skip navigation

The Turkey, Taiwan and Mexico City Earthquakes: Some Questions and Answers

Introduction | Testimony | PowerPoint Presentation

Our ultimate goal in seismic monitoring, research, and development is to be able to predict with some degree of certainty when a seismic event will occur. How well are we able to correlate the events leading up to a seismic event to the actual event?

We can quantify the long-term seismic potential, but currently we cannot predict specifically when, where, and how severe future earthquakes will be. We assign probabilities of occurrence of specific earthquakes over a broad time frame. For example, in California seismologists estimate the probabilities of occurrence within the next 30 years of specific magnitude earthquakes on specific segments of the San Andreas Fault. This information is very helpful for regional planning, engineering lifeline systems (e.g., transportation systems, water supply, etc.),organizing and implementing good engineering practices, and increasing public awareness. However, it has not been possible to identify a particular event or phenomenon as a reliable precursor of an imminent earthquake.

The difficulties in predicting earthquakes encourage the development of engineering knowledge and practices as a principal means of reducing seismic risk. Simply stated, the application of earthquake engineering and planning improves the safety and reliability of our built environment, irrespective of our difficulties in forecasting a specific seismic event.

How well are we able to track the migration of stress along fault lines?

It is virtually impractical to track the migration of stress, but highly advantageous to track the migration of strain. Strain can be tracked by dense GPS networks, and is an indicator of where stresses are accumulating. To know the stress, we have to know both the strain and certain mechanical properties of the crustal rocks experiencing strain. These mass properties can be inferred from modeling past earthquakes and acquiring high quality data during future earthquakes. It is important to recognize that crustal rocks have highly variable material properties that are not well understood nor satisfactorily modeled currently.

Following so-called "mainshocks", there are usually a series of aftershocks that can range from light to moderate. What can we learn from aftershocks?

Aftershocks indicate where stress concentrations remain after an earthquake, and help to delineate where the causative fault is located. Using existing and portable seismic strong motion instruments, aftershocks provide data from which site response can be evaluated.

Site response is the way the local soil, ground water, and rock conditions change the incoming seismic waves, frequently increasing the amplitude of acceleration and altering the frequency content of the waves. There is a need to understand site response better. Locations vulnerable to the undesirable effects of ground motion through amplification of acceleration and changes in frequency content need to be identified. These locations also need to be characterized with respect to the potential earthquake performance of buildings and lifelines that are situated there.

How is the likelihood of aftershocks computed?

They are estimated through the statistics of past earthquake sequences.

What is the average size of an aftershock for a magnitude 7-plus earthquake?

The largest one would be magnitude 6 plus, and there would be many smaller ones.

Is it possible to determine how much of the damage reported is due to aftershocks? If so, how much of the damage reported in the Turkish, Taiwanese, and Mexican earthquakes was due to aftershocks?

Determining the damage from aftershocks is frequently difficult. Often, aftershocks affect structures that were damaged during the main shock. A precise basis, therefore, is generally lacking for quantifying how much damage was added to the main shock damage. Because the preponderance of aftershocks have substantially lower magnitudes than the main shock, the amount of damage caused is usually much less than that caused by the main shock. I entered the city of Gulcuk about three weeks after the August 17, 1999 Turkish main shock just after a 5.8M aftershock (a large and significant aftershock) occurred with an epicenter very near Gulcuk. About ten previously damaged buildings collapsed completely or partially, and one building slid into the Bay of Izmit. I was told that about 6 people were killed. Although these effects are significant, they are substantially less than the hundreds of buildings that collapsed and the thousands of people who were killed in the Gulcuk area as a consequence of the main shock.

How does one effectively use a portable seismic network in a post-earthquake situation? Are there advantages to portable networks over "stationary" or "pre-positioned" monitoring networks? If so, what are they?

Portable networks have significant advantages because they are deployed at critical locations throughout the earthquake-affected area, and large amounts of data can be collected. The stationary network cannot be moved, and must "wait" for the earthquake to occur at a nearby location.

What are the largest obstacles to collecting accurate and timely seismic data? What will it take to overcome these obstacles?

The most significant obstacles are obtaining the appropriate funding and implementing effective management. The proposed Advanced National Seismic Network will help significantly to overcome these obstacles.

How much more do we know now than we did five-years ago about tectonic processes?

Our conceptual framework has not changed, but new data from GPS networks and strong motion records from large earthquakes are providing new insights into the physics of earthquakes and tectonic processes.

In your opinion, what are the most effective methods currently in place to reduce seismic hazards? What are the most promising initiatives on the horizon?

This is the most important question in this sequence and the one of greatest consequence for citizens of the United States.

To start, seismic hazards are primarily active faults and the earthquakes that occur along them. Hazards also involve locations prone to ground failure during an earthquake. In some instances we can stabilize these sites, if we can identify them.

Risk is the likelihood that a building or facility will collapse or be damaged during an earthquake. We can do little to reduce the hazard embodied in an active fault or a major earthquake, but we can do alot about the risk to the structures that we design and build. It is important to remember the frequently quoted observation that earthquakes do not kill, but collapsed buildings and facilities do.

The most effective methods now in place to reduce seismic risks are the design codes, technologies for building and lifelines rehabilitation, planning policies and emergency response measures, siting guidelines, and site stabilization technologies to reduce the hazards of ground failure. Each of these methods needs to be nourished by support for basic and applied research so that improved technologies will continue to be developed and applied to reduce the potential for loss of life, property, and community.

