A Block Ramp Errors Correction Method of Planet Subpixel Offset: Application to the 2018 Mw 7.5 Palu Earthquake, Indonesia

Abstract

Planet optical images have a short revisit period and high spatial resolution, so they have great potential in surface offset monitoring. However, few studies have focused on the error source analysis and accuracy improvement in the offset field of Planet images. In this study, we investigate the optimal band selection of the Planet images and the error sources for subpixel offset monitoring, using the 28 Sept. 2018 Mw7.5 Palu earthquake as an example. The results show that Band 2 is the optimal band for surface offset monitoring and ramp errors are the dominating error source in the offset field. Considering the imaging characteristics of Planet satellites, we propose a block ramp errors correction method. Our proposed method can improve the accuracy of the Planet subpixel offset by 25% in comparison with the traditional method. We also compare the Planet images subpixel offset result with the Sentinel-2 subpixel offset result and find they are in good consistency. We combine the corrected Planet images offset and SAR images offset to solve the complete 3D deformation field of the Palu earthquake, which shows that the ruptured fault is a left-lateral strike-slip fault accommodated with normal component in the Palu-Koro fault. The earthquake was dominated by south-north and vertical deformation, with the maximum deformation of 6 m and 4 m in the horizontal and vertical directions, respectively. This study also has set a good example for using the Planet images for natural hazards monitoring (such as earthquakes and landslides).