Phytoplankton primary productivity (PP) varies significantly over environmental gradients, particularly in physically-dynamic systems such as estuaries and coastal seas. As the Changjiang River runoff peaks during summer time, large environmental gradients appear in both the Changjiang estuary and adjacent East China Sea (ECS), likely driving significant variability in PP. As satellite models of PP often underperform in coastal waters, we aimed to develop a novel approach for net PP estimation in such a dynamic environment. Parallel in situ measurements of Fast Repetition Rate (FRR) fluorometry and carbon (C) uptake rates were conducted for the first time in this region during two summer cruises in 2019 and 2021. A series of 13C-incubations (n=31) were performed, with measured PP ranging from ~6 - 1700 mgC m-3 d-1. Net PP values were significantly correlated with salinity (r = 0.45), phytoplankton chlorophyll a (Chl-a, r = 0.88), Photosystem II (PSII) functional absorption cross-section (, r = -0.76) and maximum PSII quantum yield (, r = 0.59). Stepwise regression analysis showed that Chl-a and were the strongest predictors of net PP. A generalized additive model (GAM) was also used to estimate net PP considering nonlinear effects of Chl-a and . We demonstrate that GAM outperforms linear modelling approaches in predicting net PP in this study, as evidenced by a lower root mean square error (~140 vs. 250 mgC m-3 d-1). Our novel approach provides a high resolution means to examine carbon cycling dynamics in this important region.