Controlled Experiments on Dark-Matter Halo Structure and Galaxy Morphology I: What Sets Galaxy Sizes?

Using controlled simulations of isolated galaxies, this study identifies halo concentration as the dominant predictor of galaxy size at the massive-dwarf mass scale, while also quantifying the systematic influences of halo spin, inner density profile, and baryon fraction on stellar and cold baryon radii.

The Gist

Imagine you are a master chef trying to bake the perfect cake. You know that the size of the cake depends heavily on the size of the oven (the Dark Matter Halo). But what if you have two ovens of the exact same size, yet one bakes a fluffy, giant cake, and the other bakes a tiny, dense puck? Why does the same "oven" produce such different results?

This is exactly the puzzle astronomers Guangze Sun, Fangzhou Jiang, and Jing Wang tackled in their new paper. They wanted to figure out what specific "ingredients" and "settings" inside a galaxy's invisible home (its dark matter halo) determine how big the galaxy itself will be.

Here is the story of their experiment, explained simply.

The Experiment: A Controlled Kitchen

In the real universe, it's hard to study this because every galaxy is unique, and they are all mixed up with other galaxies. It's like trying to figure out how sugar affects a cake by only looking at cakes that were baked in different kitchens, with different ovens, and different chefs.

So, these scientists built a virtual laboratory. They created 132 "perfect" galaxies in a computer simulation.

• The Constant: They kept the "oven size" (the total mass of the dark matter) exactly the same for every single run.
• The Variables: They tweaked four specific "knobs" on the halo:
  1. Spin: How fast the halo is spinning.
  2. Concentration: How tightly packed the dark matter is in the center (like a dense fruitcake vs. a fluffy sponge).
  3. Inner Slope: How steep the "walls" of the gravity well are in the very center.
  4. Baryon Fraction: How much "real stuff" (gas and stars) they put in compared to the invisible dark matter.

They then let these virtual galaxies evolve for billions of years using super-advanced physics (called FIRE-3) that mimics how stars form, explode, and push gas around.

The Results: What Makes a Galaxy Big or Small?

After the simulations finished, they measured the size of the resulting galaxies. Here is what they found, using some fun analogies:

1. The Spin: The Ice Skater Effect 🌪️

Finding: The faster the halo spins, the bigger the galaxy.
Analogy: Think of an ice skater. If they start with their arms out wide and pull them in, they spin faster. In a galaxy, if the "halo" (the invisible container) has a lot of spin, the gas inside wants to spread out to conserve that momentum, just like the skater's arms. This creates a large, sprawling disc.

• Result: High spin = Big galaxy. Low spin = Small galaxy.

2. The Concentration: The Heavy Anchor ⚓

Finding: The more concentrated the dark matter is in the center, the smaller the galaxy.
Analogy: Imagine a magnet. If the magnet is super strong and concentrated in the middle, it pulls everything tight. A "concentrated" halo acts like a super-strong anchor, pulling the gas and stars in tight, preventing them from spreading out. A "loose" halo lets the gas drift further out.

• Result: High concentration = Small galaxy. Low concentration = Big galaxy.
• Surprise: The scientists found that concentration is actually the most important factor, even more than spin!

3. The Inner Slope: The Shape of the Bowl 🥣

Finding: The shape of the very center matters, but only if it's extremely steep.
Analogy: If you pour water into a bowl, the shape of the bottom matters. If the bowl has a very sharp, steep point at the bottom (a "cusp"), the water (gas) gets sucked right into the center and forms a tiny, dense core. If the bottom is flat, the water spreads out.

• Result: Unless the center is extremely steep, this doesn't change the size much.

4. The Baryon Fraction: The "Too Much Stuff" Problem 🌪️🔥

Finding: This was the most tricky one. Adding more gas didn't just make the galaxy bigger; it changed how the other factors worked.
Analogy: Imagine you are trying to build a sandcastle.

