Kinematic Analysis of Small and Slow-Moving Landslides Using Pleiades-1 Satellite Data

Although they may have small velocity values, small slope failures can cause damage to facilities such as roads and pipelines. The main goals of this paper are to detect and map, and quantify the kinematics of small and slow-moving landslides in Kutlugün, Northeastern Turkey. Object-based image analysis (OBIA) and rule-based classification techniques were utilized to detect and map the small and slow-moving landslides. The horizontal displacement of the landslides was investigated using the sub-pixel image correlation method, Cosi-Corr software, and Pleiades-1 images. Kalman filtering method and Real-Time Kinematics-Global Positioning System (RTK-GPS) observations were utilized to formulate a kinematic analysis model for the landslides. A total of 123 small landslides covering an area of approximately 413.332
$\text{m}^{2}$
were detected in the study area. The displacements determined by image correlation compare very well with the RTK-GPS measurements, with a maximum deviation of 0.86 mm. The movement rate of the small landslide from RTK-GPS results ranged from 0.80– 8.28 mm during the six-month monitoring period. The average displacement value for all the monitoring points is 9.88 mm, while the average movement rate is 3.11 mm during the monitoring period. Compared to the deformation obtained using only the RTK-GPS measurements, the optical image correlation produced a more coherent deformation pattern and more detailed information on the extent and distribution of deformation. The results of the kinematic analysis suggest that the entire body of the small landslide moved slowly during the monitoring period.