Landscape pattern identification and ecological risk assessment using land-use change in the Yellow River Basin

Abstract: Ecological risk assessment is essential to the diagnosis of the adverse environmental factors as a result of human activities for the decision-making in risk management. Since the Yellow River Basin of China has a fragile ecological environment and serious soil erosion, this study aims to analyze its landscape pattern and ecological risk factors for a better control of impact sites, and the sustainable development of agricultural production. First, the index of landscape pattern was calculated using the land-use change data on the Yellow River Basin in 2000, 2010 and 2018. Then, an assessment model of ecological risk index (ERI) was constructed using the index of landscape pattern, thereby exploring the spatial and temporal characteristics of ecological risks. The transfer laws of ecological risks were set at the grid scale (1 078 grid units divided by ARCGIS) and county scale (391 county-level units divided by administrative regions). Finally, the Moran's I and a local indicator of spatial association (LISA) were selected to clarify the spatial correlation of ecological risks. The research results show that: 1) The grass land and cultivated land were the main types of land-use area, accounting for more than 70% of the total area in the Yellow River Basin. From 2000 to 2018, the area of cultivated land had decreased by 14 243 km2, and the area of unused land had decreased by 9 410.3 km2, while the area of construction land had increased by 12 179.8 km2. It infers that the Yellow River Basin was still in a period of rapid urbanization during this stage, where the rigid demand was strong for construction land. 2) Most indicators increased, including the number of patches, patch density, largest patch index, landscape shape index, and Shannon diversity index, whereas, the splitting index had decreased in the Yellow River Basin from 2000 to 2018. The overall landscape pattern tended to be complex, fragmented, and decentralized. Specifically, the forest land and cultivated land showed the highest fragmentation, and the grass land and cultivated land had the most complex landscape shapes and structures. 3) There was an upward trend of the maximum ecological risk, increasing from 0.62 to 0.74 from 2000 to 2018. The higher-risk and high-risk areas continued to increase, and the low-risk and lower-risk areas were greatly reduced. Ecological risk levels mostly shifted from low to high levels. The most shifted areas on the grid scale were from low-risk areas to medium-risk areas (119 437 km2), and the most shifted areas on the county scale were from medium-risk to higher-risk areas (123 095 km2). 4) The Moran's I of ecological risk was greater than 0 from 2000 to 2018, showing a positive spatial correlation. But there was a downward trend of Moran's I, indicating that the spatial aggregation and spatial differentiation of ecological risks were weakened. The Moran's I on the grid scale was larger than that on the county scale, indicating that the grid-scale ecological risk had a stronger spatial positive correlation. There was a decreasing trend in the number of "high-high" and "low-low" agglomeration units, indicating that the local spatial aggregation of ecological risks decreased. Some recommendations were made on the differentiated management, where the key, strict, and general control areas were divided without crossing municipal administrative regions. The finding can provide a potential insightful reference for the prevention and control of ecological risks, and thereby promote the high-quality development of the Yellow River Basin.