Derivative-Aligned Anticipation of Forbush Decreases from Entropy and Fractal Markers

We develop a feature-based framework to anticipate Forbush decreases (FDs) in one-minute neutron-monitor records by tracking sliding-window invariants from information theory, scaling, and geometry. For each station we compute marker time series—including Shannon, spectral, approximate and sample entropy; Lempel–Ziv complexity; correlation dimension; and Higuchi and Katz fractal dimensions—smooth them with an exponentially weighted moving average, and analyze their within-station standardized first differences. Timing is referenced to an operational alignment time t 0 defined as the minimum of the smoothed count first difference, and marker leads are reported in minutes ( ℓ * < 0 indicates anticipation). Station-level detectability is defined on a pre- t 0 window using a robust z -score detector with bilateral threshold and persistence, requiring neither cross-correlation nor hypothesis testing. We apply the pipeline to two FD episodes with broad station coverage (2023-04-23 and 2024-05-10; 28 stations each). Across events, a compact CORE panel exhibits consistently high detection rates and predominantly negative lead distributions, with median leads of order several hours depending on the invariant and event. Lead dispersion across stations is substantial (interquartile ranges typically spanning a few hours), underscoring the value of station-wise criteria and distributional summaries rather than single-station inference. Representative marker trajectories confirm that early flagging corresponds to sustained pre- t 0 excursions in marker differences, not merely tabulated artifacts. The approach is reproducible from open code, operates on native station units without cross-station homogenization, and is qualitatively stable to sensitivity sweeps of windowing, smoothing, and detector parameters. These results support derivative-aligned invariant panels as practical early-warning complements to amplitude-threshold methods in space-weather nowcasting.