Ecological vulnerability assessment of a China's representative mining city based on hyperspectral remote sensing

Mining cities are clusters of communities that specialized in mining and extractive industries. The extensive mining activities in these cities have stimulated widespread and substantial ecological stresses to the surrounding environment, that significantly jeopardize the health condition of vegetation and human. Given the recent recognition of remote sensing in monitoring large-scale environmental change, we incorporated Ziyuan #1-02D, a recently released hyperspectral remote sensing data, into the ecological vulnerability assessment framework, using Panzhihua city as a case study, which is recognized as one of the most representative mining cities in China. The multi-spectral imaging data was widely applied in previous research. However, with the wide bands, multi-spectral imaging data cannot depict detailed characteristics of spectral targeted. As a result, we introduce indexes from the hyperspectral imaging to ecological vulnerability assessment proposed in this study, which can depict and monitor the growth and restoration of vegetation more accurately. We used the optimum index factor method to select bands from the satellite-based Ziyuan #1-02D data for quantifying vegetation indexes and red edge. Besides, we obtained inventory data, land-use, soil type, and typography of Panzhihua city to reconstruct its ecological vulnerability index (EVI) for 2020 and 2021. Comparing to the multi-spectral data, the ecological vulnerability results from hyperspectral imaging performed better in precision and concentration in EVI values, reaching the conclusions more directly. Specifically, the mining area and the relevant hazard types and impact areas were delineated through intensive fieldwork. Results suggested that the east and west districts, and north of Renhe district suffer great ecological stress, in which we observed intensive coal and metal-related mining industry. The central region, which occupies vanadium titanomagnetite mines, also shows substantial ecological issues, while the other mining industries, such as granite ore did not significantly influence the local environment. Although the newly released satellite-based data only have two-year periods, we still observed improving ecological conditions, with the southeast and west regions showing much lower ecological vulnerability values. The spatial autocorrelation analysis suggested that the high-high clustering region of EVI is located in the east, west, and Renhe districts, primarily due to the mining industries of variations scales. We also found that the clustering of low ecological vulnerable regions, mostly surrounds the vanadium titanomagnetite excavation industry, thanks to the local restoration projects.