Statistical properties of the gravitational force through ordering statistics

This paper utilizes order statistics to demonstrate that in a three-dimensional homogeneous random gas of point masses, the formally divergent variance of the Holtsmark gravitational force distribution arises exclusively from the contribution of the nearest neighbor, while contributions from all more distant neighbors remain finite.

The Gist

Imagine you are floating alone in a vast, infinite ocean of stars. These stars are scattered randomly, like popcorn kernels floating in a giant, invisible bowl. You are a tiny test particle (let's call you "Starlet").

Now, imagine every single one of those popcorn stars is tugging on you with gravity. The question this paper asks is: What does the total tug feel like? Is it a gentle, steady breeze, or a chaotic, violent storm?

The authors, Constantin and Vincent, use some fancy math to answer this, but here is the story in plain English, using some everyday analogies.

1. The "Holtsmark" Storm

For a long time, physicists knew the answer to the "total tug" question. It's called the Holtsmark distribution.

Think of the Holtsmark distribution like a weather report for gravity. It tells you that usually, the tug is moderate. But here's the weird part: The "average" storm is fine, but the "average" storm intensity is impossible to calculate.

Why? Because in this mathematical model, the "variance" (a measure of how wild the fluctuations are) is infinite. It's like saying the average wind speed is 10 mph, but the average of the wind speed squared is infinity. This happens because the math allows for a scenario where a single star is infinitely close to you, pulling with infinite force.

2. The "Nearest Neighbor" Rule

The authors wanted to figure out why this variance is infinite. Is it because of the billions of distant stars? Or is it just one specific star?

They used a tool called Order Statistics. Imagine you are at a party and you want to know who is closest to you.

• 1st Nearest Neighbor: The person standing right next to you.
• 2nd Nearest Neighbor: The person standing just behind them.
• 3rd Nearest Neighbor: The next one back, and so on.

The paper calculates the exact probability of finding these people at specific distances. They found that as you look further out (to the 10th, 100th, or 1,000th neighbor), the stars start to line up in very predictable, concentric rings. They become "correlated," meaning they are all roughly the same distance away, canceling each other out a bit.

3. The "One Bad Apple" Discovery

Here is the big reveal of the paper: The infinite chaos comes entirely from the very first neighbor.

Think of it like this:

• The Distant Crowd (Stars 2 to Infinity): Imagine a huge crowd of people far away, all shouting at you. Because they are so far and so numerous, their voices blend into a steady, manageable hum. Their combined pull is predictable and stable.
• The One Guy Next to You (The 1st Neighbor): Imagine one guy standing three inches from your ear, screaming.

The paper proves that the "infinite variance" of the Holtsmark distribution isn't caused by the crowd. It is caused 100% by that one guy standing right next to you.

If you remove the nearest neighbor, the math suddenly becomes calm and well-behaved. The "wild" fluctuations of the total gravitational force are almost entirely due to the random chance that one star happens to be incredibly close to you.

4. Why This Matters

In the real world, stars have size (they aren't mathematical points), so they can't actually get infinitely close. But in the idealized math of the universe, this paper explains a fundamental quirk: Gravity is a "local" game.

Even though gravity reaches across the universe, the statistical wildness of the force you feel is dictated by your immediate surroundings. The distant universe provides a background hum, but the nearest neighbor provides the roar.

Summary Analogy

Imagine you are trying to guess the total noise level in a stadium.

• The Holtsmark Distribution says: "The noise is usually loud, but if you try to calculate the 'average loudness squared,' you get infinity."
• This Paper says: "That infinity isn't because of the 50,000 people in the stands. It's because there's a 1-in-a-million chance that someone is screaming directly into your ear. If you ignore the person screaming in your ear, the math works perfectly fine."

The authors successfully separated the "screaming neighbor" from the "crowd," showing that the chaotic nature of gravity in a random universe is a local phenomenon, not a global one.

Technical Summary

1. Problem Statement

The paper investigates the statistical distribution of Newtonian gravitational forces acting on a test particle embedded in an infinite, homogeneous, and uncorrelated random gas of particles (a Poissonian distribution).

• Context: The classic result in this field is the Holtsmark distribution, which describes the probability density of the total gravitational force in 3D space.
• The Anomaly: A well-known property of the Holtsmark distribution in three dimensions ($d=3$) is that while its mean is finite, its variance is formally divergent.
• The Gap: While it is known that the divergence arises from the long-range nature of gravity and the proximity of particles, the specific quantitative contribution of individual "nearest neighbors" to this divergence has not been fully disentangled using order statistics. The authors aim to clarify the origin of this divergence by analyzing the force contributions of the $n$-th nearest neighbor.

