REPRESENTATIVE YIELD: Pick random spots in the field and follow the same procedures each time to improve the odds of getting reasonably accurate, unbiased yield estimates.

Make yield estimates as accurate as possible

Corn Illustrated: Don’t expect to be 100% accurate with preharvest checks two months out.

A farmer has a common question about conducting yield estimates. Three CCAs offer their advice.

Question: We always do yield checks in corn about mid-August. What can I do to make sure yield estimates are relatively accurate? We use the information to gauge our storage needs and make final decisions on forward-contracting.

The Indiana certified crop advisers panel answering this question includes Don Burgess, agronomist with A&L Great Lakes Labs, Fort Wayne; Jesse Grogan, agronomist with AgReliant Genetics, Lafayette; and Bryan Overstreet, Purdue University Extension ag educator in Jasper County.

Burgess: Larger sample sizes will give you more accurate estimates, and increasing the number of areas sampled will generally give better yield estimates. However, since estimating yield is a rather time-consuming process, it’s better to base your sampling intensity on the uniformity of the field that you are estimating, with more intensive sampling on more variable fields. Uniformity of a field can be assessed by visual appearance, prior yield maps, aerial imagery and soil maps.

Kernel size is important in estimating yields. Conditions during grain fill can influence kernel size dramatically. Choose a kernel size factor that best reflects the expected conditions during grain fill, with the understanding that these estimates could vary significantly if the conditions are not what was anticipated.

Keeping good records of estimates and final yields from year to year will help refine the process and make yield estimates more accurate in the future.

Grogan: Sampling is the biggest challenge. How many times and where in the field do you sample potential yield? Wait until grain has reached the dent stage of development. Sample multiple planter passes in the field past the end rows. Count off 1/1000 of an acre for a sampling set, and accurately count plants per acre. Follow the method of randomly selecting ears within a sampling set. Select at least five ears per set for an average. For ear length, count the kernels that contribute to yield — not the small tip kernels or the bottom row on the butt. Count kernels around for girth and average them. Kernel density and depth are difficult to estimate. Use a conservative number. If you’re within 20 bushels per acre of actual yield, you’re lucky.

Overstreet: Count ears in 1/1000 of an acre, and find the average kernels per ear by counting kernels on five ears in the sampling zone. Multiply these together and divide by a fudge factor based on kernel weight.

Here are five tips to improve accuracy. 

1. Use representative ears. Don’t use nubbins or odd ears.

2. Don’t count ears on lodged stalks. Don’t count dropped ears either.

3. Don’t count kernels on the extreme tip. They could be aborted later.

4. Be conservative with the fudge factor representing kernel weight. The fudge factor is based on the number of kernels in a bushel. The typical fudge factors used in the Midwest are 75, 85 or 95. The heavier the kernel, the lower the fudge factor.

Here is an example with 32,000 harvestable ears, 520 kernels per ear on average and a fudge factor of 85: 32 times 520 divided by 85 equals 195.7 bushels per acre. This gives you a ballpark estimate.

5. Pay attention to weather patterns and disease issues after you do your estimate. These things can influence the final yield.

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.