It takes a lot of computer power to cuts costs with the variable-rate application of inputs. Can cotton producers afford it?
There's no doubt that VRA technology has already cleared the mechanical and electronic engineering hurdles needed to start saving cotton producers as much as $50 an acre and improving yields by 30 to 50 pounds per acre.
There's even a rough outline of how various groups should work together to deliver the technology to the farm. Basically, the farmer will be responsible for gathering GPS (global positioning system) information from yield monitors and aerial imagery.
He will turn that information, stored on a disk, over to a precision ag specialist who will then process the information, develop a prescription for the VR application of inputs and give that information back to the farmer or applicator on a disk.
The aforementioned precision ag specialist will be the backbone of VR technology in Mid-South cotton. Unfortunately, a farmer with a few hundred acres of cotton can't afford to hire him along with the hundreds of thousands of dollars worth of software, hardware and computer processors that are also required. And few businesses or cooperatives are willing to risk resources for a technology that is yet to be field-proven.
JSCC precision agriculture professor Tim Sharp says it's going to be a huge challenge.
“Handling all the precision agriculture information is like trying to drink out of a fire hose,” Sharp said. “Right now, we're just diverting a little bit of that stream. But we have to develop techniques for capturing the whole fire hose, manage everything that's coming out and turn it into useful information in 48 hours. The challenge is, how do we ramp it up?”
Sharp stresses that universities with precision agriculture programs like JSCC and Mississippi State University are already involved in educating students in VR technology/remote sensing. These universities also have much of the computer hardware and software required, so it's logical that they should be an integral part of bringing the technology to the farm.
Sharp suggests VR pilot programs of 20,000 acres or so would be a logical next step. While private industry has expressed some trepidation about the involvement of government in the process, Sharp doesn't see any other way to go, at least for now, given the economic environment.
Here is one possible application of VR technology in cotton.
A cotton field is divided into as many as five geo-referenced management zones. Each of these zones represents a yield potential ranging from very high to very low. The technology for creating these zones, normalized differential vegetative index (NDVI) is on the market. The NDVI is generated from multi-spectral imagery shot from aircraft and is a very good pre-harvest indicator of potential yield.
In the very low-yielding zones, growers will want to lower their inputs to match the yield potential of the zone and vice-versa in the higher-yielding areas. Of course, not all inputs can be varied. For example, you don't want to lower the rate of a herbicide in a low-yield zone. But you likely would want to apply a lower rate of Pix or nitrogen.
All of the decisions for varying rates are made by the computer, but the equations and formulas that drive them are based on sound agronomic principles. In fact, these decisions may be similar to ones the farmer makes on a larger scale.
For example, a farmer doesn't want to spend more money than he can possibly make on any one of their fields. So he looks at the yield goal of a particular field and makes the necessary adjustments in fertilizer, nitrogen, etc. VRA farming takes that philosophy and applies it to individual management zones within each field.
Sharp says the role of the farmer is “to adopt variable-rate equipment, the controllers and the technical ability to physically apply variable-rate applications. He doesn't have time to learn the rest.
“Beyond that, there is going to have to be an organization in place to make it transparent to the grower. The support organization needs to be a coalition of federal agencies and other state agencies to pool different bits of information to make a data processing organization that has the computers, the specialists and access to the information.
“There has to be a field organization arm, too,” Sharp added. “That can be the role of private industry. Somebody has to be one-on-one at the grower level. The insect scouting still needs to be done. Once that disk is created, private industry will make it transparent to the grower.”
“In the long run, VR application of inputs will likely be a private enterprise venture,” said Alex Thomasson, associate professor in agricultural and biological engineering at Mississippi State University.
“In the short run, I think the Extension folks are going to be heavily involved. I see my role working with the producers as trying to validate the technology in a commercial application.
“Beyond that, I would think that Extension will begin to teach the producers how to use the technology. But I don't know how deep into the technology your average farmer is going to get.”
One job of any pilot program is to correct as many inefficiencies as possible.
For example, “The NRCS, FSA, DOT are all making camera passes (doing aerial photography) for their own purposes,” Sharp said. “Why not pool all this information that is being spent by the government agencies and do one set of shots that everybody can use.
“It makes a whole lot more sense. You could have one processing center that everybody draws from. Growers would have to pay for the images, but it would be a lot cheaper.”
e-mail: [email protected]ness.com