Evaluating State-Of-The-Art Methodologies for Post-earthquake Structural Reconnaissance

Abstract

Following an earthquake, there is a need to rapidly collect perishable data to evaluate the seismic performance of structures and aid in decision making regarding structural integrity and safety. Traditional reconnaissance methods, such as visual inspection and photographing of damage by individual researchers, can be time consuming and dangerous, particularly during periods of heightened aftershock activity. These reconnaissance methods are also limited in their effectiveness, as there are often gaps in damage documentation and visual records, resulting in challenges to correlate reconnaissance data with other layers of information such as instrumentation records or structural models. Newer reconnaissance techniques, such as terrestrial laser scanning, structure from motion\multi-view stereo (SfM\MVS) photogrammetry, and unmanned aerial systems photography/videography, hold substantial promise to mitigate the aforementioned shortcomings of traditional post-earthquake reconnaissance methods. However, to date, these technologies have been minimally exploited relative to their potential. This paper explores the capabilities and limitations of these technologies with application to a case study consisting of a reinforced concrete building with precast concrete floors damaged during the 2016 Kaikoura New Zealand Earthquake. The structure was extensively surveyed using several state-of-the-art methodologies to document the severity and distribution of damage resulting from the earthquake. Key factors that influenced the quality of data gathered are discussed along with limitations discovered for the utilized methods. Finally, procedures and challenges to combine data to create a detailed 3D model from the different survey methods (e.g. 3D laser scans and SfM/MVS) are presented. This model provides detailed damage documentation with limited visual gaps and can be used in conjunction with other layers of reconnaissance data to aid in the validation of detailed structural analysis models. The presented work shows that terrestrial laser scanning and SfM\MVS have significant merit for capturing structural damage when compared to traditional methods. At the level of detail of the surveys completed, these techniques successfully captured cracks with widths less than 1 mm using SfM\MVS and cracks with widths of approximately 2 mm using high-resolution terrestrial laser scanners, allowing for 3D visualization of structural damage that cannot be seen using conventional means, providing the chance for better understanding of structural performance and proves its ability to aid in decision-making post an earthquake event.