Resources
|
Research Highlights

Research Highlights
The Sky’s the Limit: Predicting Oil and Protein Levels Within Soybean Fields

In this article, you’ll find details on:

  • A three-year NCSRP project finalizes with a comprehensive database of cropping information from 400 fields across the Midwest
  • Collected data inform spatial field maps where soybeans with higher levels of protein and oil are located
  • The project springboards to a collaboration with NASA Acres

Soybeans being harvested by combine
Photo: United Soybean Board

By Carol Brown

A recently completed multi-state soybean research project is literally growing sky-high.

The three-year project began with the goal of establishing a database of information from farmers across the Midwest, and creating a tool to predict areas within a soybean field that contain higher levels of seed oil and protein. The project began in 2022 under the leadership of Ignacio Ciampitti, formerly a farming systems professor at Kansas State University. He is now a professor of quantitative agronomy and co-director of the Institute for Digital and Advanced Agricultural systems at Purdue University. 

Ciampitti and his team of more than a dozen researchers obtained information from farmer surveys on nearly 400 fields in 10 midwestern states and three southern states, then combined the data with satellite imagery to establish a comprehensive database. The project is funded through checkoff support from the North Central Soybean Research Program (NCSRP).

He recently wrapped up the project, but the research has progressed into something a little larger.

“After reading a peer-reviewed journal article about this project, NASA became interested,” Ciampitti says. “We are growing this work into the first pilot project under a new program within NASA Acres that aims to expand predictive research with our database and their satellites.”

Founded in 2023, NASA Acres is NASA’s consortium focused on advancing the value and adoption of top-quality satellite data and tools by U.S. farmers and other agri-food system stakeholders. The program has funded a number of projects involving farmers and at the recent Commodity Classic in Denver, it launched a formal program linking farmers and their farms directly into NASA Acres research and with one another. 

Ciampitti co-leads this new NASA Acres program: Farm Innovation Ambassador Team (FIAT). With the large amount of in-field-level data available from the NCSRP project, he will grow this quality work through the inaugural FIAT program. 

“A key advantage of satellite data is scale, but we can’t scale what doesn’t already work well on at least one site,” says Alyssa Whitcraft, NASA Acres Executive Director and Ciampitti’s FIAT co-lead. “And we can’t create something that works well for any farmer without farmers’ co-ownership of the project and process. That’s what is great about Ignacio’s project. There is so much on-the-ground research, and it’s already linking to NASA data. Ignacio also brings Kansas farmer Ray Flickner, a strong collaborator and one of the most innovative farmers out there.”

Predicting Oil and Protein

“For the 2022, 2023 and 2024 growing seasons, we collected yield data and crop management practices from several points within a field, along with weather information. We also collected soybean samples and processed them for oil and protein traits,” Ciampitti explains. “We’re now building on the science behind this information to understand the parameters that can help predict yield, protein and oil within a field. We are developing spatial variability maps for the possibility of segregating areas in a soybean field with higher levels of protein and oil.”

Figure 1. (A) Location of surveyed farmer fields (circles) colored according to the cluster classification, showing the total per state. (B) Histogram displaying yield, protein, and oil concentration for each cluster. 

After reviewing and aggregating the collected data, the research team is finding that predicting oil levels is slightly easier than predicting protein. Soybean oil levels relate to yield, Ciampitti says, but protein amounts are highly influenced by environmental factors, therefore a bit harder to predict. The amount of nitrogen the plant is forming is relevant to defining the final protein target.

The team clustered the data into three production regions based on a range in the differences in amounts of seed oil and protein (Figure 1), which correlated to the variables that influence prediction accuracy. Among the variables are growing days, precipitation, and temperature. Overall, the team found that protein levels were higher in the northern region (39.5%) compared to the central region (38.2%), and seed oil levels in the north were limited by low temperatures during seed filling.

A Prediction Tool for Farmers

The team also developed a prediction tool so farmers can see the economic value of higher protein and oil levels. The online tool “Soybean Quality Economic Simulator” can be used to illustrate a return on investment if soybeans were paid premium prices according to quality. 

“The tool can show whether it makes sense to pay for the quality,” he says. “The tool is related to the price for protein and oil. The logistics and the economic risk are connected to the final steps of commercialization. The tool can help envision a way to position soybeans in the market for a competitive advantage.” 

The Soybean Quality Economic Simulator can help farmers see if a premium price has a positive return on investment on their farm. The simulator was developed from the NCSRP project and is housed with Iowa Soybean Association. 

Ciampitti sees the potential for market development with higher-priced exports of soybeans with more protein and oil, especially as the biofuels industry continues to develop sustainable fuel made from soybeans. Farmers could see premium payments, he says, but the industry needs to adapt to handle protein and oil seed segregation, similar to the organic farming market.

“We want farmers to be ready and have a tool to help see higher quality in their fields for seed segregation. This is just the first step,” Ciampitti remarks.

The research project has come far in these three years — from finding farmers to work with, collecting their field data and cropping information, to amassing and analyzing the many locations and seed samples. The project has gone from boots on the ground all the way to satellites in the sky.

“Being part of a connection between farmers and NASA — it sounds a little crazy,” he says. “But when we can connect the people developing the tools with the people who will be using the tools, it’s not crazy at all.”

Additional Resources:

On-Farm Soybean Seed Protein and Oil Prediction Using Satellite Data – “Computers and Electronics in Agriculture” journal

Soybean Quality Economic Simulator

NASA Acres website

Planning for the Future: Mapping Soybean Fields for Protein and Oil Quality – SRIN article

Meet the Principal Investigator – Ignacio Ciampitti

Published: Mar 31, 2025

The materials on SRIN were funded with checkoff dollars from United Soybean Board and the North Central Soybean Research Program. To find checkoff funded research related to this research highlight or to see other checkoff research projects, please visit the National Soybean Checkoff Research Database.