Keywords: structural modeling, city modeling, 3D GIS, disaster mitigation, seismic response, dynamic analysis.

Evaluating vulnerability of city infrastructure, especially building structures, is essential to estimate casualty and economic losses during severe earthquakes. It is estimated that over 75% of fatalities in earthquakes are caused by building collapse [1]. As many uncertainties exist in estimating losses of earthquakes and related disasters, a statistic based approach is usually employed. The well-known HAZUS system [2], for instance, gives estimating results on average upon inputted scenarios.

Earthquakes are low frequency and high consequence natural disasters. Though their occurrence is still difficult to be precisely predicted, its consequence, such as response of structures, can be analyzed with sufficient accuracy on providing enough data. A major obstacle, however, is the difficulty to obtain huge amount of data of buildings in a city in conducting simulations for disaster mitigation. To deal with this problem, HAZUS system categorizes buildings into 36 types according to building materials (wood, masonry, steel, concrete), structural types (frame, shear walls etc.) and building height (low, mid and high-rise) [3]. Each building type is studied using seismic spectra [3,4]. Probabilities of damage, known as fragility curves, are deduced from these spectra. To categorize each building by these types, however, users of HAZUS need to make an inventory of buildings. This is a task that requires huge amount of manual jobs.

On the other hand, recent techniques of Geographic Information System (GIS) can process information of cities at level of individual buildings. Technologies have been developed to recognize architectures (shapes of build structures) and to reconstruct 3D scene of architectures for a block of a city [5,6]. Applications based on such technologies, such as the car navigators and 3D digital maps, are getting popular. Though only architecture models are constructed, the information yield (shapes of building etc.) can be further utilized to generate structural model of buildings.

This study makes a maximum utilization of 3D GIS data in constructing structural models. The available GIS data are a profile of a building in 2D and a height of this building. Profiles of buildings are given in polygon. To build a simplest Single Degree-Of-Freedom (SDOF) structural model, a total mass and shear stiffness of a building need to be obtained. Natural periods, the most important dynamic feature of building structures, can be estimated using empirical equations with height and dimension of buildings. Original GIS data don't give the information of number of stories of buildings. As a building is designed according to engineering codes with predefined usages, number of stories can be obtained by a reasonable guess using the height and profile of this building. The mass density of each story can also be obtained using the same method. This way, the total mass of one building can be computed out. Stiffness can therefore be calculated using a basic equation of dynamic relation. A well adopted simplified structural model for buildings is a Multi-Degree-Of-Freedom model, which has a condensed mass at each story of a building. Using superposition method a MDOF system can be condensed into a SDOF system by providing a mode shape related a natural frequency.

This approach has been implemented by constructing building models of Bunkyo district of Tokyo using given 3D GIS data [7]. With the merit of no need (or less need) of city building inventory that consumes huge man power, the proposed approach can be further improved for conducting analysis of urban disaster mitigation.

1

Department of Humanitarian Affairs (DHA), "Disaster Mitigation", (2nd edition.), 1994.

2

HAZUS System by the Federal Emergency Management Agency (FEMA), http://www.fema.gov/hazus.

3

HAZUS 99 (SR2) Advanced Engineering Building Module, Technical and User's Manual, http://www.fema.gov/hazus.

4

R.W. Clough, J. Penzien, "Dynamics of Structures", (second edition), New York, McGraw-Hill, Inc., 1993.

5

P. Debevec, C. Taylor, J. Malik, "Modeling and rendering architecture from photographs: A hybrid geometry- and image-based approach", In H. Rushmeier, editor, Computer Graphics, Proc. SIGGRAPH 96, Addison Wesley, 11-20, 1996.

6

C. Brenner, N. Haala, "Automated Reconstruction of 3D City Models", In Abdelguerfi, M. (ed), 3D Synthetic Environment Reconstruction. Kluwer Academic Publishers, 75-101, 2001.

7

P. Zhu, M. Abe, J. Kiyono, "Managing 3D GIS Data for Simulating Urban Earthquake Disasters", Proceedings of JSCE Annual Conference, Japan, 2004.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £135 +P&P)