Incoprea Research · Hypothesis tests III

Pulsar glitches are avalanches

The posted claim Glitch sizes are avalanche events in a pinned vortex lattice, so their size distribution should follow a power law (P1) and its exponent should be universal across pulsars (P3).

What we did

733 positive glitches across 226 pulsars from the Jodrell Bank catalogue (Basu et al. 2022). A power law was fitted by maximum likelihood over x ≥ x_min and compared with a lognormal; the fit was repeated at nine thresholds. The five pulsars with ≥25 glitches were fitted separately to test universality.

What we found

P1 (conditional). The distribution is broad but not a single power law: the local slope steepens from α ≈ 1.21 (x ≥ 1) to 2.46 (x ≥ 5000), and below about 3000×10⁻⁹ a lognormal fits better. Only the largest ~4% of glitches (28 events above 5000×10⁻⁹) favour the power law (log-likelihood ratio +7.7, bootstrap p = 0.003).

P3 (not supported). The five best-sampled pulsars have different exponents — Vela α ≈ 1.19, the Crab α ≈ 1.63 — with a homogeneity p < 10⁻⁴.

glitch size CCDF and per-pulsar exponents — Left: glitch-size complementary CDF with the fitted power-law tail. Right: per-pulsar exponents with 95% confidence intervals.

Verdict: Conditionally supported (P1) and Not supported (P3).

Working results, 30 October 2026. Not peer reviewed. Not medical, investment or policy advice. Prepared with an AI system (DeepSeek Harness); no independent human review.