Thermal imagery, unmanned aerial vehicles, infrared sensors and machine learning once were considered government technology used to identify enemy activity. Today, it’s the next generation of precision agriculture that helps farmers stay profitable and meet the consumption needs of a growing population.
While it may be evolving faster with today’s digital advances, some of the technology has been under wraps for decades.
Michigan State University Foundation professor Bruno Basso with the Department of Earth and Environmental Sciences said some of the first applications of remote sensing imagery and crop modeling were used by the U.S. government during the Cold War to monitor and predict Russian wheat yields.
Today, he says it’s being used for more peaceful proposes to quantify food security across the globe and environmental changes. Basso’s research, along with several other precision ag resources, was discussed July 26 during MSU’s Agriculture Innovation Day.
Don’t fertilize low-yielding areas
Aerial imagery has applications for crop scouting and identifying field conditions, but Basso says so much more can be done. “The technology is now at a stage where it is able to convert what an image detects into the amount of nitrogen that has already been used by the plant,” he explains.
Aerial images from drones, airborne planes or satellite can detect areas in fields doing well and areas struggling. The simulation model he developed provides the amount of nitrogen available to plants from the soil, such that the difference between supply from soil and demand from the plants can be varied across the field with a prescription map, just like doctors would prescribe medicine.
The Basso Lab focuses on research designed to understand and predict the effect of management decisions on profitability and environmental outcomes using remote sensing imagery and process-based crop simulation systems.
“We’re trying to help farmers clarify how this technology works and how they can make more informed decisions to use fertilizer more efficiently, so they can increase profitability and reduce losses to the environment,” Basso explains.
Basso has been able to break yield variation in fields into three categories: high and stable, low and stable, and unstable. The high and stable zones consistently produce about 200-bushel corn yields and provide growers with $210 in profits. The low and stable zones consistently underperform, yield about 130 to 140 bushels and cost growers about $56 per acre.
The unstable zones inconsistently have high and low yields but averaged about 145 bushels per acre and only provided one or two dollars in profit. The analysis was based on $3.80-bushel corn with total costs averaging $550 per acre.
While it’s usually recommended to apply more nitrogen in low-yielding areas, Basso says the research is finding it doesn’t work that way. “If these areas always have low yields, that shows fertilizer wasn’t the problem because they’ve been putting on the same amount over the years and never got a bigger yield,” he says.
He suggests pinpointing low-yielding areas and applying less while adding more nitrogen to the higher-yielding zones. “We need to start managing for profit and not for yield because there are areas that have just not been gifted as much as the other areas, so you can make more profit by reducing fertilizer applications where there is no response,” Basso says.
Basso’s remote sensing and crop modeling research has compiled remote sensing images from more than 70 million corn acres in 10 Midwest states to better understand nitrogen use. “Farmers could reduce up to 30% in economic values basically by applying fertilizer in areas where plants respond to fertilizer application,” he says.
The technology also can apply to seeds and other inputs, but nitrogen remains the largest cost for farmers, which is why Basso is studying its effect. His lab also is exploring how low-producing areas could be converted to other crops (native vegetation, bioenergy biomass crops, hemp, hay etc.) for greater returns and positive environmental benefits.
Light sensors for nutrient monitoring
Technology that monitors what goes in the front end of cows is now being used to manage fertilizer applications. At harvest, John Deere’s Harvest Lab uses infrared sensor technology mounted on forage harvesters to document constituents in forage such as neutral detergent fiber, acid detergent fiber, starches, and proteins for forage quality analysis.
Solution specialist Nathan Jenkins with John Deere says manure constituent sensing is now being used to measure nutrients in manure as its being applied and adjusts to meet fertilizer needs. “Not only are we mapping, but we can control the flow based on what the sensor sees poundage wise or gallons per acre,” he explains.
