The Marmara, Turkey Earthquake of August 17, 1999: Reconnaissance Report
Early in the morning of Tuesday, August 17, 1999, a magnitude 7.4 earthquake struck along the Anatolian fault in the northwestern region of Turkey. Epicentered approximately 11 km southeast of the industrial city of Izmit, the earthquake lasted 45 seconds and was felt over thousands of square miles in Turkeys most densely populated region. Commercial and residential buildings from Adapazari to Istanbul collapsed, resulting in large-scale loss of life. According to official government estimates (as of October 19, 1999), the earthquake killed over 17,000, and injured almost 44,000 people. Estimates of property losses (as of September 14, 1999) according to the World Bank range from $3 to $6.5 billion, which is equivalent to 1.5 to 3.3 percent of the Gross National Product of Turkey. It was the most devastating earthquake to strike Turkey since the 1939 Erzincan earthquake, which killed 30,000 people. According to official Turkish government estimates, the earthquake displaced more than 250,000 people. Approximately 120 tent cities were required for emergency housing. About 214,000 residential units and 30,500 business units were lightly to heavily damaged.
Within days, MCEER dispatched several researchers to the region - three of them simultaneously serving as part of the Earthquake Engineering Research Center (EERI) reconnaissance team - to examine the earthquakes impact. Their initial observations and impressions are reported in two publications, MCEER Response by M. Bruneau, J. Mander, W. Mitchell, A. Papageorgiou, C. Scawthorn and N. Sigaher, and in a Preliminary Report by C. Scawthorn. Both reports can be accessed from our web site at http://mceer.buffalo.edu/research/turkeyeq/default.asp.
MCEER sponsored a second reconnaissance trip to Turkey together with the Earthquake Disaster Mitigation (EDM) Research Center in Miki, Japan. Team members visited Turkey from September 28 to October 4 to conduct high level reconnaissance using satellite imagery, differential global positioning systems and in-field GPS-GIS interfaces. In addition, restoration activities already underway were observed and documented.
This report includes observations from both these reconnaissance trips. It is the product of many authors representing several disciplines and, while not a final assessment of the topics addressed, represents an interim earthquake engineering evaluation of the natural, built and social environments. As noted by several of the authors, the analogies between the North Anatolian Fault Zone in Turkey and the San Andreas Fault in the United States are strikingly similar. The observations and conclusions herein form a springboard for future collaborative research efforts, which will advance societys ability to better withstand the destruction caused by earthquakes throughout the world.
By Charles Scawthorn
EQE International, Inc.
The August 17, 1999 Mw 7.4 Marmara earthquake is a devastating catastrophe and great human tragedy for the Turkish people. Approximately 17,000 fatalities and 44,000 injuries occurred, with an estimated 20,000 collapsed buildings displacing more than 250,000 people, making it one of the worst natural disasters in recent decades.
The affected region around Izmit Bay is heavily industrialized and accounts for perhaps 10% of Turkeys GDP. Combined with other economic problems, the earthquake is expected to be a severe burden on the national economy, reducing national GNP by 0.6~1.0 % (World Bank, 1999).
The earthquake should have come as no surprise, since the long history of earthquakes is well-known (Ambraseys and Finkel, 1995), Table 1-1. Additional evidence for this events potential was the clear pattern of sequential segmented rupturing of the North Anatolian Fault Zone (NAFZ) as pointed out by Toksöz et al. in 1979 and Stein et al. in 1997, (discussed by Papageorgiou in Section 2).
The approximately 125 km of fault rupture on the North Anatolian Fault Zone is clearly analogous to situations in other parts of the world, most notably with the San Andreas fault in the San Francisco Bay Area of California. The strong ground shaking due to this fault rupture, combined with soft soils around the perimeter of Izmit Bay and other areas (e.g., Adapazari), resulted in significant geotechnical effects and permanent ground deformations (discussed by Mitchell and Holzer in Section 3). These geotechnical effects were consistent with those associated with other recent major earthquakes, and resulted in streets and buildings on the bayshore being submerged 1~2 meters in this event, and Adapazaris water distribution system being virtually destroyed.
However, the most dramatic damage and greatest contributor to the disaster was the widespread collapse of numerous multi-story reinforced concrete apartment blocks. Almost the only building type in the region is non-ductile reinforced concrete frames with hollow clay tile infill which, combined with soft stories, results in a pancake type of collapse (discussed by Bruneau in Section 4). Requirements for proper earthquake-resistive construction exist in the Turkish building code, which is a very modern code. Why werent these requirements adhered to? One important factor has been the rapid development of Turkey in general, and particularly the Marmara region. From 1990 to 1997 for example, the province of Kocaelis population grew 26%. Rapid development of the Marmara region overwhelmed local governments ability to monitor construction, and led to unregulated building, resulting in inadequate lateral force systems in buildings.
This lesson is further emphasized by the performance of structures designed and constructed by more centralized organizations with access to modern engineering, such as the transportation systems (discussed by Mander in Section 5), industrial facilities (discussed by Johnson in Section 6) and lifelines (Section 7). In these cases, relatively little damage occurred, and the major motorways, water treatment and transmission systems, gas systems, and national power grid, were all functional within hours of the earthquake. Industrial facility performance was more mixed, with some dramatic damage, such as at the Tüpras refinery (site of a major fire), but many facilities performed very well.
The human dimensions of the August 17 earthquake continued for many days, as Turks and rescuers from around the world struggled to find and save those trapped in the literally thousands of collapsed buildings. This task, which re-played similar efforts seen in Mexico City in 1985, Armenia in 1988 and elsewhere, is simply overwhelming. As Mitchell discusses in Section 8, the organization and technology does not currently exist to perform this task with any real effectiveness, so that prevention of the problem, via effective retrofitting, is the solution. The cost of disasters is further increased by the resources that must be devoted to tent cities and more durable temporary housing, debris removal and other necessary tasks, as discussed by Webb in Section 9. Both sections 8 and 9 also offer excellent insights into the social and political ramifications of such a trauma to the social fabric.
Very interesting in this earthquake was the application of new technologies for rapidly assessing and reacting to the disaster, in near real-time. Remote sensing, GPS, GIS and emergency decision support systems offer the promise of efficiently employing available resources in a timely manner, thus in the future, potentially saving those who are currently lost. Eguchi and co-workers in the final chapter discuss current efforts at applying and understanding these technologies, which are an extremely promising area for further research.
In a sense, the August 17 Marmara earthquake was a narrow-banded event. That is, considering the entire spectrum of the built environment, the damage resulting from the event, while substantial, was generally within the resources of Turkey to manage and even tolerate, with one exception. The exception was the dismal performance of the reinforced concrete frames, virtually ubiquitous in the region. The collapse of thousands of these buildings transformed this earthquake from a damaging event to a catastrophe. Within the spectrum of the built environment, only this aspect was a spike. Design and construction of reinforced concrete frames to withstand strong earthquake motions is possible, and the principles are well understood by Turkish engineers. Unfortunately, the rapid development of the region overtaxed the ability of the society to assure that these principles were followed. The result was inadequate buildings, when there need not have been, and a tragic catastrophe. The ultimate lesson therefore is that building and development is simply not a physical process - governmental institutions and social processes must develop in parallel, to keep up with the physical demands and assure minimum acceptable standards of construction and public safety. The alternative is seen in Figure 1-1, thousands forced to stand by, while victims die in the rubble.
For Further Information:
University at Buffalo, State University of New York
107 Red Jacket Quadrangle
Buffalo, New York, USA 14261