"There are a lot of microorganisms in the soil," he says, scooping up a tablespoon or two of dirt. "In a clod this size, there are probably 100 million bacteria, and nitrogen is very important to them in the same way it is to corn. They make all of their proteins with it. No one has come up with a chemical analysis that really measures how much nitrogen is going to be available to the plant."

A chemical analysis of soil, for instance, can show the levels of nitrogen present today, but it won't tell the farmer how much was available to the plant yesterday, or will be available tomorrow. Nor can a soil test even tell farmers how much nitrogen is present from one row to the next.

That's why the MU and USDA-ARS researchers have forsaken soil testing and turned to a more reliable indicator of nitrogen levels: the appearance of the plants themselves.

Corn getting plenty of nitrogen grows greener, taller and fuller than corn that is getting too little. To measure just how big and green the plants are growing, the researchers have begun using palm-sized sensors called GreenSeekers, manufactured by NTech Industries, and Crop Circles, sensors manufactured by Holland Scientific.

GreenSeekers and Crop Circles both utilize light-emitting diodes, or LEDs, to come up with what might be called a crop's "greenness index," a measure of crop health based on color. When the MU researchers first began experimenting with controlled nitrogen application, their sensors relied solely on sunlight reflected from the crop's leaves, or canopy. As a result, the time of day and cloud cover affected their readings. LED-driven sensors, on the other hand, use light from the device itself, so that canopy readings at midnight should exactly match those taken at noon.

Although the technology is sophisticated, data collection is simple. The sensors, mounted on a tractor with an on-board computer, take readings from each of the corn plants they encounter. Plants with sufficient nitrogen tend to score high on the greenness scale. Less green or yellow plants score lower, indicating more fertilizer is needed.

Researchers such as Ken Sudduth, a sensor engineer with the USDA-ARS, begin by driving tractors with sensor mounts over an area where they believe nitrogen is plentiful. "The idea is to have a reference in the same field, between the same hybrid of corn at the same growth period, where the corn is as green and healthy as it can be," Sudduth says. "We run sensors over that well-fertilized area and take a reading. The computer stores that reading. Then we drive across the rest of the field and take in data along the way."

Before sensors, farmers interested in precision agriculture resorted to less efficient ways to evaluate corn's appearance. One common method involved taking aerial photos of their fields, doing a comparative color analysis with programs such as Adobe's Photoshop, and then entering the data into their tractor's on-board computer. If all went well the computer, using global positioning system technology, would vary the nitrogen application rate according to the plants' fertilizer needs.

This method didn't work very well, chiefly because an analysis of aerial photos can't be done accurately until the corn is about waist high. At that point few farmers have applicators that can traverse their fields without destroying plants.

Published by the Office of Research.

©2006 Curators of the University of Missouri. Click here to contact the editor.