[Objective] The land use change pattern and its carbon effect in the main grain producing areas were investigated in order to provide a basis for restructuring land use and for low carbon economic development.[Methods] We used multi-period land use status data supported by the grid sampling method, a land use dynamic attitude model, a carbon emission coefficient method, and a spatial autocorrelation analysis model to determine the spatial heterogeneity characteristics of land use change patterns and their carbon effect in the Dongting Lake basin since 1980.[Results] ① Land use in the Dongting Lake basin changed in stages, with a comprehensive dynamic attitude of 0.02% from 1980 to 2000. During this time, the largest dynamic attitude was construction land, followed by unused land. The comprehensive dynamic attitude increased to 0.18% from 2000 to 2020, with accelerated growth in the area of construction land. ② The net carbon effect was manifested as a carbon sink, from 5.93×10⁷ t in 1980 to 2.82×10⁷ t in 2020, while the carbon effect caused by land use change showed that the change in carbon emissions was greater than the change in carbon sinks, leading to an increase in net carbon emissions of 6.08×10⁵ t. High value net carbon emissions were distributed in an "H" shape, and the low value emissions area expanded gradually. ③ The spatial autocorrelation of net carbon emissions in the Dongting Lake catchment area was significant. The main aggregation types were high-high and low-low from 1980 to 2000. The low-high type was scattered, and the high-high type was more concentrated and contiguous from 2000 to 2020, while the low-low type was mainly found in the northern part of the Xiangjiang River basin.[Conclusion] In Dongting Lake basin, people should maintain the orientation of "increasing carbon sinks and reducing carbon emissions", maintain the stable state of carbon sinks in forest land, scientifically guide the development and use of land with high carbon emissions, and give attention to the "assimilation" of carbon emissions in different catchment areas according to the characteristics of spatial autocorrelation.