Economy continues to grow so that carbon emissions are getting higher and higher, which has led to global temperature changes, so it is necessary to analyze the spatial differences in its characteristics. In this study, the Kriging interpolation method was used to estimate the carbon emission in the unobserved area. By selecting suitable interpolation parameters, including semi-variance function type and radius parameter, carbon emission can be accurately estimated. At the same time, verification index analysis, spatial consistency analysis and relative error analysis are reliability of the Kriging interpolation method. The research results show that the root mean square error of the Kriging interpolation method is 0.03–0.08. The Kriging interpolation method can effectively reflect the spatial variability of carbon emissions and provide a detailed spatial distribution map. Proper selection of the semivariogram type and radius parameters can obtain a better interpolation effect.