Soil arching is a phenomenon describing stress redistribution due to relative movement between adjoining portions. It commonly exists when soil interacts with structure elements, such as tunnels, retaining walls, buried structures, and piles in pile-supported embankments. A yielding or ultimate state of soil (i.e., the shear stress in soil is equal to its shear strength) is mostly assumed in analytical models for soil arching. In reality, the level of shear stress and the shape of the slip surface in the soil depend on the magnitude of relative movement of the soil. In this study, a custom-made trapdoor test box was used to investigate the progressive development of soil arching with the trapdoor displacement. The dimensions of the test box was 1.76 m long, 0.46 m wide, and 1.48 m high, whereas the trapdoor had dimensions of 0.36 m wide and 0.46 m long. Earth pressures were installed on the top and sides of the trapdoor to monitor the development of soil arching with the displacement of the trapdoor. Furthermore, data from five references available in the literature were adopted to evaluate the progressive development of soil arching with the displacement for different geotechnical applications. Terzaghi’s method was used to calculate the two-dimensional soil arching ratios (i.e., the measured earth pressure divided by the overburden pressure) in all cases in this study using three different values of lateral earth-pressure coefficient (K), and the results were compared with the measured results. Terzaghi’s method calculated the soil-arching ratios reasonably well when the trapdoor displacement was equal to or larger than 10% of the trapdoor width. To facilitate the application of progressive development of soil arching with displacement in engineering practice, a simplified ground reaction curve (GRC) was proposed as three straight lines to represent the progressive development of two-dimensional soil arching with relative displacement. The methods for determining four required parameters were provided, and the proposed method was verified using three trapdoor test results from both this study and the literature.

A geosynthetic-reinforced pile-supported (GRPS) embankment that consists of embankment fill, geosynthetic, piles, and foundation soils is a complex soil–structure system. Its key load transfer mechanisms include soil arching and tensioned membrane effects and subsoil resistance. Type of embankment fill (cohesive or cohesionless) and type of pile (end-bearing or floating) are expected to affect these load transfer mechanisms; however, their influence has not been well investigated. Six scaled model tests were conducted in this study to investigate the influence of the embankment fill properties, the clear spacing of pile caps, and the pile type on soil arching and tensioned membrane effects. This study used cohesive and cohesionless embankment fills and end-bearing and floating piles. The test results show that the cohesive embankment fill strengthened the soil-arching effect, increased the pile efficacy, and reduced the settlements of the subsoil between pile caps and the embankment crest under the same load as compared with the cohesionless embankment fill. The soil arching-effect was inversely proportional to the clear spacing of pile caps. Soil arching initiated at a low ratio of the embankment height to the clear spacing of pile caps (i.e. 0.5 to 0.7) and became stable at a higher ratio (i.e. 1.1 to 1.5). The embankment height when the soil arching becomes stable is also referred to as the critical height, at which full soil arching is formed. The measured vertical earth pressures at the edges of the pile caps were higher than those in the middle of the pile caps in all six model tests. When the end-bearing or floating piles were used, the loads on the piles (i.e. the pile efficacy) increased during the construction of the embankment. However, when the floating piles started to penetrate into the underlying soil under a higher load; the pile efficacy decreased with the embankment and the surcharge load. Floating piles resulted in less soil arching and larger settlement.

A special device was designed, manufactured, and used to create a three-dimensional (3D) layout of rigid supports to investigate the spatial tensile behavior of the geosynthetic reinforcement material over rigid supports. A polyester (PET) sheet was used to simulate the geosynthetic reinforcement material. The results from the model tests using this device were used to evaluate the spatial distribution of tension and the deformed shape of the simulated geosynthetic reinforcement material. A couple of analytical solutions currently available in the literature were selected for the comparison of the calculated tension and deflections with the measured data. This study found that the maximum tensile force in the simulated geosynthetic reinforcement material occurred at the corners of rigid supports, followed by that at the edges of rigid supports. The tensile forces along the simulated geosynthetic reinforcement material were nearly uniform in the central area and increased toward the rigid supports. The test results showed that the three-dimensional deformation pattern of the simulated geosynthetic reinforcement material should not be ignored. A second-order power law formula predicted the deformed shape of the simulated geosynthetic reinforcement material satisfactorily for practical applications. The strain compatibility methods gave good prediction of the maximum tensile force and vertical deflection of the simulated geosynthetic reinforcement material between two adjacent piles. Further research is needed to improve analytical methods by considering the 3D nature of deformation and the spatial distribution of tension in the reinforcement.

High-energy impact rollers (IR) have been increasingly used for earthworks by compacting on-site soils. The IR applies high energy to the ground and densifies deeper soils than conventional rollers and plate-type compactors. A trial program was implemented on the third runway of the Shanghai Pudong International Airport in China to validate the performance of the IR technology. Field monitoring and in-situ testing were undertaken during or after the IR compaction. The cone penetration data indicated that significant improvement in soil properties was measured up to a depth of 4 m after 20 to 25 passes of compaction. However, the improvement of the properties of the very soft silty clay at the depth of 1–2 m was limited. The vibration monitoring data suggested that the IR compaction with 12-t impact module would not cause damage to buildings at a distance of 9.5 m away from the boundary of the compaction path.