From a Scoop to the Big Picture: Big Data Analysis of Diverse Zircon U-Pb Datasets in Northeast China

The Phanerozoic geological evolution of Northeast China (Eastern Central Asian Orogenic Belt) exhibits complexity， reflecting the superimposition of multiple tectonic domains from the Paleo-Asian Ocean through the Mongol-Okhotsk regime to the Paleo-Pacific domain. Through a data-driven approach using zircon geochronology， we constructed comprehensive databases of zircons from modern river sands and bedrock samples within the Heilongjiang (Amur) River drainage system—a natural sampling network penetrating various tectonic units in NE China. This big data integration reveals precise age constraints (differences <2 Ma between datasets) at 493 Ma， 298-296 Ma， 253-251 Ma， 185-183 Ma， and 132-131 Ma， documenting the sequential amalgamation of microcontinental blocks and transitions between tectonic regimes. These peaks capture the three-stage closure of the Paleo-Asian Ocean (Erguna-Xing'an， Songnen， and North China Craton amalgamations) and two subsequent stages of Paleo-Pacific subduction. Significantly， zircon Th/U ratios exhibit a clear shift around 260-250 Ma， marking a transition from compressional orogenesis to post-orogenic extension that coincides with the final Paleo-Asian Ocean closure. Notably， the initiation of Mongol-Okhotsk Ocean activity (～320 Ma)， its termination (～230 Ma)， and the contemporaneous onset of Paleo-Pacific tectonics (～230 Ma) are not resolved as distinct peaks in our age spectra， probably because these signals were overshadowed by more dominant， contemporaneous tectono-magmatic events. Our multi-scale watershed analysis demonstrates that river sand zircon populations reflect qualitative aspects of regional geology across all scales， while quantitative representativeness exhibits strong scale dependence—being unachievable in small watersheds but showing quantitative trends in larger systems. This study establishes a paradigm for transforming data from individual “scoops” of zircon analyses into “the big picture” of geological understanding， demonstrating that data-driven approaches are powerful tools for overcoming traditional sampling limitations in regional geological studies.