In order to solve the problem of traffic meteorological service in areas with a few observation stations, three regions in northern China (Beijing-Tianjin, Shaanxi-Gansu, Qinghai three typical regions) were selected. Traffic meteorological observational data, combined with the surface data from China Meteorological Administration Land Data Assimilation System (CLDAS), China Meteorological Administration Multisource Precipitation Analysis System (CMPAS), and surface short wave and long wave radiation retrieval products from FY-4A, were used to analyze the characteristics of road surface temperature and the relationship with the environmental meteorological factors in different regions, different seasons and different times. Three methods (linear regression, random forest, and deep neural network) were employed to construct a 1-hour updated road surface temperature real-time model. The model effectiveness was tested, and the effects of different methods, data combinations, and spatial generalization ability of the model were explored. Results show that: (1) The road surface temperature is significantly correlated with the environmental meteorological factors, but different regions, seasons, and time periods exhibit different detailed characteristics; (2) The independent test shows that there is no much difference among the model results based on different methods. They can all well reproduce the daily changes of road surface temperature at 14:00 (BJT) in summer and road surface temperature at 02:00 (BJT) in winter. The error in winter is significantly lower than that in summer. The application of satellite radiation products has a significant improvement effect on the road surface model results in summer. (3) The model has good spatial adaptability. However, compared with the model constructed by using local observations, the error of the model using data from nearby traffic stations in the same climate region shows increasing with different degrees, with the smallest increase in error in the Beijing-Tianjin region.