Factors controlling the spatial distribution of coseismic landslides triggered by the Mw 6.1 Ludian earthquake in China

Understanding the spatial distribution regularity of coseismic landslides is important for mitigation in future earthquakes. Unveiling the controlling factors of the distribution of coseismic landslides is a long-standing unsolved problem. On 3 August 2014, a Mw 6.1 earthquake occurred in Ludian County, Yunnan Province, China, and resulted in a great number of landslides and substantial losses. The coseismic landslide distribution in this earthquake, which was triggered by an unmapped fault, shows a different pattern. This paper conducted a spatial analysis to reveal the controlling factors in the Ludian earthquake. A spatial database of landslides has been built by interpreting the ZY-3 pre-earthquake, unmanned aerial vehicle (UAV), GeoEye-1 and GF-1 post-earthquake. A total of 1414 landslides were identified after a combination of image interpretation and field investigation in an area of 704.7 km(2). The geological setting of the study area, including the tectonics and lithology, has been mapped in detail. The digital elevation model (DEM) of the study area with a resolution of 30 m x 30 m was presented to address the topographic characteristics. A geographic information system (GIS) was used to analyse the spatial characterization of the landslides. Correlations between the occurrences of landslides and factors, i.e., slope gradient, elevation, slope aspect, intensity, distance to coseismic fault, distance to the epicentre, lithology, distance to non-coseismic faults, distance to rivers and distance to road, were characterized with a spatial analysis in GIS. It can be concluded that the slope gradient and the distance to the coseismic fault are the two most important factors for determining the spatial distribution of coseismic disasters. Some other factors play a role by influencing the slope gradient to a certain extent, and the movement of the coseismic fault is the prone direction of coseismic landslides. Slope cutting fractures the rock mass, increases the slope gradient, and significantly increases the occurrence of coseismic landslides. A statistics analysis indicates that the landslide concentration (LC) generally has an obvious exponential decrease with an increasing distance to the coseismic fault, while it has a good Weibull cumulative distribution with the slope gradient.