Correlation Analysis between Temperature Indices and Flexible Pavement Distress Predictions Using Mechanistic-Empirical Design

Document Type


Publication Date



Department of Civil, Environmental, and Geospatial Engineering


Temperature is the most influential climatic variable for flexible pavement design using a mechanistic-empirical approach. While the effect of climatic variables on pavement designs has been previously investigated, few studies went further to seek potential temperature indices to correlate with distress predictions of flexible pavement. This study aims to explore the correlations between distress predictions and various potential temperature indices. The distresses investigated include total rutting, asphalt concrete rutting, thermal cracking, top-down cracking, bottom-up cracking, and the international roughness index (IRI). The temperature types in this study primarily include the pavement surface temperature and the air temperature. The temperature indices investigated include the annual average temperature, average temperatures in hot months and cold months, and standard deviation of annual temperature. Pavement distresses were obtained using the climatic information collected from the 39 weather stations throughout the state of Michigan. The pavement surface temperature was obtained by the enhanced integrated climate model (EICM). The correlation coefficient (R-values) between each distress prediction and each temperature index was obtained to determine the strength of correlation. Several valuable findings were obtained, including that the surface temperature in hot months exhibited very strong positive correlations with asphalt concrete (AC) rutting and total rutting. Annual average temperature showed correlations with thermal cracking as strong as the average temperature in cold months. The average annual temperature showed moderate correlations with bottom-up cracking but very weak correlations with top-down cracking. In addition, the optimum combination of the high temperature index and the low temperature index can have a strong correlation with predicted IRI. The paper also shows an example of climate zone creation based on the temperature indices, which were selected from the correlation analysis. The findings in this study can help further understand the effects of temperature indices on pavement designs. They are also beneficial for the parameter selection for climate zone creation, especially for pavement designs in the future.

Publication Title

Journal of Cold Regions Engineering