Can a billion points of data from around the globe predict drought and famine? John Corbett, co-founder of aWhere, says unequivocally yes. The Colorado-based company gathers and analyzes more than a billion data points from around the globe each day to create a full agricultural meteorological data set every nine kilometers across the planet.
aWhere’s agricultural intelligence can predict where civil unrest will start because the system knows which areas are failing to produce crops, how many calories a family of four needs to survive, and when the family will migrate if it can’t find food at home.
“It’s an early warning sign,” Corbett said, noting that the algorithm is predictive enough to show, for example, where Boko Haram activity will next occur. The extremist group won’t, after all, go to places where there is no food to steal.
Advances in precision agriculture and data analytics have enabled governments and NGOs to predict food and water shortages like never before. The Famine Early Warning System Network (FEWS NET), which was created by USAID after the 1985 famine in Ethiopia, uses satellite images and geospatial data to create monthly reports detailing projected food insecurity. The U.S. Geological Survey’s (USGS) EROS Data Center, USAID, NASA, NOAA, and Chemonics International, run the system, which aggregates data from tools such as the Normalized Difference Vegetation Index and RainFall Estimate across the African continent.
The Group on Earth Observations’ Global Agricultural Monitoring (GEOGLAM) initiative, a partnership of governments and organizations including the U.S. Department of Agriculture and NASA, trains countries to use remote sensing satellite data and existing agricultural monitoring programs to create global forecasts. The program’s goal is to better coordinate satellite monitoring observation systems in different regions of the world to improve the international community’s ability to produce and share crop yield projections and weather forecasting data.
Croplands, a five-year partnership between USGS, NASA, Google, and academia that will conclude in 2018, creates maps using data from Earth observation systems such as Landsat as well as machine learning and cloud computing. It combines that data with reference data to classify the world’s cropland by factors such as type, watering method, and intensity. Prasad Thenkabail, a USGS research geographer who heads up the program, said mapping global croplands is critical to understand water and security challenges—both of which contribute to global security.
“One of the first things you need is a good information system for policymakers,” he said. “When we have such information consistently, repeatedly, and routinely, we can compute cropland statistics and make them available on a public domain site for policymakers, international organizations, academics, and individual farmers. They can look at the history of how crops have grown year after year and plan around food shortages.”
NASA’s Soil Moisture Active Passive (SMAP) satellite measures the amount of water in the top layer of the soil for every spot on the planet every three days. Soil moisture is the most direct and earliest indicator for drought, so SMAP’s observations—unprecedented in resolution, accuracy, global coverage, and frequency—help predict and monitor drought and global food production.
A new early warning product developed by NOAA and the University of Maryland provides North American drought warnings weeks before other indicators such as the U.S. Drought Monitor. The Evaporative Stress Index (ESI) detects advance signals of plant stress, pairing satellite data from NOAA’s Geostationary Operational Environmental Satellite with vegetation data. Scientists behind ESI hope to expand it globally.
To help detect early warning signs of social unrest and conflict, PeaceTech Lab’s Noel Dickover uses climate change-related data sources combined with agricultural and weather data, social media conversations, and reports of protests. The resolution of aWhere’s weather data is so fine, Dickover said, it can show “pocket droughts”—small but intense droughts that can influence migration patterns and can make some populations more vulnerable to extremist influence.
Corbett said monitoring the weather provides clear insights about the impact of famine and drought. “What’s happening in the Middle East and the refugee problem in Europe will not go away,” Corbett said. “The conditions to produce food across North Africa and the Middle East will continue to deteriorate,” he said, referencing maps that show the poorest areas of risk. “Knowing this, you can intervene before full-scale civil unrest.”
Top photo caption: Agricultural service providers use the SoilCares Scanner to provide Kenyan farmers with soil analysis. Photo courtesy of SoilCares Foundation.
Return to feature story: Autonomous Agronomy