Abstract: Taking Ningxia as the research unit, based on land use remote sensing interpretation data, nighttime light data, socio-economic statistical data and energy consumption data, this study constructed a land use carbon emission accounting model using the carbon emission coefficient method, calculated direct carbon emissions from cultivated land and other land types, indirect carbon emissions from construction land, and then obtained net carbon emissions. Mathematical statistics, GIS spatial analysis and spatial autocorrelation analysis were comprehensively used to systematically reveal the temporal evolution characteristics, spatial pattern and spatial correlation of land use carbon emissions in Ningxia. The carbon emission economic carrying coefficient(ECC) and carbon sink ecological carrying coefficient(ESC) were introduced to carry out carbon balance zoning. The results showed that from 2005 to 2025, the net carbon emissions in Ningxia continued to grow with an increase of 360.90%. Carbon sources were highly concentrated in construction land, and carbon sinks were mainly forest land and grassland, showing a spatial pattern of “high in the north and low in the south”. At the county scale, carbon emissions had a significant positive spatial correlation. The high-high agglomeration areas were concentrated in the industrial and urban dense areas along the Yellow River in the north, and the low-low agglomeration areas were stably distributed in the southern ecological conservation areas. The spatial differentiation of ECC and ESC was significant, and the results of carbon balance zoning were highly coupled with regional industrial structure and ecological background. This study divided the study area into four functional zones: lowcarbon maintenance zone, economic development zone, carbon sink development zone and comprehensive optimization zone. It can provide a scientific basis for county-level carbon emission reduction, carbon sink enhancement and low-carbon optimization of territorial space in Ningxia.