Abstract: Nitrous oxide (N2O) has made a strong contribution to the climate change, particularly on the global warming potential 265 times greater than that of CO2 on a 100-year time horizon. Agricultural soil is an important source of N2O emission, accounting for approximately 33% of the global N2O. Furthermore, nitrogenous fertilizer plays a crucial role in N2O emission in agriculture, accounting for approximately 70% of the agricultural soil N2O emissions. Therefore, the reduced fertilizer can be used to alleviate the soil N2O emission, but it simultaneously can decrease the crop yield. A tradeoff between crop yield and ecological environment has become urgent in the planting link of crops. An application of deep fertilization has the potential to increase the crop yield and the use efficiency of fertilizer. However, there is still lacking the effect of reduced fertilizer on soil N2O emission under deep fertilization. In this study, a field experiment was conducted using the static chamber method and gas chromatography to determine the impacts of fertilizer treatments on soil N2O emission, soil Water-Filled Pore Space (WFPS), soil temperature, and yield of winter rapeseed, from October 2019 to May 2020. Five treatments included: 100% (DF100), 80% (DF80), and 60% (DF60) of the local recommended slow-release fertilizer (N-P2O5-K2O=187.5-52.5-60 kg/hm2) under deep fertilization, broadcast with the local recommended slow-release fertilizer (BF100), and no fertilizer (F0). The results showed that the soil N2O emission under DF100 was 13.3% greater than that of BF100, but the difference was not significant (P>0.05). The DF100 treatment significantly increased the yield of winter rapeseed by 20.1%, Partial Factor Productivity (PFP) by 20.1%, and Agronomic Efficiency (AE) by 31.9% (P<0.05). The DF100 treatment reduced the yield-scaled N2O emissions (Y-SN2O) by 5.0%, indicating that the increasing rate of rapeseed yield was greater than that of N2O emission under deep fertilization. In addition, the reduced fertilizer significantly decreased the soil N2O emission, winter rapeseed yield, and fertilizer use efficiency (P<0.05). The soil N2O emission under DF100 increased by 22.7%, 42.5%, and 153.7%, compared with the DF80, DF60, and F0 treatment, respectively. The yield of winter rapeseed under the DF100 treatment was 1.30, 2.24, and 3.24 times greater than that of DF80, DF60, and F0, respectively. The PFP under DF100 treatment increased by 3.8% and 34.5% than that of DF80 and DF60, respectively. Meanwhile, the AE under DF100 treatment increased by 19.7% and 201.3% than that of DF80 and DF60, respectively. A trend of an initial increase then decrease in the value of Y-SN2O was observed with the increase of fertilizer amount. The maximum Y-SN2O was 0.293 g/kg under the DF60 treatment, increasing by 23.1%, 48.8%, and 57.4% than that under the F0, DF80, and DF100 treatment, respectively. To balance environmental and economic conditions, the application amount of slow-release fertilizer can be reduced appropriately, according to the local recommendation, but it should be greater than 600 kg/hm2. The optimal combination on the amount of slow-release fertilizer and deep fertilization can be used to increase the yield of winter rapeseed and use efficiency of fertilizer, as well as reduce soil N2O emission. This finding can provide a sound reference for the reduction of N2O emission in winter rapeseed fields, and rational fertilization in mechanized direct-seeded rapeseed.