• If you have a little bit of sand (low gas), the wind (spin) can easily blow it into a wide, flat shape.
• But if you dump a massive pile of wet sand (high gas fraction) all at once, it gets heavy and clumps together in the middle. The wind can't spread it out. Instead of a wide castle, you get a tall, compact tower.
• Result: When there is too much gas, the galaxy collapses into a compact ball, regardless of how much the halo is spinning.

The Big Takeaway: It's Not Just One Thing

For a long time, scientists thought Spin was the main boss of galaxy size. This paper says, "Not so fast!"

They used a "statistical detective" tool (like a smart AI) to rank the importance of these factors. The ranking came out as:

1. Concentration (The most important!)
2. Spin (Second place)
3. Inner Slope (Third place)
4. Gas Amount (The "wildcard" that changes the rules)

Why Does This Matter?

This study helps us understand why the universe is so diverse. It explains why some dwarf galaxies are huge, fluffy discs, while others are tiny, compact blobs, even if they live in halos of the same mass.

It also tells us that our old theories (which mostly focused on spin) were missing a huge piece of the puzzle. To predict how big a galaxy will be, we can't just look at how fast it's spinning; we have to look at how "packed" its invisible home is and how much gas it has to work with.

In short: The size of a galaxy isn't just about how fast it spins; it's about how tightly its invisible home is packed, and whether it has too much "stuff" inside that forces it to collapse into a compact shape.

Technical Summary

Here is a detailed technical summary of the paper "Controlled Experiments on Dark-Matter Halo Structure and Galaxy Morphology I: What Sets Galaxy Sizes?" by Sun, Jiang, and Wang.

1. Problem Statement

In the $\Lambda$CDM paradigm, while dark matter (DM) halo mass is the primary driver of galaxy morphology and size, significant scatter exists in galaxy sizes at a fixed halo mass (particularly at the massive-dwarf scale, $M_{vir} \sim 10^{11} M_\odot$). Previous cosmological simulations have yielded conflicting results regarding which secondary halo properties (spin, concentration, inner density slope, baryon fraction) regulate this scatter.

• Limitations of existing studies: Cosmological simulations often rely on instantaneous halo properties rather than progenitor properties, suffer from ambiguous definitions of "halo formation time," and lack the statistical power to systematically vary halo structural parameters due to limited simulation volumes or zoom-in constraints.
• Goal: To isolate and quantify the specific influence of four key halo parameters—concentration ($c_2$), inner density slope ($s_1$), spin ($\lambda$), and baryon fraction ($f_b$)—on galaxy size at a fixed halo mass, free from the complexities of cosmological assembly history.

2. Methodology

The authors employed a suite of controlled, isolated galaxy simulations to create a factorial parameter grid.

• Simulation Setup:

  • Code: GIZMO (mesh-free finite-mass mode) with the FIRE-3 physics model (including updated stellar feedback, metal cooling, and radiation pressure).
  • Halo Mass: Fixed at $M_{vir} = 10^{11} M_\odot$.
  • Initial Conditions: Generated using the Dekel-Zhao (DZ) profile, which allows independent control of the inner density slope (unlike standard NFW profiles). Gas was initialized in hydrostatic equilibrium and allowed to relax for 3 Gyr before full physics were turned on for an additional 3 Gyr.
  • Parameter Grid: A nearly factorial design varying four parameters:
    • Concentration ($c_2$): 3, 10, 20.
    • Inner Density Slope ($s_1$): 0.0 (flat core) to 1.5 (steep cusp).
    • Spin ($\lambda$): 0.02 to 0.08.
    • Baryon Fraction ($f_b$): 0.03, 0.07, 0.14.
  • Sample Size: 132 distinct simulation runs.

• Analysis Techniques:

  • Metrics: Galaxy size defined as the 3D stellar half-mass radius ($r_{1/2, \star}$) and the cold baryon half-mass radius ($r_{1/2, \star+cg}$).
  • Statistical Modeling:
    1. Quadratic Response-Surface Method (RSM): Fitted a second-order polynomial to quantify linear, quadratic, and interaction effects of the parameters.
    2. Random Forest (RF) Regression: Used non-parametric feature importance ranking (MDI, permutation, SHAP) to validate results without assuming a specific functional form.
  • Comparisons: Results were benchmarked against the classical Mo et al. (1998) analytic model and empirical predictors from Jiang et al. (2019).

