Ag sector leans on AI tools to better inform crop, nutrient decision making

When 2012 brought a drought to the state, it had corn growers feeling a bit skittish about their yield potential, which caused corn prices to peak.

If a similar scenario were to play out today, modeling software can provide clarity through more accurate yield forecasting.

“The model could have looked at that and said, ‘You know, you guys are probably not going to be average or have record yields, but you’re not as bad as you think you are’ and that’s the way it turned out,” said Jeff Sandborn, a fourth-generation cash crop grower located in Portland, referring to the 2012 panic.

“You would have felt a little better about taking advantage of some of those crazy market prices,” he said. 

The power to collect and analyze data is not only essential for growers, but it’s now easier and far more advanced thanks to artificial intelligence.

For decades, farmers have collected and archived data from their fields in both digital and analog ways, which can be loaded into modeling software to provide historical context. Now, through components like yield monitors, sensors on planters and even aerial imaging with drones, farmers are able to collect more extensive data.

Sandborn has diligently collected and archived data from his farm and now is working to apply that data with the help of Bruno Basso, an earth science professor at Michigan State University who specializes in crop modeling and land use sustainability.

“This is where data is important — bad data can give you bad results,” Sandborn said.

Not only does Basso’s modeling software take into consideration the history of the field to plot out stability zones — identifying areas of traditionally low, high, average and inconsistent yield — but also he uses 40 years of weather data to provide yield projections for various weather patterns.

Ultimately, the information is used to grow more crops per acre using fewer nutrients.

Another common use of precision agriculture technology is GPS grid sampling, in which farmers take samples of soil throughout multiple grids of fields to identify exactly which nutrients and the quantity of them that is needed.

Optimizing fertilization leads to cost savings for farmers, uniform crops and uniform yield while also benefiting the environment, said Dwayne Ruthig, CEO of the Caledonia Farmers Elevator Co.

Despite a perceived cultural divide between farmers that heavily utilize AI technology and traditional farming purists, Ruthig estimated that around 90 percent or more of farmers are routinely using technology such as GPS grid sampling.

“The ones that are not are typically our organic farms. Obviously, they don’t use commercial fertilizers in the same way,” he said. “They use different processes to get their nutrients. They may be grid sampling to see what they need to do in certain areas, but they are not (using) commercial fertilizers.”

With self-driving equipment, computer modeling and forecasting, and yield imaging conducted by drones already in existence, many farmers might be left wondering what could be next.

“I think we’re heading down the road to full autonomy,” Sandborn said.