Susceptibility of existing and planned Chinese railway system subjected to rainfall-induced multi-hazards

Abstract

Rainfall-induced flooding, landslides, and debris flows have been the main natural hazards that impact on the safety and operation of the Chinese railway system. With the rapid increase of railway length and passengers over the past 20 years, a better understanding of susceptibility of railway infrastructures to rainfall-induced hazards is urgently needed, in particular because extreme precipitation has increased in many parts of China. This paper provides a quantitative approach for susceptibility assessment of transportation system subjected to natural hazards at a relatively large scale. The susceptibility of both existing and planned Chinese railway system subjected to rainfall-induced multi-hazards is quantitatively analyzed using a machine learning method (Random Forest) and historical disaster events triggered between 1980 and 1998. The main environmental and climatic factors that influence the safety of the railway system are identified and ranked. Susceptibility maps illustrate the hotspot zones of both existing and planned railways in China that need significant attention and preventative action to reduce potential losses and disruption of operation.