• Accepted Paper

Relaxation timescales as diagnostics of feedback topology and mechanism in adaptive self-organized critical systems

Lock Yue Chew, Ning Ning Chung, John H. Miller, Sean S. Downey, Marc F. Müller, and J. Stephen Lansing

Phys. Rev. E - Accepted 12 November, 2026

DOI: https://doi.org/10.1103/h8rr-3xw1

Abstract

Power-law statistics alone cannot distinguish self-organized from externally-tuned criticality. Here we investigate whether relaxation timescale τ—an observable intrinsic to all critical dynamics—can provide this diagnostic power by reflecting the structure of effective energy landscapes. Analyzing four models within a unified computational framework that renders relaxation times commensurate across architecturally distinct systems (Ising, Kéfi vegetation, Daisyworld, Balinese subak irrigation), we find that systems with indistinguishable steady-state statistics (α ≈ 1) span two orders of magnitude in τ, appearing to cluster by feedback topology and its associated mechanism: τ ≈ 376 (Ising at L = 50, local positive feedback, fitted value), τ ≈ 15 (Kéfi vegetation, local positive and global positive feedback), τ ≈ 4 (Daisyworld, local positive and global negative feedback), and τ ≈ 3 (Balinese subak at b = 3, local positive and global negative feedback). This hierarchy reflects how feedback topology and mechanism shape energy landscape geometry. Three independent methods (Hamiltonian model, optical satellite imagery, and SAR analysis) converge on consistent power-law exponents in Balinese rice terraces, confirming that these spatial signatures are measurable in empirical systems; measurement of the fast temporal dynamics (τ ≈ 3) awaits high-frequency satellite time series. The τ hierarchy, if validated across additional model systems, would provide a second independent observable that substantially enhances the diagnostic power of SOC analysis.

Export citation

Export citation

Choose format for download:

Download Citation

If the author has provided any supplemental materials with this article they will be available upon publication of the version of record.

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation