Suppose you could test multiple management decisions across five growing seasons to discover the best recommendations for growing soybeans on your farm in 2025? Shawn Conley, University of Wisconsin soybean specialist, is working with Extension staff and commercial interests to test such a system.

“Our goal was developing a method for rapid and accurate environment-specific identification of cropping systems with the greatest yield potential,” Conley says. “This is difficult because of the complexity of interactions between soils, weather and management practices for fields across the country.

“By using data science and machine-learning algorithms, we developed a method to evaluate thousands of potential cropping systems and identify those with the greatest yield and profit potential in each farm across the U.S.”

His database consists of thousands of data points for both corn and soybeans spread across the major corn and soybean states. Each point represents a genetics-by-environment-by-management (G-by-E-by-M) yield-specific result. Commercial interests developed a tool to turn input for any field into specific recommendations for top yield or maximum revenue, depending on your goals. The tool is called Agroptimizer.

Conley offers hypothetical scenarios to illustrate how the system works.

Related:Crop residue: Should you take it or leave it?

Soybean example

Suppose you grow soybeans near Arlington, Wis. Here are your management choices for 2025:

“That’s 128 management combinations,” Conley explains. “Spread across five years, that is 640 possible G-by-E-by-M yield-specific points.”

Data from other studies indicate May 1 as the preferred planting date vs. June 1, Conley notes. So, for May 1, the program determined the combination that should produce the highest yields and the lowest yields.

High-yield recommendation: Select a high-yielding Group 2 variety and seed 160,000 seeds per acre in a 15-inch-row no-till system following corn. Apply fungicide.

Low-yield recommendation: Select a Group 1 variety and seed 140,000 seeds per acre in a 30-inch-row no-till system following soybeans. Don’t apply fungicide.

“Getting the right soybean variety is important,” Conley says. “In Wisconsin soybean variety trials in both 2023 and 2024, there was a 17-bushel difference between the top and lowest-yielding variety in southern Wisconsin. Even at $9 per bushel, that is a $153 swing.”

Corn example

Suppose you’re growing corn near Lincoln, Neb. Here are your management choices for 2025:

That’s 432 management combinations multiplied by five years to get 2,160 G-by-E-by-M specific yield combinations.

Related:6 common traits of top-yielding fields

“It’s impossible to set up one research trial to test all those factors,” Conley says. “Yet through machine-learning, artificial intelligence and a huge database, we get relevant answers.”

For the highest yield, the program suggests selecting a 110-day hybrid and plant April 30 following soybeans at 33,000 seeds per acre in 15-inch rows; apply 225 pounds of N. For highest net return, drop seeding rate to 28,000 and N rate to 150 pounds.

VALIDATING RECOMMENDATIONS: You have a chance to test technology-generated recommendations on your farm against your normal practices. Read on to see how to do it.

Your chance to test tech recommendations

Are your choices for variety, planting date and seeding rate optimum? Are you willing to pit your current system against custom recommendations from Agroptimizer? It’s a computer model driven by artificial intelligence that is backed by a huge database, using algorithms to determine what should work best on your farm.

“We’re looking for farmers to be our ‘boots on the ground’ and test recommendations versus their current program,” Conley explains. He is director of the Cool Bean research effort in Wisconsin. “We also want to validate sophisticated scouting technology to see if we can scout better faster.”

Agroptimizer is a commercial program. Farmers within the North-Central Soybean Research Program region can use it for free to obtain site-specific crop recommendations. States within NCSRP include Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, Pennsylvania, South Dakota and Wisconsin.

Related:How to improve odds when double-cropping corn, soybeans

Special project, Part 1

Extension in 10 states are participating in a project to determine how program recommendations fare in the real world, Conley says. Those include Illinois, Indiana, Iowa, Michigan, Missouri, Nebraska, North Dakota, Ohio, Pennsylvania and Wisconsin.

“We’re asking farmers chosen to participate to set aside at least 5 acres in a fairly uniform field and provide cropping history,” Conley says. “You will use Agroptimizer to plan recommendations, then plant the trial.”

Follow your normal practices for soybeans on the rest of the field, Conley says. Then, program staff will provide scouting and soil sampling to collect various layers of data, sharing what they find with you. After harvest, the staff will analyze data. Your confidentiality will be protected, and you will receive a small honorarium.

“The biggest benefit is that you will help us validate results and improve accuracy,” Conley says. “The tool will provide insights for best management practices in your fields that can help increase yield and profit in the future.”

Special project, Part 2

The Cool Bean research team also wants to test a scouting alert system, developed around a new tool that uses Sentinel-2 satellite images to automatically calculate a plant health index called the Normalized Difference Vegetation Index. NDVI will be automatically generated every five to 10 days. These grids can be used to guide precision scouting efforts.

“We would like to scout a field on your farm this season,” Conley says. “We’ll do all the work! Just grow your soybean crop normally.

“We will come in and scout your field every two to three weeks, looking for insects, weeds, diseases, growth stages and abiotic stressors. After harvest, we will ask for your yield monitor data.”

As before, confidentiality is paramount, and a small honorarium will be offered. To learn more about either project, contact John Gaska at UW at [email protected], or call 608-220-2693.

You can also contact the project lead Extension person in your state: