The adverse combination of excessive mining practices and the resulting land subsidence is a significant obstacle to the sustainable growth and stability of regions associated with mining activities. The Lakhra coal mines, which contain some of Pakistan’s largest coal deposits, have been overlooked in land subsidence monitoring, indicating a considerable oversight in the region. Subsidence in mining areas can be spotted early when using Interferometric Synthetic Aperture Radar (InSAR), which can precisely monitor ground changes over time. This study is the first to employ the Small Baseline Subset (SBAS)-InSAR and stacking-InSAR techniques to identify land subsidence at the Lakhra coal mines. This research offers critical insights into subsidence mechanisms in the study area, which has never been previously investigated for ground deformation monitoring, by utilizing 150 Sentinel-1A (ascending) images obtained between January 2018 and September 2023. A total of 102 deformation spots were identified using SBAS-InSAR, while stacking-InSAR detected 73 deformation locations. The most extensive cumulative subsidence in the Lakhra coal mine was −114 mm, according to SBAS-InSAR, with a standard deviation of 6.63 mm. In comparison, a subsidence rate of −19 mm/year was reported using stacking-InSAR with a standard deviation of 1.17 mm/year. The rangeland covered 88.8% of the total area and exhibited the most significant deformation values, as determined by stacking and SBAS-InSAR techniques. Linear regression showed that there was not a strong correlation between subsidence and topographic factors. As detected by optical remote sensing data, the subsidence locations were near or above the mines in the research area, indicating that widespread mining in Lakhra coal mines was the cause of subsidence. Our findings suggest that SAR interferometric time series analysis is helpful for proactively identifying and controlling subsidence difficulties in mining regions by closely monitoring activities, hence reducing negative consequences on operations and the environment.

Understanding the distribution of rock glaciers provides key information for investigating and recognizing the status and changes of the cryosphere environment. Deep learning algorithms and red–green–blue (RGB) bands from high-resolution satellite images have been extensively employed to map rock glaciers. However, the near-infrared (NIR) band offers rich spectral information and sharp edge features that could significantly contribute to semantic segmentation tasks, but it is rarely utilized in constructing rock glacier identification models due to the limitation of three input bands for classical semantic segmentation networks, like DeeplabV3+. In this study, a dual-encoder DeeplabV3+ network (DEDNet) was designed to overcome the flaws of the classical DeeplabV3+ network (CDNet) when identifying rock glaciers using multispectral remote sensing images by extracting spatial and spectral features from RGB and NIR bands, respectively. This network, trained with manually labeled rock glacier samples from the Qilian Mountains, established a model with accuracy, precision, recall, specificity, and mIoU (mean intersection over union) of 0.9131, 0.9130, 0.9270, 0.9195, and 0.8601, respectively. The well-trained model was applied to identify new rock glaciers in a test region, achieving a producer’s accuracy of 93.68% and a user’s accuracy of 94.18%. Furthermore, the model was employed in two study areas in northern Tien Shan (Kazakhstan) and Daxue Shan (Hengduan Shan, China) with high accuracy, which proved that the DEDNet offers an innovative solution to more accurately map rock glaciers on a larger scale due to its robustness across diverse geographic regions.

In recent years, coal mining has shifted from surface to underground multi-seam and multi-panel operations, leading to enhanced ground deformation and elevated risks of secondary geo-hazards. However, the deformation mechanisms and spatiotemporal evolution of mining-induced ground movement in high-intensity repeated mining areas require further investigation. To gain further insight, this study focuses on elucidating the deformation mechanisms and hazard-chain evolution induced by downward multi-seam and multi-panel mining in the Hongyan coal mine, located in the loess gully region. An integrated InSAR-based multi-source monitoring and modeling framework was adopted, systematically combining InSAR, historical satellite imagery, UAV-based surveys, and ground observations with numerical simulations to characterize the spatiotemporal evolution of mining-induced deformation and examine the coupling processes within the hazard chain. The monitoring results show a strong spatiotemporal correlation between mining activities and ground deformation: subsidence basins and temporal variations correspond closely to the mining sequence, and the spatial distribution of fissures aligns with the advancing working faces. The analysis indicates mining-induced stress redistribution and stratum instability are the root causes of subsidence. Subsidence characteristics are affected by topography, mining sequence, and the cumulative impacts of multi-seam mining, leading to stepwise subsidence and subsidence basins. The overlying loess’s topography and characteristics affect the subsidence distribution. The “stress arch” formed in the goaf evolves with the multi-panel mining process, gradually collapsing during continuous mining and leading to stratum instability. Initially spreading stress and preventing rock movement, the upper residual pillars aggravate stratum damage following critical stratum failure. Mining exerts spatiotemporal control over hazard development, with the hazard chain evolving upward from the mining horizon, driven by fissure propagation and subsidence as the core processes, and reinforced by a bottom-up chain reaction and feedback among successive hazards. This study provides scientific insights for the planning and hazard prevention of multi-seam mining in loess gully regions.