By Laura Lindsey, Peter Thomison
Ohio's corn and soybean crops experienced exceptional growing conditions in 2019, including record rainfall in May and June followed by drier than normal August and September conditions in many areas. As a result of the early season saturated soils, corn and soybean planting was delayed across most of the state. For soybean, planting date is the most important cultural practice that influences grain yield. Planting date is also a major factor affecting crop performance and profitability in corn. The persistent rains and saturated soils caused localized ponding and flooding. These conditions resulted in root damage and N loss that led to uneven crop growth and development between and within fields. Agronomists often question the value of test plot data when adverse growing conditions severely limit yield potential.
With corn, is data from test plots planted in June of questionable value since corn is typically planted by mid-May for optimal crop performance? According to USDA-NASS estimates, 50% of Ohio’s corn acreage was planted after June 9, 2019. When selecting corn hybrids to plant in 2020, using May planting dates is preferable especially when comparing hybrids of similar relative maturity (and GDD requirements). Nevertheless, if hybrids have performed well in June as well as in May, they demonstrate resiliency that should be considered in hybrid selection. Major planting delays and replanting due to erratic weather conditions (excessive spring rainfall) occur about every three to four years in Ohio (https://agcrops.osu.edu/newsletter/corn-newsletter/2019-12/delayed-planting-effects-corn-yield-%E2%80%9Chistorical%E2%80%9D-perspective), so hybrids that perform well when planted on both normal and late planting dates should not be overlooked.
The validity of test plot results depends primarily on whether effects of the varied stress conditions are uniform across test plots. If not, test plot data may be questionable. To be certain that effects of stress were fairly uniform, it would be necessary to monitor test plots on a regular basis to determine crop response to the various stresses as they occurred; however, such monitoring was probably unlikely in many test plot fields.
Another problem with test plot results is that the various yield limiting factors may accentuate the natural "variability" already existing in the field, and may thereby further "mask" the true treatment effects that are being compared. Stress conditions like the ponding and saturated soils this year coupled with slight differences in soil organic matter, drainage, weed control, etc. across a field may magnify differences in crop performance. If test plot results include a coefficient of variance (CV) value, the CV can be used to help understand the variability among test plots. CV is an indicator of data uniformity. Larger CVs indicate that the data were less uniform possibly due to environmental variability. Lower CVs indicate that the data were more uniform.
If one assumes that the varied stress conditions affected test plots uniformly within a field, then interpretation of test plot data becomes an issue. This issue can be especially relevant when evaluating results of hybrid and cultivar performance trials affected by excessive soil moisture. Did a hybrid or cultivar yield well under saturated soils because it genuinely possessed some flooding tolerance or because it was planted in better drained areas of the field? This year we had more than 30 bu/A differences in plot yield between hybrid entries planted at different locations within a field that are related to soil drainage and N loss. Usually there are striking visual differences between such plots associated with plant height and overall plant health but differences are not always pronounced.
Test plot information this year can still be very useful but take precautions. Results from single on-farm strip tests should not be used to make a decision on adoption of a treatment or variety. Even replicated data from a single test site should be avoided, especially if the site was characterized by abnormal growing conditions. Use test plot data from multiple sites (and preferably from at least 2 years of testing) and inquire about the weather patterns and conditions associated with the results. Look for consistency in a product or cultivar's performance across a range of environmental conditions.
Source : osu.edu