Skip to main content

Crop Yield and Condition Forecasting from Field to Global Scales

What We Do

We monitor and forecast agricultural yields and conditions globally, ranging from single fields to entire countries, by applying machine learning algorithms to earth observation data.


United States of America, Vietnam, Japan, Philippines, Brazil, Mexico, Australia, South Africa, India, Canada, Pakistan, Indonesia, Paraguay, Turkey, Argentina, China, Italy,  Romania, Kazakhstan, France, Germany, U.K., Ukraine, Russia, Spain, Hungary

How Satellites Make This Work

Access to timely Earth Observation (EO) data can ensure effective monitoring of agricultural conditions globally. We use EO datasets (NDVI, Temperature, Precipitation, Evaporative Stress Index, Soil Moisture, Growing Degree Days), GEOGLAM crop calendars, GEOGLAM crop masks to create dashboards of crop conditions across a range of crops and countries. These dashboards are updated weekly and provide a useful source of information in producing the GEOGLAM crop monitor reports. These EO datasets are used to create machine learning models of crop yields that are used for in-season yield and condition forecasting.

Brian Barker, University of Maryland
Inbal Becker-Reshef, University of Maryland
Ritvik Sahajpal, University of Maryland
Team Members

GEOGLAM Crop Monitor teams

Learn More