In-crop optical sensing, also called canopy sensing, is boosting yields and cutting costs for Robert Goettl, his brother Jerome and nephew Justin. Their GreenSeeker crop sensing system allowed the Le Center, Minn., farmers to shift away from reliance on all fall nitrogen (N) application. For the past four years they have combined a base fall rate with variable in-season applications using a coulter cart. Yields were up 10 bu./acre or more in the drier-than-normal 2012 season.
“We like to try for a base of 130 lbs. of nitrogen and then top off with the in-season liquid,” says Robert. “In the past we might have put on 200 lbs. of N, but not get 200 bu. of corn. With the GreenSeeker system, it is almost a one-to-one relationship.”
Micah Eidem, GreenSeeker market manager for Trimble, suggests adoption of optical sensing is finally pushing past early adopters. It has been two decades since GreenSeeker first hit the market, with OptRx from Ag Leader and CropSpec from TopCon being introduced more recently. Reaching this tipping point hasn’t been easy. Even the terminology, such as NDVI (Normalized Difference Vegetation Index) sensing, can be challenging. Add to that the technical differences in how the sensors work.
“Educating potential users is the biggest challenge,” says Eidem. “The university guys understand what the technology does and that there is solid data behind it. Research continues to come back positive, and now the growers are starting to understand it, as well.”
Technical know-how needed
John Sawyer, Extension fertility specialist, Iowa State University, and Ken Sudduth, agricultural engineer, USDA ARS, are two researchers who have worked with the technology. “Use requires an understanding of what the sensors do and how to properly implement them,” says Sawyer. “Through plant canopy sensing, they provide an indication of biomass and plant stresses. Sensing results can be related to nitrogen or various other stresses like other nutrient deficiencies, low population and moisture stress.
“The challenge is trying to identify what the stress is, when to collect the information and at what stage,” he continues. “Using the concept of relative sensing, by comparing to known non-stressed field areas, helps identify the true stress-causing plant canopy differences.”
Sudduth reports that the sensors offer fairly comparable accuracy. “They differ in sensed wavelengths, size of sensed area and sensing orientation, with GreenSeeker and OptRx pointing straight down over the corn row. The CropSpec is mounted at an angle and looks at a combination of the top and sides of the crop,” he says. “Yet, if you look at the data comparing them, all three have strong, straight-line relationships, indicating they are responding to crop variability similarly.”
Perhaps more important than design is the algorithm used. It encompasses the process that turns a sensor reading into an N application rate. There are different algorithms for different crops, states and even regions within a state. An informal working group is trying to develop a common algorithm for corn that can be used across the board.
“The sensors produce different numerical data, with GreenSeeker and OptRx being more similar, so you need to take that into account when using an algorithm developed with a specific sensor,” warns Sudduth. “The same algorithm could work with a different sensor; however, the numbers may need to be adjusted a bit.”
Hard data plus intuition
Robert Goettl admits that adopting the technology requires flexibility and a certain level of intuition. A week to 10 days ahead of field application, the brothers create an N-rich strip by applying a minimum 10-25 gal. N, depending on field history, to an average part of a field.
The corn’s V5 stage or a little earlier is Goettl’s optimum window. An initial dry run through the field across the N-rich strip and base area lets him set parameters for minimum and maximum rates. This is where field knowledge comes into play.
“Some fields may indicate a range of 0-35 gal., while others show 10-25 gal.,” says Goettl. “We find a broader range is needed on clay hills than on flat, black ground. And the more variability in the field, the broader the range you want to set.”
The Goettls say they are now putting more N down where it is needed, and the timing of the application allows the crop to use it more efficiently. Overall, they are using much less N than they did before using GreenSeeker, though their base rate has increased and their in-season 28% and 32% applications have remained constant overall.
“We average around 16-18 gal./acre, but it can vary from zero to 35 gal./acre,” says Goettl. “We know we are making better yields and making more money.”
A Nebraska farmer, Matt Helmke, relies on a customer with an Ag Leader OptRx system to variable-rate apply N to his corn at the V6 stage. The precision ag specialist for a local retailer also raises corn and soybeans near Pickrell. “The system is mounted on a Hagie sprayer with drop nozzles on the boom,” says Helmke. “We saw more than a 10-bu./acre increase in 2011 using the same amount of N, just rearranging where we put it.”
Helmke found that corn planted on ground with lower organic matter and lower soybean yields the previous year matured too quickly due to N deficiency. “Taking out the nitrogen stress allowed the plants to increase seed-fill time and increase yield,” he says. “Environmentally, we are now more efficient with our N because there is less chance of runoff or leaching, compared to fall anhydrous ammonia applications. We also avoid disturbing the soil and losing moisture, which can happen when knifing in N.”
With few growers in his area side-dressing their corn, Helmke isn’t surprised at the slower rate of optical sensing adoption, compared to other precision ag technologies, noting that it requires a change in operation. “It’s like any other new technology,” he says. “Interest is high, with everyone looking at the early adopters who have implemented it in their systems.”
Easier adjustments for weather
Optical sensors may also help growers deal with extreme weather patterns. Doug Weist, who farms in north-central Montana near Choteau, uses his CropSpec sensor system to better match N applications to soil types, but also to available moisture on his wheat, barley, dry pea and canola fields. In coming years, he plans to add corn and expects sensors to help make that crop more economically viable.
A dryland farmer in an arid region, Weist has transitioned from a grain/fallow rotation to a 4:1 or greater crop/fallow mix. More efficient use of N through in-crop, multiple VRT applications have helped make the transition pay.
“We can control everything but the weather,” says Weist. “Traditionally, wheat and barley growers put on all their fertilizer in January and February and hope for rain. We try for a base rate and then apply in-crop with the sensors, depending on soil moisture and likelihood of rain. If warranted, we’ll make three to four variable-rate applications over the season.”
CropSpec allows Weist to use dynamic calibration to adjust rates on a relative basis across the field; no N-rich strip needed. He varies application amounts from 5 to 15 gal./acre based on an average reading for the first quarter mile. His truck-mounted SprayFlex spray system lets him react quickly to rain forecasts.
“If there is an 80% chance of rain, I can cover the whole farm (up to 2,500 crop acres) in two days,” says Weist.