The results from eight cruises that were conducted between 2000 and 2003 are used to elucidate the spatial and temporal dynamics of both nitrate-based new production (NP) and primary production (PP) in the northern South China Sea (SCS). The spatial variation study, which was based on the results of three cruises, compares the differences among the shelf, slope and basin of the SCS in the spring and fall. Temporal (seasonal) variation was studied during eight cruises at a basin site and two shelf sites in the SCS. Nutrient enrichment experiments were conducted concurrently with field biochemical surveys to evaluate the role of nitrogen, phosphorus and iron in limiting phytoplankton growth. The results clearly show that winter was the most productive season of the year in the basin and possibly on the shelf as well. Integrated primary production (IPP) and nitrate-based new production (INP) of the basin were 0.55 and 0.26 g C m −2 d −1, respectively, in winter; 0.26 and 0.07 g C m −2 d −1, respectively, in spring; 0.19 and 0.03 g C m −2 d −1, respectively, in summer; and 0.28 and 0.05 g C m −2 d −1, respectively, in fall. The f-ratio (INP/IPP) was as high as 0.47 in winter, decreasing to 0.29–0.36 in early mid-March and 0.14–0.20 from late March to October, and fluctuated coincidentally with abundance of nitrate. The presence of nutrients in the surface layer, supporting the winter productivity, could be attributed to the shallow nitracline, which made nutrient mixing possible when a cold and strong northwest monsoon prevailed, reflecting the vital role of nitrate availability in modulating new production in the SCS. Spatial variation of INP or IPP among the basin, slope and shelf are not as obvious as the seasonal variations because of great local variability. In general, however, the shelf was more productive than the slope and the basin. In the spring, mean IPP and INP in the shelf, slope and basin were 0.72 and 0.16, 0.34 and 0.10, 0.49 and 0.15 g C m −2 d −1, respectively. In the fall, IPP and INP in the shelf and basin were 0.45 and 0.14 g C m −2 d −1, and 0.34 and 0.09 g C m −2 d −1, respectively. There exists a positive correlation between INP and IPP for the spatial data set in either the spring or the fall. f-Ratios are not statistically different between spring (0.28) and fall (0.29). While both nitrate availability and irradiance significantly modulated the spatial variations of production, effects of nitrate availability were more prevalent than light intensity. Diatoms and coccolithophores prospered in winter in the basin or on the shelf in both winter and summer, when surface water was rich in nitrate, and were more abundant than in summer in the basin when nitrate was scarce. The results of the nutrient enrichment experiments also supported the role of nitrogen availability in productivity of the SCS water. In these experiments, chlorophyll a concentration and primary productivity responded to nitrate supplementation, while iron and phosphorus supplementation did not stimulate phytoplankton growth. The predicted ratio of nitrate-based NP to N 2-fixation-based NP from Trichodesmium is greater than 50:1, indicating that nitrate derived from the subsurface water was the main source of new nitrogen, which supported phytoplankton growth in the SCS.