Every second, more than 4,000 data points are being taken to analyze the makeup of manure being spread. Jenkins said farmers can gain a better understanding of what nutrients fields are receiving from manure applications and reduce the need for commercial fertilizers. The manure constituent sensing technology is available for purchase starting in August.
Saving on the edges and in curves
Product line planning manager Jay Witkop with John Deere said farmers now manage fields by zones but eventually will make decisions on a per-plant basis. “We need to be as precise as we can to truly optimize not only every acre, but every plant,” he says.
To get there, he believes machine learning technology will be the next step. It compiles images, and through algorithms, is comparable to facial recognition for plants. “We’re taking thousands and thousands and thousands of pictures of crops and weeds in different situations, and then we have people who are actually sitting at computers and looking at those pictures to identify if it’s a cotton plant or a weed.” Witkop says.
The equipment also needs to physically be able to treat individual plants. “We are focused very heavily with our Blue River Technologies Co. on really building this out and understanding where some of the value and capabilities are,” he says.
Right now, Witkop says, sprayer technology is taking that approach with individual nozzles, which can provide an additional 2% to 5% savings, especially on field edges and in curves. “The machine is going at one speed, but the speed at the boom tip at the inside verses the boom tip on the outside can be drastically different and, in that case, we’re significantly overapplying our product on the inside of that curve and underapplying on the outside,” he explains.
For herbicides, that can damage plants and not kill weeds, which can lead to herbicide resistance in some cases. With fertilizer, Witkop said farmers are paying to reduce their yields in areas overapplied while shorting product on the outside of turns. The same is true with planters — overpopulating insides of curves and underpopulating on the outsides.
Long term, he sees ag equipment adding more electrical components to control inputs. “The shift has been going from mechanical to hydraulic to electrical,” he says. “That electrical transformation is happening because of the precision control it provides. That’s going to give us the variability to change things across an entire spray boom or planter.”
There’s an app for that
Sometimes solutions are simple and free. Field crops Michigan State University Extension educator Monica Jean says finding the low-hanging fruit of ag technology can help farmers solve some problems for free with little effort.
“I don’t have a lot of time, and I think a lot of farmers are in the situation where they don’t have a lot of money,” she says. Jean and field crops team member Eric Anderson presented “There’s an Ag App for That” to help farmers search for free apps and how to evaluate if they’re working. Their session compared the xarvio app to other free weed ID apps. Xarvio also uses the camera app to identify pests and diseases.
Their first recommendation is to learn who owns the data and identify the management area farmers want to improve. “Would you like this more on a phone?” Jean asks. Do you want it integrated on your farm, so it can be on a desktop, on your iPad or phone in the cab with everything feeding into each other? Or do you just want it to be an app?”
Common free apps to look for, she said, can help with sprayer calibration, map the spread of pests or diseases and identify weeds. “You can download it; you can try it and if you don’t like it, you can delete it,” she suggests.
If cellphone or internet service is an issue, Jean recommends trying apps to see what functions are limited or if they can work offline. Both Jean and Anderson, who work in the Upper Peninsula and southwest Michigan, say lost service is an issue. “You need to know where your dead spots are, what apps you’re using and whether they’re going to allow you to work offline or not,” Anderson says.
Other questions to ask include: Do you need an app or a data logger? Is there a free version out there, or do you already own the software to a different comprehensive program such as Encirca, which may be a service you’re paying for with the capabilities you want?
Anderson recommends speaking with ag service providers to take inventory of what you’re already paying for before searching for new apps. “For mobile apps, whatever you’re going to be doing in the field, like scouting, would be what you want an app for rather than something that can be done with better software from a computer,” he explains.
MSU Extension also is working with specialists in Ohio, Kansas and Indiana to help farmers understand emerging mobile technology, big data, drones and prescription field data. “We’re not good at everything. There’s always one or two things on the farm you struggle with, apps can help,” Jean says.
Heslip works as the Michigan anchor/reporter for Brownfield Ag News.