Abstract
In terms of the dynamic dependence between icing-inducing factors, this study is to explore the risk distribution of highways when icing events occur in the study area. A joint distribution considering the dynamic correlation of inducing factors was first constructed employing the Copula theory, which then yielded the possibility of icing events. Meanwhile, hazard zones and intensities of icing were proposed under different exceeding probabilities. After finishing the vulnerability analysis of highways, the risk matrix was used to conduct the icing risk for the highway, which was then applied to the construction of the risk zoning map. The results showed that there was an upper-tail dependence between extreme precipitation and temperature in the study area in winter, which could be well captured by the Gumbel Copula function. Indeed, the constructed joint distribution can express the possibility of icing under different intensities of precipitation and temperature. Besides, the highway with the tallest vulnerability in the study area was the Hegang-Yichun line. The case application showed that during March 2020, the traffic lines with a high icing risk were distributed around Fujin, Jiamusi, Hegang, and Qitaihe cities, and the Hegang section of the Hegang-Yichun line was at the highest icing risk. The low-risk lines were concentrated in the western part of the study area. This study is of great significance for the prevention and control of ice-snow disasters on the highway in cold regions.
Dynamic Correlation or Tail Dependence Hedging for Portfolio Selection
