Soil moisture plays an important role in maintaining ecosystem stability and sustainable development, especially for the upper reaches of the Yellow River region. Therefore, accurately and conveniently monitoring soil moisture has become the focus of scholars. This study combines three machine learning algorithms: random forest (RF), support vector machine (SVM), and back propagation neural network (BPNN)—with the traditional monitoring of soil moisture using remote sensing indices to construct a more accurate soil moisture inversion model. To enhance the accuracy of the soil moisture inversion model, 27 environmental variables were screened and grouped, including vegetation index, salinity index, and surface temperature, to determine the optimal combination of variables. The results show that screening the optimal independent variables in the Xijitan landslide distribution area lowered the root mean square error (RMSE) of the RF model by 16.95%. Of the constructed models, the combined model shows the best applicability, with the highest R² of 0.916 and the lowest RMSE of 0.877% with the test dataset; the further research shows that the BPNN model achieved higher overall accuracy than the other two individual models, with the test set R² being 0.809 and the RMSE 0.875%. The results of this study can provide a theoretical reference for the effective use of Landsat satellite data to monitor the spatial and temporal distribution of and change in soil water content on the two sides of the upper Yellow River basin under vegetation cover.

The upper reaches of the Yellow River are critical ecological barriers within the Yellow River Basin (YRB) that are crucial for source conservation. However, environmental challenges in this area, from Longyangxia to Lijiaxia, have emerged in recent years. To assess the ecological environment quality (EEQ) evolution from 1991 to 2021, we utilized remote sensing ecological indices (RSEIs) on the Google Earth Engine (GEE) platform. Spatial autocorrelation and heterogeneity impacting EEQ changes were examined. The results of this study show that the mean value of the RSEIs fluctuated over time (1991: 0.70, 1996: 0.77, 2001: 0.67, 2006: 0.71, 2011: 0.68, 2016: 0.65, and 2021: 0.66) showing an upward, downward, and then upward trend. The mean values of the overall RSEI are all at 0.65 and above. Most regions showed no significant EEQ change during 1991–2021 (68.59%, 59.23%, and 55.78%, respectively). Global Moran’s I values (1991–2021) ranged from 0.627 to 0.412, indicating significant positive correlation between EEQ and spatial clustering, and the LISA clustering map (1991–2021) shows that the area near Longyangxia Reservoir shows a pattern of aggregation, dispersion, and then aggregation again. The factor detection results showed that heat was the most influential factor, and the interaction detection results showed that greenness and heat had a significant effect on regional ecosystem distribution. Our study integrates spatial autocorrelation and spatial heterogeneity and combines them with reality to provide an in-depth discussion and analysis of the Longyangxia to Lijiaxia Basin. These findings offer guidance for ecological governance, vegetation restoration, monitoring, and safeguarding the upper Yellow River’s ecological integrity.

Background: The Xining Basin, located at the confluence of the northeastern Qinghai–Tibet Plateau and the western Loess Plateau, where extensive loess deposits are found, is densely inhabited. Major transport corridors, encompassing the Qinghai–Tibet Railway and the Lanzhou–Xinjiang High‒Speed Railway, traverse the region. The area is rendered highly susceptible to geohazards such as shallow landslides, collapses, and debris flows, especially during extreme rainfall, by its fragile geology and sparse vegetation. Posing serious threats to the local ecology, major infrastructure, and community safety, these hazards are manifestly concerning. Performing an investigation into how dominant flora enhance the stability of shallow landslides is of significance for the prevention and control of slope disasters. Methods: In this study, the loess region of the Xining Basin was selected as the area of research. Three dominant herbaceous species— Elymus nutans , Elymus trachycaulus , and Elymus sibiricus —were chosen for single‒root and multiroot tensile testing. Combined with direct shear tests on root–soil composites, the enhancement effect of their root systems on soil shear strength was systematically evaluated. Results: At the single‒root scale, there exists a salient increase in the tensile resistance for herbaceous roots with diameter, whereas the tensile strength manifests a decrease. Marked interspecific differences were observed: E. trachycaulus achieved the highest single‒root tensile resistance (17.10 N) and ultimate elongation rate (42.28%), whereas E. nutans exhibited the highest tensile strength (97.53 MPa). At the scale of multiroot, an increase in the number of roots increased the tensile resistance, whereas the tensile strength decreased. E. trachycaulus manifestly had the highest tensile resistance and strength in its multiroot (243.35 N, 36.43 MPa), whereas E. sibiricus had the lowest (46.14 N, 17.17 MPa). The tensile resistance and strength of multiroot were lower in measurements than in predictions on single‒root properties (14.86‒48.13%), which provides a demonstration of their mechanical behaviour entailing interactions of nonlinear complexity rather than superposition of linear simplicity. Direct shear tests on root–soil composites demonstrated that all three herbaceous species significantly enhanced soil shear strength. Compared with those of bare slopes, the cohesion ( c ) values of their root–soil composites increased by 157.99%, 104.54%, and 78.34%, respectively. Conclusion: This study examined the mechanical characteristics of herbaceous roots and their slope protection mechanisms across scales from single roots to multiroot systems. These findings provide a theoretical and practical basis for the selection of vegetation and prevention of soil disasters in ecological slope protection projects.