Forward-Only Temporal-Coherence Occupancy Regimes in Atlantic Tropical Cyclone Intensity Evolution (1851–2024)

Nathan Howell 

Rapid intensification (RI) remains one of the most consequential and difficult-to-characterize aspects of tropical cyclone intensity evolution, commonly defined as an increase in maximum sustained wind speed of at least 30 kt within 24 h (Kaplan and DeMaria 2003; Kaplan et al. 2010; DeMaria et al. 2021). This study evaluates whether Atlantic best-track intensity records contain forward-detectable temporal occupancy regimes. A minimal diagnostic is constructed from HURDAT2 using causal rolling means of maximum sustained wind speed across multiple advisory-window lengths and a directional-agreement activation rule. The framework is evaluated under replay-constrained execution and frozen parameter governance using only current and prior intensity values from 1,991 Atlantic tropical cyclones during 1851–2024.

Storms occupying the active regime for longer cumulative durations form a subset with elevated conditional RI occurrence. Conditional RI occurrence increases from a basin-wide baseline of approximately 26% to about 60% for storms with at least 140 h of active occupancy and to about 67% for storms with at least 168–270 h of active occupancy. Diagnostic behavior is additionally characterized using occupancy climatology, lifecycle-relative onset structure, and lead-time survival analysis. Alignment significance is evaluated against autocorrelated block-shuffle and phase-constrained null frameworks designed to preserve storm lifecycle evolution.

Occupancy behavior remained stable across parameter neighborhoods, preserved timing organization under constrained null frameworks, and retained separable timing structure under tested conditioned formulations, although advisory-level occupancy coefficients attenuated substantially in shared-sample models. These results support interpretation of the diagnostic as a forward-only occupancy framework for temporal organization in tropical cyclone intensity evolution.