Make yield results make sense
Like other farmers this fall, Jim Rouse and his crew saw a lot of variability in yields they harvested in the Iowa Crop Performance Test for 2011. These trials are scattered at 36 locations around Iowa, and farmers use the data each year to help decide which corn hybrids and soybean varieties to plant. The crop performance test results are online at www.croptesting.iastate.edu.
The program is administered jointly by Iowa State University and the Iowa Crop Improvement Association. Rouse is executive director of the ICIA and is in charge of the testing. The corn hybrids and soybean varieties are submitted by seed companies; most of the plots are on cooperating farms, and some are on ISU research farms. The plots are planted and grown using the same fertilizer program and management a farmer would use.
• ISU/ICIA crop performance trials evaluate varieties entered by seed companies.
• These tests provide helpful, objective third-party comparisons of varieties.
• Look at yield data, but also consider maturity, traits and other factors.
“During the growing season, we walk our fields, taking notes,” says Rouse. “Just as farmers in some areas had windstorms that struck their fields, we also had a lot of lodging in some of our plots this year. Whether it was root lodging or stalk lodging, we saw more downed corn than usual and more green snap, too. The lodging data we gathered was pretty high in places, and that always causes a debate on whether we should harvest such plots.”
Generally, Rouse and his staff go ahead and harvest the plot, and then evaluate the results statistically to determine if they can make a meaningful yield comparison among the hybrids or varieties. In a couple such locations this year, they were able to go ahead and use the yield data.
There are also a few sites where they’re not publishing data. “You can’t always tell just by looking at the crop,” says Rouse. “The reason we’re not using data from some sites is because the variability in yield is so high within entries in a test, due to the storm damage. We don’t have confidence the yield number is an accurate representation of the genetic performance ability of the entries in those tests.”
Comparing yield results
Key to keep in mind when viewing yield trial results is that the actual yield numbers don’t matter. Whether the average for a location is 210 or 150 bushels, don’t be misled, says Rouse.
What matters most is the relative performance — how the hybrids compare to each other in the list of results. That’s because environment has a big impact on yield. Some fields got nailed by weather, some didn’t.
“As a farmer, you should make your selection decision from the corn hybrids and soybean varieties that come to the top in the list of results,” he says. “That’s true whether the hybrid or varieties come from a list that starts with a high yield of 200 bushels per acre, or only 150.”
Also, keep in mind district results are more important than a single location, because the district average for the hybrids and varieties entered combines the results over several different environments. For example, in southwest Iowa this year, the average corn yield in two tests was only 140 and 135 bushels per acre.
But statistically each entry performed pretty close to itself across the replications. The data is sound statistically, so it’s usable. “That wasn’t a high-yield environment in 2011, with the weather problems they had. But it is a fair comparison,” says Rouse.
This year Rouse ended up having to discard three corn locations and two soybean locations in Iowa. “That’s disappointing, considering the time and effort we put into those plots,” he says. “But you can’t publish numbers if you’re not confident they are an accurate representation of the potential yield of the hybrids or varieties.”
Along with lower yields at some locations, this year’s tests in Iowa had some pleasant surprises, too. Yield results at Ames, averaging 191 to 201 bushels of corn per acre over four tests, helped boost the central Iowa district average. Also better than expected, the southeast Iowa test had corn averaging 170 bushels per acre and soybeans ranging from 50 to 55 bushels. That’s surprising, as southeast Iowa had areas of drought, and some farmers suffered very disappointing yields.
In northern Iowa, the 2011 test results show early-season soybeans outyielded full-season-maturity beans. Usually, the yields are close to the same between early- and full-season beans, or maybe the full-season beans have a slight edge. “But this year northern Iowa got dinged by the early freeze,” notes Rouse, “and that shows up in the yield average.
The full-season, later-maturing beans there in 2011 produced consistently lower yields than earlier-maturity beans. A 4- to 5-bushel difference between our tests at a location is unusual.”
Choosing hybrids to plant
When deciding on which corn hybrids and soybean varieties to plant next year, pay close attention to the district yield averages, not single-location results. And two-year averages are more meaningful than single-year results.
“Hybrids and varieties that rise to the top in multiple environments are those that will likely do well next year, regardless of weather,” says Rouse. “People naturally want to look at test results closest to their farm. What the average yield was, which hybrid won, etc. Was it the same hybrid you planted? That’s fine to see if your hybrid was near the top or in the middle of the pack. But remember, pay attention to the district averages, rather than put emphasis on a single location’s results.”
Goss’s wilt, a corn leaf disease, hit many Iowa fields this year. The ISU/ICIA trials don’t rate hybrids or varieties for resistance to diseases. Farmers should ask the seed dealer for that information, says Rouse. The companies are evaluating their hybrids for disease resistance.
This article published in the November, 2011 edition of WALLACES FARMER.