NEESR-SG: Seismic Performance Assessment in Dense Urban Environments

Investigators: Andrew Whittaker

Funding Source: National Science Foundation

Start Date: 2008  |   End Date: 2014

Abstract: In our cities, buildings are constructed in clusters (the city block). Ideally, they should be designed to resist earthquake forces as clusters of buildings, because the response of one building can affect the response of neighboring buildings. However, the interactions between densely spaced buildings are not captured in current design practice, because buildings are typically designed as isolated structures. Soil-structure interaction (SSI) effects on closely spaced low- and medium-rise buildings are poorly understood. For example, it is not clear how basements of different sizes affect how the ground shakes these buildings. Recent earthquakes have damaged groups of buildings in cities, but it is difficult to learn from these observations due to the lack of documentation of the ground motion and building performance. However, a comprehensive program of realistic scaled centrifuge experiments, where the input motion, ground conditions, ground response, and structural response can be carefully tracked, followed by back-analyses of these model tests, can be employed to enhance the profession?s understanding of SSI effects of buildings in a dense urban environment.

The unique capabilities of the NEES UC Davis centrifuge will be used to advance our understanding of SSI effects for clusters of buildings so that reliable assessments can be made. In a centrifuge, a box containing soil and model buildings is spun at a rotational acceleration of 50 g so that a 2 foot thickness of soil has the same stresses as a 100 foot thickness of soil. Building models are scaled similarly so realistic responses are measured in these experiments. The testing program will develop a database of well-documented model case histories of building performance within a dense urban environment at sites undergoing moderate and severe ground shaking with and without ground failure. Researchers can then use these experimental results to advance our understanding of these phenomena and our ability to analyze them. Physical experiments followed by numerical simulations will allow us to develop guidance for designers and policy makers on how clusters of buildings perform during earthquakes.

This project will advance fundamental science and knowledge in engineering with substantial intellectual benefits to both geotechnical and structural engineering disciplines. Both disciplines will contribute to and benefit from the development and deployment of an integrated performance-based seismic design and a robust loss-estimation methodology. The project will also train Ph.D. students, bring in undergraduate students from a teaching university, engage under-represented students, impact building code development and performance-based seismic design, and outreach to a broad spectrum of end-users by emphasizing web-accessed media. We will also translate the challenge of solving this realistic problem to undergraduates (and others) via a Shaking of a City Block shaking table competition to have students consider the effects of adjacent structures and soil on seismic performance. Data from this project will be made available through the NEES data repository.