2014 Forecasted Corn Yields Based on July 20 Hybrid Maize Model Simulations

2014 Forecasted Corn Yields Based on July 20 Hybrid Maize Model Simulations

Locations of Hybrid Maize Model Models
Figure 1. Hybrid-Maize Model locations.

Interpreting Yield Forecasts

When the range of possible (i.e., simulated) end-of-season yield potential for the current year is compared to the long-term average yield potential (Long-term Yp, fourth column from right in Tables 1 and 2), it is possible to estimate the likelihood for below-average (75%), average (median, 50%), or above-average (25%) yields.

Comparing estimated 2014 yield potential versus the long-term average gives the yield difference from the mean. At a given site, the 75% scenario is most likely if weather conditions are harsher than normal (for example, hot and dry weather or early killing frost) from July 20 until crop maturity, whereas the 25% scenario could occur if weather is more favorable (cool, adequate rainfall, no late season frost) than is typical for this period. Thus, there is roughly a 50% probability that final yield potential will fall between the 75% and 25% scenarios. As the season progresses, the range of yield outcomes will shrink as the estimated yield levels for the 75% and 25% scenarios converge toward the median value.

So, how reliable are these estimates? In well-managed, timely planted fields with good crop establishment that have not been damaged by hail, flooding, diseases, weeds, and insect pests, past experience indicates that estimates of yield potential using Hybrid-Maize are robust. (See How did 2013 Corn Yields Fare in Nebraska?) In fields with poor establishment, high disease or pest pressure, or those affected by hail or flooding, we expect Hybrid-Maize yield forecasts to be considerably higher than actual yields from such fields. Likewise, the model will tend to overestimate yields in crops that suffered large kernel abortion due to severe heat and water stress during pollination.

July 24, 2015

Corn is pollinating throughout Nebraska or will be shortly. Questions are being asked about how this year's weather to date has impacted expected yield and the most likely weather scenario for the rest of the 2014 growing season. Since 2011, we have run in-season corn yield predictions using the Hybrid-Maize Model developed by researchers in the Department of Agronomy and Horticulture at UNL.

The Hybrid-Maize model simulates daily corn growth and development and final grain yield under irrigated and dryland conditions. This model estimates "yield potential," which is the yield obtained when the crop is not limited by nutrient supply, diseases, insect pressure, or weed competition — conditions that represent an "optimal management" scenario. It also assumes a uniform plant stand at the specified plant population and no problems from flooding or hail. Because weather and management factors are "location-specific," Hybrid-Maize simulations are based on actual weather data and typical management practices at the location being simulated.

The Hybrid-Maize model can be used during the current crop season to forecast end-of-season yield potential under irrigated and dryland conditions. To do so, Hybrid-Maize uses observed weather data until the date of the yield forecast and historical weather data to predict the rest of the season, which gives a range of possible end-of-season yields. This simulated range by Hybrid-Maize narrows as corn approaches maturity. Hybrid-Maize helps us understand

  • how current in-season weather conditions affected corn growth up to the date of the simulation and
  • the most likely scenarios for completing the growing season.

Hybrid-Maize then compares these projections with actual data from previous years.

Simulations of 2014 end-of-season corn yield potential at 25 locations across the Corn Belt were performed on July 20 (Figure 1). Separate yield forecasts were performed for irrigated and dryland corn for those sites in Nebraska and Kansas where both irrigated and dryland production are important. Input data used for the simulations include weather data provided by the High Plains Regional Climate Center (HPRCC), the National Weather Service (NWS) station network, the Illinois Water and Atmospheric Resources Monitoring Program (WARM), the Ohio State University, Ohio Agricultural Research and Development Center (OARDC) Weather Service, and the University of Wisconsin Extension Ag Weather.

Site-specific information on soil properties and typical crop management (hybrid maturity, plant populations, and historical and 2014 planting dates) was provided by local extension educators and agronomists in each state: Jennifer Rees, Keith Glewen, Charles Shapiro, Greg Kruger  (all at UNL);  Central Valley Ag agronomists; Mark Licht (Iowa State University), Ignacio Ciampitti (Kansas State University), Peter Thomison (Ohio State University), and Joe Lauer (University of Wisconsin). (See contributors).

We would have liked to have run yield forecasts for locations in the main corn growing areas of Minnesota, but daily weather data of acceptable quality was not available. We will try to identify weather data for running the corn yield projections next year for Minnesota, as well as for other states with significant corn production such as Indiana, Missouri, and South Dakota.

