Predicting Resilient Modulus of Unsaturated Subgrade Soils Considering Effects of Water Content, Temperature, and Hydraulic Hysteresis

The resilient modulus, M-R, of subgrade soils is an important parameter in design and analysis of pavements. Subgrade soils are often unsaturated and can experience a wide range of suctions due to changes in water content and temperature induced by seasonal variation and climatic events. Laboratory- and field-measured data show that M-R is affected by stress level, water content, temperature, and hydraulic hysteresis. However, none of the existing models explicitly accounts for the effect of temperature nor can they accurately predict M-R in high suctions. In this study, a generalized model is developed that can predict M-R while incorporating water content, temperature, changes in deviatoric stress, and hydraulic hysteresis. A base model is first presented to predict the variation of M-R in regard to water content and is dictated by two distinct water retention mechanisms-capillarity and adsorption. Accordingly, a two-part model is employed to separately account for changes in M-R with water content under capillary and adsorption mechanisms. This feature allows the proposed model to accurately capture the characteristics of M-R over the entire range of suction. The base model is then extended to incorporate the effects of temperature, changes in deviatoric stress, and hydraulic hysteresis. The proposed model exhibited an excellent performance upon validation against a total of 218 experimentally measured M-R values reported in the literature spanning 14 different test sets on nine different soils tested under different conditions. The predictive errors are significantly lower than that from four alternative models, including the model in the Mechanistic-Empirical Pavement Design Guide (MEPDG). The presented model is straightforward and can be used in practice to predict the M-R of subgrade soils considering concurrent changes in water content and temperature.