Setting Up Side X Side Comparisons With A Yield Monitor

Crop Insight Vol. 7, No. 10
by Todd Peterson, Precision Farming Specialist

Summary

• The availability of new genetic and seed/herbicide technology has increased the need for effective and accurate on-farm evaluations of the new products available.
• When evaluating new products, growers should not rely on results from just one location or a small number of environments. Instead, they should combine their data with similar comparisons to improve the predictability of future results.
• Yield monitors and mapping systems can make it very easy to gather more yield and performance data from side by side hybrid and variety comparisons.
• Using a split-planter design along with a yield mapping system will allow later mapping of differences between adjacent hybrid strips across variable conditions within a field.
• Pioneer is working with software developers to evaluate prototype products that create difference maps from side x side and split-planter hybrid/variety comparisons.

The adoption rate for new genetic and agronomic technology in corn and soybean is astonishing. Just a few years ago, biotechnology was a futuristic promise that many thought would never play a major role in the seed business. Today, crops protected from insect damage by a bacterial gene and crops with specific herbicide resistance are significant forces in the market. Perhaps as important as seed products from biotechnology are new hybrids and varieties from conventional breeding programs, which continue to deliver improved agronomic and end use traits to seed customers. Progress in developing new genetics decreases the average life span for new hybrids and varieties, so the rapid identification of superior hybrids or varieties is increasingly critical to profitable farm operation.

Farmers often want to test a new hybrid for a year or more before committing significant acres to the new product. The bottom line for farmers (and those who advise them), is the need to make the best hybrid and variety choices appropriate for their farm conditions and operation. The rapid development of seed/herbicide systems complicates the decision, and increases the importance of actively testing these products on farmer's fields. This Crop Insights will revisit some of the principles of comparing hybrids and varieties in on-farm evaluations, and describe how to best use yield monitors and mapping systems to get the job done.

Combining data across locations important

An on-farm hybrid or variety test plot provides valuable information about how products perform at that location and during that growing season only. The weather and conditions during the growing season have a tremendous impact on the relative yield or performance of each hybrid or variety tested. The ability to predict how a hybrid or variety will perform the next year is severely limited unless results from a number of environments (or locations) are combined. Put another way, combining results from different locations increases confidence that the way the products performed was not just due to chance, nor was it due to the unique conditions at a few locations. As a general rule, comparisons of hybrids or varieties should include at least 10 different locations in order to make a prediction about how that product may perform in the future. The following table points out how the probability of successfully predicting which hybrid is "best" increases dramatically as you increase the number of locations or environments tested for a given measured yield difference.

Probability of Selecting the Best Hybrid

Predicting future hybrid performance from results obtained at one location is only slightly better than a coin toss. Note that to correctly detect a 6 bu/acre yield difference between two hybrids 19 out of 20 times, it is necessary to make the comparison at 200 locations. However, to detect a 12 bu/acre difference with the same frequency (95% of the time), only 30 locations would be required. Increasing the precision of the comparison (i.e., detecting smaller differences) requires more locations to assure that the yield differences are not due to chance alone.

Limit the number of comparisons

An extraordinary number of new seed products are available each year, which complicates comparison testing and decision-making. Increasing the number of entries in a test plot can increase the amount of variability in the study to the point where prediction of future performance is diminished. Large mega-plots organized by some grower groups or clubs may be good for field days and enable many farmers to "see" new seed technology and products. However, they may not be the best way to pick hybrids and varieties to plant the following year. Growers are better off limiting the number of entries, and combining data from similar comparisons at other locations. The most efficient and powerful way to compare hybrids is to simply check their performance when planted in side x side strips at numerous locations.

Using a yield monitor to compare hybrids

Yield monitors can greatly simplify the process of gathering data at harvest time. All of the commercially available systems can keep track of individual "plots" or "loads" determined and defined by the operator. Each strip in the trial is usually assigned a unique load number (or name), and estimates of total pounds of grain harvested, average grain moisture, area harvested, and calculated bu/acre are stored by the monitor for each strip or plot.

While there have been concerns that different hybrids may react differently when yields are estimated using a weigh wagon versus a yield monitor, there is no consistent evidence that this occurs. There have been a few cases reported in which differences in test weight or grain moisture were so extreme that the yield monitor calibration may not have been adequate to handle the comparison. In general, these were cases with circumstances well outside the normal range of harvest conditions. For example, we recommend not using yield monitors to compare hybrids that vary more than 5 lbs/bu in test weight, or 5 % in grain moisture (see 1996 Guidelines for Using Yield Monitors to Collect Pioneer Strip Trial Data, Crop Insights, Aug. 30, 1996). Pioneer is continuing to study this issue in cooperation with major yield monitor manufacturers. Yield monitoring systems are improving, and we are quite comfortable overall using yield monitor data to compare hybrid and variety performance. Our goal is to generate a valid comparison regardless of the weighing device used to estimate yield. Proper plot design and harvest technique can help insure data quality when the weighing device is a yield monitor.

