Broken Expectations: The Effects of Modelling Assumptions on the Inferred Dark Matter Distribution in the Milky Way's Satellites

This study uses tailored simulations of Milky Way satellites to demonstrate that while Galactic tides have minimal impact on dark matter density inference, the standard Jeans analysis method tends to underestimate inner densities and J-factors due to model limitations, and the consistency of results with cosmological predictions suggests either a light Milky Way or restricted tidal effects.

The Gist

Here is an explanation of the paper "Broken Expectations" using simple language and creative analogies.

The Big Picture: The Cosmic Detective Game

Imagine the Milky Way (our home galaxy) is a giant, invisible city. Floating around it are tiny, dim "satellite towns" called dwarf galaxies. These towns are made mostly of stars, but they are held together by a massive, invisible scaffolding called Dark Matter.

Scientists want to know what this invisible scaffolding looks like. Is it a dense knot in the center (a "cusp") or a fluffy, spread-out cloud (a "core")? This matters because if we know the shape of the dark matter, we might figure out what dark matter is (a new particle? a different kind of physics?).

To solve this, astronomers act like detectives. They look at how the stars in these dwarf towns are moving (their "kinematics") and use a mathematical tool called the Jeans Equation to work backward and guess the shape of the invisible scaffolding holding them up.

The Problem: The "Perfect World" Assumption

The problem is that the mathematical tool (the Jeans Equation) was built for a "perfect world." It assumes:

1. The galaxy is a perfect sphere (like a beach ball).
2. The galaxy is calm and settled (in "dynamical equilibrium").

But our universe isn't perfect. Our Milky Way is a giant bully. It has a massive gravitational pull that stretches and squeezes these dwarf galaxies as they orbit around it. This is called tidal force. It's like a giant hand stretching a piece of taffy.

The Big Question: If we use a tool designed for perfect spheres on galaxies that are being stretched and squashed by the Milky Way, will our detective work be wrong? Will we get the wrong answer about the dark matter?

The Experiment: Building a Virtual Universe

The authors (Kristian and Anna) decided to test this. Instead of looking at real galaxies (where we can't see the "true" answer), they built a virtual universe inside a computer.

1. The Setup: They created 5 specific dwarf galaxies (Fornax, Draco, Sculptor, Carina, and Ursa Minor) that look and act exactly like the real ones we see today.
2. The Twist: They started these galaxies with a "perfect" dark matter shape (a dense knot in the center, known as an NFW profile).
3. The Simulation: They let these virtual galaxies orbit around two versions of the Milky Way:
   • The "Heavy" Milky Way: A massive, strong bully that stretches the dwarfs a lot.
   • The "Light" Milky Way: A smaller, gentler bully that stretches them less.
4. The Test: They ran the standard detective tool (called pyGravSphere) on their virtual galaxies to see what it "inferred" about the dark matter. Then, they compared the tool's guess to the actual truth they knew from the simulation.

The Findings: What Went Wrong (and Right)

Here is what they discovered, translated into everyday terms:

1. The "Stretching" Didn't Break the Tool (Mostly)

Analogy: Imagine trying to guess the weight of a person by watching them walk. If they are walking normally, you guess right. If they are being pulled by a rope (tidal force), they might stumble.
Result: Surprisingly, the "stumbling" caused by the Milky Way's gravity didn't ruin the detective's guess about the inner shape of the dark matter. Even though the galaxies were being stretched, the tool still figured out the general shape reasonably well. The assumption that the galaxy is "calm" wasn't the biggest problem.

2. The Tool Had a "Blind Spot" (The Real Culprit)

Analogy: Imagine trying to draw a map of a coastline using a ruler that only knows how to draw straight lines. No matter how hard you try, you can't draw the jagged, curvy edges of the real coast.
Result: The tool (pyGravSphere) uses a specific mathematical model called a "broken power-law." It's like a ruler that can only draw straight lines. The real dark matter distribution in the outer edges of these galaxies gets weird and steep because of the Milky Way's stretching. The tool couldn't handle this "weirdness."

• The Consequence: Because the tool couldn't fit the outer edges, it got confused and guessed that the center was less dense than it really was. It thought the dark matter was fluffier than it actually was.

3. The "Mass Estimator" is Sensitive to Mood

Analogy: There is a simpler, quick way to guess the weight of a galaxy (the Wolf Estimator). It's like guessing a person's weight just by looking at how fast they are running.
Result: This quick guess works well (within 10% accuracy), but it depends on where the galaxy is in its orbit.

