the researchers caution that the biggest uncertainty in their predictions comes from our incomplete knowledge of how many dark subhalos actually exist in the Milky Way. Depending on how these clumps are distributed and how they are stripped of their mass by the galaxy's gravity, the actual number of impacts could be a few times higher or lower than their central estimate.

Ultimately, this work provides a clear roadmap for the next decade of dark matter hunting. It tells astronomers exactly where to look and what to expect when they turn their most powerful instruments toward the sky. By focusing on these specific streams and using the new statistical tools developed in this study, scientists will be able to test whether the dark matter surrounding us is made of the cold, clumpy particles predicted by standard theory, or something entirely different. The detection of even a single strong impact would be a monumental discovery, offering the first direct evidence of the invisible building blocks of our universe.

Technical Summary: Detectable Subhalo Impacts in Milky Way Streams

Problem Statement
Dark matter (DM) subhalos below the $10^8 M_\odot$ scale are poorly constrained by current observations, yet their abundance and internal structure offer critical tests for DM theories beyond the standard Cold Dark Matter (CDM) paradigm, including Warm DM (WDM), Fuzzy DM (FDM), and Self-Interacting DM (SIDM). While stellar streams are sensitive probes of low-mass subhalos ( $10^5 - 10^6 M_\odot$ ), previous studies have largely focused on a few well-known streams (e.g., GD-1, Palomar 5) or relied on simplified analytic models. A systematic assessment of the expected number of individually detectable strong impacts across the growing catalog of Milky Way streams, accounting for modeling uncertainties and observational capabilities of upcoming surveys (LSST and Via), has been lacking. Furthermore, defining a "detectable" impact is complicated by the fact that distant subhalo encounters produce large-angle perturbations that are degenerate with uncertainties in smooth-stream modeling (e.g., mass-loss history and the Galactic potential).

Methodology
The authors developed a statistical pipeline to forecast the number of detectable subhalo impacts on 14 promising Milky Way streams selected from a catalog of 131 candidates (Bonaca & Price-Whelan 2025).

Stream Selection and Modeling:
- Streams were selected based on length ( $>20^\circ$ ), known progenitor mass, and the ability to be well-modeled by a particle spray algorithm (Fardal et al. 2015).
- Stream ages ( $T_{\text{form}}$ ) were determined by fitting simulated stream lengths to observed data, varying progenitor mass and mass-loss rates ( $\gamma$ ).
- Selection criteria required the simulated stream width to match observed widths within a factor of 2, filtering out streams with complex morphologies or uncertain origins.
Proxy Model for Smooth Streams:
- To address the divergence of detectable impacts with impact parameter ( $b$ ) and uncertainties in smooth-stream modeling, the authors introduced a polynomial proxy model.
- Smooth stream observables (density $\rho$ , transverse angle $\phi_2$ , and radial velocity $v_r$ ) are fitted with a third-order polynomial.
- A detectable impact is defined as a statistically significant deviation from this best-fit polynomial, rather than from a specific smooth simulation. This approach absorbs large-scale variations caused by distant encounters or modeling uncertainties, ensuring the number of detectable impacts converges.
Impact Simulation Pipeline:
- Sampling: Subhalo encounters were sampled over the stream lifetime using an analytic approach (Erkal et al. 2016) with a CDM subhalo mass function and radial profile.
- Three-Layer Filtering: To manage computational cost while maintaining accuracy, a three-tier selection process was applied:
  1. Analytic Cut: A conservative estimate of signal size based on circular orbits.
  2. Orbit Integration (OI): A moderate-cost simulation accounting for elliptical orbits and the actual Milky Way potential.
  3. High-Statistics Simulation: Full particle spray simulations for impacts passing previous cuts to generate accurate signal profiles.
- Detectability Criterion: An impact is considered detectable if the signal-to-noise ratio squared ( $SNR^2$ ), calculated from the residuals of the polynomial fit across $\rho$ , $\phi_2$ , and $v_r$ , exceeds a threshold of 15.51 (corresponding to a 95% confidence level for 8 degrees of freedom).
Observational Scenarios:
- Forecasts were generated for two scenarios: present-day data (Gaia + DESI) and near-future data (LSST 10-year + Via spectroscopic survey).

Key Results

Expected Impact Rates: Under the near-future LSST+Via scenario, only 5 of the 14 studied streams have an expected number of detectable impacts ( $N_{\text{det}}$ $N_{det}$ ) greater than 0.2:
- Jet: $5.15^{+1.10}_{-0.95}$ (the most promising target).
- Orphan-Chenab: $1.40^{+0.62}_{-0.47}$ .
- ATLAS-Aliqa Uma: $1.25^{+0.60}_{-0.44}$ .
- GD-1: $0.55^{+0.43}_{-0.28}$ .
- Palomar 5: $0.40^{+0.39}_{-0.23}$ .
- With present-day Gaia+DESI data, only three streams show non-zero expected detections, with rates significantly lower (0.1–0.45).
Convergence via Polynomial Subtraction: Without the polynomial proxy subtraction, the number of detectable impacts grows with impact parameter $b$ and subhalo mass $M$ , failing to converge. The subtraction method successfully regulates this growth, concentrating detectable impacts within $b < 5r_s$ (where $r_s$ is the subhalo scale radius).
Observable Contributions: The total $SNR^2$ is shared across observables. On average, density ( $\rho$ ) contributes ~49%, transverse angle ( $\phi_2$ ) ~36%, and radial velocity ( $v_r$ ) ~16%. Including $\phi_2$ and $v_r$ is crucial for recovering detection rates that would otherwise be underestimated in density-only analyses.
Dependence on DM Models:
- WDM and FDM: At current mass bounds ( $m_{\text{WDM}} = 5.9$ keV, $m_{\text{FDM}} = 1.4 \times 10^{-20}$ eV), the number of impacts is suppressed by a factor of $\sim 4$ compared to CDM.
- SIDM: An extreme SIDM scenario with core-collapsed halos ( $M < 10^8 M_\odot$ ) results in an $O(1)$ enhancement in impact numbers due to higher central densities.
Systematic Uncertainties: The predictions rely heavily on the assumed subhalo population. Variations in the subhalo radial distribution (e.g., based on MW satellite measurements vs. FIRE-2 simulations) can alter the predicted impact rates by a factor of a few.

Significance and Claims
The paper claims to provide the first systematic forecast of individually detectable subhalo impacts across a broad sample of Milky Way streams, moving beyond case studies of single streams. By introducing a polynomial proxy model, the authors resolve the theoretical issue of divergent impact counts at large impact parameters, offering a robust, data-driven definition of detectability.

The authors emphasize that their results represent "best-case estimates" under idealized assumptions (static Milky Way potential, analytic CDM mass function). They explicitly state that distinguishing between DM models (CDM vs. WDM/FDM/SIDM) based solely on the number of impacts will be challenging due to Poisson fluctuations and systematic uncertainties in the subhalo population. Instead, the work identifies specific streams (notably Jet) as high-priority targets for future surveys to constrain DM properties and potentially reconstruct the properties of individual perturbers (mass, orbit, impact time). The study serves as a guide for allocating observational resources and highlights the necessity of improved stream modeling to fully leverage stellar streams for dark matter physics.

More like this