UC Davis Institute of Transportation Studies

One of the benefits of using mechanistic-empirical (ME) design methods for pavements is the ability to calculate pavement response to various loading and climate conditions, and then in turn to model the entire damage process that is expected to occur over the pavement lifetime. One property that is currently not accounted for within California’s ME design software (CalME) is the change in stiffness of unbound materials that may occur due to seasonal moisture patterns. The engineering properties of unbound material may change due to a variety of factors, such as fluctuations in water content, changes in suction during wetting or drying periods, changes in overburden stress, and they are also dependent on geologic setting. Before moving to develop and implement more complex relationships to model assumed changes in the properties of unbound layers due to seasonal moisture changes, the University of California Pavement Research Center (UCPRC) evaluated the extent of variation that is observed in the field. The goals of this research are to evaluate whether or not these speculated seasonal changes in unbound material properties warrant further design optimization, and if so, how research to characterize such optimization should proceed in the future.

In this research, an experiment was performed to evaluate if noticeable changes in subgrade stiffness can be identified and explained using available, pertinent, and easy-to-use pavement monitoring equipment. Testing was performed twice on sections across California, once in the wet season and once in the dry season, to get a broad picture of the types of materials present and their corresponding properties during wet and dry seasons. Monitoring of a test section at UC Davis was also performed more frequently to observe changes in stiffness occurring after rainfall events and during wetting and drying cycles.

The literature and various laboratory experiments investigating the influence of moisture and suction on the resilient response of unbound materials strongly suggest that a large degree of variability in stiffness should be encountered in different moisture conditions. However, the results of the study revealed that a majority of the unbound material tested experienced minor, if any, changes at all in stiffness between the two rounds of testing seasonal testing. The most susceptible materials to stiffness variation were not necessarily the compacted subgrade material, but were the stabilized and unstabilized granular materials directly underlying the asphalt surface. While changes in moisture content and penetration resistance were observed between the two rounds of testing, they did not necessarily correspond to significant fluctuation in the field-tested stiffness of the unbound materials; rather, other factors such as spatial variability, drainage conditions, soil type, and influences from overlying layers tended to have a much larger influence on the resilient response of these materials than did seasonal moisture change.

It is therefore recommended that CalME’s current assumption of constant stiffness for unbound layers continue to be used, except in cases where the designer identifies issues with drainage, irrigation, or other likely causes of seasonal variation of stiffness. Performing FWD testing for backcalculation of unbound layer stiffnesses after the rainy season, or at other times of highest moisture contents where rainfall is not the main source of moisture, will impart some conservatism into designs.

UC Pavement Research Center Research Report UCPRC-RR-2017-11