When employing remote sensing images, it is challenging to classify vegetation species and ground objects due to the abundance of wetland vegetation species and the high fragmentation of ground objects. Remote sensing images are classified primarily according to their spatial resolution, which significantly impacts the classification accuracy of vegetation species and ground objects. However, there are still some areas for improvement in the study of the effects of spatial resolution and resampling on the classification results. The study area in this paper was the core zone of the Huixian Karst National Wetland Park in Guilin, Guangxi, China. The aerial images (Am) with different spatial resolutions were obtained by utilizing the UAV platform, and resampled images (An) with different spatial resolutions were obtained by utilizing the pixel aggregation method. In order to evaluate the impact of spatial resolutions and resampling on the classification accuracy, the Am and the An were utilized for the classification of vegetation species and ground objects based on the geographic object-based image analysis (GEOBIA) method in addition to various machine learning classifiers. The results showed that: (1) In multi-scale images, both the optimal scale parameter (SP) and the processing time decreased as the spatial resolution diminished in the multi-resolution segmentation process. At the same spatial resolution, the SP of the An was greater than that of the Am. (2) In the case of the Am and the An, the appropriate feature variables were different, and the spectral and texture features in the An were more significant than those in the Am. (3) The classification results of various classifiers in the case of the Am and the An exhibited similar trends for spatial resolutions ranging from 1.2 to 5.9 cm, where the overall classification accuracy increased and then decreased in accordance with the decrease in spatial resolution. Moreover, the classification accuracy of the Am was higher than that of the An. (4) When vegetation species and ground objects were classified at different spatial scales, the classification accuracy differed between the Am and the An.

The ecological security pattern is an important way to coordinate the contradiction between regional economic development and ecological protection and is conducive to promoting regional sustainable development. This study examines Guangxi, a karst region in China. The ecosystem service function and ecological environment sensitivity were both selected to evaluate the ecological conservation importance, and based on the results of the ecological conservation importance evaluation, suitable patches were selected as ecological sources. Meanwhile, resistance factors were selected from both natural factors and human activities to construct a comprehensive resistance surface, circuit theory was used to identify ecological corridors, ecological pinch points, and ecological barrier points, and ecological protection suggestions were then proposed. The results show that there are 50 patches of ecological sources in Guangxi, with a total area of 60,556.99 km²; 115 ecological corridors, with the longest corridor reaching 194.97 km; 301 ecological pinch points, whose spatial distribution is fragmented; and 286 ecological barrier points, most of which are concentrated in the central part of Guangxi. The results of this study provide a reference for the construction of ecological security patterns and ecological conservation in developing countries and karst areas.

Analysis and prediction of urban ecological risk are crucial means for resolving the dichotomy between ecological preservation and economic development, thereby enhancing regional ecological security and fostering sustainable development. This study uses Nanning, a Chinese landscape garden city, as an example. Based on spatial granularity and extent perspectives, using 30 m land use data, the optimal scale for an ecological risk assessment (ERA) and prediction is confirmed. This study also explores the patterns of spatial and temporal changes in ecological risk in Nanning on the optimal scale. At the same time, the Patch-generating Land Use Simulation model is used to predict Nanning’s ecological risk in 2036 under two scenarios and to propose ecological conservation recommendations in light of the study results. The study results show that: a spatial granularity of 120 m and a spatial extent of 7 km are the best scales for ERA and prediction in Nanning. Although the spatial distribution of ecological risk levels is obviously different, the overall ecological risk is relatively low, and under the scenario of ecological protection in 2036, the area of high ecological risk in Nanning is small. The results can provide theoretical support for ERA and the prediction of landscape cities and ecological civilization construction.

