Boreal forest is an important component in the global carbon balance and has been a focus of study for a long time. In China, about 30% of forested areas are boreal forests, which play a key role in the country's carbon budget. Fire is a dominate forest landscape process in the boreal forests of northeastern China. Because of the stochastic nature of fire and forest succession, reliable prediction of aboveground forest biomass for boreal forests is challenging. Thus, predicting the dynamics of boreal forest biomass requires accounting for fire's effect. The effect of fire on the dynamic of forest aboveground biomass is a long-term process that occurs at various spatial and temporal scales. It would be difficult to capture the fire process with traditional field experiment research. In order to better understand the ecological processes related to fire, a spatially explicit forest landscape model based on our prior knowledge of biology, ecology, and computer science became a valuable tool for studying the forest structure and biomass prediction, at various spatial and temporal scales. Therefore, model simulation can help us to better understand the complex interactive effects of forest landscape processes and vegetation on forest biomass. In this study, we used a forest landscape model (LANDIS PRO) to investigate the effect of fire on landscape-level predictions of the tree component of biomass in a boreal forest landscape in the Great Xing'an Mountains. We first selected five major forest types (larch, Larix gmelinii; pine, Pinus sylvestris var. mongolica; spruce, Picea koraiensis; birch, Betula platyphylla; and aspen, Populus davidiana) in our study area, and treated the succession-only scenario as the reference scenario. We then calibrated and validated the simulated results of the LANDIS PRO model. We predicted the tree biomass over three time intervals (0-50 years, 50-150 years, and 150-300 years), and quantified the effect of fire on predictions of total biomass and spatial distribution over short-, mid-, and long-term intervals. The simulation results showed that the initialized forest landscape constructed from the forest inventory data from the year 2000 adequately represented the forest composition and structure of that year. The simulated density and basal area of the year 2010 adequately represented the forest inventory data of that year at the landscape scale. Compared to the succession-only scenario, the predicted biomass decreased by 1.7-5.9 t/hm² in fire-only scenarios across all simulation periods. Compared to the succession-only scenario, the effect of fire on aboveground biomass differed significantly among the three intervals (short-, medium-, and long-term) (P < 0.05). Under the succession-only and fire scenarios, the spatial distribution of biomass differed significantly (P < 0.05) among simulation periods. The evidence from our study indicates that fire strongly influences the spatial distribution of forest biomass and that the fire scenario reduced more biomass in subalpine land types than in others. These results have significant implications for forest managers interested in designing management systems for long-term forest sustainability.