Abstract: The one-gravitational (1-g) shaking table model test is a frequently-used method to study the seismic responses of underground structures. However, past studies show that the test results were distinct. In some tests, no damage of the underground structures was observed under extremely strong earthquake motions. There were problems when one tried to observe the damage pattern of model structures. The seismic responses of underground structures can be influenced by many factors including the buried depth of structures. In the study, we investigate how the buried depths of a structure influences the test results using the 1-g shaking table tests of free-field sand and the Integral Response Displacement Method. The seismic responses of the free-field soil in the depth direction were acquired through the tests, including the distributions of the soil deformation, shear strain, shear stress and acceleration. The seismic responses of a structure at different buried depths were obtained using the Integral Response Displacement Method. The results show that the variation of the buried depth would change the soil deformation, shear force and inertia force loaded on the structure. The change was not dependent on the scaling ratio of the buried depth, but was mainly determined by the soil properties and seismic motions. Therefore, when conducting the experimental design, the buried depth of the model structure should be carefully determined with respect to the seismic responses of the site. In the 1-g shaking table test, the deformation of the soil played the most critical role for the seismic responses of the model structure. The effect of the deformation of the soil accounted for 70% to 80% of the overall seismic responses of the structure. The model structure would have the greatest seismic responses when it was buried at the depth with the largest relative soil deformation.