Abstract

The recent advent of terrestrial interferometric synthetic aperture radar (InSAR) has greatly enhanced the ability of monitoring slope deformation. However, the displacements obtained are one-dimensional, offering little insight into the underlying deformation mechanism. This study summarizes an approach for obtaining three-dimensional slope displacement vectors through the integration of InSAR and two-dimensional image feature tracking (FT) technologies. The method, referred to as InSARTrac, uses a single digital camera oriented in the InSAR line of sight (LOS) generating time-lapse imagery, from which FT extracts (sub-) pixel shifts of pixel clusters. The 1D LOS InSAR measurements are vectorially combined with the 2D normal to the LOS FT measurements to obtain the 3D displacement vector. Bench-scale target displacement tests using a high precision translation for displacement and reference gave a 3D accuracy of 0.05 mm at a distance of 13 m, which corresponds to 1.3 mm at 500 m, assuming linear behaviour. These initial results indicate that InSARTrac can provide a reliable means for obtaining accurate 3D slope displacement vectors remotely and without the use of reflectors. Current studies are focused on implementing InSARTrac in a number of different field environments to investigate outdoor measurement accuracy and the range of potential applications.

Introduction

Terrestrial interferometric synthetic aperture radar (InSAR) has greatly enhanced the ability to monitor slope deformations in real-time and gather time-series of displacements at incredible resolution. Some of the powerful aspects of InSAR monitoring relate to its ability to operate independent of lighting and most atmospheric conditions, fully remote (reflectorless) operation, sub-millimetre measurement resolution and full domain coverage. These attributes have proven valuable for tracking the temporal development of slope deformations and for making early warning predictions concerning potential collapse events. However, the displacements obtained with InSAR are one-dimensional in the line-of-sight (LOS) direction, and therefore offer little insight concerning the causative deformation mechanism and subset of geologic structures involved. In order to relate the measured deformations to subsurface structural geologic boundary conditions, the full 3D displacement vector pattern is needed. One method for obtaining additional dimensional information involves combining the results from multiple simultaneous InSAR devices at different positions [e.g. 1–3], but this approach significantly increases equipment costs.

This content is only available via PDF.
You can access this article if you purchase or spend a download.