June 24, 2016
Big data experts to share green ideas at World Economic Forum
Three Cornell experts who use big data to make the world a little greener will share their innovative ideas at one of the most prestigious gatherings of thought-leaders in the world.
Faculty members Harold van Es, Carla Gomes and Joshua Woodard will present their research at the intersection of computation, food and computational sustainability, a new field pioneered at Cornell, at the World Economic Forum (WEF) June 26-28 in Tianjin, China.
The WEF, a Swiss nonprofit foundation, is best known for its annual winter meeting in Davos, Switzerland. The WEF’s “summer Davos” in China, called the Annual Meeting of the New Champions, convenes 1,700 business leaders, policymakers and experts from 90 countries for a summit on the latest in innovation, science and technology.
“I felt this was a really unique opportunity, not only for us to show our work but also to interact with other people who are leaders. It’s an honor to be invited and to be part of that,” said van Es, professor of soil and water management, who canceled a trip to eastern Europe to attend. “It’s just something that you don’t turn down. You’re willing to move mountains to get there.”
Van Es and Gomes, professor of computer science, and Woodard, assistant professor of applied economics and management, will each give two five-minute presentations at the WEF’s Ideas Lab.
Gomes, director of the Cornell Institute for Computational Sustainability, is not able to attend in person but will appear in a videotaped presentation. She’ll talk about how computational sustainability techniques applied in a variety of areas can manage natural resources much more effectively.
“If Wall Street can do very smart things, we should be able to use these smart algorithms to protect our environment,” she said.
For example, Gomes and her colleagues are developing advanced models of how birds are distributed across landscapes over time. The team has given the data to the Nature Conservancy, which will pay Sacramento Valley farmers to keep water in their rice fields during migration to offer birds a resting place while en route.
“This is a very effective way of generating habitats for the birds,” Gomes said. “It’s not very expensive. You’re just renting the smallest rice fields for a short window of opportunity, and this is only possible with advanced computational techniques.”
Van Es will talk about a technology he and a team of entrepreneurs and Cornell researchers developed that precisely predicts how much fertilizer U.S. farmers should apply to their corn fields, reducing their tendency to over-fertilize.
Nitrogen is an essential agricultural nutrient that has prevented widespread hunger through its use in fertilizer. But nitrogen also produces a potent greenhouse gas, negatively affects water quality and uses a lot of energy in its manufacturing.
The tool his team developed, Adapt-N, uses complex computational models to examine soil, weather, crop and field management data to provide the farmer with a fertilizer recommendation for a particular field. The tool can even predict nitrogen needs for a 60-by-60-foot section of land. Now 1.5 million acres are cultivated with Adapt-N.
“It’s an interesting story, because it’s about a nutrient that’s needed to grow crops but also has these environmental impacts, and it’s about how we addressed it,” van Es said.
Woodard will talk about how to use big data to make crop insurance more sustainable. Although the U.S. crop insurance market is the largest in the world, with over $100 billion in liabilities annually, the government’s pricing is fairly crude, Woodard said. For example, it doesn’t take into account the type of soil on a given farm. That can translate into billions of dollars in premium error, potentially higher taxpayer costs and perhaps even incentives to farm on sensitive land.
Woodard’s research shows integrating several large soil data sets into the pricing equation would dramatically improve pricing accuracy. Related research integrating genetic data and remotely sensed data, such as vegetation information gathered by satellite, could further facilitate building more sustainable policies, he said.
“Today we have the capabilities to price and manage agricultural risk not only at the genetic level and the soil level, but also from space, which is really exciting,” Woodard said. “Better remote sensing and analytics in the future will only further enable developments in this area.”