Telegraphic Semantic Compression Applied to Open Scientific Data: A Multi-Domain Subtraction-Based Residual Anomaly Survey

We present the Presignal Telegraphic Semantic Compression (TSC) pipeline, a four-layer data reduction and synthesis framework applying subtraction-based residual analysis to open-access scientific datasets across six independent domains. The pipeline ingests raw data in any format, extracts statistical signal via structured semantic compression, synthesizes anomaly characterizations using large language model analysis, and appends structured entries to a persistent Chronicle research ledger. Applied to 15 Chronicle targets spanning NASA photometry (TESS and Kepler), CERN particle collision data, USGS seismic events, LIGO gravitational wave catalogs, and Swift gamma-ray burst populations, the subtraction-first methodology consistently surfaces residual structure invisible to domain-specific standard pipelines. Key findings include: aperiodic dimming burst signatures across six TESS targets surviving full PDC pipeline correction; four distinct flux depletion events on KIC 8462852 (Boyajian's Star) including a 20.72% depth episode spanning 61 hours; systematic Gutenberg-Richter law violations in M5.0+ seismic data coinciding with the three most catastrophic earthquake years in the study period; a 4.0-sigma excess at 115.28 GeV in CMS dimuon data correlated with the historical LEP2 unexplained anomaly region; 51 gravitational wave merger events exceeding 100 solar masses total challenging stellar evolution pair-instability constraints; and a 9.7% GRB population residual resisting standard bimodal classification. All source data drawn from public archives. Pipeline code and Chronicle ledger available on request.