Seismic and geological controls on earthquake-induced landslide size
地震诱发滑坡规模的地震与地质控制
Earth and Planetary Science Letters, Volume 506, 15 January 2019, Pages 268-281
A. Valagussa, O. Marc, P. Frattini, G. B. Crosta
摘 要：Landslide size controls the destructive power of landslides and is related to the frequency of occurrence, with larger landslides being less frequent than smaller ones. For this reason, the analysis of landslide size is essential for landslide hazard assessment. We analyse six earthquake-induced landslide inventories with earthquake magnitude ranging between 6.6 and 7.9 Mw (Papua New Guinea, 1993; ChiChi 1999; Northridge, 1994; Niigata–Chuetsu, 2004; Iwate–Miyagi Nairiku, 2008; Wenchuan, 2008). For each inventory, we developed magnitude–frequency curves to analyse the size distribution of landslides as a function of ground motion, distance from the seismic source (both fault trace and epicentre), local relief, and lithology. For three earthquakes, we observed a clear relationship between the landslide size and ground motion, with larger landslides associated with higher ground motion. We investigate different possible causes for such observation, and propose that the main mechanical reason is that stronger shaking induces higher stresses that may overcome the strength, which increases with depth, triggering larger landslides. We also show that landslide size decreases with distance from the fault trace, whereas, this trend is not clear for distance from the epicentre. Local relief does not seem a first order control on landslide size for the earthquake-induced landslides considered here. Some lithologies do influence landslide size, but we were unable to identify a general behaviour for different lithologies.
Distribution and genetic types of loess landslides in China
中国黄土滑坡分布与发育类型
Journal of Asian Earth Sciences, Volume 170, February 2019, Pages 329-350
Jianbing Peng, Shaokai Wang, Qiyao Wang, Jianqi Zhuang, Penghui Ma
摘 要：In this study, 14,544 loess landslides collected from the Loess Plateau of China are used to reveal their distribution and origins. Main factors considered include landslide density, regional tectonics, soil properties, geomorphic structure, rainfall distribution, seismic activity, and human activity. Based on these factors, the Loess Plateau is divided into eight zones with high possibilities of landslides that are classified into six genetic types. The first genetic type of clustering distribution of loess landslides is related to several main regional active faults. The second type is related to broken pieces of loess geomorphic structures. The third type is considered to be due to unloading of the geomorphic peripheral slope that can cause the internal structural plane to open and thus controls the prototype and scale of the loess landslide. Change of hydrogeological conditions in the slope softens the loess pedestal and the slippery layer, forming a softening zone and thus controlling the shear crack and slip model of the fourth type of loess landslides. The fifth type is due to seismic activities that can generate additional stress to the Earth's surface and landform slopes, causing the opening of joints and seismic liquefaction. The last type of landslides is caused by human activities by changing the geological structure, hydrogeological conditions, and the inner stress balance of slopes.
New method for landslide susceptibility mapping supported by spatial logistic regression and GeoDetector: A case study of Duwen Highway Basin, Sichuan Province, China
基于空间逻辑回归与GeoDetector的滑坡易发性制图新方法：中国四川都汶公路盆地实例研究
Geomorphology, Volume 324, 1 January 2019, Pages 62-71
Jintao Yang, Chao Song, Yang Yang, Chengdong Xu, Lei Xie
摘 要：Landslides are destructive not only to property and infrastructure but also to people living in landslide-prone regions. Landslide susceptibility mapping (LSM) is critical for preventing and mitigating the negative impacts of landslides. However, many previously proposed LSM modeling techniques included only the attribute information of spatial objects and ignored the spatial structural information of spatial objects, which led to suboptimal LSM. In addition, the selection of condition factors was not objective to such an extent that it may have reduced the reliability of LSM. To address these problems, a new method based on GeoDetector and a spatial logistic regression (SLR) model is proposed. GeoDetector is used to select condition factors based on the spatial distribution of landslides. The SLR model is used to make full use of the structural and attribute information of spatial objects simultaneously in LSM. The GeoDetector-SLR model is validated using data from the Duwen Highway Basin, which includes the epicenter of the May 12, 2008 Wenchuan earthquake in southwestern China. Prediction accuracy of the GeoDetector-SLR model is found to be 86.1%, which is an 11.9% improvement over the traditional logistic regression model, indicating an improved and reliable solution for evaluating landslide susceptibility.
