The global increase in coastal hypoxia over the past decades has resulted from a considerable rise in anthropogenically-derived nutrient loading. The spatial relationship between surface phytoplankton production and subsurface hypoxic zones often can be explained by considering the oceanographic conditions associated with basin size, shape, or bathymetry, but that is not the case where nutrient-enriched estuarine waters merge into complex coastal circulation systems. We investigated the physical and biogeochemical processes that create high-biomass phytoplankton production and hypoxia off the Changjiang (Yangtze River) Estuary in the East China Sea (ECS). Extensive in situ datasets were linked with a coupled Regional Ocean Modeling Systems (ROMS) and carbon, silicate, and nitrogen ecosystem (CoSiNE) model to explain the temporary decoupling of phytoplankton production and hypoxia. The CoSiNE model contains two functional groupings of phytoplankton—diatoms and “other”—and the model results show that diatoms were the major contributors of carbon export and subsurface hypoxia. Both observations and simulations show that, although surface phytoplankton concentrations generally were much higher above hypoxic zones, high-biomass distributions during the summer–fall period did not closely align with that of the bottom hypoxic zones. Model results show that this decoupling was largely due to non-uniform offshore advection and detachment of subsurface segments of water underlying the Changjiang River plume. The near-bottom water carried organic-rich matter northeast and east of the major hypoxic region. The remineralization of this particle organic matter during transit created offshore patches of hypoxia spatially and temporally separated from the nearshore high-biomass phytoplankton production. The absence of high phytoplankton biomass offshore, and the 1–8 weeks’ time lag between the surface diatom production and bottom hypoxia, made it otherwise difficult to explain the expanded core hypoxic patch and detached offshore hypoxic patches. The findings here highlight the value of developing integrated physical and biogeochemical models to aid in forecasting coastal hypoxia under both contemporary and future coastal ocean conditions.