Fractional vegetation coverage (FVC) is an important indicator of ecosystem change. At present, FVC products are mainly concentrated at low and medium spatial resolution and lack high temporal and spatial resolution, which brings certain challenges to the fine monitoring of ecological environments. In this study, we evaluated the accuracy of four remote sensing inversion models for FVC based on high-spatial-resolution Sentinel-2 imagery and unmanned aerial vehicle (UAV) field-measured FVC data in 2019. Then the inversion models were optimized by constructing a multidimensional feature dataset. Finally, the Source Region of the Yellow River (SRYR) FVC product was created using the best inversion model, and the spatial-temporal variation characteristics of the FVC in the region were analyzed. The study’s findings revealed that: (1) The accuracies of the four FVC inversion models were as follows: the Gradient Boosting Decision Tree (GBDT) model (R² = 0.967, RMSE = 0.045) > Random Forest (RF) model (R² = 0.962, RMSE = 0.049) > Support Vector Machine (SVM) model (R² = 0.925, RMSE = 0.072) > Pixel Dichotomy (PD) model (R² = 0.869, RMSE = 0.097). (2) Constructing a multidimensional feature dataset to optimize the driving data can improve the accuracy of the inversion model. NDVI and elevation are important factors affecting the accuracy of machine learning inversion algorithms, and the visible blue band is the most important feature factor of the GBDT model. (3) The FVC in the SRYR gradually increased from west to east and from north to south. The change trajectories of grassland FVC from 2017 to 2022 were not significant. The areas that tend to improve were mainly distributed in the southeast (1.31%), while the areas that tend to degrade were mainly distributed in the central and northwest (1.89%). This study provides a high-spatial-resolution FVC inversion optimization scheme, which is of great significance for the fine monitoring of alpine grassland ecological environments.

Fractional vegetation cover (FVC) products provide essential data support for ecological environmental monitoring and simulation studies. However, the lack of validation efforts of FVC products limits their applications. Based on Sentinel-2 data and intensive multi-scale measured FVC data, the accuracies of two FVC products (GEOV3 and MuSyQ) in alpine grassland ecosystems were validated through direct validation and multi-scale validation. Based on the heterogeneity of the underlying surface (HUS) of the monitoring plots, the impact of the HUS of the monitoring plots on the product validation was analyzed. The results showed that: (1) the measured data directly validated that the GEOV3 FVC product performed better than the MuSyQ FVC product; (2) the multi-scale validation method based on high-resolution reference FVC map of Sentienl-2 satellite images validated the accuracy of these two FVC products, which was higher than the accuracy directly validated by FVC measured data, leading to overestimation of the validation results; and (3) the HUS of the monitored plots has a significant impact on the FVC product validation. By quantifying the HUS of the monitored plots and removing the heterogeneous monitoring plots, the uncertainty of the validation results can be reduced. It is necessary to consider the impact of validation methods and the HUS on the validation results in future product validation.

Long-time series global fractional vegetation cover (FVC) products have received widespread international publication, and they supply the essential data required for eco-monitoring and simulation study, assisting in the understanding of global warming and preservation of ecosystem stability. However, due to the insufficiency of high-precision FVC ground-measured data, the accuracy of these FVC products in some regions (such as the Qinghai–Tibet Plateau) is still unknown, which brings a certain impact on eco-environment monitoring and simulation. Here, based on current international mainstream FVC products (including GEOV1 and GEOV2 at Copernicus Global Land Services, GLASS from Beijing Normal University, and MuSyQ from National Earth System Science Data Center), the study of the dynamic change of vegetation cover and its influence factors were conducted in the three-rivers source region, one of the core regions on the Qinghai–Tibet Plateau, via the methods of trend analysis and partial correlation analysis, respectively. Our results found that: (1) The discrepancy in the eco-environment assessment results caused by the inconsistency of FVC products is reflected in the statistical value and the spatial distribution. (2) About 70% of alpine grassland in the three-rivers source region changing trend is controversial. (3) The limiting or driving factors of the alpine grassland change explained via different FVC products were significantly discrepant. Thus, before conducting these studies in the future, the uncertainties of the FVC products utilized should be validated first to acquire the fitness of the FVC products if the accuracy information of these products is unavailable within the study area. In addition, more high-precision FVC ground-measured data should be collected, helping us to validate FVC product uncertainty.