Acta Geographica Sinica ›› 2021, Vol. 76 ›› Issue (1): 73-86.doi: 10.11821/dlxb202101006

• Land Surface Pattern • Previous Articles     Next Articles

Risk assessment of sandstorm diffusion and landscape pattern optimization in southern Xinjiang

SHI Nana(), HAN Yu, WANG Qi, HAN Ruiying, GAO Xiaoqi, ZHAO Zhiping, LIU Gaohui, XIAO Nengwen()   

  1. Chinese Research Academy of Environmental Sciences, Beijing 100012, China
  • Received:2019-12-15 Revised:2020-10-27 Online:2021-01-25 Published:2021-03-25
  • Contact: XIAO Nengwen;
  • Supported by:
    The Biodiversity Survey and Assessment Project of the Ministry of Ecology and Environment (2020)


Based on the three dimensions of "hazard-formative factors, hazard-formative environments, hazard-affected bodies", we assessed the risk of sandstorm diffusion in the southern Xinjiang using the Spatial Principal Component Analysis (SPCA) method. A multi-level ecological network was built and components of key landscape patterns were optimized by the Minimum Cumulative Resistance (MCR) model. (1) The risk of hazard was high in this region by single dimension factor analysis. Under the action of wind, eight sandstorm source areas and Tarim Basin were prone to sandstorm diffusion because of the existence of abundant sand materials. The hazard-formative environment was relatively fragile. Hejing and Aheqi counties had relatively good ecological background. In the Tarim Basin and its southern counties, the sandstorm tends to spread, especially in Hotan Prefecture, and counties of Qiemo and Ruoqiang. Oasis population and agricultural production were highly concentrated and vulnerable to environmental factors. (2) Some 46.53% of the area had high risk of sand diffusion due to abundant sand sources and poor site conditions. The most important factors affecting the wind sand diffusion were vegetation coverage and soil types of the underlying surface. The main causes of regional sand source diffusion were the intensity of wind field. (3) Based on the MCR model, 20 ecological corridors were constructed to connect ecological source areas, including five river types, nine road types and six green belt types of corridors. Among them, corridors 1 and 4 connected via Taklimakan Desert, and the rest of the corridors presented a circular distribution pattern at the outer edge of Tarim Basin. At the same time, 30 ecological nodes were identified, including seven class-A ecological nodes and 23 class-B ecological nodes, which were mainly distributed in Hotan and Bayingolin prefectures. The risk of sandstorm diffusion can be reduced by constructing shelterbelts and improving vegetation coverage. Results provide technical support for the prevention and control of sandstorm diffusion and the optimization of landscape patterns in the arid and semi-arid areas of northern China.

Key words: risk assessment of sandstorm diffusion, landscape pattern, spatial principal component analysis, Minimum Cumulative Resistance model, southern Xinjiang