Dynamic Geo-Temporal Analysis and Assessment: How a time rail learned where

Generative Temporal Referencing (GTR) places records on a single line through time before anything else is done with them. This report records what happened when location was attached to that time-keyed record. Over nine days in October 2026, a working newsroom of one editor and two AI sessions began giving its records a place. Each location was set at the precision its source supports and staked to a fixed reference point. It was refined by appended steps as better evidence arrived, and the coarser placements were kept. The resulting practice is named Dynamic Geo-Temporal Analysis and Assessment (DGTA). The report covers: • the development of DGTA, step by step, from a hash-linked record • its standards, at concept level: one frame, stated precision, appended refinement, placement by reference, rings in place and time, proximity across stories, measurements joined by place and time, and maps drawn only as queries • a measurement case: – NOAA tide gauges at La Jolla and Los Angeles ran 15 to 16 inches above the predicted tide over October 6–9, 2026. – On the morning of October 9, both exceeded the National Weather Service minor flood threshold. – On the predicted tide alone, both would have stayed more than a foot below it. • a narrative case, in which locations spoken on a live broadcast without coordinates were recovered from the USGS earthquake catalog • its relation to prior work in temporal GIS, geoparsing, georeferencing, event data and temporal databases • what the practice shows, and what it does not yet show The report sets the 2026–27 king-tide windows as dates of interest for a multi-station test. Appendix A gives the data the newsroom created and the scripts that reproduce every result from NOAA and USGS data. DGTA is a practice of Generative Temporal Referencing, as Latent Value Query is. This is a concept disclosure; the implementation is withheld.