August 2020 Corn Forecast
Dr. Eric Hunt, Staff Scientist
Dr. Judah Cohen, Director of Seasonal Forecasting
The Bottom-Line Up Front
Atmospheric and Environmental Research is applying our world-class seasonal forecasting to give the agricultural community information to make better-informed decisions with actionable yield projections for the upcoming season. Based on the season-to-date and our current seasonal forecast model, we are currently projecting the 2020 national corn yield to be 189.6 bushels per acre (bpa) with a total production of 15.98 billion bushels (Figure 1, below). This is an increase of 2.1 billion bushels compared to 2019.
Figure 1. Distribution of corn production (top; billions of bushels) and yield (bottom; bpa) based on the weighting of our forecast. The most likely range for production is between 15.6 and 16.4 billion bushels and 186-195 bpa for yield. The worst (best) case scenarios based on our forecast are 15.2 (16.8) billion bushels for production and 181 (200) bpa for yield.
The Details
Record corn yield and production at the national level is a near lock this year. While this season is unlikely to set records against trend, there is a chance of yields finishing over 190 bpa, which would translate into corn production over 16 billion bushels. This season hasn’t been perfect (if there is such a thing) but there haven’t been many major weather-related problems across the major corn producing locations in the U.S. Here’s a short list of why we are bullish on yield:
- While the north central U.S. has been warmer than average most of the summer, there has been a noticeable lack of temperatures above 95F for the entire Corn Belt. Temperatures are not likely to be cool enough in August for corn yields to finish in the 195-200 bpa range but the current conditions (as of today-August 3rd) are very favorable for grain filling. The current outlook for the first half of fall is for above average temperatures, which will aid in ensuring maturity before the first freeze and with harvest.
- Precipitation has mostly been favorable this season for most corn producing locations. Exceptions to this over the past six weeks include much of western Iowa, parts of northeast Nebraska and parts of Indiana, Michigan, and Ohio. While it is certainly probable the drought has knocked down yield potential in these locations, significant yield losses are not expected, especially in western Iowa. Some places in western Iowa also caught some decent moisture later last week, which may be all that crop needs in some cases.
- Excessive moisture this spring was not a widespread problem. That coupled with the expectation of a mostly drought-free Corn Belt is one of the reasons we were rather bullish on corn yields even back in mid-May. As shown in a special report released earlier this year, some of the most highly productive corn producing regions in the Corn Belt actually suffer more from too much spring precipitation than from more moderate levels of drought.
- The median SMI of the Corn Belt has stayed in a “Goldilocks range” for most of the season. While that metric alone is not necessarily sufficient for projecting higher yields, the percentage of locations in the Corn Belt that have been abnormally wet or dry has been consistently low this season.
- Neither the Evaporative Stress Index (ESI) nor Quick Drought Response Index (QuickDRI) have been suggestive of extreme water stress or excessive moisture anywhere in the Corn Belt over the past few weeks.
While above trend yields seem a near-guarantee, there are a few reasons the yield projection of 189.6 bpa could be a bit too bullish:
- Temperatures from the middle of August into September are warmer than currently expected, which could lead to some reduction in yield potential.
- The ongoing flash drought in western Iowa worsens and/or spreads into adjacent regions. An onset of flash drought in mid-August is much less harmful than one that starts in June but still would have some impact.
Forecast Yield for Crop Reporting Districts:
While we fully expect an above trend year for corn nationally and for the Corn Belt overall, this is not the case for every district, as we show in Figure 2. First though, a brief description of how to interpret the map. We have broken the map into five categorical possibilities based on season-to-date conditions, the forecast, and the NASS historical yield for each district:
- Above (A, blue): We are highly confident of above trend for the district, with only isolated locations experiencing below trend yields.
- Likely Above (LA, light blue): Past precedent would suggest above trend conditions but below trend is still possible if the season’s finish is unexpectedly poor.
- Near Trend (NT, gray): We expect yields to be within a few points of trend either way.
- Likely Below (LB, orange): Past precedent suggests below trend is likely for the district, though individual fields may finish above trend
- Below (B, red): We are highly confident of below trend for the district, with only isolated locations experiencing above trend yields.
Figure 2. Categorical projections of corn trend. A- Above trend (blue); LA- Likely Above trend (light green); NT- Near Trend (gray); LB- Likely Below trend (orange); B- Below trend (red). Asterisks show the 15 districts that produced the most corn over a 20-year period from 1999-2018, as shown in this recent paper.
