Abstract: Applying statistics in geo-ecological process researches is an important manifestation of the combination of human activities and the geo-ecological process. Spatialization of agricultural statistics is a means to restore and reconstruct the spatial features of agricultural statistics. The resulting raster maps can provide data support for a wide range of geo-ecological studies. In support of land use, a regression model was constructed by taking population density as the dependent variable and grain yield per unit of arable land as independent one. The model was then applied to spatially distribute provincial-level grain output statistics, resulting in a 1 km×1 km grid for China’s grain output in 2000. The validation of the resulting maps by using grain output statistics at province-, municipal- and county- administrative units showed that the method can correctly map the spatial distribution of grain output, especially for the major grain producing areas, with a high accuracy of the data reproduction. The results can be used as a data source for geo-ecological studies in agricultural areas. The accuracy of gridded grain output was observed to decline with the downscaling of the administrative region. The paper concluded that a higher accuracy grid would be generated if more ancillary factors associated with grain outputs were incorporated.