19 March 2015

Harvesting “big data” to help farmers

Posted by Nanci Bompey

By Nanci Bompey

A farmer and tractor tilling soil. Scientists are using big data  to develop new tools farmers can use to grow more food without harming the environment.  Source: USDA

A farmer and tractor tilling soil. Scientists are using big data to develop new tools farmers can use to grow more food without harming the environment.
Source: USDA

Scientists are using massive amounts of information about the climate, weather and land to develop new tools farmers can use to grow more food without harming the environment.

The need to feed growing populations, coupled with the effects of climate change, means greater demands on farmland all over the world, said David Fischhoff, chief scientist at The Climate Corporation. Farmers need to be able to grow more food on the same amount of land in a way that is sustainable for the long term. In addition, agriculture itself has an impact on the environment – from the energy and water used to the greenhouse gases that are produced, Fischhoff said.

“All that taken together has put a lot of pressure on agricultural productivity, even while the need for increased productivity is there,” Fischhoff said last month during a panel discussion at the American Association for the Advancement of Science annual meeting in San Jose, Calif. He and others panelists discussed how processing large amounts of data may help society deal with a range of potential effects from climate change. Those include greater loads on the electrical grid and increased flooding as well as food-production challenges.

In agriculture, The Climate Corporation is developing new applications farmers can use to address problems posed by climate change and population growth by figuring out, for example, the optimal number of seeds to plant or just the right amount of fertilizer to apply.

Underlying these new software tools are measurements of the atmosphere, weather, soil and plants taken from satellites, on-the-ground sensors, drones and other methods. The data is then used to develop different types of models that, taken together, can make predictions which farmers can factor into their decisions and which can enable them to increase yields in a way that is sustainable, Fischhoff said.

“The goal is to increase productivity and do it as sustainably as possible,” he said.

For example, climate and crop-growth models can be combined to make predictions about the amount of moisture in a grain at a particular point in the growing season. That can help a farmer determine how much water to use. A better understanding of how much nitrogen is in the soil can aid a farmer to understand exactly how much fertilizer to apply, helping to cut down on the amount that ends up in streams or in the atmosphere, Fischhoff explained.

Other scientists on the panel discussed using electricity bills to predict how much air conditioning usage will increase as the climate warms, and developing applications warning users about imminent flooding based on information from a variety of sources.

“We want to go from simply capturing data to doing analytics, providing insight and ultimately making recommendations,” Fischhoff said of the work his company is doing to help farmers.

— Nanci Bompey is a public information specialist/writer in AGU’s public information office.