Wouldn't you like to cut your herbicide bill in half? Christy Sprague, assistant professor of crop science at the University of Illinois, has been working over the past two years to make that a reality.

Sprague and her team of researchers have used airplanes, satellites and variable-rate sprayers to delineate weeds from soybeans, calculate weed densities and reduce herbicide application rates.

The remote sensing project is part of AG 20/20, a joint USDA and NASA program. They're hoping that in a few short years, their research will have commercial applications that could reduce your chemical rates and costs.

The researchers apply a reduced rate over much of the field. But Ken Copenhaver, Spectral Visions, Champaign, IL, speculates that they could use even less herbicide.

“Right now we're erring on the side of caution,” he says. “Based on what we've seen, once we perfect the process, it's realistic that we can eventually get down to using 50% of the labeled rate.”

Spectral Visions takes photographs of soybean plots by plane and satellite using a range of light wavelengths, from visible to near infrared. These images show researchers that weeds can be distinguished from soybeans in conventional and narrow rows as well as differences between grasses and broadleaves.

Based on the compiled data, a field map is made. In areas with high image reflectance, or more weeds, researchers use a higher herbicide rate. “Applying Roundup UltraMax, we used a 100% rate for the highest weed infestations, 66% for a medium reflectance level and for low infestations we used a 33% rate,” says Sprague. “We pretty much had 100% control on the different patterns that we used, based on the types of infestations that were picked up from the imagery.”

Picking out problem spots using current image resolutions has been a challenge for the research team to overcome. On a satellite image, one pixel is 4 meters square, while an airplane image is three times larger at 1 meter square.

While the higher resolution of the aerial images makes them more practical for now, the project's economist, Gary Schnitkey, is looking forward to the day when the satellite resolution catches up with the current aerial resolution.

“You've got to be able to process the imagery en masse for this to have potential,” he says. “If the flyover technology would be replaced by satellite technology — where you shoot a whole county or state at one time — then it would be economical. You've got to be able to take that data and process it at a reduced cost for this to work.”

While the soybean project is still in the research stages, its potential is real.

“The postemergence herbicide application in soybeans has a high potential for success,” says Copenhaver. “We go out at a stage where soybeans don't show up very well in the imagery, but the weeds show up nicely.”

This all adds to savings for farmers. Schnitkey estimates savings between $1.70-2.40/acre for 2001. The estimate includes image costs, but not equipment costs.