Starting at wheat harvest in southern Texas in June and continuing at least until the last double-crop soybean is pulled from the pod in November, thousands of growers will be experiencing their first ride with a yield monitor at their side. Like any computer the monitor screen will produce a changing array of numbers, the result of hundreds of statistical calculations. That’s right, calculations. A yield monitor is not like a scale — dump it in, weight, and go on. Here’s a glimpse of how they “know” what they know.
“Yield is the result of a mathematical calculation using several measurable components, and information relative to the commodity,” that’s “researcheze” from Tom Krill, Ph.D., Ohio State University. Krill makes an excellent point in a presentation he calls, Yield Monitoring: the beginning or the end? His message is the “act” of yield monitoring as we know it in precision agriculture is a complex task requiring mathematical computations and geographical positioning. In other words, carrying the Phillips screw driver and crescent wrench isn’t likely to solve a yield monitor problem.
Problem? You mean problem, there are problems with yield monitors? Actually, the most common monitors in the fields today reportedly worked extremely well during the 1996 harvest season. But nothing mechanical works prefectly, right? “The yield monitor is a sophisticated piece of equipment and is only as good as its installation, calibration, and combine operator,” says Krill. Here are the pitfalls to watch out for as the harvest season heats up next year.
- Improper installation can result in improper sensor operation — get qualified help when a “do-it-yourself” installation is planned.
- A wrong calibration table can cause distortion on any mathematical calculation performed by the monitor’s black box.
- The combine operator must determine and maintain a consistent width of cut or swath.
Even though Krill gives high marks to the accuracy of today’s yield monitors he reminds us that the combine was designed to be a “smoothing machine” and attempts to maintain an even continual flow while in operation. Krill says, “this smoothing can hide some of the variation that may be occurring in the actual field.” The combine may also be unable to remove some foreign matter from the clean grain flow. “This,” says Krill, “is another factor that can affect the performance of the sensors.”
When we look at yield monitor accuracy Krill recommends that we judge it in three arenas: field, strip and spatial. The field information can be the sum of the actual net weights from weigh tickets for a particular field. An estimated weight is the sum of all calculations performed by the yield monitor for a given field.
Krill’s research with the AgLeader 2000 yield monitor and strip yield measurement indicates that this monitor was found to do a “respectable job” of estimating the quantity of grain harvested in two years of strip trial data.
“The final book on yield monitoring accuracy has not been written,” say Krill. “Little, if any, research is available regarding spatial accuracy of yield monitors.” Spatial accuracy is the calculation of yield at a specific point in the field. That’s where practical application is running ahead of research.
Most farmers with yield monitors address areas or points in the field where yields vary from the “norm,” of the measured field. These growers are asking questions and learning about their fields by relying on their observations of spatial variability noted on yield maps. Yet, research is suggesting we haven’t really documented the occurrence of spatial information from the current yield monitor. This fact is nothing to panic over according to most yield monitor owners. Most growers compare point-to-point yield monitor data for relative differences. Actual yield at any one spot is always a curiosity, but probably not a requirement for “precise” farming.
Editor’s note: This article first appeared in the Premiere Issue 1997 of PrecisionAg Illustrated.