Cyberinfrastructure Poised To Revolutionize Environmental Sciences And Other Disciplines

The convergence of information and communication technologies into a national “cyberinfrastructure” is poised to revolutionize the environmental sciences and many other disciplines in the coming years, according to researchers presenting at the AAAS Annual Meeting in Seattle. The two Feb. 13 sessions on cyberinfrastructure were organized by the heads of two National Science Foundation (NSF) directorates.

The speakers will describe a very near future in which computing capabilities will provide better forecasts of when and where earthquakes are likely to occur and how the ground will shake as a result. Global climate models will simulate complex chemical, biological and geological processes in the Earth’s air, oceans and land over thousands of years. Robotic sensors will monitor ecosystem health or track pollutants in urban watersheds in real- time.

“New instrumentation, data-handling and computation capabilities will expand the horizons of what we can study and understand about the environment,” said Margaret Leinen, head of NSF’s Geosciences directorate and co-organizer of the two AAAS symposia. “Cyberinfrastructure is empowering a new generation of environmental researchers in their quest to unravel how the world around us works.” Cyberinfrastructure has become a common theme throughout NSF, and every directorate has funded or is exploring cyberinfrastructure-related projects.

In environmental science, cyberinfrastructure combines computation, information management, networking and intelligent sensing systems into powerful tools that permit scientists to investigate the natural world and the humanbuilt environment in their full complexity, from the molecular scale to the planetary. This complexity requires collecting and analyzing large volumes of data, performing experiments with computer models rather than just in laboratories and bringing together collaborators from many disciplines.

The challenges of vast amounts of data and complex processes across many scales are faced by many, if not all, scientific disciplines. The NSF’s larger goal for a national cyberinfrastructure is to provide the information technology and knowledge management resources needed to tackle the problems at the frontiers of all science and engineering disciplines, and make those resources as reliable and easy to use as the electricity and water in our homes.

“From the Internet to the Extensible Terascale Facility, the emerging cyberinfrastructure NSF supports is a product of the scientific community’s demands for and reliance on information and communications technologies,” said Peter Freeman, head of the NSF’s Computer and Information Science and Engineering (CISE) directorate and the symposia’s other co-organizer. NSF’s Extensible Terascale Facility is a multiyear effort to deploy a comprehensive infrastructure of computation, information and instrumentation resources for academic research and education.

New CISE division director for Shared Cyberinfrastructure, Sangtae Kim, will co-chair the symposia. Kim is the Donald W. Feddersen Distinguished Professor at Purdue University, an endowed chair for research at the intersection between information technology and engineering, and was vice president and information officer of Lilly Research Laboratories.

Last February, a report from the NSF Advisory Committee for Cyberinfrastructure noted that cyberinfrastructure is “essential, not optional, to the aspirations of research communities” and that success would require collaboration between the physical and life sciences, computer science and the social sciences.

The AAAS symposia bring together computer and environmental scientists, many collaborating on NSF awards, to describe research at the frontiers of computer science that is leading to cyberinfrastructure and the groundbreaking research in environmental science that will be possible when tapping into vast computation and data resources becomes as easy as turning on a light switch.

Deborah Estrin, director of the NSF-funded Center for Embedded Network Sensing at UCLA, will describe how networks of smart sensors are being deployed to monitor and collect information on endangered species, soil and air contaminants and medical patients, as well as buildings, bridges and other man-made structures. Estrin is also slated to deliver an AAAS topical lecture on “Instrumenting the World with Wireless Sensor Networks.”

To better predict earthquake occurrence and the resulting ground motion, NSF is supporting the Community Modeling Environment project, led by the Southern California Earthquake Center (SCEC). The project’s goals are to better understand earthquakes and to provide information crucial to designing civil infrastructure and to disaster planning in regions such as Southern California. SCEC’s Thomas Jordan will discuss both the scientific advances and the cyberinfrastructure from the project, including smart modeling tools and the shared computing environment and virtual community created between SCEC, the University of Southern California, the San Diego Supercomputer Center and the Pittsburgh Supercomputing Center.

Research on the weather, the climate and the whole-Earth system will also benefit from the capabilities of a comprehensive cyberinfrastructure. Bob Wilhelmson of the National Center for Supercomputing Applications at the University of Illinois, Urbana- Champaign, will describe how the NSF-funded Linked Environments for Atmospheric Discovery project will integrate many real-time data streams with customized weather models and on-demand computing to provide timely severe weather forecasts in unprecedented detail.

Jeffrey Kiehl of the NSF-supported National Center for Atmospheric Research (NCAR) will describe the work done with the Community Climate System Model, one of the world’s most sophisticated climate models. Developed by a consortium of climate and computer scientists, this experimental tool integrates global models of the atmosphere, ocean, land and sea- ice to study the Earth’s climate. And Timothy Killeen, director of NCAR, will discuss whole-Earth system modeling and the multi-agency Earth System Modeling Framework collaboration.

Ecologists and biodiversity researchers face challenges in accessing and integrating the data needed to ask groundbreaking questions and to help scientists, policymakers and the public make informed decisions about the environment. William Michener of NSF’s Long-Term Ecological Research (LTER) Network Office will discuss how the NSFsupported Science Environment for Ecological Knowledge (SEEK) project is tackling the challenges of integrating data collections. When these integrated collections are combined with modern reasoning software, the computer can become a scientist’s “intelligent assistant.”

Other symposia speakers include Dan Reed of the University of North Carolina and a well-known leader in cyberinfrastructure; William Swartout of the University of Southern California’s Institute for Creative Technologies; Jeffrey Naughton of the University of Wisconsin; and Tom Anderson of the University of Washington.

The two AAAS symposia on cyberinfrastructure are scheduled for Friday, Feb. 13, 9 a.m. – 12 p.m. and 2:30 p.m. – 5:30 p.m. Estrin’s topical lecture on sensor networks is 1:30 p.m. – 2:15 p.m. the same day.

Cyberinfrastructure activities are the most recent evolution of NSF’s long history of leadership in providing the most advanced information technologies for the U.S. academic community. NSF supported campus computing centers in the 1960s, established national supercomputer centers in the 1980s and supports the Extensible Terascale Facility and many other cyberinfrastructure projects today. In parallel, NSF established NSFnet in the mid- 1980s, which evolved into today’s commercial Internet, and in the 1990s helped connect hundreds of institutions to advanced research networks.

Also in the late 1990s, NSF established the Partnerships for Advanced Computational Infrastructure (PACI), which have nurtured and supported the growing demand by the science and engineering community for cyberinfrastructure. The Extensible Terascale Facility and the seeds of many cyberinfrastructure-related projects for specific disciplines, supported by NSF and other agencies, can be found in the PACI program, the NSF Middleware Initiative and projects funded through the NSF’s Information Technology Research priority area.