Identifying the spatiotemporal characteristics of individual red bed landslides: a case study in Western Yunnan, China

Strata in red bed areas have typical characteristics of soft-hard interbedding and high sensitivity to water. Under the comprehensive action of internal stratigraphic structure and external hydrological factors, red bed landslides have highly complex spatiotemporal characteristics, presenting significant challenges to the prevention and control of landslide disasters in red bed areas, especially for slope and tunnel engineering projects. In this study, we applied an interdisciplinary approach combining small baseline subset interferometric synthetic aperture radar(SBAS-InSAR), deep displacement monitoring, and engineering geological surveying to identify the deformation mechanisms and spatiotemporal characteristics of the Abi landslide, an individual landslide that occurred in the red bed area of Western Yunnan, China. Surface deformation time series indicated that a basic deformation range developed by March 2020. Based on In SAR results and engineering geological analysis, the landslide surface could be divided into three zones: an upper sliding zone(US), a lower uplifted zone(LU), and a toe zone(Toe). LU was affected by the structure of the sliding bed with variable inclination. Using deep displacement curves combined with the geological profile, a set of sliding surfaces were identified between different lithology. The groundwater level standardization index(GLSI) and deformation normalization index(DNI) showed different quadratic relationships between US and LU. Verification using the Pearson correlation analysis shows that the correlation coefficients between model calculated results and measured data are 0.7933 and 0.7577, respectively, indicating that the DNI-GLSI models are applicable. A fast and short-lived deformation sub stage(ID-Fast) in the initial deformation stage was observed, and ID-Fast was driven by concentrated rainfall.

Predicting the severity of traffic accidents on mountain freeways with dynamic traffic and weather data

Traffic accident severity prediction is essential for dynamic traffic safety management.To explore the factors influencing the severity of traffic accidents on mountain freeways and to predict the severity of traffic accidents,four models based on machine learning algorithms are constructed using support vector machine(SVM),decision tree classifier(DTC),Ada_SVM and Ada_DTC.In addition,random forest(RF)is used to calculate the importance degree of variables and the accident severity influences with high importance levels form the RF dataset.The results show that rainfall intensity,collision type,number of vehicles involved in the accident and toad section type are important variables influencing accident severity.The RF feature selection method improves the classification performance of four machine leaming algorithms,resulting in a 9.3%,5.5%,7.2% and 3.6% improvement in prediction accuracy for SVM,DTC,Ada_SVM and Ada_DTC,respectively.The combination of the Ada_SVM integrated algorithm and RF feature selection method has the best prediction performance,and it achieves 78.9% and 88.4% prediction precision and accuracy,respectively.