Development of Asphalt Pavement Temperature Prediction Models for Sub-Saharan Tropical Climate: The Case of Ghana

The use of models for predicting asphalt pavement temperature tends to have limited applicability in regions with environmental conditions significantly different from those under which the models were developed. This study developed two asphalt pavement temperature prediction models for the West African tropical country of Ghana. The study locations were Kumasi and Tamale, in the Forest and Savannah climatic zones, respectively. Mid-depth asphalt layer and surface temperature data were measured in both cities for 1 year (May 2022–April 2023). Two non-linear regression models were calibrated using data collected on two newly rehabilitated asphalt roads and validated using an independent dataset collected on two different roads in the same cities. The validated models predicted asphalt pavement temperatures with a high level of accuracy, as indicated by the low model errors (root mean square error ranged from 1.924 to 2.679 °C and mean percentage error from 0.037 to 0.295%) and high adjusted coefficient of determination values, which ranged from 0.919 to 0.920. Future research efforts in improving the proposed models should include data collected at additional locations and over a longer duration.