<figcaption> Credit: COURTESY OF MIKE TUCKER, UNIVERSITY OF GEORGIA</figcaption>
Credit: COURTESY OF MIKE TUCKER, UNIVERSITY OF GEORGIA

Three years ago, George Vellidis, an agricultural engineer at the University of Georgia in Tifton, was speaking to an electrical engineering colleague who was working for the military to develop radio frequency identification (RFID)-equipped "smart dust" that could be sprinkled on battlefields to gain information. She planted a seed in his head: Could the same technology used to find wandering pets and cattle and keep tabs on cars in the EZ Pass lane be used to help farmers water their crops?

To decide irrigation schedules, "most farmers look for the weather forecast on television, see rain, [and say] 'I'll hold off for a day or two ... or look for signs of crops being wilted,'" says Vellidis. But such imprecision often translates into less-than-optimal yield, especially with the climate changes in recent decades.

"All over the world," says Vellidis, "weather patterns are...

The laborious process of manually gauging soil moisture gave way to slicker methods when agricultural engineers "said 'this is ridiculous, we're going to automate this,'" says Vellidis. They installed cables to connect sensors in fields, but those could be cut by plows or struck by lightning. When the technology switched to standard wireless methods, a single node in a field went for $700, and few growers who weren't researchers could afford more than one, let alone enough of them to represent the full range of soil types.

So Vellidis and his assistant attached an active RFID tag - a silicon chip plus antenna - to two soil-moisture sensors and two temperature sensors placed at different depths, and a 9-volt battery. Every five minutes, each tag can gather and broadcast data from the sensors into a central receiver up to half a mile away. Growers can then download the compiled information each morning on a handheld personal digital assistant, using graphical representations to figure out when and where to water the crops. "If you have a set point of soil dryness at which you want to irrigate, you can see that point," Vellidis explains.

At the moment, putting together the contraptions by hand "is tedious," he says. "We have to do all the soldering ourselves."

The slashed price on each tag ($70 each) means that farmers can scatter numerous tracking devices to account for variable soil types, with an optimal number of 20 devices spread over 60 acres. Still, Vellidis says he is working on lowering the cost of the central receiver even more so that the system is affordable for users without research grants. "A farm is on a razor's edge in terms of profitability," he says. Growers "won't make an investment unless they can make a reasonable return."

If all goes well, Vellidis hopes the low-maintenance system will be available to an already eager consumer base - tens of thousands of irrigation-dependent farms per state, most of them still checking moisture levels manually - by the time the next growing season rolls around. A paper on the new RFID application is now under review by the journal Computers and Electronics in Agriculture.

Corzine, a fifth-generation farmer who grows corn and soybeans in Illinois, calls farming "a guessing game" in many respects. Though he "learned from my dad and granddad [to tell] when the soil is dry enough or warm enough ... this helps take a lot of that guesswork out."

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!