1521-9615/04/$31.00 © 2004 IEEE
Published by the IEEE Computer Society
Guest Editor's Introduction: A Pathway to Decisions on Earth's Environment and Natural Resources
Earth is the only planet in our solar system that has maintained relatively stable climate conditions over geological timescales, a feat that has allowed highly diversified life to evolve. Why is this, and what makes the Earth system so unique?
Although we don't know the complete answers to these questions, we do know that a planetary mechanism controlled by a combination of carbon and water cycles fueled by the sun's energy has created a unique climatic condition on Earth that is conducive to carbon-based life on this planet. However, we don't know exactly how this planetary mechanism works. Analogous planets in our solar system with similar water or carbon cycles ceased at some point, thus creating suboptimal conditions for carbon-based life to emerge and evolve. Mars represents the best example of an absent carbon cycle; Venus is the best example of how the lack of a water cycle fails to control the build up of atmospheric greenhouse gases, resulting in a very hot surface unsuitable for diversified life. As we pursue the search for the origin and signs of life beyond Earth, understanding how life emerged on this planet and what it takes to sustain it is equally compelling.
Earth's unique nature and the power of space-based remote sensing gave rise to Earth exploration's prominence in the Space Act and in NASA's new vision and mission. NASA's vision is to improve life here, to extend life to there, to find life beyond. Its mission is to understand and protect our home planet, to explore the universe and search for life, to inspire the next generation of explorers…as only NASA can.
Over a decade ago, the global Earth science community defined a goal of understanding this planetary mechanism. It required a comprehensive scientific program for characterizing, understanding, and ultimately predicting Earth's behavior as a system of atmosphere, oceans, continents, and life that interacts over a vast range of space (local, regional, and global) and time (seconds, hours, days, and decades) scales. This scientific grand challenge is the NASA Earth science program's focus because of the unique vantage point space provides for viewing and understanding the entire Earth and its planetary system.
NASA's goal is to understand how the Earth system is changing and what the consequences are for life on Earth. Five main and 24 subsidiary scientific questions guide NASA's investments in its scientific research, technology, and satellite development (see the " Hierarchy of Science Questions
" sidebar). The logical progression in the scientific understanding we gain from answering these questions will determine our conceptual Earth system models' ability to mimic Earth's planetary system behavior and ultimately predict it. Such models and their output are also useful to the decision makers responsible for managing Earth's natural resources or establishing environmental-policy decisions.
NASA identified five key program elements required to support its goal:
• a comprehensive, integrated, and sustained Earth-observing system;
• a comprehensive data and information management system to acquire, process, archive, and distribute the information and knowledge from the observing system to scientists and decision makers;
• an Earth system modeling and computational framework for developing, integrating, testing, verifying, and ultimately routinely operating Earth system models in predictive mode;
• advanced technology research and development to support the evolution of observing, modeling, and computing elements; and
• decision-support system components for translating scientific knowledge and Earth system predictions to decision makers for managing Earth's natural resources and environmental policy decisions.
Over the past decade, NASA has invested about 10 percent of its overall annual budget in supporting its Earth Science Enterprise portfolio. Roughly half of the Earth science budget is invested in developing the satellites and data and information management systems; the other half is invested in scientific research, advanced technology development, and modeling and developing decision-support systems.
As a result of this investment, 20 Earth-observing research satellites are orbiting our planet (see Figure 1
). These satellites have remote-sensing capabilities ranging from conventional imaging radiometers, spectrometers, and interferometers to the latest in hyperspectrometers, lasers, and radar, all of which allow a deeper characterization and understanding of Earth's atmosphere, oceans, and continental surfaces and their interactions. A comprehensive data and information management system complements the space-based systems: it commands and controls the satellites; acquires, processes, archives, and distributes more than 3 Tbytes of data each day; and serves more than two million users each year. The challenge ahead is to create systems and processes that routinely transform petabytes of data into megabytes of knowledge—an image or a document that a nonscientist can act on. NASA also sponsors more than 2,000 science and technology-focused research, development, and application projects at US universities and private and federal research laboratories.
Figure 1. NASA's existing Earth-observing research satellites.
NASA's investment in Earth system science is complemented by a comparable amount from its domestic partners—10 US federal agencies under the auspice of the US Climate Change Science Program—and its more than 60 international partners. These arrangements provide a solid foundation for using the program's current capabilities effectively and for its evolution in the future. NASA now has the fundamental building blocks in place to pursue the grand challenge of understanding and predicting Earth's planetary system.
Access to high-end computational technology is key to this grand challenge's success. Such computational capability makes it possible to manage the very large volume (petabyte-scale) of data from the observing systems and run the complex Earth system models. Current data-assimilation techniques under development aim to combine observations from multiple sources (space, ground, and air) and create seamless records of Earth's atmospheric, oceanic, and terrestrial parameters in space and time, which requires two to three orders of magnitude improvement in computational power, archive, and access capabilities. Developing and routinely operating the complex Earth system models over the next decade will require another three to four orders of magnitude improvement in computational capabilities.
This special issue's purpose is to describe in greater detail this grand challenge's high-end computational aspects and to invite the experts from disciplines of computational physics and computer sciences to join Earth system scientists in understanding and predicting Earth's planetary system. The first article describes a notional Earth system modeling framework that is currently the focus of research and development among a consortium of several US federal agencies and more than 15 universities. The next article describes the current state of Earth system modeling for short-term weather forecasts. The third article establishes a solid foundation for understanding and ultimately predicting geophysical natural hazards such as earthquakes and volcanoes. This article also illustrates a completely different computational requirement: distributed "grid"-type computing. The last article describes the task of translating the best available Earth system scientific knowledge and predictive skills for decision makers managing Earth's natural resources.
We hope the publication of this special issue will encourage experts from biology, chemistry, physics, and computational science to join Earth system scientists in overcoming the grand challenge of understanding and predicting how our Earth system changes and what the consequences will be for our generation and those that will follow.
Ghassem R. Asrar
is the Associate Administrator for Earth Science at NASA. His technical interests include interactions of the land surface with the atmosphere and the use of new remote-sensing techniques to derive new information about them. He received his graduate degrees in civil engineering and environmental physics from Michigan State University. He is a member of the American Geophysical Union and the Geoscience and Remote Sensing Society. He is a fellow of the IEEE and the American Meteorological Society. Contact him at Code Y, NASA Headquarters, Washington, DC 20546; firstname.lastname@example.org.