Open-pit coal mining plays an important role in supporting national economic development; however, it has caused ecological problems and even seriously impacted regional ecological stability. Given the importance of maintaining ecological stability in semi-arid coal mining areas, this study used a coupling coordination degree approach based on the structural and functional state transition model (SFSTM) to evaluate the spatio–temporal variation of ecological stability from 2002 to 2017 by using MODIS and Landsat datasets in the semi-arid open-pit coal mining area. Besides, random points were created for different ecological stability levels (containing natural areas, coal mining areas, and reclamation areas) and segment linear regression was conducted to determine the structural change threshold for negative state transitions based on mining and positive state transitions based on reclamation. Furthermore, the impact factors of ecological stability were analyzed. Results showed that ecological stability fluctuated significantly over 16 years, showing a trend of first increasing and then decreasing. It was found that precipitation and temperature were the key natural factors affecting ecological stability, and mining activities constituted the dominant factor. The average perturbation distances to ecological stability from mining activities in the west, southwest, and east mining groups were 7500, 5500, and 8000 m, respectively. SFSTM is appliable to the coal mining ecosystem. Quantitative models of ecological stability response can help resolve ambiguity about management efficacy and the ecological stability results facilitate iterative updating of knowledge by using monitoring data from coal mining areas. Moreover, the proposed ecological structural threshold provides a useful early warning tool, which can aid in the reduction of ecosystem uncertainty and avoid reverse transformations of the positive state in the coal mining areas.

The open-pit coal mine dump in the study area contains many low-concentration heavy metal pollutants, which may cause pollution to the soil interface. Firstly, statistical analysis and geostatistical spatial interpolation methods described heavy metal pollution’s spatial distribution. The mine dump heavy metal pollution distribution is strongly random due to disorderly piles, but it is closely related to slope soil erosion. Furthermore, the soil deposition area is where pollutants accumulate. For example, all heavy metal elements converge at the bottom of the dump. Usually, the pollution in the lower part is higher than that in the upper part; the pollution in the lower step is higher than the upper step; the pollution in the soil deposition locations such as flat plate and slope bottom is higher than the soil erosion locations such as slope tip and middle slope. Finally, the hyperspectral remote sensing method described heavy metals pollution’s migration characteristics, that the pollutants could affect the soil interface by at least 1 km. This study provides a basis for preventing and controlling critical parts of mine dump heavy metal pollution and pollution path control.

Accurate monitoring of plant dust retention can provide a basis for dust pollution control and environmental protection. The aims of this study were to analyze the spectral response features of grassland plants to mining dust and to predict the spatial distribution of dust retention using hyperspectral data. The dust retention content was determined by an electronic analytical balance and a leaf area meter. The leaf reflectance spectrum was measured by a handheld hyperspectral camera, and the airborne hyperspectral data were obtained using an imaging spectrometer. We analyzed the difference between the leaf spectral before and after dust removal. The sensitive spectra of dust retention on the leaf- and the canopy-scale were determined through two-dimensional correlation spectroscopy (2DCOS). The competitive adaptive reweighted sampling (CARS) algorithm was applied to select the feature bands of canopy dust retention. The estimation model of canopy dust retention was built through random forest regression (RFR), and the dust distribution map was obtained based on the airborne hyperspectral image. The results showed that dust retention enhanced the spectral reflectance of leaves in the visible wavelength but weakened the reflectance in the near-infrared wavelength. Caused by the canopy structure and multiple scattering, a slight difference in the sensitive spectra on dust retention existed between the canopy and leaves. Similarly, the sensitive spectra of leaves and the canopy were closely related to dust and plant physiological parameters. The estimation model constructed through 2DCOS-CARS-RFR showed higher precision, compared with genetic algorithm-random forest regression (GA-RFR) and simulated annealing algorithm-random forest regression (SAA-RFR). Spatially, the amount of canopy dust increased and then decreased with increasing distance from the mining area, reaching a maximum within 300–500 m. This study not only demonstrated the importance of extracting feature bands based on the response of plant physical and chemical parameters to dust, but also laid a foundation for the rapid and non-destructive monitoring of grassland plant dust retention.