Distribution and morphology of landslides in northern Finland: An analysis of postglacial seismic activity
芬兰北部滑坡分布与形态学：后冰川地震活动分析
Geomorphology, Volume 326, 1 February 2019, Pages 190-201
Antti E. K. Ojala, Jussi Mattila, Mira Markovaara-Koivisto, Timo Ruskeeniemi, Raimo Sutinen
摘 要：The ages and sizes of landslides occurring in seismically active areas can be used to reconstruct the seismic history of the area and estimate the maximum moment magnitudes of past earthquakes. Here, we present a data set of 121 landslides discovered in northern Finland that were analyzed for their morphometric characteristics. We show that 89 debris slide type landslides in the data set are clustered close to known postglacial faults (PGFs) and thus provide information on the characteristics of postglacial paleoseismic events. By using empirical correlations between the landslide volume–area data and earthquake moment magnitude, we estimate maximum moment magnitudes Mw ≈ 6.9–7.7 for postglacial earthquakes in the Suasselkä, Isovaara–Riikonkumpu, Venejärvi, and Vaalajärvi areas, where earlier estimates based on fault length and displacement have yielded magnitudes varying between Mw ≈ 6.5 and 7.5. We also show that the landslides in northern Finland are located within a radius of 35 km from the closest known PGF and that sizes of the landslides decrease as a function of distance from PGFs, hence providing strong empirical evidence for their seismic origin. As far as we are aware, this is the first use of landside data in quantifying postglacial seismicity within the Fennoscandian Shield area.
Distribution of landslides caused by heavy rainfall events and an earthquake in northern Aso Volcano, Japan from 1955 to 2016
1955至2016期间日本阿苏火山北部暴雨和地震引发的滑坡分布
Geomorphology, Volume 327, 15 February 2019, Pages 533-541
Atsuhisa Yano, Yoshinori Shinohara, Haruka Tsunetaka, Hideaki Mizuno, Tetsuya Kubota
摘 要：A new landslide cannot form until the soil has recovered to the critical depth for the recurrence of a landslide through the weathering of bedrock and soil transportation from adjacent areas. In volcanic areas with tephra deposits, landslides expose tephra and not bedrock. Therefore, the immunity of landslides in volcanic areas may be different from that of landslides in non-volcanic areas. Herein, we developed landslide inventory maps (LIMs) for 6 periods during 1955–2016 using aerial photographs and digital elevation models in northern Aso Volcano. In this area, landslides were found to continuously occur due to rainfall events and a large earthquake. Using the 6 LIMs, we examined the terrain attributes (i.e., slope angle, slope aspect, and normalized distance to ridge) and the overlap of landslides. Among the terrain attributes, slope angle was a dominant factor affecting the occurrence of landslides caused by both rainfall events and an earthquake. The total landslide areal density in 2016 was 50% for a slope angle of 35%–45%. 2 atypical events (a rainfall in July 2012 and an earthquake in April 2016) caused landslides to occur on slopes that were relatively resistant to landslides by typical amounts of rainfall, resulting in high landslide density in 2016. The intensity of rainfall for an event in July 2012 was considerably higher than that for other rainfall events. The type of landslides caused by an earthquake in April 2016 was different from that of landslides caused by rainfall. The depths of some landslides caused by this earthquake were deeper than those of landslides caused by the rainfall in July 2012. The overlap ratio was <2.3% for all combinations of the 6 LIMs. The small overlap ratio in the study area suggests that the immunity of landslides continued during the 60-year period we examined. Further research clarifying the process leading to subsequent landslides would be useful to better understand the immunity of landslide and the associated landslide susceptibility in volcanic areas.
