Beyond labels: multistation coupling regimes discriminate natural, industrial, and SpaceX Starship launch events (TAMC-FRANJAMAR v3)

F. Javier Martínez SánchezAuthor of an article published in Astronomy and Computing DOI: https://doi.org/10.1016/j.ascom.2026.101155 A temporary Elsevier Share Link is also available for free access to the final published version: https://authors.elsevier.com/a/1nHpr7tDLPb2TB Please use the DOI as the permanent citation link. The Elsevier Share Link provides temporary free access to the final published article. ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------- TAMC-FRANJAMAR v3 — Reproducible Multistation Analysis Framework This repository provides a fully reproducible pipeline for analyzing collective structure in multistation waveform data across heterogeneous extreme events and engineered launch systems. Core idea:events are not best characterized only by labels such as earthquake, explosion, collapse, or launch, but by their underlying coupling structure across the sensor network. The framework extracts this structure directly from data using fixed parameters and identical processing across all events. Key capabilities: Multistation synchrony and coherence analysis Event-centered temporal structure Robust anomaly and extreme-event extraction Null-model validation Unsupervised clustering of coupling regimes Cross-event comparison of compact, delayed, and fragmented responses Validated across heterogeneous datasets including:underground tests, industrial explosions, atmospheric events, structural failures, tectonic earthquakes, and SpaceX Starship flight tests. SpaceX Starship case study:Starship IFT-1 to IFT-8 are included as a controlled technological testbed for evaluating whether TAMC-FRANJAMAR separates events by structural coupling morphology rather than by externally assigned outcome labels. This analysis is methodological and does not attempt to infer vehicle performance or diagnose engineering failure mechanisms. Quick start:conda create -n tamc python=3.10 -yconda activate tamcpip install -r requirements.txtpython full_pipeline_franjamarv3.py Outputs are generated under /outputs/. Live monitoring dashboard, experimental and non-predictive:https://franjamar-monitor.streamlit.app/ The same fixed pipeline is applied in continuous monitoring mode using a T–3 h delayed window, enabling near-real-time exploration of multistation coherence patterns across multiple regions. Reproducibility:All results are fully reproducible under fixed parameterization. Related project: GNGS A complementary project from the same research line is GNGS — Geomagnetic Network Geometry and Synchrony, a local, reproducible and traceable framework for studying the spatial and temporal organization of geomagnetic disturbances across distributed ground-magnetometer networks. GNGS analyses network synchrony, hierarchical structure, Peak–Valley asymmetry and multistation geomagnetic behaviour, providing both HISTORICAL and RECENT operational modes. GNGS v1.0:https://doi.org/10.5281/zenodo.22021222 GNGS is descriptive and diagnostic and does not provide earthquake or seismic-event predictions.