Objectives Zhouqu County in Gansu Province is highly prone to landslides. On July 12, 2018, a landslide blocked the Bailong River near the county, posing a serious threat to the life and property of local residents and the safety of infrastructure. The early identification of potential landslides around this area is of great significance to protect the life and property of local residents and the security of its infrastructure.
Methods Firstly, small baseline subset interferometry technology (SBAS-InSAR) was adopted to identify the potential active landslides in the surrounding area of Zhouqu County, with the Sentinel-1A satellite ascending and descending images from October 2017 to December 2018. Then, a combination of optical remote sensing image interpretation and field investigation is used to determine and identify potential landslides. The visual interpretation method can be used to identify landslides that have occurred or are occurring on the optical images. The field investigation was used to verify the landslide results identified by InSAR and optical imagery to increase the effectiveness and accuracy of landslide detection.
Results A total of 23 active landslides were detected in the study area, with 16 active landslides detected on the ascending images and 11 on the descending images, with 4 active landslides having valid deformation information detected on both the ascending and descending images. Combined with the visual interpretation of optical remote sensing images and field investigation, the deformation characteristics of 4 typical landslides, including Jiangdingya, Mentouping, Suoertou, and Xieliupo Landslides, were analyzed. It was found that the deformation rates of Jiangdingya, Mentouping, and Xieliupo Landslides were significantly higher during the rainy season than during the non-rainy season, and the Suoertou Landslide did not show a significant increase in the deformation rate during the rainy season but showed fluctuations in the deformation rate during the rainy season.
Conclusions The use of single-orbit radar satellite data for landslide detection can easily lead to the missed detection of landslide hazards, and the combined use of ascending and descending radar images can improve the accuracy of landslide detection in steep mountainous terrain. The deformation rates of active landslides in the study area have a good correlation with rainfall. The results can provide supports for the geohazard mitigation and prevention in Zhouqu County. The strategy which combines optical imagery, field investigation, and time-series InSAR can efficiently and accurately detect and monitor landslides over a wide area.
DownLoad: