高原(数学)
气候变化
腐蚀
环境科学
自然地理学
风积作用
风险评估
水文学(农业)
地质学
地理
地貌学
数学分析
数学
海洋学
计算机安全
岩土工程
计算机科学
作者
Qilong Tian,Xiaoping Zhang,Jie He,Haijie Yi,Liang He,Yang Qin-ke
标识
DOI:10.1016/j.ecolind.2023.110669
摘要
With global warming and increasing anthropogenic activities, the ecosystems on the Tibetan Plateau are becoming increasingly fragile, exacerbating the risk of soil erosion in the area. However, the interacting effects induced by numerous erosion processes occurring at different times and spaces on the Tibetan Plateau are challenging to quantify using the conventional single-process assessment models (e.g., water erosion, wind erosion, and freeze–thaw). Consequently, our understanding of the complicated state of the plateau with respect to the potential for soil erosion is limited. Therefore, we created a methodological framework using multi-criteria decision-making (MCDM) to evaluate the potential risk of various soil erosion processes driven by climate change and human activity under current soil and topographical circumstances in this region. The results showed that the assessment model was reliable, and the estimated accuracy using the receiver operating characteristic curve for the training and validation datasets was 0.721. The majority of the areas of the plateau (60.69%) were at very-low or low-risk levels, while 17.55% of the plateau (mainly in the southeast and along the surrounding high mountains) was at high risk. In general, the average potential risk of soil erosion significantly increased during 1990–2020, and climate change exerted more pressure on the land surface than human activities did. The potential risk increased dramatically in 28.15% of the total plateau area, and was found to be concentrated in the plateau’s southern, eastern, central, and northern regions. The findings of this study provide a basis for local ecological environmental protection and land resource management on the Tibetan Plateau and propose a new protocol to forecast and prevent soil erosion worldwide.
科研通智能强力驱动
Strongly Powered by AbleSci AI