Prediction of the spatiotemporal evolution of vegetation cover in the Huainan mining area and quantitative analysis of driving factors 

The prediction of the spatiotemporal dynamic evolution of vegetation cover in the Huainan mining area and the quantitative evaluation of its driving factors are of great significance for protecting and restoring the ecological environment in this area. This study uses the Landsat time-series data to estimate the vegetation cover using pixel dichotomy and uses the transition matrix to analyze the spatiotemporal transfer of vegetation cover from 1989 to 2004, 2004 to 2021, and 2021 to 2030. In addition, a structural equation model (SEM) was established in this study to assess the driving factors of vegetation cover. The quantitative analysis and the Cellular Automata (CA)-Markov model were performed to predict the future vegetation cover in the Huainan mining area. The results: 1) a significant transfer among various vegetation types over the 1989-2004 period. During this period, the high-cover and medium-cover types revealed the most significant transfer-in and transfer-out, covering total areas of 738.5211 and 527.2884 km 2 , respectively. Whereas from 2004 to 2021, the high cover types showed the most significant transfer-in and transfer-out, covering total areas of 295.8993 and 205.3845 km 2 , respectively. The predicted land cover from 2021 to 2030 showed that the high cover type was the most transferred out, covering a total area of 540.7317 km 2 and 555.5709 km 2 , using the CA-Markov and multi-criteria evaluation-CA-Markov (MCE-CA-Markov) models, respectively; 2) among the three dimensions, path coefficients of human activities were -0.11 and -0.39, respectively, while the path coefficients of topographical factors were 0.63 and 0.71 using SEM in 2015 and 2020, respectively. Human activities are the key factors affecting the vegetation growth, while topographical factor is the main factor promoting the vegetation growth; 3) Highly consistent CA-Markov and Multi-criteria evaluation (MCE) predicted results of vegetation cover in 2030 compared to that in 2021. The bare soil and low cover types were mainly concentrated in the mining area, showing connected patches. In addition, the bare soil type revealed a continuous expansion pattern, more particularly in the northwest direction. Large-area bare soil and low-cover patches were observed in the Zijing field, Banji field, and Yangcun exploration areas. The prediction of the spatiotemporal evolution of vegetation cover in the Huainan mining area and the quantitative change in driving factors are of significant importance for the restoration of the ecological environment in mining areas.