Farmers are prone to tinker and experiment with standard agricultural equipment, and it was probably inevitable many would do the same when computers came to the farm.

Terry Griffin, assistant professor of economics at the University of Arkansas Division of Agriculture, has seen and worked with adventurous, technologically curious farmers for years. He discussed with Delta Farm Press his past and current work, how farmers are using technology to conduct their own experiments, and what’s in the technological pipeline. Among his comments:

On the early use, and evolution, of on-farm technology…

“Through the USDA, I got access to the ARMS (Agriculture Resource Management Survey) data set, which among other things, showed how farmers are using technology, their adoption rates of yield monitors and GPS.

“There were really good data for soybeans (sampled in 2002), cotton (sampled in 2003) and corn (sampled in 2005). The information suggested that for soybeans and corn, the third-highest use for yield monitors with a GPS unit was to conduct field experiments.

“In 2003, for cotton, conducting field experiments was the highest use for yield monitors with GPS units.

“From 2003 through 2006/2007, I worked with a group of farmers and spent a lot of time with them during my graduate program. I went to their farms and observed, asked a bunch of questions and saw how they do things.

“I came away with the understanding that, one way or another, all of them were doing on-farm experiments — ranging from formal tests to something as simple as trying to plant a new variety in half a field to compare with another variety. Some were looking at multiple varieties in the same field, others were checking variety/fungicide interactions.”

Where was this work done?

“It was at my last location, at Purdue, as research for my dissertation. But, the farmers I worked with were from all over. One was from Arkansas, another was in Canada. It was a broad group.


For a free subscription to Delta Farm Press Daily, click here .
“I looked closely at five farms and seven on-farm experiments. One farmer told me that not only did the new technology allow him to make decisions with more confidence, but he also made decisions more quickly. Instead of pondering issues through the winter like they used to do, they were able to make decisions in November, right after harvest. The harvest data were available and could be analyzed quickly.”


On the importance of yield monitors…

“Prior to yield monitors, farmers still wanted to do harvest experiments, but doing anything like that slowed harvest terribly. Now, with yield monitors and GPS, they’ve figured out that good data don’t have to come at the expense of speed. They’ll check Variety A versus Variety B across a field — all kinds of things that can work in simpler designs and are easier to implement.

“Some farmers I’ve worked with are interested in doing their own data analysis, and we did some workshops in 2005 and 2007 to help them. In all, 31 farmers attended for use on their own operations and, potentially, to help other farmers in the winter season. I’m still in contact with some of them and they’re doing elaborate on-farm experiments, analyzing data and making decisions based on their findings.”

What’s the specific computer program being used?

“There are several options for farm software, and growers have different reasons for choosing each. FarmWorks, MapShots EAiSuite, SMS from AgLeader are all commonly used, and MapShots’ Ag Studio is just being released. Farmers may choose one over the other based on their yield monitor manufacturer or their desire for the best financial record-keeping package.

“In the workshops, I was training them how to use GeoDa, which is now based out of Arizona State University after moving from the University of Illinois. GeoDa is a free download and several of our grad students are using only this program for their thesis work. It has a point-and-click interface, so it doesn’t take any special skills to use the software.

“We’ve taught farmers how to bring data into the program, analyze it and interpret the results. It’s no longer hard to send data through software — the real opportunity we have is to teach people how to interpret the results.

“It’s much like a soil test: Everyone understands the process and need to pull a soil test, packing it up and sending it to the lab. They may not know how to run the lab equipment, but they know how to read the pH level and figure out what to do from there.

“We try and do the same thing with farmers. They may not know how the computer software and this equipment actually work, but they can learn how to interpret results of their on-farm experiments.”

On Griffin’s more recent work with cotton…

“I’ve been in Arkansas three years, and most of the work I’ve done has been with cotton farmers, who are really interested in learning more about site-specific management. Cotton Incorporated and the Arkansas State Support Committee for cotton have funded several projects on which I’ve worked.

“An ongoing test is on seeding rates. Three farms — one in the north, one in the south, one in east-central Arkansas — are collaborating on this three-year project. They’re planting different rates of cottonseed: one very low and one very high, along with three rates that are in ‘normal’ ranges.

“The really neat thing is, the farmers are so interested in knowing the answer on seeding rates, they’re willing to take a yield hit. That shows how badly they want to know the optimum rate for their local conditions and management practices.

“We’re also looking at soil variability and how seeding rates may actually need to change, not only across a farm but in a field. We’re trying to make seeding rate recommendations based on soil texture. How much sand or clay is in the soil? What is its water-holding capacity?”

On several other projects…

“Another three-year project in three locations involves nitrogen and potassium. We’ve been testing university recommendations to see if they’re high enough for the farms. We’re evaluating the data and doing follow-up sampling to see how the fields have changed or remained stable.

“Another project involves nematodes and nematicides, again, based on soil texture with in-field variability.

“We’re doing more on-farm work for several reasons. One, we want to know how these systems and products respond in a ‘natural’ environment, as opposed to a more-controlled research environment. Two, we’re interested in site-specific changes — often there’s not a constant soil type: going from sand to silt-loam in the same field, for example. In that sort of field, there may be three results from the same product. Now, we have the ability to measure and analyze factors involving soils and elevation differences with yield.

“Even in fields without a planned variety or fertilizer experiment, farmers are using similar tools to determine how yield responds in each soil type or other management zone. There’s a potential experiment in every field that contains multiple management or productivity zones. “Using the proper analysis, farmers can determine the differences that exist in each field when all management inputs are constant.”

New technology farmers should watch for?

“There are several things. One thing I’m excited about is that when yield monitors came on the scene, it allowed us to collect a lot of yield data at a relatively low cost. Once we had GPS and automated controllers, it lowered the cost of implementing experiments. We could design an experiment in the winter and, come spring, we could just take it into the field and do it.

“Right now, the hardest part of the whole process is analyzing the data. There are things coming down the pipe that will make this even easier for farmers. I hope it’ll also mean the analyses will be automated and won’t need a human to handle the data. Right now, there are very few people who are able to do analysis.

“Other technologies are interesting. Scott Monfort, Arkansas Extension plant pathologist, is working with an on-the-go soil temperature reader. He has an infrared thermal coupler that he plunks into the soil by the disk opener.

“Getting those readings is important from a pathology standpoint, and he’s working with it on a site-specific basis.

“I’ve got a couple of Ph.D. students doing some great work with precision agriculture. One is a weed scientist looking at spatial drift of herbicide-resistant plants across the field. Another is a pathologist looking at spatial dispersion of pathogens within a field and what’s causing it.”

e-mail: dbennett@farmpress.com