Abstract: From October to November 2018, the Baige landslide in the upper reaches of Jinsha River collapsed twice, causing a barrier lake, and there is still significant creep deformation. This paper proposed an improved TS-InSAR (ITSI) approach to acquire the landslide area's time series creep deformation measurements. Based on the SAR image data obtained by Sentinel-1A satellite from September. 2017 to March. 2021, the pre-disaster and post-disaster time series processing are carried out, and the annual average creep velocity field (the pre-disaster extreme value is 9.71cm/a, and the post-disaster extreme value is 25.42cm/a) and cumulative creep deformation measurements (the extreme value is over 13cm in the pre-disaster period, and it reaches 61cm within two years after the disaster) are extracted then. Combined with the deformation time series measurements obtained, the creep deformation's time series evolution characteristics are emphatically analyzed subsequently. The results indicated that the Baige landslide was gradually stable in the two years after the disaster. In addition, the monitoring results shew that the Xiaomojiu landslide in the 5km upstream of the Baige landslide is also active. The relevant results would like to provide a reference for the research in disaster prevention and reduction in this region.

Key words: Baige landslide, TS-InSAR, creep deformation, deformation evolution characteristics, Xiaomojiu landslide