Key points are not available for this paper at this time.
• Joint decomposition of InSAR and laser scanning improved NS displacement by 90 %. • WLS method integrates InSAR with laser scanning and other 3D displacement data. • Approach monitors complex deformations, ensuring robust and accurate results. Three-dimensional displacement estimation is essential for understanding surface changes caused by natural and anthropogenic processes, as it allows for a more complete interpretation and improved risk assessment. Interferometric Synthetic Aperture Radar (InSAR) is widely used for displacement monitoring due to its high resolution and accuracy. However, it measures displacements only in the satellite line-of-sight (LOS) direction. Some of the processing techniques enable LOS displacement decomposition. However, the accuracy of the north–south (NS) component is low. To overcome this limitation, this study presents an approach that integrates LOS displacements with 3D displacement data obtained from laser scanning (airborne − ALS and UAV-borne − ULS), using the Weighted Least Squares (WLS) method. Validation was performed against reference surveying measurements. The integration with ULS data achieved RMSE values of 0.040 m (NS), 0.031 m (EW), and 0.031 m (Up), while ALS integration yielded 0.102 m, 0.075 m, and 0.066 m, respectively. The most significant improvement was observed in the NS component, with an accuracy gain 90 %. The study also compared weighting strategies within WLS, showing the best performance when accuracy estimates of input data were available. The proposed decomposition model was tested at a challenging site with high-magnitude and complex 3D deformations, which made only the DInSAR method effective. However, products from other InSAR processing methods could be potentially used if they are reliable. Similarly, surface models created from point clouds obtained through various methods, such as dense image matching, can be utilized, as demonstrated in this investigation.
Wielgocka et al. (Thu,) studied this question.