Assessment of land degradation risks in the Loess Plateau

Abstract

Human activity and climate change are degrading the environmentally fragile Loess Plateau in dry and semiarid regions. Land deterioration threatens human and ecological existence. To prevent additional land degradation and ensure the ecological development and quality of arable land in the region, China launched “Grain for Green” in the late 1990s. This effort greatly boosted Loess Plateau vegetation. However, land degradation is complex, and so we must also examine natural and social variables to prevent additional degradation. Thus, this study presents a comprehensive index system to quantify land degradation on the Loess Plateau and uses machine learning to anticipate high‐risk locations. The project improved land degradation, and the spatial distribution of degradation risk is high in the northern and low in the eastern and southern regions of the Plateau. Gross Domestic Product and population density are the main drivers of land degradation. Industrialization and urbanization have raised the risk of land degradation, which now accounts for 1%–2% of the area. This study emphasizes sustainable land management in the Loess Plateau, a critical area for sustainable development in China. The integrated assessment indicator system and random forest modeling machine learning help grasp the current status and future preventive measures. The outcome of this study advances the Loess Plateau land degradation research and sustainable land management. The research findings possess significant scientific reference value in terms of mitigating and managing land degradation in environmentally vulnerable regions worldwide.

Ecological Risk Assessment of Land Use Change in the Tarim River Basin, Xinjiang, China

“…Another method involves ecological risk assessment based on landscape ecology theory. This method emphasizes the impact of landscape patterns on specific ecological functions or processes and focuses on the overall loss of landscape in providing ecosystem services and ecological functions [30,31]. Evaluation indicators include landscape fragility, resilience, and stability.…”

Analysis of the spatial differentiation and driving force of arable land abandonment and non-grain in the hilly mountainous areas of Gannan

“…In terms of research methods, scholars have mostly used spatial autocorrelation analysis [ 18 ] to analyze the spatial and temporal distribution of the phenomena of arable land abandonment and arable land non-grain in the context of arable land abandonment and arable land non-grain. At the same time, scholars mostly use Tobit model [ 19 ], probit model [ 20 ], random forest model [ 21 ], spatial error model [ 22 ], multiple linear regression method [ 23 ], logistic regression [ 24 ] and other ways to explore the influencing factors of two kinds of arable land phenomena. From the perspective of research scale, the problems of arable land abandonment and non-grain have been involved at the provincial scale [ 25 ] and city and county scales [ 26 ].…”