SAC Phase 1 Analytical Studies of Building Performance

Project Title:

Interpretation of SAC Survey Data on Damaged Welded Steel Moment Frames following the Northridge Earthquake

Sub-contractors:

William E. Gates, S.E., Dames & Moore, Inc.

Project Summary:

This report summarizes the key finding from the interpretation of data collected on welded steel moment frame (WSMF) buildings surveyed for damage following the 1995 Northridge earthquake. The report also identifies the areas of uncertainty and makes recommendations for future data gathering and analysis to confirm the trends identified in this study.

Under SAC Task 2, building-specific data was collected and analyzed by a series of investigators. Bonowitz and Youssef (1995) collected data on 89 WSMF's from engineers that investigated the buildings for damage to the welded moment connections. Eight of the buildings included in this database were studied in detail with analyitical models and simulated ground motions by an independent group of selected engineers under SAC Task 3. Data from these 8 buildings was more extensive, including both earthquake demand as well as connection capacity.

Ground motion data was analyzed and developed into contours for plotting by Somerville, Graves and Saika (1995). This data included peak ground accelerations, velocities, and building spectral response. The contour data was plotted on a series of maps along with the location and degree of damage for each of the buildings surveyed by the engineering firms. A second database on damaged buildings was developed by Michael E. Durkin (1996) from an inital set of 1,284 buildings which was narrowed down to 150 WSMF buildings by using a random selection process. This database of damaged steel frame buildings was also superimposed on the contour maps along with the buildings surveyed by the engineering firms. These maps provide a basis for studying the effects of earthquake demand on reported building damage based on geographic location. The maps were plotted by The California Office of Emergency Services, Geographic Information Systems Group.

Gates and Morden (1995) conducted interviews with six engineering firms, two testing laboratories, and Building and Safety officials of the City of Los Angeles to document the lessons learned from the WSMF damage in the Northridge event. This survey provided qualitative checks on the database collected by Bonowitz and Youssef (1995). The engineers interviewed were the same ones responsible for reporting on more than 50% of the buildings in the database collected by Bonowitz and Youssef.

The statistical database collected by Bonowitz and Youssef (1996), as well as the maps of earthquake intensity versus building damage, provide valuable insight into the extent and location of earthquake damage to the WSMF buildings in the Northridge event. Due to time and budget limitations, certain shortcuts had to be made in the database that was collected and in the method of statistical analysis for interpretation. The results that have been gleaned from this database indicate significant trends of importance both to code writers, as well as building officials.

The areas of further data gathering and analysis that could fruitfully be implemented include the following:

1) The building database should be expanded to include the 400+ buildings currently being investigated under the City of Los Angeles Mandatory Inspection Ordinance for WSMF's.

2) The information on each building that is now bieng gathered includes the type of damage and the location of each connection. It is strongly recommended that this information be retained on a connection-by-connection basis rather than on a floor frame basis.

3) The data that has been collected on the 89 buildings should be expanded to include details on the type of foundation or base support for the steel moment frame. There are indications from identical buildings located on the same site that moment frames founded on flexible foundations experience little or no damage while adjacent frames supported on rigid or basement walls experience heavy damage.

4) All strong-motion instrumented WSMF buildings that have experienced damage or not should be modeled and analyzed in detail to develop the demand-capacity relationships useful in completing the database and assessing the basic factors leading to damage.

5) Multivariate statistical methods should be used in assessing the building database, as well as the earthquake demand database. The two-dimensional statistical approach that has been followed to date does not provide a direct correlation between the many variables that play a role in the overall damage statistics for the buildings.

6) A simple statistical model should be constructed and analyzed using the multivariant approach to predict, on the basis of recorded ground motions and observed building characteristics, whether certain classes of steel moment frame buildings will be damaged.

7) Further data needs to be developed on the earthquake demand relationships. How does the soil stiffness and depth to bedrock relate to the damage observed in the buildings? How does the groundwater level and soil instability due to liquefaction affect the damage to the WSMF? Is spectral velocity or spectral acceleration a better measure of earthquake damage potential than peak ground velocity or peak ground acceleration? These are examples of a few of the key areas that require study.

Finally, it is suggested that SAC review has zone of building damage identified in the maps and compare it with the zone of inspection specified by the City of Los Angeles and its neighboring communities. It appears that the zone of potential damage may need to be extended further east towards downtown Los Angeles in the area south of the Santa Monica Mountains to capture all of the buildings that may have been damaged in the Northridge earthquake.

SAC Home

SAC Steel Project
c/o Earthquake Engineering Research Center
1301 South 46th Street
Richmond, CA 94804
(510) 231-9477
FAX: (510) 231-5664
sacsteel@sacsteel.org