Why exactly do farmers' yields vary, and how can they capitalize on those variables are two questions William Batchelor has worked toward answering since about 1994.
For several years, the new ag and biological engineering department head at Mississippi State University directed his precision farming work on soybean fields throughout the Midwest. Now the former Iowa State University official hopes to help Mississippi farmers through similar models.
Crop growth models are computer programs that calculate the growth of corn, soybean, cotton and rice plants on a daily basis, Batchelor explained to an audience recently at the Land Bank of Mississippi's stockholders' meeting.
“These models calculate the daily growth of a plant and how much growth is reduced by the stress it endures daily,” he said. The models account for sunshine, rainfall, temperature maximum and minimum ranges, management practices, genetic characteristics and soil properties.
Farmers know that insects, weather, disease and weeds interact over the course of a season to affect plants, but they have difficulties pinpointing the effects.
Batchelor said creating models involves three basic steps. “The first application is to understand why yields vary, the second is to find management zones in the field and the third is to develop optimum prescriptions,” he said.
“That is the question we have received from soybean and corn growers for many years. They know the yields vary, but they ask, ‘How do I develop an optimum nitrogen prescription or herbicide application to capitalize on that variability?’
“Now we can identify areas in the field where water stress does not explain the year-to-year variability.”
These kinds of models emerged during the 1970s, he said, partly in Mississippi, during an era when the federal government was eager to stabilize domestic wheat and corn prices. To do so, the USDA needed to predict how much of the commodities the USSR might be purchasing from U.S. farmers.
In Iowa, researchers have put theoretical models into practice, dividing certain soybean fields into 1- to 2-acre grids, which enable them to minimize the error between simulated yields and measured yields over a multi-season analysis. Scientists then compare and contrast data.
“One thing we used the precision models for is yield analysis. We can estimate the yield loss associated with different field stress factors,” Batchelor said.
“Our farmers were very interested in knowing how much soybean yield they are losing from soybean nematodes, and whether or not it would be economical to treat them. (They ask) ‘How much yield loss did I get from weeds this year? From water stress?’
“These are shown in a spatial pattern across the field.”
Models, Batchelor said, have thus far been able to explain more than 72 percent of variables in crops in Iowa, northern Illinois, South Dakota and central Missouri.
Not only are researchers working toward identifying causes of variability, but they are also creating associated economic models.
Batchelor said scientists have developed a technique within the model's framework that can estimate what a farmer's payoff would be in transitioning from a uniform rate management to a variable rate management.
“We have done a lot of work trying to determine whether a variable rate prescription is a $1-per-acre return, a $10-per-acre return or a $30-per-acre return,” he said.
While scientists can never predict weather conditions with absolute certainty, Batchelor said, they have devised a method that simulates the profit associated with different populations for 20 to 30 varying seasons, based on different weather data.
“As it turns out, if you come up with the average yield response as a function of population over a long period, you can pinpoint the breakeven costs of moving from single to variable rate population management for soybeans,” he said.
Batchelor noted that growers in the Mid-South and Southeast have an advantage in weather forecasts that many other regions don't have. “We have the advantage of knowing something about what weather may occur based on whether we are in an El Niño or La Niña or a neutral weather year pattern,” he said. Knowing in December what weather pattern to expect can make produce dividends using these precision models.
Recently, Batchelor has monitored a cornfield in Germany, where farmers are interested in precision model data linked to environmental impacts. Germany, he said, is interested in decreasing the amount of nitrogen deposits in the public's drinking water. The country offers farmers substantial subsidies based on how much they limit such nutrients.
“In fact, our analysis has shown that dominates their decision-making. They are better off putting out minimum nitrogen and living off the credit than they are trying to sell corn in today's open market prices,” Batchelor said.
Such incentives in Europe is a trend American farmers should take note of, he said, because the federal government is slowly moving its farm subsidy program in a similar direction.
“From a farm bill standpoint, it is moving much more from a subsidy-based bill to one more focused on conservation efforts,” Batchelor said.
“My perspective is finding how much it is going to cost you, the farmer, so we can provide information to the agriculture committees as they negotiate standards and thresholds. We have never had to a way to calculate the direct cost of an environmental policy, and crop models give us a way to calculate that so we can price environmental credits that may come up.”
Batchelor added that legitimizing precision crop models could be a future benefit to farmers in regards to securing crop insurance policies and other farm-related financial loans.