Abstract: To reveal the spatiotemporal evolution patterns of ecosystem services in the cold-temperate primeval forest region after an extraordinarily severe lightning-induced fire, this study focused on the fire-affected area in the primeval forest region of northern Daxing'an Mountains in Inner Mongolia, which was hit by the July 28, 2002 fire. Based on multi-source remote sensing imagery from 2001 to 2020, methods such as fire severity classification, land use transition analysis, and the InVEST and CASA models were employed to systematically evaluate the post-fire ecosystem response and recovery processes. The results indicated that the total fire-affected area was 182.5 km², which was dominated by medium-high and high fire severity levels, and spatially exhibited a concentric distribution pattern around high-severity cores. Land use changes showed a clear two-phase pattern of initial degradation followed by recovery: forest continuously degraded into grassland from 2001 to 2015, with slow recovery beginning after 2015. The depth of forest degradation was positively correlated with fire severity, while recovery potential decreased significantly with increasing fire severity. Although all types of ecosystem services underwent a three-phase evolution of impact, recovery, and stabilization, their recovery trajectories and endpoints differed markedly. Water conservation function was the most sensitive to fire, declining by over 56% after the event, while habitat quality exhibited the strongest resilience and the most stable recovery. By 2020, water conservation and soil retention functions had recovered to levels exceeding pre-fire baselines, and Normalized Difference Vegetation Index (NDVI) had recovered to over 96% of the baseline. In contrast, Net Primary Production (NPP) remained the only indicator not restored to pre-fire conditions, with a recovery rate below 90%. This study confirms that fire severity is the key driver governing the post-fire ecological recovery process and pattern. Areas with low and medium-low severity maintain better ecological resilience and can recover effectively through natural processes. In contrast, areas with high and medium-high severity, due to severe damage to ecosystem structure, face difficulties in natural recovery and require targeted artificial assistance for restoration. This study clarifies the response relationship between extraordinarily severe lightning-induced fire disturbance and ecosystem service restoration in cold-temperate primeval forest regions. The research results can provide a scientific basis and data support for hierarchical fire management, formulation of post-fire ecological restoration schemes, and sustainable improvement of ecosystem services in fire-prone forest areas of cold-temperate zones in China, as well as practical guidance for hierarchical precision restoration, synergistic enhancement of ecological services, and the construction of regional ecological security barriers in cold-temperate primeval forest areas.