August 15, 2012

Yield projections improve slightly at some sites; some Nebraska irrigated sites near average.

The 2012 corn growing season has been unusually hot and dry. To evaluate the impact on potential production at 12 sites across the Corn Belt (Figure 1), we used the Hybrid-Maize model to estimate end-of-season yield potential based on actual weather up to August 13 and historical long-term weather data thereafter. (Data  from each of the past 30 years was used.) This approach gives a “real-time,” in-season estimate of expected yield potential (the median value shown in Table 1), and the most probable range (25th to 75th percentiles) depending on weather conditions from August 13 until the corn crop reaches maturity.

By comparing this range of possible simulated end-of-season yield potential against the long-term average (long-term Yp, fourth column from right in Table 1), it is possible to estimate the likelihood for below-average (25th percentile), average (median), or above-average (75th percentile) yields. Comparing estimated 2012 yield potential versus the long-term average gives the size of the expected yield difference. While the 25th percentile projection is most likely if weather conditions are harsher than normal from August 13 until crop maturity, the 75th percentile scenario is more likely if weather is more favorable than is typical at a given site. There is roughly a 50% probability that final yield potential will fall between the 25th and 75th percentile levels, a 75% chance that yield will be at or below the 75th percentile, and a 25% probability that it will be at or below the 25th percentile value.

Simulations were run for dryland corn in Iowa, Illinois, and South Dakota, and for both irrigated and dryland corn in Nebraska. Simulations were based on the typical planting date, hybrid relative maturity, plant population, and soil properties at each location. Underpinning data used in these simulations are provided in Table 1. Details about the Hybrid-Maize model and our simulation forecast methods can be found in a previous CropWatch article.

As the season progresses, the range of yield outcomes shrinks and the 25th and 75th percentile values converge toward the median value. Indeed, August 13 projections give a much narrower range than our projections two weeks earlier based on July 30 simulations. The good news is that projected yield potential since July 30 has stabilized or even increased slightly at 7 of 12 sites as weather has improved, especially during the most recent week. The bad news is that projections of final yield potential are below the long-term average at all but two sites.

Dryland Corn

Even in hot, dry years like 2012, parts of the Corn Belt escape untouched and catch adequate rainfall. This appears to be the case in the northern tier of the Corn Belt (e.g. Brookings, South Dakota) and near the Great Lakes (e.g. Dekalb, Illinois) (Figure 1) where projected dryland yield potential is within 2% of the long-term average. In contrast, there is moderate yield loss of 26-33% for dryland corn in south central Nebraska (Clay Center), central and northeast Iowa (Gilbert and Nashua), and west central Illinois (Monmouth). Severe yield loss of 40-65% is projected for dryland corn in eastern and northeastern Nebraska (Mead, Concord), northwest Iowa (Sutherland), and south central Illinois (Bondville).

Irrigated Corn

In contrast to large loss of yield potential in these dryland systems, drought years like 2012 highlight the value of irrigated agriculture and the stability it provides to our food system. Although hotter than average temperatures have shortened the grain filling period at all irrigated sites, which reduces yield potential somewhat, projected decreases are modest at about 5% in south central Nebraska (Clay Center, Holdrege), and 10% in east and northeast Nebraska (O’Neill, Concord, Mead). High grain prices are likely to offset the impact such losses will have on profits from irrigated corn.

Model Reliability

Given the severity of reductions in yield potential at some locations, and the apparent lack of negative impact at others, the question arises as to how reliable these projections are? In areas with relatively little heat or water stress, past experience indicates that predictions of yield potential using Hybrid-Maize are robust. In contrast, we would expect predictions of yield loss to be underestimated by Hybrid-Maize in areas where there was high temperature stress during the critical two to three day period of pollination, or where there were large water deficits that severely reduced development of the leaf canopy before tasseling. Both phenomena are not well accounted for in the current version of the model although we plan to release an improved version of Hybrid-Maize later this year that addresses these deficiencies.

Summary

The bottom line is that 2012 will be a difficult year in terms of U.S. corn production. Although irrigated yields will be somewhat lower than long-term averages, dryland corn yield potential in much of the Corn Belt will be moderately (25-33% below normal) to severely reduced (40-65% below normal). Where both prolonged drought and high temperature stress at pollination occurred, yields could be reduced by 65% or more. The final outcome will be determined by weather conditions until maturity. Fortunately, predicted weather patterns indicate a trend toward more normal temperatures and rainfall in many places.

While 2012 will certainly be a significant drought year, episodic droughts of this magnitude have occurred at regular intervals in the U.S. Corn Belt over the past 100 years of recorded weather data. Nebraska is fortunate that about 70% of total corn production comes from irrigated systems, and that improved agronomic management practices such as conservation tillage and more stress-tolerant hybrids can significantly reduce dryland corn yield losses under moderate drought. But there is little that can be done to mitigate the impact of severe, prolonged drought especially when coupled with high temperature stress at critical growth periods.

Patricio Grassini, Research Associate Professor, Department of Agronomy and Horticulture
Jenny Rees, UNL Extension Educator
Haishun Yang, Associate Professor, Department of Agronomy and Horticulture
Kenneth G Cassman, Professor, Department of Agronomy and Horticulture