2. Methodology

The authors employ Order Statistics to analyze the spatial distribution of particles and their resulting gravitational forces. The methodology proceeds in three main stages:

A. Spatial Statistics of Nearest Neighbors

The authors derive the statistical distributions for the distances to the $n$-th nearest neighbor in a $d$-dimensional homogeneous medium with mean density $\rho$.

• Single Neighbor Distribution: They derive the probability density function (PDF) $w_n(r)$ for the distance $r$ to the $n$-th nearest neighbor. This is derived from the Poisson distribution of particle counts in a volume $V(r)$.
• Joint Distribution: A key methodological advancement is the derivation of the joint spatial distribution $w^{(n)}(r_1, r_2, \dots, r_n)$ for the first $n$ nearest neighbors. They show that for ordered distances ($r_1 \le r_2 \le \dots \le r_n$), the joint PDF factorizes into a product of derivatives of the volume function, multiplied by an exponential term dependent on the largest radius $r_n$.
• Correlation Analysis: They analyze the correlation between successive radial coordinates ($r_n$ and $r_{n+1}$), finding that as $n$ increases, these coordinates become highly correlated, implying that distant neighbors accumulate in thin spherical shells.

B. Transformation to Force Space

Using the spatial distributions derived above, the authors transform the distance variables into force variables.

• Since the gravitational force magnitude scales as $F \propto r^{-2}$ (in 3D), they map the distance PDFs $w_n(r)$ to force PDFs $W_n(F)$ for the $n$-th nearest neighbor.
• They derive explicit analytical expressions for the PDF of the force modulus contributed by the $n$-th neighbor, $W_n(F)$.

C. Analysis of the Holtsmark Distribution

The authors decompose the total force $F = \sum F_i$ into contributions from individual neighbors. They analyze the moments (mean and variance) of these individual contributions and sum them to reconstruct the properties of the total Holtsmark distribution.

3. Key Contributions and Results

A. Derivation of Joint Spatial Distributions

The paper provides a rigorous derivation of the joint PDF for the $n$ nearest neighbors (Eq. 5):
$$w^{(n)}(r_1, \dots, r_n) = e^{-N(r_n)} \prod_{i=1}^n N'(r_i)$$
where $N(r)$ is the expected number of particles in a sphere of radius $r$. This result generalizes previous work and allows for the calculation of correlations between neighbors.

B. Force Distribution of the $n$-th Neighbor

The authors derive the force distribution $W_n(F)$ for the $n$-th nearest neighbor.

• Mean Force: The average force $\langle F_n \rangle$ decreases as $n$ increases, scaling roughly as $n^{-2/3}$.
• Variance: For $n > 1$, the variance of the force contribution is finite. However, for the first nearest neighbor ($n=1$), the second moment (and thus the variance) is infinite.

C. Origin of the Holtsmark Divergence

The central finding of the paper is the resolution of the Holtsmark variance divergence:

• The total variance of the Holtsmark distribution is the sum of the variances of individual neighbor contributions (plus cross-terms which are negligible due to symmetry).
• The authors demonstrate that the sum of variances for $n \ge 2$ converges.
• Conclusion: The divergence of the Holtsmark distribution's variance is entirely caused by the contribution of the single nearest neighbor ($n=1$).
• Tail Behavior: They show that for large forces ($F \to \infty$), the tail of the Holtsmark distribution ($W_H(F) \sim F^{-5/2}$) is mathematically identical to the tail of the distribution of the force from the nearest neighbor alone. This confirms that extreme force fluctuations are dominated by the closest particle.

D. Dimensional Dependence

The paper notes that this divergence is specific to dimension $d=3$.

• In $d=1$, the variance is finite.
• In $d=2$, the variance diverges logarithmically with the number of particles $N$.
• In $d=3$, the divergence is driven strictly by the $n=1$ term.

4. Significance

• Theoretical Clarification: The paper moves beyond the "black box" of the Holtsmark distribution to provide a microscopic explanation for its statistical anomalies. It proves that the "heavy tail" of the force distribution is not a collective effect of many distant particles, but a local effect dominated by the single closest neighbor.
• Methodological Utility: The derived joint distributions for nearest neighbors provide a powerful tool for future studies in astrophysics and plasma physics, particularly for analyzing systems where local clustering or specific neighbor interactions are critical.
• Implications for Stability: The authors suggest that understanding these fluctuations is crucial for studying the stability of gravitationally bound gases and the formation of coherent structures (like filaments), as the nearest neighbor exerts a disproportionately large and fluctuating force compared to the rest of the system.

In summary, Payerne and Rossetto use order statistics to deconstruct the Holtsmark distribution, proving that the infinite variance of gravitational forces in a 3D random medium is a direct consequence of the statistical properties of the nearest neighbor, effectively isolating the source of the divergence.