3. Key Results

A. Dependence of Galaxy Size on Halo Parameters

• Halo Concentration ($c_2$): Identified as the most dominant predictor. Higher concentration leads to systematically smaller galaxies ($r_{1/2, \star} \propto c_2^{-0.7}$). This effect is stronger for cold baryons than for stars.
• Halo Spin ($\lambda$): Higher spin leads to larger galaxies, but the relationship is non-linear and saturates at high spin values ($\lambda > 0.06$).
• Inner Density Slope ($s_1$): Generally weak sensitivity for $s_1 \lesssim 1$. However, in highly cuspy haloes ($s_1 = 1.5$), galaxy sizes are significantly reduced.
• Baryon Fraction ($f_b$): Exhibits a non-monotonic and complex impact.
  • At low $f_b$ ($\lesssim 0.07$), galaxies form extended, settled discs.
  • At high $f_b$ ($\approx 0.14$, near cosmic mean), the system undergoes centrally concentrated star formation, resulting in compact galaxies.
  • Crucially, $f_b$ acts as a modulator: it changes the sensitivity of galaxy size to halo spin and concentration. High $f_b$ amplifies the dependence on structural parameters.

B. Feature Importance Ranking

Using both RSM and Random Forest metrics, the authors established a consistent hierarchy of importance for regulating galaxy size:

1. Halo Concentration ($c_2$): Most informative (contribution $\sim 30-40\%$).
2. Halo Spin ($\lambda$): Second most important ($\sim 20-25\%$).
3. Inner Density Slope ($s_1$): Significant but weaker ($\sim 20-25\%$).
4. Baryon Fraction ($f_b$): Least direct, acting primarily through interactions ($\sim 15\%$).

C. Evaluation of Existing Models

• Mo et al. (1998) Model: The model captures the general scaling but systematically underpredicts galaxy sizes by a factor of $\sim 5$. This is attributed to low angular momentum retention factors ($f_j = j_d/m_d \ll 1$) in the simulations, suggesting star-forming gas is drawn from low-angular-momentum tails and feedback redistributes angular momentum.
• Jiang et al. (2019) Concentration Predictor: The empirical relation $r_{1/2, \star} \propto c_2^{-0.7}$ successfully captures the qualitative trend. The controlled simulations confirm this dependence arises from internal virialized dynamics, not just cosmological assembly history.
• Single-Parameter Limitations: Predictors based solely on spin or concentration leave large scatter, confirming that a multi-parameter approach is necessary.

D. Definition of Galaxy Size

The authors found that defining the galaxy as stars + cold gas ($T < 10^4$ K) yields systematically larger radii than stars alone. Including cold gas improves the agreement with the Mo98 model by shifting systems closer to the $j_d = m_d$ relation, indicating that cold gas retains more angular momentum.

4. Significance and Conclusions

• Primary Driver: The study definitively identifies halo concentration as the primary secondary parameter controlling the scatter in galaxy sizes at fixed mass, surpassing the traditional focus on halo spin.
• Mechanism: The results suggest that the internal structure of the dark matter halo (specifically concentration) imprints itself on the galaxy size through the depth of the potential well and the efficiency of gas collapse, independent of cosmological merger history.
• Baryon Role: The baryon fraction is not a simple linear regulator but a complex modulator that determines whether a halo forms a compact or extended system, heavily influencing how other parameters manifest.
• Implications for Modeling: Semi-analytic and empirical models of galaxy formation must incorporate halo concentration and baryon fraction, rather than relying solely on mass and spin, to accurately reproduce the observed diversity of dwarf galaxy sizes.

This work provides a controlled baseline for understanding galaxy-halo connections, isolating physical mechanisms that are often degenerate in full cosmological simulations. Future papers in this series will extend these findings to other halo masses and morphological features like clumpiness.