Prediction method of soil–water characteristic curve and suction stress characteristic curve based on void ratio: a case study of Yan’an compacted loess

Soil–water characteristic curve (SWCC) and suction stress characteristic curve (SSCC), as the core of unsaturated soil mechanics, are important constitutive relationship curves for unsaturated soils, which can accurately describe the stress state in the soil. This paper examines the pore size distribution curves (PSD) of loess collected from Yan’an, China, by performing nuclear magnetic resonance (NMR) tests. In this paper, the Young–Laplace theory and the fractal model are combined with the PSD to establish the relationship between matric suction and volumetric water content. Finally, it establishes a mathematical model for predicting SWCC and SSCC. In addition, the evolution of SWCC and SSCC under different dry densities of compacted loess was studied based on the predictions. The study found that pore volumes with pore diameters smaller than the critical pore diameter could be used to predict the residual water content. Additionally, the water-holding capacity can be expressed in terms of the fractal dimension, which is controlled by the PSD. Based on the rapid characteristics of the NMR test, this paper makes a rapid prediction of SWCC and SSCC. It gives a new idea for the subsequent rapid testing of SWCC and SSCC.