With the definition of the freezing process under the influence of the road traffic, an early warning model is built based on the gray correlation method for monitoring the hazard from the freezing processes in the road traffic in the south area. For each process, the four indexes, including the average daily precipitation and air temperature, the overall rainy days and frozen days, contain the metric to quantify its hazard degree. With the analysis of the meteorological factors and the road-blocking disaster of the freezing process in south China during January 21-25, 2016, it is concluded that the extremely high risk area (Grade 5) and the road block area have a very good coincidence relationship, and as the level of the hazard degree rises, the ratio of the hit road block rises accordingly. The relationship is applicable for the state highways, the national highways and the provincial highways, while the peak value of the ratio is at the high hazard degree for the state highways but it is at a lower hazard degree for the provincial highways. These characteristics show that the early warning model proposed in this paper not only gives a good indication, but also discriminates the roads of different grades, to provide a sound reference for the traffic risk control.

road traffic  /  highway freezing  /  early warning