Abstract: In order to effectively reveal the spatial distribution characteristics of soil heavy metals and provide scientific basis for environmental quality assessment in Beijing cultivated soils, geo-statistics was used to explore the scale effect of spatial structure of heavy metals in Beijing cultivated soils. The results showed that spatial structure of Cr, Ni, Zn and Hg had distinct dependence on spatial scale. According to the optimality test of theoretical semi-variance function, the spatial structure of Cr, Ni, Zn and Hg could be described fully by multi-scale nested model. In addition, the spatial interpolation accuracy of Cr, Ni, Zn and Hg using the multi-scale nested model was higher than that using the ordinary Kriging at a single scale. Therefore, the multi-scale nested model can estimate the spatial distribution characteristics of soil heavy metals more effectively than the ordinary Kriging method.