[Paper Review] Mean-Field Theory and Sandpile Models

Matthew Stapleton, Kim Christensen|arXiv (Cornell University)|Oct 24, 2005

Theoretical and Computational Physics2 references3 citations

TL;DR

This paper develops a mean-field theory for sandpile models using a one-dimensional random walker with absorbing and reflecting boundaries to model avalanche dynamics. It shows that scaling and universality emerge from bulk conservation and symmetry, with avalanche size distribution governed by critical exponents τ and Δ; crucially, τ depends on driving location while Δ is universal, and finite driving rates introduce an exponential cutoff, confirming SOC requires zero driving rate in the thermodynamic limit.

ABSTRACT

We review and refine the concept of a mean-field theory for the study of sandpile models, which are of central importance in the study of self-organized criticality. By considering the simple one-dimensional random walker with an absorbing and reflecting boundary we are able to construct a complete mean-field picture which we can solve in detail for different types of driving. Using this theory, we are able to clarify the effect of finite driving rate on sandpile models, as well as the observed sensitivity of certain universal quantities on the driving.

Motivation & Objective

• To clarify the role of mean-field theory in understanding self-organized criticality (SOC) in sandpile models.
• To investigate how finite driving rates affect scaling behavior and critical exponents in SOC systems.
• To explain the observed numerical sensitivity of the exponent τ to driving location while Δ remains constant.
• To develop a general scaling theory for avalanche size moments in sandpile models, incorporating driving rate and system size.

Proposed method

• Model avalanche dynamics as a random walker on a line with an absorbing boundary at x=0 and reflecting at x=L.
• Use the Fokker-Planck equation with drift and diffusion to describe the probability density φ(x,t;x₀) of the walker’s position.
• Define avalanche size T as the first passage time to x=0, with P(T;L) derived from the flux at the absorbing boundary.
• Analyze the scaling behavior of P(T;L) in the limit L≫1 and T≫1, identifying power-law scaling with exponents τ and Δ.
• Introduce a finite driving rate q as a negative drift velocity −v to model external particle injection, leading to an exponential cutoff in P(T;q).
• Derive moment scaling laws ⟨Tⁿ⟩ ∝ L^{γₙ} with γₙ = Δ(1 + n − τ), and construct a general scaling function incorporating x₀/L, qL/D, and lattice scale a.

Experimental results

Research questions

• RQ1 How does mean-field theory explain the emergence of scaling and universality in sandpile models without explicit fluctuations or medium interactions?
• RQ2 Why does the critical exponent τ depend on the driving location x₀ while Δ remains invariant, contrary to expectations from renormalization group theory?
• RQ3 What is the effect of a finite driving rate on the scaling behavior of avalanche size distributions in SOC systems?
• RQ4 How do conservation laws and time-reversal symmetry constrain the functional dependence of observables on the initial position x₀?
• RQ5 Can a general scaling theory be formulated for avalanche size moments that incorporates system size L, driving rate q, and initial position x₀?

Key findings

• The mean-field model exhibits power-law scaling P(T;L) ≈ aT^{-τ}G(bT/L^Δ) with τ = 3/2 and Δ = 2 in the zero-drift limit, consistent with mean-field SOC universality.
• The exponent τ depends on the initial driving position x₀, while Δ remains constant, due to linear dependence of P(T;L) on x₀, which arises from conservation and time-reversal symmetry.
• Finite driving rate q introduces an exponential cutoff in the avalanche size distribution, with characteristic time T_q = 4D/q², implying scaling only emerges in the q→0 limit.
• The scaling of moments ⟨Tⁿ⟩ is governed by ⟨Tⁿ⟩ ≈ ΓₙL^{Δ(1+n−τ)}, confirming that the gap between exponents γₙ is proportional to Δ.
• The scaling function Ψₙ(x₀/L, qL/D) in the moment scaling law ⟨Tⁿ⟩ ∝ x₀L^{2n−η(n)−1}a^{η(n)}/Dⁿ shows that x₀/L and qL/D are the key control parameters, with a dependence on lattice scale a only through an exponent η(n).
• The model confirms that SOC scaling requires zero driving rate, as finite q breaks scale invariance via the T_q cutoff, aligning with the hypothesis that SOC is a critical state only in the thermodynamic limit of slow driving.