Stability analysis of a loess landslide considering rainfall patterns and spatial variability of soil

• Published in: Computers and geotechnics Vol. 167; p. 106059
• Main Authors: Ma, Jianhua, Yao, Yunqi, Wei, Ziran, Meng, Xingmin, Zhang, Zonglin, Yin, Hailong, Zeng, Runqiang
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.03.2024
• Subjects: Preferential flow; Probability of instability; Random rainfall patterns; Slope stability; Soil variability; Random rainfall patterns; Probability of instability; Slope stability; Soil variability; Preferential flow
• ISSN: 0266-352X; 1873-7633
• DOI: 10.1016/j.compgeo.2023.106059

Accurate analysis of the stability of rainfall-induced landslides is an important and urgent task. Under complex geological conditions, traditional stability analysis methods cannot meet stability evaluation requirements. To study the impact of the rainfall pattern on the probability of instability of loess slopes, the intensity-frequency-duration curve and breakdown coefficient methods were introduced to analyse the effects of the rainfall extremum distribution and time scale, and bounded random cascade simulation were used to generate random rainfall patterns (RRPs). The obtained RRPs under several classical rainfall patterns were compared in terms of the safety factor, rainfall threshold (IaS/IaF), and annual failure probability (AFP) of loess slopes. Finally, soil preferential flow and spatial variability combined with the RRPs were introduced to analyse the change in the stability of rainfall-induced loess landslides under different theoretical frameworks. It was revealed that the use of a uniform rainfall pattern produced conservative stability evaluation results. Considering the RRPs, the calculation of IaS/IaF and AFP should include the common characteristics of various rainfall patterns. By comprehensively evaluating slope stability considering soil spatial variability and rainfall pattern variability, this study revealed that neglecting certain factors may result in incorrect slope stability estimates. Incorporating multiple assumptions in failure probability evaluation could provide more comprehensive slope safety assessment.