In this paper, we use the InVEST model and five periods of land use data from 1980 to 2020 to assess the habitat quality of the Ningxia Hui Autonomous Region in Western China, which has characteristics of a typical fragile ecosystem. We further analyse the spatial and temporal characteristics of habitat quality evolution and its relationship with land use and landscape pattern indices to explore the close relationship between regional habitat quality changes and human natural resource conservation and utilization. The research results show that the overall habitat quality of Ningxia Hui Autonomous Region was stable and at a moderate level (0.57–0.60) during the 40 years from 1980 to 2020; Habitat patches (2020) with low (24.89%), high (22.45%) and very high (29.81%) quality occupy a larger proportion of the area, followed by very low (13.31%) and moderate levels (9.54%). Over the past 40 years, there have been 275 sample sites in Ningxia where habitat quality has deteriorated, 1593 sample sites where the habitat quality has remained stable, and 184 sample sites where the habitat quality has increased. From 1980 to 2020, the Mean Patch Area of landscape types in Ningxia decreased by 25.9 hm². The Patch Density increased by 0.06 /hm². The Largest Patch Index decreased by 15.69%. The Edge Density increased by 2.5 m/hm². The Contagion Index decreased by 2.99%. The Area‐Weighted Mean Patch Fractal Dimension remained basically unchanged (0.01). The Landscape Shape Index showed a trend of first increasing and then decreasing, increasing by 13.94. The Area‐Weighted Mean Shape Index has been reduced by 9.45. The Shannon Diversity Index and Shannon Evenness Index both show an increasing trend, but the amplitude is relatively small, with 0.09 and 0.04, respectively. There was a significant spatial aggregation of high and low habitat quality in Ningxia, with high values usually distributed in the northern and southern areas with good natural conditions and low values distributed in areas with frequent human activities and poor natural conditions. The decrease in habitat quality in Ningxia was mainly due to the expansion of cultivated land and construction land, the increase in landscape fragmentation and the resulting decrease in connectivity. On the other hand, due to the implementation of ecological protection measures, such as the project of returning farmland to pasture and grass to forest, the quality of habitats in Ningxia increased. The conclusions of this study support the idea that the conservation of habitat quality in ecologically fragile areas should fully preserve the original natural habitats and reduce the interference of human activities to increase the habitat suitability of the landscape and the habitat connectivity between patches. At the same time, targeted ecological protection policies should be developed to restore the areas where the habitat quality has been damaged and ultimately maintain the stability of biodiversity and ecosystems in ecologically fragile areas. Meanwhile, for ecologically fragile areas with similar ecological characteristics to those of Ningxia, our research supports the idea of increasing the protection of the stability of the original habitats, increasing the proportion of ecological restoration projects, financial investment and seeking cooperation with local community managers and residents will help to improve the quality of the regional habitats and the enrichment of the biodiversity, and ultimately promote the harmonious coexistence of human beings and nature in the modernized sense of the word.

Recognizing the spatial effects of regional poverty is essential for achieving sustainable poverty alleviation. This study investigates these spatial effects and their determinants across three distinct administrative levels within Hubei Province, China. To analyze the spatial patterns and heterogeneity of multi-scale regional poverty, we employed various spatial analysis techniques, including the global and local Moran’s I statistics, the Lineman, Merenda, and Gold (LMG) method, as well as Multiscale Geographically Weighted Regression (MGWR). We found that: (1) Regional poverty exhibits significant spatial dependence across various scales, with a higher level of spatial dependence observed at higher administrative levels. (2) The spatial distribution of poverty is primarily influenced by geographical factors, encompassing first-, second-, and third-nature geographical elements. Notably, first-nature geographical factors make substantial contributions, accounting for 36.99%, 42.23%, and 23.79% at the county, township, and village levels, respectively. (3) The influence of geographical factors varies with scale. Global effects of various factors may transcend scales or remain confined to specific scales, while the local impacts of different factors also exhibit variations across scales. These results underscore the necessity for collaborative efforts among government entities at different levels with the anti-poverty measures tailored to local contexts.

Landscape ecological risk assessment can reflect the impact of landscape pattern on ecological processes and functions, and has become an important method for predicting and measuring the quality and dynamic evolution of the ecological environment. Studying the trend of landscape ecological risk evolution is important for optimizing the regional landscape pattern and maintaining the sustainable development of the ecological environment in ecologically fragile areas. Ningxia Hui Autonomous Region is a typical agro-pastoral region in northern China, which is facing prominent ecological and environmental problems such as soil erosion and land desertification. With the environmental problems becoming more and more serious, the ecological risk also poses increasing potential danger, which becomes one of the bottlenecks restricting regional economic development. This paper selects Ningxia Hui Autonomous Region as the study area. Based on land use data in 2000, 2005, 2010, and 2015, the landscape ecological risk assessment model is built from the perspective of landscape ecology, and combines the spatial statistical analysis method and the gravity shift model to explore the spatial–temporal evolution of the spatial distribution characteristics of landscape ecological risk. The results showed that: (1) During the study period, the area of farmland and grasslands has decreased, and the area of forest, water, and settlement has increased; (2) The transformation of land-use types is mainly distributed in the central northern plain and the south valley, where human activities are intensive. The transfer relationship among different land-use types is mainly between grasslands and farmland; (3) The hotspots and aggregation area of landscape ecological risk are mainly distributed along the city belt and irrigated agricultural areas along the Yellow River in the northern Ningxia Plain; (4) From 2000 to 2005, the landscape ecological risk level decreased, and from 2005 to 2015, the landscape ecological risk index continued to rise. We conclude that the natural environment of Ningxia determines the characteristics of the spatial distribution of landscape ecological risk, while the use and modification of land through human activities are the main driving factors of landscape ecological risk change. The results of landscape ecological risk assessment at different scales based on the landscape index will provide support for regional environmental management.