Study on Early Recognition of Loess Landslides Based on Field Investigation

Authors

  • Dalei Peng State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology
  • Qiang Xu State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology
  • Xing Qi State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology
  • Xuanmei Fan State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology
  • Xiujun Dong State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology
  • Shu Li State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology
  • Yuanzhen Ju State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology

DOI:

https://doi.org/10.15273/ijge.2016.02.006

Keywords:

loess landslide, early recognition, landslide potential, rising underground water level, static liquefaction, field investigation

Abstract

For the purpose of agriculture irrigation, river water has been introduced into Heifangtai region from the Yellow River several times a year since 1968, which ensued a steady rise in underground water level by a yearly rate of 0.18 m with a cumulative total rise of up to 20 m over Heifangtai catchment. This ponding effect of water injection triggered landslides (3~5 events per year) at the edge of Heifangtai tableland. Failures of loess slopes have caused high causalities, and inhibited the local economy. This study aims at investigating the distribution patterns of loess landslides and their formation conditions in Heifangtai region, so as to establish approaches to early recognition and prediction of underlying landslides. Landslides in Heifangtai were delineated through visual interpretation of high resolution images obtained by unmanned aerial vehicle photogrammetry. Digital Elevation Model with high resolution 10 cm was acquired by 3D laser scanning technology and close-range photogrammetry. With field investigation, landslide distribution patterns and characteristics, as well as the local geological structure, were examined and two early recognition criteria for landslides are established. The results indicate that an intact slope section between two old landslide sections is the most possible locations for new landslides and areas with rising groundwater level are prone to landslides. For example, in 2015 a landslide occurred at the gaps of a 2014 landslide body. The rising of local groundwater level was induced either by blockage of underground water outlets by landslide deposits and freeze in winter or by heavy rainfall. Early recognition of loess landslides would protect local communities and land resources from landslide hazards.

Downloads

Issue

Section

Articles