In flooded paddy fields, peak greenhouse gas nitrous oxide (N₂O) emission after rewetting the dry soils is widely recognised. However, the relative contribution of biotic and abiotic factors to this emission remains uncertain. In this study, we used the isotope technique (δ¹⁸O and δ¹⁵N^SP) and molecular-based microbial analysis in an anoxic incubation experiment to evaluate the contributions of bacterial, fungal, and chemical denitrification to N₂O emissions. We collected eight representative paddy soils across southern China for an incubation experiment. Results show that during the 10-day incubation period, the net N₂O emissions were mainly produced by fungal denitrification, which accounted for 58–77% in six of the eight investigated flooded paddy soils. In contrast, bacterial denitrification contributed 6–15% of the net N₂O emissions. Moreover, around 11–35% of the total N₂O emissions were derived from chemical denitrification in all soil types. Variation partitioning analysis (VPA) and principal component analysis (PCA) demonstrated that initial soil organic carbon (OC) concentrations were the primary regulator of N₂O source patterns. Soils with relatively lower OC concentration (7–15 mg g⁻¹) tend to be dominated by fungal denitrification, which accounted for the net N₂O production at the end of the incubation period. Overall, these findings highlight the dominance of the fungal denitrification pathway for N₂O production in flooded paddy soils, which predominates in soils with relatively lower OC content. This suggests that fungal contribution should be considered when optimizing agricultural management system timing to control N₂O emissions in flooded paddy soil ecosystems, and for the relevant establishment of predictive numerical models in the future.