• If the galaxy is at its closest point to the Milky Way (Pericentre), it's being squeezed.
• If it's at its farthest point (Apocentre), it's relaxed.
  The tool's accuracy wiggles slightly depending on this "mood," but generally, it's still a good guess.

4. The "Dark Matter Annihilation" Score (J-Factors)

Analogy: Scientists want to know how much "dark matter signal" we might detect from space. This is called the J-factor. It's like calculating how much light a bulb emits.
Result: Because the tool underestimated the density in the center, it also underestimated the J-factor. It told us there would be less signal than there actually is. However, the error bars (the "confidence interval") were also too small. The tool was too confident in its wrong answer.

The "Tidal Stirring" Mystery

There is a popular theory that dwarf galaxies were once flat disks (like pizza dough) that got stretched into round balls by the Milky Way's tides. This is called "Tidal Stirring."

• The Tension: The authors found that for their simulations to match reality, the Milky Way would have to be relatively "light" (not too massive). If the Milky Way were too heavy, the tidal forces would have stripped the dwarfs too much, making them look nothing like what we see today.
• The Implication: This might mean the "Tidal Stirring" theory needs a rethink, or perhaps the Milky Way is lighter than we think.

The Takeaway

"Broken Expectations" means that while we thought the main problem with our galaxy models was the "stretching" from the Milky Way, the real problem is actually our mathematical tools.

• The Good News: The Milky Way's tides aren't ruining our ability to see the dark matter shape.
• The Bad News: Our current math tools are too rigid. They can't handle the complex shapes of the outer edges of these galaxies, which tricks them into thinking the center is less dense than it is.

The Solution: We need better, more flexible mathematical tools (like using a flexible ruler instead of a straight one) to accurately map the invisible dark matter in our cosmic neighborhood. Until then, we should be a bit humble about how confident we are in our measurements of the universe's invisible scaffolding.

Technical Summary

Here is a detailed technical summary of the paper "Broken Expectations: The Effects of Modelling Assumptions on the Inferred Dark Matter Distribution in the Milky Way's Satellites" by Tchiorniy and Genina (2026).

1. Problem Statement

The nature of dark matter (DM) is often investigated by analyzing the density profiles of dwarf spheroidal (dSph) satellites of the Milky Way. A central debate in cosmology is the "core-cusp problem": $\Lambda$CDM simulations predict cuspy DM halos ($\rho \propto r^{-1}$), while some observations suggest cored profiles ($\rho \propto r^0$).

To infer these profiles, astronomers rely on spherical Jeans equation modeling, which assumes:

1. Spherical symmetry of the system.
2. Dynamical equilibrium (time-invariant distribution function).

However, Milky Way satellites are subject to Galactic tidal forces, which can deform the satellites and drive them out of equilibrium. While previous studies have addressed tidal stripping of unbound stars, there is a lack of consensus on how tides affect the inference of the inner DM density and the J-factor (a measure of DM annihilation flux) when using standard Jeans modeling codes. This paper aims to quantify the systematic errors introduced by tidal perturbations and modeling assumptions in the recovery of DM distributions.

2. Methodology

2.1 Tailored Simulations

The authors created a suite of high-resolution $N$-body simulations for five classical dSphs: Fornax, Draco, Sculptor, Carina, and Ursa Minor.

• Initial Conditions: Dwarfs were initialized with Navarro-Frenk-White (NFW) DM halos (cuspy) and Plummer stellar distributions. They were set up to match observed stellar masses, half-light radii, and velocity dispersions.
• Milky Way Potentials: Simulations were run in two static potentials from the galpy library:
  • "Light" MW: $M_{vir} = 0.8 \times 10^{12} M_\odot$.
  • "Heavy" MW: $M_{vir} = 1.6 \times 10^{12} M_\odot$.
• Orbits: Orbits were derived from Gaia EDR3 proper motions, ensuring the dwarfs reached their present-day positions consistent with observations.
• Resolution: Each halo contained $10^7$DM particles to avoid artificial disruption (softening length$\epsilon = 7$ pc).
• Evolution: Simulated using GADGET-4 for 10 Gyr. Only bound particles (identified via SUBFIND) were used for analysis to isolate the effects of tidal deformation from contamination by unbound stars.

