Ecosystem services (ESs) are irreplaceable natural resources, and their value is closely related to global change and to human well-being. Research on ecosystem services value (ESV) and its influencing factors can help rationalize ecological regulatory policies, and is especially relevant in such an ecologically significant region as the Yellow River Basin (YRB). In this study, the ecological contribution model was used to measure the contribution of intrinsic land use change to ESV, the bivariate spatial autocorrelation model was applied to investigate the relationship between land use degree and ESV, and the geographical detector model (GDM) and geographically weighted regression (GWR) were applied to reveal the impact of natural and socio-economic factors on ESV. Results showed that: (1) The total ESV increased slightly, but there were notable changes in spatial patterns of ESV in the YRB. (2) Land use changes can directly lead to ESV restoration or degradation, among which, conversion from grassland to forest land and conversion from unused land to grassland are vital for ESV restoration in the YRB, while degradation of grassland is the key factor for ESV deterioration. (3) According to GDM, NDVI is the most influential factor affecting ESV spatial heterogeneity, and the combined effect of multiple factors can exacerbate ESV spatial heterogeneity. (4) GWR reveals that NDVI is always positively correlated with ESV, GDP is mainly positively correlated with ESV, and population density is mainly negatively correlated with ESV, while positive and negative correlation areas for other factors are roughly equal. The findings can provide theoretical support and scientific guidance for ecological regulation in the YRB.

An increased land use intensity due to rapid urbanization and socio-economic development would alter the structure and function of regional ecosystems and cause prominent environmental problems. Revealing the impact of land use intensity on ecosystem services (ES) would provide guidance for more informed decision making to promote the sustainable development of human and natural systems. In this study, we selected the Hanjiang River Basin (HRB) in Hubei Province (China) as our study area, explored the correlation between land use intensity and ecosystem Services’ Value (ESV), and investigated impacts of natural and socio-economic factors on ESV variations based on the Geographical Detector Model (GDM) and Geographically Weighted Regression (GWR). The results show that (1) from 2000 to 2020, land use intensity in HRB generally showed an upward trend, with a high spatial agglomeration in the southeast and low in the northwest; (2) the total ESV increased from 295.56 billion CNY in 2000 to 296.93 billion CNY in 2010, and then decreased to 295.63 CNY in 2020, exhibiting an inverted U-shaped trend, with regulation services contributing the most to ESV; (3) land use intensity and ESV had a strong negative spatial correlation, with LH (low land use intensity vs. high ESV) aggregations mainly distributed in the northwest, whereas HL (high land use intensity vs. low ESV) aggregations were located in the southeast; (4) natural factors, including annual mean temperature, the percentage of forest land, and slope were positively associated with ESV, while socio-economic factors, including GDP and population density, were negatively associated with ESV. To achieve the coordinated development of the socio-economy and the environment, ES should be incorporated into spatial planning and socio-economic development policies.

The extent to which landscape spatial patterns can impact the dynamics and distribution of biodiversity is a key geography and ecology issue. However, few previous studies have quantitatively analyzed the spatial relationship between the landscape pattern and habitat quality from a simulation perspective. In this study, the landscape pattern in 2031 was simulated using a patch-generating simulation (PLUS) model for the Yellow River Basin. Then, the landscape pattern index and habitat quality from 2005 to 2031 were evaluated using the Fragstats 4.2 and the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model. Furthermore, we analyzed the spatial distribution characteristics and spatial spillover effects of habitat quality using spatial autocorrelation analysis. Finally, the spatial association between the landscape pattern index and habitat quality was quantitatively revealed based on a spatial lag model. The simulation results showed that: (1) from 2005 to 2031, the landscape of the Yellow River Basin would be dominated by grassland and unused land, and the areas of construction land and water body will increase significantly, while the area of grassland will decrease; (2) patch density (PD) and Shannon’s diversity index (SHDI) show significant increases, while edge density (ED), landscape shape index (LSI), mean patch area (AREA_MN), and contagion index (CONTAG) decrease; (3) from 2005 to 2031, habitat quality would decrease. The high-value areas of habitat quality are mainly distributed in the upper reaches of the Yellow River Basin, and the low-value areas are distributed in the lower reaches. Meanwhile, both habitat quality and its change rate present positive spatial autocorrelation; and (4) the spatial relationships of habitat quality with PD and COHESION are negative, while ED and LSI have positive impacts on habitat quality. Specifically, landscape fragmentation caused by high PD has a dominant negative influence on habitat quality. Therefore, this study can help decision makers manage future landscape patterns and develop ecological conservation policy in the Yellow River Basin.