Comparing hybrids with a yield mapping system

Side x side hybrid or variety comparisons can provide even more information when the combine yield monitor is connected to a differentially-corrected GPS receiver and equipped to store yield map data. The advantage of a yield map comparison over a stand-alone yield monitor is that it provides information on spatial differences in hybrid performance in the field. This does not replace the need for comparisons at multiple environments or locations, but does provide more information about hybrid performance than simply having one estimate of yield for a hybrid strip.

Two hybrids that produce equivalent yields when weighed over an entire strip may differ significantly within the strips due to different responses to soil, topography, or past management practices. A properly planned hybrid strip comparison that maps the difference between adjacent strips across variations in the field can provide better information about the product's performance and suggested management in the future.

Some farmers have developed the technology to change hybrid or variety on-the-go as they travel across a field. This technology may be best utilized in very specific situations, such as planting iron-chlorosis tolerant soybean varieties or corn hybrids in areas with high soil pH. Mapping the performance difference between adjacent hybrid strips across a field may support the need for this technology, or help identify circumstances in which changing hybrids or varieties on-the-go makes sense.

Setting up split-planter hybrid/variety comparisons

A very simple and effective way to set up meaningful hybrid or variety comparisons is by using a split-planter design. Ideally, the width of the combine harvest head is exactly one-half of the planter width, which is fairly common in the Corn Belt. In this case, two hybrids are each randomly assigned to one-half of the planter before planting the field, and the appropriate seed placed in the planter boxes. The farmer then plants the field as normal, resulting in pairs of adjacent hybrid strips across the field.

This design avoids the need to stop and clean out planter boxes, which saves valuable time during the busy planting season. To guard against any bias toward one hybrid or another, consider making at least six passes through the field with the split-planter configured one way, then switching hybrids to the alternate side of the planter for a similar number of passes. Take care to insure that the seeding rate and planter adjustments are appropriate to achieve the same stand of each hybrid or variety. Logging GPS locations during planting using a laptop PC, a data-logger, or yield monitor can help track where each hybrid strip is planted. Your Pioneer sales representative can help identify effective hybrid or variety pairs to compare depending on grower objectives and field conditions.

Testing hybrids with the Bt gene side x side

There has been some concern about testing hybrids with the Bt-gene alongside normal corn hybrids, but present research does not support this concern. The questions arise because the Bt protein may be expressed in the pollen grains, which may drift to the normal corn hybrid strips. If Bt-pollen grains on the normal hybrid reduces corn borer feeding and damage, the yield of the normal hybrid may be artificially increased due to its proximity to the hybrid that expresses the Bt protein.

Little is known about the so-called "halo" effect, but many researchers will be studying it this summer. We will support side x side comparisons of Bt- versus normal corn hybrids until we learn more about the effects of Bt-protein expression in drifted pollen. If the Bt pollen does reduce corn borer feeding in adjacent normal hybrids, it only means that the comparative performance of hybrids with Bt is conservative, and that the Bt-gene may add even more value than previously believed.

Harvesting the split-planter comparison

Data from the split-planter hybrid comparison can be collected using a yield monitor, a yield mapping system, or a weigh wagon. Weighing pairs of adjacent strips several times through the field can provide an estimate of the variability involved in the comparison, and provides confidence in the final results. Using a weigh wagon along with a yield monitor can help assure the calibration and function of the yield monitor. Care must be taken at harvest to save each combine pass as a separate "load" on the yield monitor, and to record which hybrid was measured in each load.

If the split-planter comparison will be harvested with a yield mapping system, care should be taken to harvest each pair of adjacent strips while traveling in the same direction. This helps reduce errors in handling and assigning yields to a location, particularly if the strip traverses a slope. Also, the yield monitor should be set to record data once each second, which will provide a nearly continuous yield comparison between the two hybrids across the field. This will allow construction of a map of the difference between adjacent strips.

Pioneer is working with developers of mapping software to provide a means of mapping the difference between adjacent hybrid strips, and hopes to pilot these programs in time for the 1997 harvest. Contact your Pioneer sales representative or field sales agronomist for further instructions and assistance in setting up these comparisons.

References:

Using test plots to improve hybrid/variety selection. Crop Insights, Oct. 16, 1992.

Conducting on-farm strip tests with a yield monitor. Crop Insights, March 29, 1996.

1996 Guidelines for using yield monitors to collect Pioneer strip trial data. Crop Insights, Aug. 30, 1996.