Issue No. 04 - July/August (2011 vol. 13)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MCSE.2011.59
Rudolf Eigenmann , Purdue University
Ayhan Irfanoglu , Purdue University
Understanding and mitigating the effects of disasters requires the integration of research and practice in multiple science, engineering, technology, and social science fields to create a world that's more resilient in terms of public safety, economic strength, and collective welfare. The January/February 2011 issue of CiSE explored hurricane prediction. 1 Hurricanes are one of many natural disasters threatening our populations and infrastructure. Recent events have made it clear that earthquakes—and the large tsunamis that they can cause—also represent significant threats against which we must prepare ourselves.
The earthquakes in Haiti (2010), Chile (2010), New Zealand (2011), and Japan (2011) have shown not only what colossal damage earthquakes can cause, but also the importance of good design, construction, and preparedness. The magnitude 7.0 Haiti earthquake in January 2010 devastated the vulnerable infrastructure of this Caribbean island nation and claimed more than 200,000 lives, making it "the most destructive event a country has ever experienced when measured in terms of the number of people killed as a share of the country's population." 2 By contrast, the February 2010 magnitude 8.8 Chile earthquake caused comparably little damage in a country known for its good seismic design and construction practices. The March 2011 magnitude 9.0 earthquake in Japan—the fourth largest earthquake ever recorded—caused limited damage from ground shaking, but the induced tsunami caused colossal destruction on a broad scale, reminding us of the 2004 Sumatra-Andaman tsunami that killed nearly 250,000 people in 14 countries around the Indian Ocean.
In both Chile and Japan, design and construction are subject to modern codes that consider the structural safety of the built environment when impacted by an earthquake. These codes are based on earthquake engineering research that investigates the behavior of structures and soil during earthquakes. In the US, the George E. Brown Jr. Network for Earthquake Engineering Simulation (NEES; www.nees.org) maintains an experimental- and cyber-infrastructure for such research. NEES includes 14 research laboratories that provide equipment for large-scale testing, such as large earthquake simulating platforms (shake tables) to generate realistic earthquake-like motions and centrifuges to simulate soil conditions during ground shaking. One of the research laboratories includes a wave basin to study tsunamis.
These experimental facilities are linked through a robust, user-requirements-driven cyberinfrastructure, the NEEShub. A key component of this cyberinfrastructure for collaboration is the Project Warehouse, the NEES repository for earthquake engineering research data. Furthermore, computational earthquake and tsunami research complements the physical experiments of NEES projects. Computational simulations let researchers explore parameters that can't be identified efficiently through physical tests because of the cost of such large-scale, destructive tests, the time needed to build specimens, or limitations of the current testing equipment and experimental setups. Furthermore, some large-scale experiments simply can't be conducted as physical tests—for example, it's impossible to test the earthquake response of the entire infrastructure of a metropolitan area.
In This Issue
The six articles in this special issue describe earthquake and tsunami research as well as the NEES infrastructure.
Three articles describe the latest state of technology in computational strong-motion seismology and earthquake engineering simulation and its need for high-performance computing methods. In "Large-Scale Earthquake Simulation: Computational Seismology and Complex Engineering Systems," Ricardo Taborda and Jacobo Bielak describe and illustrate their state-of-the-art, large-scale earthquake simulations, which include simulation of seismic wave propagation from the source fault to distant sites, nonlinear soil response, and interaction of the urban built environment with the surrounding ground during the earthquake shaking. In "Rupture-to-Rafters Simulations: Cyber-Enabled Unification of Science and Engineering for Earthquake Hazard Mitigation," Swaminathan Krishnan and his colleagues combine fault rupture, seismic wave generation and propagation, and nonlinear building response simulations to study earthquake hazards and building vulnerability in urban areas. In their article, "Petascale Computation for Earthquake Engineering," Muneo Hori, Takuzo Yamashita, and Koichi Kajiwara describe how they combine simulations of not only ground shaking and structural response but also human response to shaking, and then describe their results for simulations of the Tokyo metropolitan area.