Corn Yield Potential (Yp) forecasts, as well as the underpinning data used for the simulations, are shown in Tables 1 and 2, linked here. Estimates of the long-term yield potential are based on 25+ years of weather data (fourth column from the right) at each location and a simulated Yp for each of those historical years, which allows calculating the probability of yield outcomes. This range of historical yield outcomes can then be compared to the range of forecasted yield outcomes for the current 2014 growing season (three columns on the right). These include the yield potential simulated under the most likely scenario of weather expected for the rest of the season (median) and for relatively favorable and unfavorable scenarios for the rest of the season (25% and 75% scenarios). These scenarios are based on historical weather data from the date of the simulation to end of the growing season.

In general, when comparing the median estimated yield for 2014 to the long-term average yield potential based on 25+ years of weather data, 2014 irrigated yields are near-average in Nebraska and Wisconsin and trending higher than the long-term yields in Kansas (Table 1). In fact, there is already a high probability of above-average irrigated yield in Kansas due to below-average temperature so far this season. In the case of irrigated corn in Nebraska, forecasts indicate a median yield potential 5-6 bushels per acre below the long-term average at North Platte and Holdrege. However, this is not consistent throughout the state; estimates of irrigated corn yield potential are near-average at Mead or even above the long-term average at Clay Center and in northern Nebraska (Concord and O'Neill).

In the case of dryland corn, above-normal rainfall, coupled with low rates of daily water use due to low daytime temperature, are the leading factors contributing to the above-average yield potential forecasts across the entire Corn Belt (Tables 1 and 2) . Remarkably, there is already a high chance of above-average dryland yields in eastern Nebraska (Mead and Concord), Kansas, and almost all simulated locations in the central and eastern Corn Belt (Iowa, Illinois, and Ohio). In Wisconsin, where the growing season is shorter than in the other simulated locations, the median dryland yield was near the long-term average because the positive effect of above-normal rainfall was offset by the negative impact of late sowing in 2014 (about 10 days later than an average year) and the higher probability of an early killing frost during grain filling. In the case of dryland corn in Nebraska, the model is estimating yields of +14 (Clay Center), +45 (Mead), and +36 (Concord) bushels per acre above the long-term average dryland yield potential for each location.

Factors in 2014 that may cause lower yields than these forecasts, even with optimal management, include hail or flood damage as well as a greater likehood of foliar diseases. Likewise, given the large amount of rain in some areas, nitrogen leaching and/or denitrification may limit yields due to nitrogen deficiency if additional nitrogen was not applied to affected areas.

Conclusions

Yield forecasts from 25 locations across the Corn Belt indicate above-average yield potential for well-managed dryland corn in fields unaffected by hail or flooding. However, there is still a chance (though small) of having an average or below-average dryland corn yield potential at several locations if weather conditions are unfavorable (e.g., dry) during the rest of the season. (See 75% yield potential in Tables 1 and 2).) Irrigated corn yield potential looks to be about average. However, if adequate rainfall and temperatures continue through the end of July and into August, we would expect both dryland and irrigated yield forecasts to go up.

We will follow up with further Hybrid-Maize estimates later in the season. Cooler than normal weather, however, could increase the probability of an early killing frost at locations that were planted late, resulting in yields lower than currently forecast.

Patricio Grassini, UNL Assistant Professor of Agronomy and Horticulture, Extension Cropping System Specialist and Robert B. Daugherty Water for Food Institute Fellow
Haishun Yang, UNL Associate Professor of Agronomy and Horticulture and Robert B. Daugherty Water for Food Institute Fellow
Roger Elmore, UNL Professor of Agronomy and Horticulture, Extension Cropping System Specialist and Robert B. Daugherty Water for Food Institute Fellow
Ken Cassman, UNL Professor of Agronomy and Horticulture and Robert B. Daugherty Water for Food Institute Fellow
Jenny Rees, UNL Extension Educator
Charles Shapiro, UNL Extension Soils Specialist  and Professor of Agronomy and Horticulture
Keith Glewen, UNL Extension Educator
Greg Kruger, UNL Assistant Professor of Agronomy and Horticulture and UNL Extension Cropping System Specialist
Mark Licht, Extension Cropping System Agronomist, Iowa State University
Ignacio Ciampitti, Crop Production and Cropping System Specialist and Assistant Professor of Agronomy, Kansas State University
Peter Thomison, Extension Specialist and Professor, Ohio State University
Joe Lauer, Professor, University of Wisconsin-Madison