Investment in corn research has lead to yield improvements once thought unattainable.

Blog by Scott Johnson, Spectrum Corn Product Manager

Every five years or so it is important to examine corn yield trends. Historically, there have been only two instrumental developments leading to leaps in yield levels. At about the same time, nitrogen fertilizers and hybrid corn converged, creating the increases in corn production shown on the graph below in the late 1930’s. And since, there has been no similar revolutionary event to explain such changes, because they have not occurred. Even in the era of biotechnology for the past 20 plus years.

So, let’s break down the graph plotting yield changes over time, according to national corn yield averages summarized each year by the USDA…

On the left hand side of the graph, left of the dotted line, is the open-pollinated era of corn production where farmers were saving seed and replanting it each year. Interestingly, for the 70 years where there was record keeping, there was no measurable improvement. As a matter of fact, yields were slowly shrinking over this period of time. Then, around 1937, double cross hybrid corn was introduced to the farm; the culmination of nearly four decades of breeding research at the Land Grant universities. As the inbred-hybrid concept and double crosses were born, nearly all US corn acres were being planted to these hybrids within five years. In the mid-1960’s double cross hybrids were replaced by single crosses, and around the year 2000, single crosses were being grown in conjunction with biotechnology.

There are two key attributes of the (linear regression) equations explaining the trend lines and how well these lines fit the data. First, the slope of the trend lines represent the yield gain in bushels per acre per year. For example, yield gains in the double cross era averaged 1.6 bushels per acre per year as shown on graph. Studies have shown two-thirds of these gains to be attributed to genetic improvements, while the remaining third to improved management. Second, the R-squared values shown represent how well the data fit the linear regression model. In other words, if R-squared values are 1.00, then every data point is directly on the trendline, suggesting maximum predictability; the smaller the number, the more actual data moves above and below the trendline, and thus resulting in less predictable outcomes year-to-year.

If you want a bird’s eye view view of corn production in the United States for the past 150 years, PLEASE STUDY THIS GRAPH. It tells a remarkable story of where we’ve been, and likely where we’re going. In the scientific period of plant breeding, or the past 80 years, we can see how investment in corn research has lead to yield improvements once thought unattainable.

In the Double Cross era, from about 1937 to 1967, yield steadily increased (1.5688 bu per acre per year) with not a lot of “excitement” from year to year. The R-squared value was 0.88. Relatively uneventful. As the name implies, double crosses were the result of crossing two hybrids together to produce seed for the farmer. Because there were four parents as components of the hybrid, as opposed to just two in single crosses, there was inherently more genetic diversity and ability to withstand more environmental extremes. Had single crosses not been introduced in the late 1960’s, we would currently be growing 147.5 bushels per acre per year, instead of 165.0. The last 50 years of breeding, biotechnology, and management advances have given us 17.5 bushels over what could have occurred had double crosses continued to dominate. One might argue it may have resulted in less anxiety year to year when there was weather extremes, compared to single crosses.

Which brings us to the single cross era and the era of biotechnology.

The Single Cross era, from about 1967 to current, brought with it increasing yields at a clip of 1.7 bushels per acre per year…and more yield volatility. Yield potential was increasing over the double crosses, however, there were much larger swings from year to year. As evidenced in the graph and the diminishing R-squared value. Had biotechnology and improved management practices not been introduced in a significant fashion around the year 2000, we would be growing an average of 161.8 bushels per year, instead of 165.0. The graph shows trendlines projecting forward for the double and single cross eras. It is clear the double cross trendline slowly ventures away from the single and biotech era lines, however, the single cross trendline runs virtually parallel to the biotech line. Again, one might argue the 3.2 bushel increase during the biotech era may also be influenced by seed treatments and precision agriculture in the past 17 years. It is important to note, ALL USDA data were included in the graph, including 2012 drought year. If 2012 were omitted, do we also omit 1983, 1988, and 1993?

“Some question whether the yield trend line has shifted again in recent years due to the advent of transgenic hybrid technology [biotechnology] in the mid-1990’s, but the data show little evidence that a third significant shift in corn productivity has occurred,” R.L. Nielsen, August 2012.