1521-9615/05/$31.00 © 2005 IEEE
Published by the IEEE Computer Society
Guest Editor's Introduction: iSERVO—The International Solid Earth Research Virtual Observatory
The international Solid Earth Virtual Observatory (iSERVO; www.iservo.edu.au) is a planned partnership of primarily Asia Pacific economies interested in studying, forecasting, and mitigating the damaging effects of large and great earthquakes as well as the tsunamis they produce. The iSERVO concept is founded on the use of modern information technology to simulate the nonlinear dynamical earthquake and tsunami processes at all relevant scales, and to organize, collate, and interpret the massive new data sets that will be available in the partner countries beginning within the next three years.
iSERVO is envisioned to be an open partnership of many nodes in many regions. It will be organized under the auspices of the Asia-Pacific Economic Cooperation's highly successful APEC Cooperation for Earthquake Simulation (ACES) organization, which currently involves four economies: Australia, China, Japan, and the US. This multilateral grand challenge science research cooperation involves the leading international earthquake simulation and prediction research groups.
The collaboration's goal, at its core, is to improve earthquake forecast methods (applicable to the Pacific Rim) through the development of physically based numerical simulation models for earthquake nucleation, generation, and wave-propagation processes. Interaction and collaboration occurs through biennial workshops, working group meetings, and overseas visits by individual scientists. ACES is intended to build on the complementary strengths in earthquake research and observation that exist throughout APEC member economies. The four founding partner countries formally endorsed ACES at the APEC Industrial Science and Technology Working Group's October 1997 meeting in Singapore.
iSERVO is built on a foundation of modern Web services and Semantic Web software technology; it's the latest example of the virtual observatory concept developed with modern Grid computing technology, which has proven highly useful in the high-energy particle physics and astronomical communities. In the iSERVO concept, each economy is represented by one or more iSERVO nodes, which will all be linked to a series of databases, data sources, simulations, collaboratories, Grid computing engines, and other resources via the iSERVO computational portal. Products of this collaboration will be
• the iSERVO node, composed of the computational portal linking all nodes together into a common Grid computing framework;
• the collection of simulation codes, data, analysis software, and collaboration and conferencing facilities, accessible through a uniform interface;
• publications arising from shared collaborations among all nodes; and
• workshops, scientific exchanges, visits, and greatly facilitated international cooperation on problems of critical scientific importance.
The iSERVO portal aims to provide disaster-scenario simulations, hazard estimates, and warning capabilities based on new earthquake-forecast technologies that are currently under development. Researchers have successfully used such technology to forecast the locations of large earthquakes on decadal time scales (see http://quakesim.jpl.nasa.gov). The iSERVO portal should also provide real-time, near-real-time, and archived earthquake and tsunami forecasts as well as warning and hazard broadcast technologies.
This issue of Computing in Science & Engineering contains a series of articles describing research conducted in the four ACES partner economies related to the implementation of the iSERVO concept. The first article, "ACceSS: Australia's Contribution to the iSERVO Institute's Development," by Peter Mora et al., describes the genesis of the ACES organization and its development within Australia into the Australian Computational Earth System Simulator (ACceSS)—a major national research facility in Australia—through to the iSERVO institute's formation. The authors describe technology for computing and simulating earthquakes at a variety of scales, based on the novel Escript scripting language, which employs a Python-based object-oriented approach. They also discuss examples of simulation results for earthquake processes.
In "The China ACES-iSERVO Grid Node," Xiang-Chu Yin et al. describe the implementation of iSERVO within China, a country that has had many disastrous earthquakes in its history—the most recent notable example is 28 July 1976 magnitude 7.8 Tangshan earthquake, which killed more than 250,000 people. Efforts in China are based on the SciGrid (Scientific Computing Grid), developed by the Supercomputer Center of the Chinese Academy of Sciences (SCCAS). One of the major forecast technologies under development in China is the Load-Unload Response Ratio method, whose goal is to quantify stress changes during the progressive failure processes leading up to large earthquakes.
In the third article, "Quest for Predictability of Geodynamic Processes through Advanced Computer Simulation," Mitsuhiro Matsu'ura summarizes research in Japan, based on an integrated view of complex Earth system processes from the physics of the Earth's core outward to the dynamics of its crustal processes. The development of this capability was intimately tied to the construction of the Earth Simulator, until recently the world's fastest supercomputer. The iSERVO effort in Japan is coordinated with a new Center of Excellence program for the "Predictability of the Evolution and Variation of the Multisphere Earth System."
The final article, "A Web Services-Based Universal Approach to Heterogeneous Fault Databases," by Lisa Grant et al., describes components of the US contribution to iSERVO in the areas of Grid computing, Web services, and database technology. The authors focus on the QuakeSim earthquake simulation system, which was developed under funding to the Jet Propulsion Laboratory for use by the seismological, crustal-deformation, and tectonic communities. The project's goal has been to develop a virtual laboratory for the study of earthquake behavior over multiple time scales.
The great Sumatra earthquake and tsunami of 26 December 2004 dramatically demonstrated the need for a research and operational e-structure such as iSERVO. Initial estimates suggested that this event, which occurred on the Sumatra subduction zone, had a magnitude M = 9.0, but further work has suggested that its magnitude might have been as large as M = 9.3. In either case, this event is one of the 10 largest earthquakes in recorded history. Hundreds of thousands of people were killed and injured, and the full extent of the devastation will probably never be known. If a system such as iSERVO had been in place at the time, local hotels, businesses, government representatives, and even local citizens might have been able to receive early notification of the approaching catastrophe via Web-based information services. Potentially, many thousands of lives across the Indian Ocean basin might have been saved as a result. The articles in this issue suggest that information technology solutions to these problems, developed and implemented across many economies, will play a critical role in solutions to the problem of societal vulnerability to extreme events and natural disasters.
is the director of the Center for Computational Science and Engineering at the University of California, Davis, where he is also a professor of physics and engineering. His research interests include complex systems, high-performance computation, and the physics of extreme events with a focus on earthquake forecasting. Rundle has a PhD in geophysics and space physics from the University of California, Los Angeles. He is a member of the American Physical Society (fellow), the American Geophysical Union, the Seismological Society of America, and the IEEE Computer Society. Contact him at email@example.com; http://cse.ucdavis.edu/~rundle/.