In an article related to tsunami research, "Numerical Simulation of Complex Tsunami Behavior," Patrick J. Lynett and Philip L.F. Liu discuss challenges in simulating the generation, propagation, and nearshore impact of tsunami waves, including their effect on waterfront structures and inundation of coastal areas and towns.
Lastly, two articles describe the NEES network's computational and data infrastructure. Frank McKenna's "OpenSees: A Framework for Earthquake Engineering Simulation" presents one of the earthquake research community's most widely used open source simulation tools for structural and geotechnical analysis. Finally, in "The NEEShub Cyberinfrastructure for Earthquake Engineering," Thomas J. Hacker and his colleagues describe the NEES cyberinfrastructure, which facilitates computational simulation as well as the collection and sharing of earthquake engineering data.
The Future of Earthquake and Tsunami Research
Today, earthquake and tsunami research using physical experiments is the main scientific and engineering method. However, computational simulation techniques have made significant progress over the past decade, as evidenced by these articles. A new "hybrid" research infrastructure is also being developed, in which physical testing and computational simulation tools are used simultaneously.
The advantages of simulation technology—which has been discussed in more than a decade of CiSE articles—for earthquake and tsunami research is clear: computational methods can overcome limitations set by space and time dimensions, human abilities, economic resources, and legal constraints. Coupled with today's powerful high-performance computer platforms, these advantages promise significantly accelerated progress in the future. To truly realize this potential, however, several challenges must be overcome.
As is true for many other disciplines, the accuracy of the models representing the complex physical phenomena involved must be improved. Although midscale computer simulations suffice for a range of research problems, petascale and exascale algorithms must be developed for the most advanced solutions. Given the nature of the elements involved in this area of science, the calibration of results with data from field observations poses another challenge. A further major obstacle—and opportunity—is the sharing of information and research data, which will increase collaboration and facilitate development of new ideas. Such data include ground motion and structural response data recorded during earthquakes, as well as data from laboratory, field, and computational experiments. The growing interdependence of societies around the globe has made development of a common understanding of earthquakes and mitigating the risk associated with these events ever more important. Improved earthquake hazard mitigation will be possible only through the open sharing of experience, information, and real data from these low-probability/high-consequence events.
This special issue covers only a small section of the earthquake and tsunami research field. One important area not touched on is geophysics—in particular, tectonics—which is essential for understanding tectonic plate behavior, and in turn is key to estimating seismic hazard. Furthermore, as the Japan earthquake has made abundantly clear, it's important to understand the relationship between ground motion, tsunamis, and other disasters initiated by earthquakes. The interaction of earthquakes and nuclear reactor safety as well as the initiation and spreading of fires following earthquakes in urban areas are just two examples of such relationships. All of these issues present challenges as well as opportunities for computational technology in science and engineering for decades to come. Providing solutions will save lives and reduce the damage that future earthquakes and tsunamis are sure to cause if we are not ready for them.
Selected articles and columns from IEEE Computer Society publications are also available for free at http://ComputingNow.computer.org.
Rudolf Eigenmann is a professor of electrical and computer engineering at Purdue University. His research interests include high-performance computing, cyberinfrastructure, and compilers for parallel processing. Eigenmann received a PhD in electrical engineering and computer science from ETH Zurich, Switzerland. He is a senior member of IEEE. Contact him at email@example.com.
Ayhan Irfanoglu is an associate professor of civil engineering at Purdue University. His research interests include structural engineering, especially earthquake engineering, structural dynamics, engineering seismology, and simulation. Irfanoglu has a PhD in civil engineering from the California Institute of Technology. He is a member of American Concrete Institute, American Society of Civil Engineers, and Purdue's Earthquake Engineering Research Institute where he is chair of the Heritage and Existing Structures Committee. Contact him at firstname.lastname@example.org.