Traditional multi-temporal InSAR (MT-InSAR) technology commonly employs a global single threshold in selecting valid pixels. In mining areas where there are significant temporal and spatial decorrelation effects or where the deformation gradient is large in a short period, the issue of sparse measuring points and inadequate spatial sampling frequently arises, further causing the incomplete acquisition of ground subsidence information. Here, we take 22 scenes of ALOS-1 images covering the Datong mining area as an example and use the adaptive distributed scatterer InSAR combined with land cover (ADSI-CLC) method to measure the long term time series of ground deformations in this area. The results indicate that the spatial-temporal distribution pattern of the deformation measurements obtained by ADSI-CLC method is similar to that of the StaMPS-SBAS method, and has a good correlation with mining facilities on high-resolution optical remote sensing images. However, the ADSI-CLC method can significantly increase the number and spatial distribution density of measuring points, identifying approximately four times as many distributed scatterers (DS) points as the StaMPS-SBAS method in this study area. Specifically, the consistency between the two methods is high in areas with zero deformation or minor deformation. In regions with large deformation, the StaMPS-SBAS method cannot effectively obtain deformation measurements, while the ADSI-CLC method can derive deformation measurement results that are consistent with the funnel-shaped spatial distribution characteristics caused by mining activities, because of the increased number of measuring points. These results indirectly verify the reliability and effectiveness of the ADSI-CLC method in the ground deformations monitoring of the mining areas. Overall, the ADSI-CLC method can provide more detailed temporal and spatial deformation information, and can be used to monitor and warn the surface stability in mining areas.

针对传统时序InSAR技术在测量点选取过程中多采用全局单一阈值, 在受时空失相干影响严重和短时间内形变梯度较大的矿区容易出现测量点稀疏且空间采样不足, 因而无法获取矿区地面沉降完整信息的问题, 以覆盖大同矿区的22景ALOS-1数据为例, 利用结合土地覆盖类型的自适应DS-InSAR方法(ADSI-CLC)获取该区域长时间序列地表形变信息。ADSI-CLC方法形变测量结果的时空分布模式与StaMPS-SBAS方法具有相似性, 且与高分辨率光学遥感影像上的采矿设施保持良好的空间相关性。ADSI-CLC方法可以显著地提高测量点的数量和空间分布密度, 在研究区识别出的DS点数量约是StaMPS-SBAS方法的4倍。两种方法的测量结果在无形变和小形变区域吻合程度较高; 在大形变区域, StaMPS-SBAS方法无法有效地获取形变结果, 而ADSI-CLC方法由于测量点数量的增加, 反演得到的形变测量结果基本上符合由采矿活动造成的漏斗状的空间分布特征, 间接地验证了该方法在矿区地表形变监测中的可靠性和有效性, 证明ADSI-CLC方法能够提供更详细的时空形变细节信息, 更好地服务于矿区地表稳定性的监测和预警。