Adam Spelhaug, agronomist, Peterson Farms Seed, Harwood, N.D. has some ideas on how to pick the best seed for 2013.
"Make your first cut in seed selection not on yield, but on the on agronomic characteristics you need for each of your unique fields," he says.
For instance, if you have soybean fields with iron chlorosis or soybean cyst, or corn fields with high salts or gravel spots, first eliminate the products that will not meet those agronomic challenges. If you have fields with significant glyphosate resistant weeds, you will want to consider alternative herbicide systems on those acres.
"On the flip side, many fields are just plain more productive than others. These are the fields for those hot new numbers," he says.
Once the agronomic needs are considered, you clearly want the highest yielding varieties and hybrids you can find, right? But how do you determine which those are?
"When you look at plot data, the first thing to keep in mind is the purpose of using that data. Do you really care which product yielded the best this year? I would suggest that the answer is no. What you are trying to determine is which products will yield the best next year. 2012 was hotter and drier than normal. 2011 was wetter and warmer than normal. 2009 was very wet and very cold in this region. What will next year be like? Since we have no crystal ball to know next year's weather now, it is important to plant most of your acres to products that have performed well in many different environments.
He suggests keeping the best performing half of the products you planted on your farm this year. Plant those on about half your acres. For the other half, he recommend gathering as much data as possible and trying something else.
Increased use of GPS and field mapping is making on-farm trials more accurate. Conducting large scale replications across a field with different hybrids/varieties is a great way to evaluate performance, while minimizing the variance of the field itself.
Replicated small plot research, similar to what is done by universities and credible seed companies, is a great source for data. These plots minimize the variance in the area the crop is raised and show the true genetic difference among the entries.
"Single side-by-side trials never give you adequate data and should not be used in decision-making. This is true for evaluating small plots as well. Use only replicated data with multiple years and locations. Side-by-side show plots are really only useful for evaluating physical differences, since yield can be affected by too many variables (soil, location, mechanical) in a small area," he says.
The more entries there are in the plot, the less valuable the results are as well.
"Your chances of finding the best hybrid or variety in a plot with two entries is 52%, in a plot of four it is 33%, and in a plot with 10 or more entries it is lower than 15%," he says.
Source: Peterson Farms Seed