Development of structural type-based physical vulnerability curves to debris flow using numerical analysis and regression model

Abstract

Urbanization has significantly increased the risk of debris flows to houses located at mountain boundaries, particularly in South Korea, where urban centers are surrounded by mountains. The degree of damage to a house strongly depends on the structural rigidity of the building, even when exposed to the same hazard intensities. However, the structural characteristics of buildings are usually disregarded in vulnerability assessments. In this study, vulnerability curves were proposed for three different structural types of buildings: brick masonry, light-weight steel frame, and wood structures, using numerical and regression analyses. A total of 22 debris flow events from 2011 to 2020 and 63 cases of damage to unreinforced structures were collected and analyzed. Back analysis of the debris flows was performed using Flow-R, and the impact pressure acting on each structure was calculated. Three different vulnerability curves were proposed using various linear and nonlinear regression analyses; the Avrami equation provided the highest statistical significance. The vulnerability curves of three different structures were similar for the slight damage range. However, over the slight damage range, the wood structure was the most vulnerable at a given hazard intensity, whereas the brick masonry structure was the least vulnerable. Validation was performed using data from recent debris flow events, revealing precise damage level predictions for 9 out of 11 buildings. Compared with previous research, the degree of damage at the same hazard intensity was found to be significantly dependent on the structural type. The developed vulnerability curve for each unreinforced structure can be used as key data for vulnerability and risk assessments.