The paper selects 30 provincial units in China as the research objects, and the input-output method is used to calculate the indirect carbon emissions from food consumption. The results showed that: (1) The indirect carbon emissions reached 1.522 billion tons, of which 19.81% came from power and heat production and supply industry, 17.19% from agriculture, forestry, animal husbandry and fishery industry, and 15.14% from food processing industry; (2) The spatial differentiation of indirect carbon emissions from food consumption was significant, showing a "center-periphery" distribution characteristic. (3) The global and local correlation characteristics are positive. Eight provinces including Henan, Hubei, Hebei, Tianjin and Shandong belonged to the "high-high" cluster type, Qinghai, Gansu, Ningxia, Guangdong and Guangxi belonged to the "low-low" cluster type, while Liaoning, Jilin and Xinjiang were classified as the "high-low" cluster type; (4) Urbanization level was the main factors affecting the indirect carbon emissions of food consumption, followed by culture level of labor force, agricultural capital investment and agricultural production efficiency. Indirect carbon emissions could be reduced by decreasing the energy consumption level of the food industry chain, promoting the green transformation of agricultural development, and guiding residents to develop reasonable dietary habits.