Increasing atmospheric nitrogen (N) deposition has profound effects on carbon (C) cycling in forest ecosystems. As an important part of belowground C dynamics, soil respiration is potentially affected by changing N availability. However, the responses of total soil respiration (RS_T) and its three components, soil respiration derived from plant roots (RS_R), root-free soil (RS_S) and the litter layer (RS_L), to such N enrichment remains poorly understood. To assess the effects of N enrichment on soil respiration components, three levels of N addition, namely low (LN, 50 kg N ha⁻¹ year⁻¹), medium (MN, 100 kg N ha⁻¹ year⁻¹) and high (HN, 150 kg N ha⁻¹ year⁻¹), were conducted over five growing seasons from 2011 to 2015 in a temperate Chinese pine (Pinus tabulaeformis) forest in northern China. A control plot without N addition (CK) was also established. The five-year mean annual rate of RS_T was 2.18 ± 0.43 μmol m⁻² s⁻¹, and the contributions of RS_R, RS_S and RS_L were 8.8 ± 3.1%, 82.2 ± 4.5% and 9.0 ± 5.5%, respectively. Compared with CK, RS_T was significantly increased by 16.5% in the HN plots, but not in the LN or MN treatments. RS_S was significantly decreased by 18.1%, 26.6% and 18.4% in the LN, MN and HN plots, respectively, due to the reduction of both microbial biomass carbon (MBC) and enzyme activity. In contrast, RS_R was increased by more than twice under the MN treatment, which promoted root growth and activity (higher fine root biomass and N concentration). A significant elevation in RS_L was only detected in the HN plots, where the increased litter input enhanced litter decomposition and hence RS_L. Our findings clearly demonstrated that N addition of different intensities had different effects on soil components. In particular, the above- and belowground components of heterotrophic respiration, RS_L and RS_R, showed contrasting responses to high level addition of N. Thus, we highlight that the response of soil respiration components to N addition should be examined individually. Our results may contribute to a better understanding of soil respiration dynamics under future N scenarios, and have important implications in forest management.