Since the late 1990s, Deborah Estrin, professor of computer science at UCLA, has researched the networking of distributed sensors. That led her to studying applications of distributed networking in environmental monitoring, which are related to instrumented environments, the smart grid, environmental assessments, and instrumented watersheds and foodsheds. In her view, the old adage, “You can’t manage what you can’t measure,” applies to how networking and sensor technology can help address problems of sustainability. Distributed sensing contributes to the rich modeling of complex natural-engineered-human systems. Both are essential to optimizing our use of resources, such as energy, water, agriculture, and transportation, says Estrin.
All are subjects of the new National Resource Council (NRC) report, “Computing Research for Sustainability,” released by the Committee on Computer Research for Environmental and Societal Sustainability, which Estrin chairs. “Sustainability is both a vitally important and incredibly rich problem space whose realization needs the technologies and methods of computer science research and the participation of computer scientists,” says Estrin. “However, we can’t possibly do it alone. As computer scientists we must co-innovate with domain experts [in energy, agriculture, and urban systems] to shape the socio-technical solutions to the pressing and complex challenges facing our planet.”
The 155-page report says one goal of applying computer science to sustainability is to inform, support, facilitate, and even automate decision-making. Four broad research areas in computer science are crucial to attaining this purpose: measurement and instrumentation; information-intensive systems; modeling, simulation, and optimization; and human-centered systems. As these areas correspond to established research areas in computer science, the research community is well-positioned to make progress, according to the NRC report.
It calls for interdisciplinary approaches to sustainability to work from the ground up. The best way to foster interdisciplinary effort is for funding agencies to embrace interdisciplinary projects and value the discipline-enmesshed nature of the intermediate results, says Estrin.
To prepare future computer scientists interested in sustainability, the report suggests that undergraduate and graduate computer science education include tracks in life-cycle analysis, agriculture, ecology, natural resource management, economics, and urban planning. “The work that will matter most cannot happen in a vacuum,” says Estrin. “If we recognize the importance of computer scientists to address sustainability needs, then we must recognize the need for creative and effective programs to support training for computer science students to become effective members of multidisciplinary endeavors.”
According to Estrin, automating decision-making can further sustainability because low-level processes like automating the control of energy capture, and high-level decisions by individuals about water and transportation systems, are “all about improving the fidelity of those processes.”
The NRC report offers three examples—smart energy grids, sustainability agriculture, and resilient infrastructure—to show the potential impact of advances in computing. The systems challenges that smart grids present, for example, will require computer science research and methodological approaches to user interfaces in addition to improved modeling and analytical tools at all levels. The report also addresses the role of computer science in responding to and recovering from natural and engineering disasters, like hurricanes and oil spills.
Based in Manhattan, Karen A. Frenkel is a freelance writer and editor specializing in science and technology.