2.2 Inference Techniques

The authors applied two primary methods to the simulated data:

1. pyGravSphere: A non-parametric Jeans modeling code using a broken power-law model for the DM density profile and a radially varying anisotropy profile. It utilizes the 4th-order projected virial theorem (Virial Shape Parameters) to break the mass-anisotropy degeneracy.
2. Wolf Mass Estimator: A simplified estimator ($M \approx 4 \langle \sigma_{los}^2 \rangle R_e / G$) tested for accuracy across different orbital phases (pericenter, apocenter, present day).

2.3 J-Factor Calculation

J-factors were calculated in two ways:

• "True" J-factor: Derived directly from the simulation snapshot by summing the squared densities of bound DM particles within a $0.5^\circ$ cone.
• Inferred J-factor: Derived from the density profiles recovered by pyGravSphere.

3. Key Contributions & Results

3.1 Impact of Tides on Density Profile Recovery

• Equilibrium Assumption: The study finds that tidal forces do not significantly degrade the quality of density profile inference for the inner regions of these dwarfs, provided the analysis is restricted to bound stars. The assumption of dynamical equilibrium remains robust for the sample studied.
• Systematic Bias in pyGravSphere: The code consistently underestimates the inner DM density and infers flatter density slopes (more core-like) for all simulated dwarfs, even though the input halos were cuspy NFW profiles.
  • Cause: This bias is attributed to the broken power-law model used in pyGravSphere. The model struggles to describe the steepening of the density slope in the outer halo caused by tides, and the "smoothness" constraints on the power-law breaks force the inner slope to flatten to compensate.
  • Implication: The apparent "cores" found in some real dSphs may be artifacts of the modeling technique rather than physical reality.

3.2 Wolf Mass Estimator Accuracy

• Orbital Dependence: The accuracy of the Wolf estimator depends on the orbital phase and the shape of the velocity dispersion profile.
  • In the "Light" MW, velocity dispersion profiles often rise in the outskirts, leading to mass overestimation.
  • In the "Heavy" MW, profiles flatten or drop, leading to mass underestimation (typically by 5–10%).
• Stellar vs. DM Tracers: The estimator is highly accurate ($\sim$10%) for stellar tracers because the stellar component is more tightly bound and less deformed by tides than the DM halo. However, for DM tracers extending to the tidal radius, the estimator shows oscillating errors correlated with the orbital phase (pericenter vs. apocenter) due to halo asphericity.

3.3 J-Factor Inference

• Underestimation: The inferred J-factors are generally underestimated compared to the true values, primarily due to the underestimation of central densities.
• Error Bars: The confidence intervals (error bars) on the J-factors are often underestimated. The authors suggest that a systematic uncertainty of at least $\sigma_J \approx 0.1$ dex should be added to J-factor estimates derived from pyGravSphere to account for modeling biases.

3.4 Mass-Concentration Relation & Milky Way Mass

• The simulated dwarfs in the "Light" MW potential generally align with the $\Lambda$CDM mass-concentration relation.
• In the "Heavy" MW potential, dwarfs like Draco and Ursa Minor appear as outliers (too low concentration for their mass) unless significant tidal stripping is assumed.
• The study suggests that a lighter Milky Way (or limited tidal action) is required to reconcile the observed kinematics of these dwarfs with standard $\Lambda$CDM NFW halos, potentially challenging "tidal stirring" formation scenarios where dwarfs are heavily processed.

4. Significance and Conclusions

• Modeling vs. Physics: The primary source of error in inferring DM cores in dSphs is not the violation of dynamical equilibrium by tides, but rather the parameterization limitations of the Jeans modeling codes (specifically the broken power-law model in pyGravSphere).
• Unbound Stars: The presence of unbound stars (tidal debris) in observational samples poses a greater threat to the accuracy of Jeans models than the tidal deformation of the bound system itself.
• J-Factors: Current J-factor estimates may be systematically biased low, and their uncertainties are likely underestimated. This has implications for constraints on Dark Matter particle physics (e.g., WIMP annihilation limits).
• Future Directions: The authors advocate for:
  1. Using more flexible density models (e.g., Zhao profiles or increasing bin counts).
  2. Accounting for orbital phase when applying mass estimators.
  3. Developing non-parametric approaches that can handle the specific density slopes induced by tides.

The paper provides a critical calibration of the tools used to study Dark Matter, highlighting that "broken expectations" regarding density profiles often stem from the models themselves rather than the underlying physics of the satellites.