Influence of human activity on landslide susceptibility development in the Three Gorges area

  • PDF / 4,757,973 Bytes
  • 37 Pages / 439.37 x 666.142 pts Page_size
  • 108 Downloads / 242 Views

DOWNLOAD

REPORT


Influence of human activity on landslide susceptibility development in the Three Gorges area Yongwei Li1,2 · Xianmin Wang1   · Hang Mao1 Received: 9 April 2020 / Accepted: 24 August 2020 © Springer Nature B.V. 2020

Abstract Human activities are important factors that trigger frequent occurrences of landslides; thus, for landslide control, it is critical to determine the influence of human activity on landslide occurrence probability. The Three Gorges area is a region in the world that typically experiences serious landslide disasters and frequent human activities. The objective of this work is to employ the Three Gorges area as an example to reveal the impact of human activity on the dynamic development of landslide susceptibility from 2010 to 2019. Some new viewpoints are suggested for the five aspects: (1) High-precision landslide susceptibility maps are generated by a combination of multiresolution segmentation and convolutional neural network algorithms. Moreover, the dynamic development rules of landslide susceptibility from 2010 to 2019 are revealed. (2) The change in landslide susceptibility in the study area from 2010 to 2019 was mainly caused by the combined action of rainfall and human activity. The fluctuation of reservoir water level had a less influence on the development of landslide susceptibility. (3) Some human activities, especially road construction, farmland appropriation for building construction, agricultural reclamation, farmland cultivation and irrigation, initiation of commercial planting, urban expansion, and large-scale deforestation, may dramatically increase landslide occurrence probability. (4) Human activities, e.g., conversion of farmland to forestry, artificial recovery of natural vegetation, and later periods of artificial planting, may obviously reduce landslide susceptibility. (5) The human activity causes and mechanisms influencing landslide susceptibility in the study area are proposed, including transpiration and anchorage of plants, slope reinforcement by plant roots, destruction of slope stress equilibrium, and soil erosion. Keywords  Landslide susceptibility · Convolutional neural network · Human activity · Multiresolution segmentation

* Xianmin Wang [email protected] 1

Hubei Subsurface Multi‑Scale Imaging Key Laboratory, Institute of Geophysics and Geomatics, China University of Geosciences, Wuhan 430074, China

2

School of Civil Engineering, Central South University, 68 Shaoshan Road, Changsha 410075, Hunan, China



13

Vol.:(0123456789)



Natural Hazards

1 Introduction Landslide susceptibility evaluation (LSE) is a key technique for the prevention and control of landslide disasters (Eeckhaut et al. 2006; Zhou et al. 2018). Human activity is an important triggering factor that results in frequent occurrences of landslides and is closely related to landslide susceptibility  (Froude and Petley 2018). With the emergence of extreme weather events and intensification of human activities, landslide susceptibility dynamically changes with time (Nefeslioglu and Gorum 2

Data Loading...