With respect to rehabilitation, research is needed on advanced materials and technologies to retrofit vulnerable buildings and lifelines. Cost-effective retrofit procedures must be stressed. With respect to siting and the control of ground failure hazards, research is needed to understand better the complexities of surface fault rupture and develop improved mapping and zoning approaches for locations with the potential for surface faulting. Research on ground failures like landslides and soil liquefaction is important, as is research on how local soil conditions modify seismic shaking and how these conditions can be identified and designed or zoned for.

From the perspective of planning and emergency response, we need substantial research from applied social scientists on how to develop effective programs at the local municipality and state levels that deal with potentially lethal buildings. We also need research on the appropriate forms of financial incentives, zoning, and information dissemination to engage community cooperation in earthquake loss reduction.

As indicated in my written and oral testimony, there is substantial risk in the United States with respect to non-ductile concrete buildings, many of which are high occupancy buildings that threaten the lives of many residents. There is an urgent need to develop an inventory of buildings in seismically active areas of the US to identify where non-ductile concrete buildings and other vulnerable structures (e.g., unreinforced masonry and open-first-story timber frame apartments) are located. All citizens should have access to knowledge about the buildings they live and/or work in, but this type of inventory is not currently available.

The Network for Earthquake Engineering Simulation (NEES) currently be funded through NSF is one of the most promising initiatives on the horizon. This network will be a tremendous boost for experimental facilities, and will link experimentation through information technologies with advanced modeling and simulation.

The NSF network of Engineering Research Centers focused on earthquake engineering plays an indispensable role in educating the next generation of engineers and scientists in the development and application of technologies to mitigate earthquake and other natural disaster effects. These centers, headquartered at strategic locations throughout the country, link industry, government, and academia to tackle earthquake risk reduction as a national problem. The centers are organized to grow basic research through technical development with industry into demonstration projects that deliver new products, standards, and procedures. The centers are a real benefit for the country, and support for them should continue.

It is important to continue strong support for the earthquake engineering research sponsored by NSF. Much of this research contributes to improved design codes and siting strategies. There is a compelling need to continue the support for earthquake reconnaissance. Learning from earthquakes, such as the recent seismic events in Turkey and Taiwan, provides invaluable knowledge and data about how the natural and built environments actually perform. This is information is critical for improving our models, design codes, and planning processes.

Although basic science is a primary concern of the House Science Committee, the implementation of that science by means of improved engineering and planning is the most important outcome that an enlightened science policy can accomplish. Strong support is also needed for FEMA to translate the new knowledge gained from research on seismology and soils and structural behavior into building provisions, codes and standards, and training programs for those in engineering, architecture, construction, planing, and emergency response. The conversion of scientific discovery into saving lives and the preservation of property and community must be the result of governmental support for research on natural hazards. Agencies like NSF and FEMA help government realize the practicalities of that goal.

Turkey reportedly has a strike-slip fault zone similar to the San Andreas Fault in California.  What we have been able to learn about this type of fault through the recent seismic activity in Turkey and past activity in California?

Some significant lessons are that earthquakes transfer stresses to adjacent segments of a causative fault and may transfer significant stress to nearby faults. In addition, the rupture of a fault in a single earthquake can skip from one segment of the fault to another, across gaps that are several miles wide. This latter effect is well illustrated by the August 17, 1999 Kocaeli earthquake in Turkey where four distinct segments of the fault participated in a rupturing event that involved step-overs of several miles from one segment to the other.

What do we know about the migration of strain along the fault in Turkey?

The most recent cycle of migrating earthquakes on the Northern Anatolean Fault began in 1939 with a severe earthquake near Erzincan. Since that time severe earthquakes have been occurring along the fault trace, migrating in a westward direction. A new earthquake occurs every one or two decades. The August earthquake was a manifestation of this westward propagation of strain. Rupture during the Kocaeli earthquake skipped a segment of the fault, leaving a seismic gap. This gap ruptured during the 12 November 1999 Duzce earthquake. The existence of seismic gaps is another important lesson we have learned from observing earthquakes on various fault sytems. The identification of a seismic gap provides the potential location and evidence for a significant earthquake in the future.

Is there any way to predict the chances of an event in Istanbul in the near future?

Predictions of earthquakes affecting Istanbul are subject to the same limitations and uncertainties that are explained above. However, the recent earthquakes on the Northern Anatolean Fault provide some clues concerning the earthquake risk affecting Istanbul. We know that the Northern Anatolean Fault is undergoing a westward propagation of strain and rupture. The progression of rupture and earthquakes has left the fault segments nearest Istanbul as the next segments to rupture. The question is what segment will rupture? There at least two candidates: the Yalova segment that is about 50-60 miles south of Istanbul and the Northern Boundary segment that is about 15 to 20 miles south of Istanbul. Aftershock concentrations on the Yalova segment suggest that this segment may have taken on the heaviest stress concentration after the August earthquake and therefore is more prone to future rupture, but this evidence is far from conclusive. Which segment ruptures is of critical importance. Rupture on the nearest Northern Boundary segment would entail substantially more severe ground motion in Istanbul. The city is heavily populated, strategically located, and poorly prepared for a major earthquake in its vicinity.

Introduction | Testimony | PowerPoint Presentation