Of the top 15 corn producing districts (Figure 2), above trend corn is expected or likely in 14. This coupled with the likelihood of above trend, perhaps record yields, elsewhere in the Corn Belt and outside the region led us to a forecast that is around 7 percentage points above trend. Much of the Corn Belt has had a solid growing season and above-trend corn yields can be expected in most districts. Exceptions to this are across parts of Ohio and southern Michigan where moderate flash conditions have developed over the past several weeks. While yields aren’t expected to be significantly below trend, we do expect below trend in southwest Michigan and central and north central Ohio. While we are more bullish than some on yield across western and central Iowa where a flash drought has developed this summer, it is likely to bring yields down a little closer to trend. A poor finish to the season could definitely bring west central Iowa below trend. If adequate moisture is realized over the coming weeks, then don’t be surprised to see above trend corn in west central Iowa, even though part of that crop reporting district is in severe drought. Some rainfed corn in northeast Nebraska will likely take a bit of a yield hit too (likely a bit more than west central Iowa) but irrigated corn should do well, hence the projection of likely above corn trend.
The Methods
Our corn forecast is based on the following: the season to date conditions, the AER forecast for the remainder of the season, historic yield, production, and acreage data from NASS, and a regression model. Let’s take a look at each of these in a bit more detail. The season conditions to date are based on subjective interpretation of the weekly Soil Moisture Index (SMI), the Evaporative Stress Index (ESI), Quick Drought Response Index (QuickDRI), and Vegetation Drought Response Index (VegDRI) maps and the objective interpretation of the condition reports. The SMI maps are produced weekly and were developed from research done on NASA grant # NNH16CT05C, awarded to AER and the University of Wisconsin in 2016.
Retrospective analysis of the SMI in the growing season going back to 1979 shows that below trend yields are typically the case for a district if either or both of the following are valid: 1) if the more extreme dry or wet ends of the index persists for more than a month and/or 2) an unusually wet spring turns into a flash drought. For most locations, neither of the two applies this season, so above trend conditions can be expected. The districts in Ohio where we are most confident of below trend yields spent have been abnormally dry for several weeks and historic data are not suggestive that this area is as drought tolerant as western Iowa.
We then evaluate the crop condition reports from the USDA every week to see how the percentage of corn nationally and at the state level is changing and how it compares to past seasons. This year’s condition has been better than average but falls short of the very best seasons. The combination of analysis of the ESI, SMI, QuickDRI, VegDRI, the USDA NASS condition reports, reports from crop scouts/tours, and historical corn yield data are used in a statistical regression model to determine the national yield. Finally, we use the seasonal forecast to determine if the season-to-date yield forecast should be adjusted up or down based on analog years.
Our acreage estimate is based on regression using historical NASS data and our harvest acreage estimate is based upon 92.0 million planted acres of corn this spring. The combination of the expected harvest acreage and yield led us to the expected production of 15.98 billion bushels. This is an increase over our previous official forecast in May, though we are more confident in this estimate than our previous one. General details on the AER seasonal forecast and verification from 2019 can be found on our website. For more specific questions, please contact Eric Hunt at ehunt@aer.com.
About the authors:
Dr. Eric Hunt is an agricultural climatologist from Lincoln, NE and has several members of his extended family actively farming in Illinois and Nebraska. Eric has been with AER since 2012 and received his Ph.D. from the University of Nebraska. Among other activities, he is currently working on NASA funded projects to study the evolution of flash drought. He routinely blogs about agriculture and weather on the AER website.
Dr. Judah Cohen is the Director of Seasonal Forecasting at AER’s home office in Lexington, MA. Judah has been with AER since 1998 and received his Ph.D. from Columbia University. Judah is perhaps best known for his NSF funded work documenting Arctic-midlatitude connections and the Arctic Oscillation, which he routinely blogs about on AER’s website.
About AER:
Founded in 1977, Atmospheric and Environmental Research is an award-winning environmental research, consulting and weather information services company with demonstrated expertise in numerical weather prediction, climate dynamics and radiation, circulation diagnostics, atmospheric chemistry, air quality and risk assessment, planetary sciences, remote sensing, satellite meteorology, and systems engineering. Consulting services are available. AER is a business unit of Verisk Analytics (VRSK). For more information, please visit our web site at www.aer.com.
Disclaimer: This report and the information and data contained herein (the Report) are wholly advisory in nature and are provided AS IS. AER makes no representations, covenants or warranties of any kind, either express or implied, with respect to the Report, including, without limitation, warranties of condition, quality, durability, suitability, merchantability or fitness for a particular purpose, or in respect of any warranty arising by statute or otherwise in law or from a course of dealing or usage of trade. The information included in the Report may be statistical samples and/or actuarial calculations and AER makes no warranties or representations, either express or implied, that the Report will accurately reflect, predict or resemble experience for an entire industry or any member or members of any industry. AER shall have no liability and shall not be responsible for business and legal conclusions, judgments and decisions made with respect to the Report. AER does not warrant and makes no representations regarding the completeness, currency, accuracy or predictive value of the Report. AER makes no representations and assumes no responsibility for the accuracy of the Report and is not responsible for errors resulting from omitted, misstated or erroneous information or assumptions.
Atmospheric and Environmental Research, Inc.
All rights reserved.