Urban areas, which take only 2% of the world's land territory, have generated approximately 75% of global carbon emissions. Therefore, calculating carbon storage based on land use type and exploring the impact of urban land use change on carbon storage can reveal the spatio-temporal change rules of carbon storage and provide scientific basis for the national spatial planning under the dual carbon target. Based on the land use data, InVEST model was used to estimate the change of carbon storage in Beijing from 1990 to 2018, and FLUS model was used to predict the land use change under three urban development scenarios:natural evolution scenario, population distribution and urban development scenario, and green intensive ecological protection scenario. Then InVEST model was used to predict the change of carbon storage under the three scenarios in 2035. Finally, the spatial autocorrelation model was used to study the zoning management of the three scenarios, and the suggestions for the future urban development and low-carbon urban construction planning of Beijing were proposed. Based on the research, the following results were obtained:(1) from 2000 to 2010, carbon storage decreased by 4.3%. After 2010, carbon loss was relatively moderate, and improved significantly after 2015. From 2010 to 2018, the carbon storage increased by 3.5%. (2) Except for the Natural Evolution Scenario(NES), the predicted value of carbon storage under the two scenarios would further increase, and the predicted value of carbon storage under the GIEPS(Green Intensive Evolution Protection Scenario) was the highest, which was 16.39×10⁶ t, 7.5×10⁶ t more than that of NES. (3) The local spatial autocorrelation analysis showed that the carbon storage values under the three scenarios were similar in spatial distribution, and the high carbon storage values were clustered in Huairou, Yanqing, Changping and Mentougou districts in the north, northwest and west of the city.