CompSustNet: Expanding the Horizons of Computational Sustainability

Our first Expedition in Computing (CNS-0832782) established the new field of computational sustainability, which aims to identify, formalize, and provide solutions to computational problems to help balance environmental, economic, and societal needs and thereby facilitate a path towards a sustainable future. This Expedition in Computing launched CompSustNet (http://www.compsust.net), a vast research network charged with significantly expanding the horizons of the field of computational sustainability and fostering the advancement of state-of-the-art computer science to achieve the scale to tackle global problems. Sustainability challenges encompass a combination of distinguishing aspects that make them unique in scale, impact, complexity, and richness, posing new challenges and opportunities to computing and information science. Research is focused on cross-cutting computational topics such as optimization, dynamical models, simulation, big data, machine learning, and citizen science, applied to sustainability challenges concerning conservation and biodiversity, balancing socio-economic demands and the environment, and renewable energy. Our Expeditions (CNS-0832782 & CCF-1522054) have advanced both computer science and sustainability fields with over 300 publications in top tier computer science and sustainability sciences journals and conference proceedings, as well as broader scientific publications such as Science, Nature, Proc. of the National Academy of Sciences, and Scientific American. Our projects have also had real-world impact in a variety of domains such as bird and wildlife conservation, poverty and social intervention, renewable energy, and materials discovery, through collaborations with The Nature Conservancy, the World Wildlife Fund, the World Bank, and the Caltech's Joint Center for Artificial Photosynthesis, and many other institutions. We have educated a large number of undergraduate and graduate students, instilling in them an appreciation for sustainability and computational thinking to address sustainability challenges. We placed 14 students in tenure-track faculty positions, each with a clear computational sustainability angle. We also created a series of conferences, and established and nurtured a broad community of researchers and practitioners around the field of computational sustainability. The startup Atlas AI (http://atlasai.co/), funded by The Rockefeller Foundation, is pioneering new ways to estimate socioeconomic indicators by combining high-resolution satellite imagery and other data with the latest advances in artificial intelligence.

Our spatio-temporal models of bird species distribution were used to produce the U.S. Department of Interior's State of the Birds reports in 2009, 2010, 2011, 2013, 2014, 2016, and 2017 based on citizen-science data collected by the Cornell Laboratory of Ornithology' eBird project on bird watcher sightings. The fine grained information about bird distributions provided by our models led to novel approaches to bird conservation, such as the Bird Returns program developed by The Nature Conservancy, which is creating thousands of additional acres of habit for migratory birds.

• Advances in the phase-map identification problem in materials discovery, including the Phase-Mapper tool and AgileFD algorithm, led to the discovery of new solar light absorbers. Both the journal ACS Combinatorial Science and AI Magazine featured these developments as cover articles.
• The World Bank is using our research on predicting poverty by combining satellite imagery and machine learning in Africa, as well as in other underdeveloped regions. It received broad press coverage, and in 2017 Scientific American selected it as one of the 10 ideas that will change the world. This work was published in the journal Science.
• The startup Atlas AI (http://atlasai.co/) is pioneering new ways to estimate socioeconomic indicators by combining high-resolution satellite imagery and other data with the latest advances in artificial intelligence. Atlas AI uses cutting-edge techniques to build an accessible analytics platform to analyze and predict crop yields, economic well-being, and other sustainable development indicators at fine resolution across the developing world. The Rockefeller Foundation funds Atlas AI.

See also the Expeditions in Computing Computational Sustainability: Computational Methods for a Sustainable Environment, Economy, and Society, CNS-0832782