|
 SCEC
has been awarded $10 million from the National
Science Foundation (NSF) to develop computing capabilities that
will lead to better forecasts of when and where earthquakes are
likely to occur in southern California, and how the ground will
shake as a result.
The five year
project will develop the ability for scientists to improve computer
models of how the Earth is structured and how the ground moves during
earthquakes. The project team includes collaborating researchers
from SCEC, the Information Sciences Institute (ISI) at USC, the
San Diego Supercomputing Center (SDSC), the Incorporated Institutions
for Seismology (IRIS), and the USGS.
These Earth
scientists and computer scientists will create an online collaborative
laboratory-- a "collaboratory"-- allowing scientists from
across the country to conduct science together in much more effective
ways than are currently possible. This facility will be called the
"SCEC Community Modeling Environment."
The collaboratory
will provide resources and tools needed to improve estimates of
how the earth will shake during particular earthquakes, which is
very important information for designing better buildings. Currently
scientists are able to predict the slow, rolling-type motion that
will be experienced during possible earthquakes very well. But predicting
the fast, violent shaking that causes the most damage is thousands
of times more complicated and time consuming.
"We have
lots of different types of data that we attempt to synthesize into
an understanding of earthquake processes," said Dr. Thomas
Jordan, director of SCEC. "But each scientist typically works
on one small aspect of the problem. In order to come up with a comprehensive
and integrative understanding of earthquakes, we have to be able
to put all of this together. The collaboratory will combine three
primary information technologies to organize the science of studying
earthquakes, and allow many scientists to be involved."
One component is
"digital library" technology, which will allow scientists
to organize and retrieve information stored throughout the country.
This requires new tools to access existing data collections and simulation
programs, as well as the ability to incorporate new collections of
data that are generated by the simulations.
Another component
is "grid" technology, which allows calculations
to take advantage of the processing power of a network of many computers.
Many computer-based earthquake simulations last several hours even
using a high-speed computer, and often scientists must wait days
or weeks for an allotted time to run their programs. Using grid
technology, scientists will design simulations that will be automatically
sent to the best computer that is available.
 The
USC School of Engineering's Information
Sciences Institute (ISI), which played a key role in developing
the Globus Toolkit that has
become the international standard in Grid Computing technology,
looks forward to playing an active role in the project. ISI's Carl
Kesselman, director of the ISI Center for Grid Technologies, said
that "advanced problems like the ones SCEC will tackle in this
project are welcome, because they extend the range of problems that
Grids can be used to address and inevitably lead us to develop new
and better Grid technology."
The third component
comprising the collaboratory is called "knowledge representation
and reasoning," which will allow computers to automate
the processing of earth science data and make it easier for scientists
to work together on complex calculations. This requires developing
a common way to describe data, computer programs, and results, and
how these components relate to each other.
"In many
cases these are highly developed technologies, but we will push
them into new arenas," added Jordan. "This project was
funded not just because it is a good thing to do in terms of applying
information technology to earthquakes, but because the problems
in information technology that need to be solved to develop the
collaboratory are themselves important."
"Forty
years ago seismology was one of the primary drivers in the development
of high-performance computing," explained Dr. Bernard Minster,
science director of SCEC and director of the Institute
of Geophysics and Planetary Physics of the University of California.
It again raises a major ITR challenge with the requirements placed
on us by quantitative earthquake science: from rock physics and
fault rupture, to regional wave propagation simulations, to regional
crustal deformation over many earthquake cycles. All these problems
are coupled, across many orders of magnitude in spatial and temporal
scales.
The San
Diego Super Computing Center at the University of California
San Diego will collaborate in developing data and knowledge management
tools that will support this research, led by Reagan Moore of SDSC's
Data-Intensive Computing Environments (DICE) group.
 The
partnership also includes the Incorporated
Research Institutions for Seismology, a 97-member university
consortium supported by the National Science Foundation. IRIS has
extensive experience in the management of very large data collections
of earthquake recordings and application of information technologies
in the collection and distribution of these data. In addition, IRIS
brings a national and international perspective that will encourage
the use of the project's results in areas and science well outside
the project's immediate focus on earthquake hazards in southern
California.
Another partner
in the project is the United
States Geological Survey (USGS), the Federal agency charged
with earthquake hazard analysis and risk reduction. "This collaboration
has the potential to revolutionize our ability to quantify the earthquake
risk in the second largest metropolitan region of the United States,
" said Dr Lucile Jones, USGS scientist-in-charge for southern
California. "The USGS looks forward to applying successful
developments from this collaboration to the reduction of earthquake
risks in other parts of the nation."
The SCEC project
is one of 309 awards that will receive more than $156 million from
NSF's Information Technology Research
(ITR) priority area, which spurs fundamental research and innovative
uses of information technology in science and engineering. The SCEC
activity is one of eight large projects that will each total between
$5.5 million and $13.75 million over five years.
"Information
technology has already transformed our daily lives, yet its most
significant impact so far may be in science and engineering,"
said Gary Strong of the National Science Foundation. "These
new NSF projects show that information technology is enabling new
types of fundamental research not previously feasible, by helping
to gather and make sense of a data avalanche that will solve countless
mysteries about the world around us."
|
