The soil-water characteristic curve (SWCC) is the primary partially saturated soil information as its behavior and properties can be derived from it. Although there have been many studies of unsaturated soils and the SWCC, there is still no combined constitutive model that can simulate soil characteristics accurately. In cases when hydraulic hysteresis is dominant (e.g. under cyclic loading) it is particularly important to use the SWCC. In the past decades, several mathematical expressions have been proposed to model the curve. There are various influences on the SWCC as a source of information, so the curves obtained from conventional tests often cannot be directly applied; and the mathematical expressions from one scenario cannot be used to simulate another situation. The effects of void ratio, initial water content, stress state and high suction were studied in this work revealing that water content and stress state are more important than the other effects; but that the influences tend to decrease when suction increases. The van Genuchten model was modified to simulate better the changes in the degree of saturation at low values of suction. Predictions were compared with experimental results to determine the simulation capability of the model.

Structural deformation and damages may occur in existing tunnels because of new tunneling underneath. There is limited research on the interaction between overlying tunnels and large-diameter slurry shields in soft clay. This paper presents a case on the response of existing metro tunnels to large-diameter slurry shield underpassing in the muddy clay of Hangzhou, China. Through long-term displacement monitoring, displacement characteristics of existing tunnels during and after new tunnel construction were analyzed. Unusual tunnel heaves were observed after the passing of the shield tail because of the buoyancy effect. The existing tunnels heaved with the shape of an inverted $V$ , which was well-described with the Gaussian curve. The location of maximum tunnel heave was affected by the skew angle between the new and existing tunnels. Because the overburden pressure decreased in the under-crossing area, increased maximum ground heaves were observed. The adopted countermeasures for controlling tunnel heave were introduced and analyzed. By controlling the uplift of the new tunnels, excessive tunnel heave can be prevented. Increasing the clearance between the new and existing tunnels and optimizing the grouting parameters can diminish tunnel heave effectively. Long-term tunnel settlement caused by grout and soil consolidation developed considerably after shield underpassing. Decreased tunnel settlement and increased differential settlement could be observed during the second shield undercrossing. Detailed monitoring data and analysis provided useful experience for similar projects.