|M.Sc Student||Morik Smadar|
|Subject||Quantification of Surface Modifications Using Three-|
Dimensional Sensing for Paleo-Seismic Analysis
|Department||Department of Civil and Environmental Engineering||Supervisor||Professor Sagi Filin|
|Full Thesis text - in Hebrew|
Most earthquakes occur along plate boundaries or near them. Earthquakes result from plate tectonics and are the expression of elastic rebound of blocks of rocks. Tectonic movements cause slow accumulation of strain in the rocks, which are ongoing until reaching a critical point, where strain release causes elastic rebound, and generates a sudden movement and displacement of blocks of rocks. Earthquakes inflict a significant detriment on society, in both human lives and property, and therefore motivate studies and allocation of financial resources for their research. An important aspect in this context is the study of past earthquakes as a means to infer about future ones. Retracing past earthquake magnitude is complex and common indicators are based mostly on subjective evidence of the extent of destruction. The common approach is based on stratigraphic studies which are intrusive and cause environmental damage. As they cannot be carried in all cases, non-intrusive approaches e.g., studying earthquakes effect on the earth’s surface, are of great value.
In recent years, data acquisition technologies for modeling 3D environments have become widely available, with laser scanning being the prominent among them. Use of laser scanners facilitates dense and accurate topographical information about the surface. Nonetheless, processing such data is complex because of volume, span, and dimensionality concerns. The study examines the feasibility of laser scanning data to analyze past earthquakes effect on the surface topography. It then studies methods for documenting and characterizing surface features which allow to quantify the expression of earthquakes and faults in a way which is non-subjective and is not site specific. The focus is on the development of autonomous methods for the extraction of morphological features, e.g., fault lines or channel paths, which indicate the existence of faults. For the characterization, the research proposes a multi-scale approach which echoes the varying magnitude of surface features and enables overcoming the inherent complexity of natural topography. The research also studies fault characterization in two different scales: large and medium, and consequently two scanning settings are examined, airborne and terrestrial. Methods which adapt to the two different scanner types are offered as the data they provide are different in characteristics. Application of the proposed methods on different sites shows successful detection and reconstruction, which provide essential information for analysis of the related earthquakes.