Reliable estimates of historical catch, fishing pressure, and stock status are fundamental to the sustainable management of China’s marine fisheries. However, assessments are frequently constrained by discontinuities among official landings, fishing-effort statistics, fishery-independent surveys, and commercial catch-per-unit-effort data. This study develops an integrated data–model–management framework to reconstruct long-term changes in fishing pressure, assess the sustainable production potential of China’s coastal fisheries, and translate scientific estimates into operational total catch controls. The analysis integrates historical trawl surveys conducted in the 1950s, a monthly fishery–environment dataset covering 346 fishing grounds and 30 major taxa during 1971–1982, commercial fishing records from 2009–2016, standardized fishery-independent surveys from 2014–2024, and recent provincial statistics on catch, vessel numbers, engine power, and fishing employment. Catch and effort series are harmonized through cross-source consistency checks and fleet-capacity standardization. Generalized additive models and machine-learning ensembles are used to standardize catch per unit effort and separate the nonlinear effects of environmental variability from changes attributable to fishing pressure. Stock status and sustainable yield are subsequently evaluated using complementary data-limited approaches, including CMSY/Bayesian Schaefer models, swept-area estimation, and length-based Bayesian assessment. The multi-method assessment indicates that the sustainable marine capture production of China’s coastal fisheries is likely to fall within approximately 8.34–12.58 million tonnes. A national control target of 10 million tonnes for 2030 is therefore broadly consistent with the estimated sustainable range. Nevertheless, substantial differences are identified among the Bohai–Yellow Sea, East China Sea, and South China Sea in resource condition, fishing intensity, fleet structure, and recovery potential. These differences suggest that allocation based solely on recent catch shares may perpetuate historical fishing-pressure imbalances.