Abstract. Ground subsidence is regarded as one of the most common
geohazards, accompanied with the rapid urban expansion in recent years. In
the last 2 decades, Wuhan, located in the alluvial Jianghan Plain, has
experienced great urban expansion with increased subsidence issues, i.e.,
soft foundation subsidence and karst collapse. Here we investigated
subsidence rates in Wuhan with 2015–2019 Sentinel-1 synthetic aperture radar (SAR) images. We found that the overall subsidence over the Wuhan region is significantly correlated with the distribution of engineering geological subregions (EGSs). We further validated the interferometric SAR (InSAR) measurements with better than 5 mm accuracy by comparing with leveling measurements. Subsidence centers in Qingling–Jiangdi, Houhu, Qingshan, and Dongxihu were identified with
displacement rates of approximately 30 mm/yr. Our results demonstrated that the dominant driving factor is ongoing construction, and the fact that the subsidence centers shifted with construction intensities. The Qingling–Jiangdi area in our study is a well-known site of karst collapse. We find that the nonlinear subsidence of this area is correlated with the seasonal rainfall.
Decorrelation is one of the main limitations for InSAR. Masking decorrelated pixels is crucial for retrieving information from SAR interferograms. However, for traditional masking methods, manually drawing masks is time-consuming and may be unfeasible when decorrelation areas are with complicated and blurred boundaries. Setting a single coherence threshold is also difficult, if not impossible, to mask out all decorrelated pixels without losing valid phases. Here, we propose a deep-learning segmentation network (Mask Net) based on Selective Kernel Res-Attention UNet, for generating decorrelation masks with applications to TanDEM-X interferograms. We conduct several experiments to determine the training strategy and parameters, including sample size, batch size, loss function and down-sampling scheme, to optimize network performance. Afterwards, we compare the performance of Mask Net with other classical segmentation networks. Our evaluation metrics show that Mask Net outperforms the best performance of other segmentation networks by IoU of 6.32% and F1 Score of 3.97%, respectively. It also possesses the fastest inferring speed, 0.4505s on sample size of 1024-by-1024 pixels, which is at least ~50% faster than other segmentation networks. We applied Mask Net to three TanDEM-X interferograms of Kīlauea crater in Hawaii, metropolitan region of Wuhan, and Muztagata Glacier in China. Our results show that comparing with coherence threshold method, Mask Net can clearly mask out all decorrelation regions, rarely causing loss of valid phases. It also exhibits better segmentation performance than other deep-learning segmentation networks, especially for those complex decorrelation boundaries, with less computational time.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.