Enhanced Methodologies for Improved Ground Kinematics Interpretation in the Context of Landslide Early-Warning Systems

• Author / Creator

  Sharifi, Sohrab

• Landslides are widespread geohazards in Canada that cause loss of millions of dollars annually directly and indirectly. Such geohazards are prevalent in Canada’s landforms which compromise the safe operation of infrastructures and public safety. Not only stabilizing an area with this scale is financially beyond feasible, but many options are also next to insufficient. A reliable risk management strategy would advise to keep observing the ground to ensure the hazard is adequately mitigated. Monitoring programs are therefore becoming indispensable tools and an integrated pillar in the modern practice of geohazards management. Early-warning systems (EWSs) are a robust tool to this end and developing such systems calls for collaborative efforts and scientific exchanges between different fields considering that kinematical inputs into EWSs come from a variety of means of measurements.
  Technology has improved landslide monitoring in many aspects by granting high spatio-temporal resolution of readings, millimetric accuracies, (near) real-time acquisition and remote data retrieval. Technology in this context comes with certain limitations as well that impede its further incorporation into EWSs. The quality of received data determines the performance of an EWS and thus effectiveness of mitigative actions. The high sampling rate of some instruments (e.g., Shape-Accel Arrays or GNSS units), subject to the presence of scatter (noise), obscures the true displacement of the ground. As a result, the slope’s kinematics will not be fully understood, and the onset of acceleration cannot be simply detected to raise the alarm. Data filtration is then a significant constituent in an EWS yet poorly addressed. In addition to in-situ methods of monitoring, remote sensing techniques such as space-borne interferometric synthetic aperture radar (InSAR) are another source of recording ground displacements. InSAR constantly illuminates a footprint by broadcasting microwaves to the ground and records the backscattered waves which can be transformed into displacement values. However, the application of InSAR in landslide monitoring suffers limitations too. SAR sensors are only sighted to displacement along its line-of-sight (LOS). As a result, the velocity map it provides is not a full reflection of the direction and magnitude of movements. The added ambiguity due to this InSAR’s limitation prohibits direct incorporation of the results in EWSs. A decomposition of LOS velocity in three dimensions is then necessary. Common decomposition methods are associated with certain assumptions. The reliability of EWSs heavily revolves around the precision of measurements but little insights on the accuracy of these methods exist.
  This research aims to devise and pursue methodologies to enhance the kinematics interpretation respecting the present limitations in developing modern EWSs. The first element of this thesis is data analysis in which the simple moving average is compared against Gaussian-weighted moving average and Savitzky-Golay filters. The impact of using each in interpreting the landslide kinematics is studied as well as detecting the onset of acceleration and forecasting failure time using the inverse-velocity method. The second element is dedicated to examining different approaches in decomposing InSAR’s LOS velocities: ignoring the northward component, surface-parallel flow method, aspect-parallel flow method and steepest terrain following method. This is carried out by quantifying their accuracy in light of LOS estimations' accuracy and the mathematical impact of assumptions associated with each method. The outcome of this research is expected to assist geoscientists with implementing alternative methodologies in EWSs which leads to an improved reliability by acquiring a more accurate and truthful kinematics of movements.

• Subjects / Keywords

  • Landslide
  • Early-Warning System
  • Kinematics

• Graduation date

  Spring 2024

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/r3-mq82-8793

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.