The seasonal monitoring program (SMP) of the long term pavement performance (LTPP) program provides an excellent opportunity to study the seasonal trends in pavement response. In 64 pavement sections in the United States and Canada, FWD testing is performed monthly every other year, subsurface temperature is measured hourly, subsurface moisture and freezing conditions are measured monthly, and climatic factors such as ambient air temperature, solar radiation, wind speed, and precipitation are collected continuously.

In this study, a number of SMP sections were used to establish some general and statistically significant trends. The relationship between the elastic modulus of the AC layer and the temperature gradient within the layer was investigated. The relationship between the frost penetration in unbound layers and their backcalculated modulus was studied. Modeling such variability using linear and nonlinear formulations was attempted, and the practical implications of such effects on the data collection practice are discussed.

The results of the analysis showed that the AC layer temperature gradient appears to influence the layer modulus. An exponential model was developed to quantify that effect and adjust the backcalculated modulus to the temperature gradient. Analysis of frost penetration trends showed that surface deflection and the backcalculated subgrade modulus are very sensitive to the existence and extent of a frozen layer within the subgrade. As expected, the thicker the frozen layer, the larger the subgrade modulus, and the further the layer, the lower the subgrade modulus. Exponential models were used to characterize this effect.