Issue 30
June 2000
 

Prairie Grains is the official publication of the Minnesota Association of Wheat Growers, North Dakota Grain Growers Association, South Dakota Wheat, Inc. and the Minnesota Barley Growers Association.

Copyright
Prairie Grains Magazine
June 2000

Five steps to estimate wheat yield potential

By Dr. Michael Peel
NDSU Extension Small Grains Specialist
mpeel@ndsuext.nodak.edu

A key question that all wheat producers would like answered be
fore making management and treatment decisions: What is this crop worth? That's a tough call for anyone to make early in the growing season, because no one knows what the weather will do or what production concerns or pests may arise.

What we can do, however, is estimate the yield potential of your wheat or durum crop. The information may be useful to aid in your decision-making process this summer, keeping in mind that this is a "rule of thumb" predictor with room for error.

Yield is a measure of how productive a crop is per unit area.  The three components that contribute to yield of wheat, and durum, are: 1) the number of heads produced in a unit area; 2) the number of kernels that are produced on a head; and 3) the weight of each of those kernels. Estimates of wheat yield potential can be made prior to harvest based on these components; final yield is determined after harvest. 

Most components of yield can be determined well before a wheat crop is ready for harvest.  The formula below—essentially the same as what is used during the Wheat Quality Council's annual preharvest spring wheat tour—uses this component information to estimate yield potential:

Bu/A = (heads/3ft of row x spikes/head x kernels/spikelet x 0.142*)¸row spacing

* Conversion factor that incorporates area, kernel weight and volume.

To use the formula

1.  Count the number of heads in three feet of row.  Do not count small heads with one or two kernels, they contribute little to yield.  If the crop was planted in banded or paired rows, count the entire band.
2.  Determine the number of spikelets per head.   This should be an average of six or more randomly selected heads.  Frequently top and bottom spikelets are aborted and rarely contribute little to overall yield; thus they should not be counted.  Other empty spikelets also should not be counted.
3.  Determine the average number of kernels per spikelet.  If the crop is nearing maturity, heads used to determine spikelet number can be hand thrashed and the total kernels divided by total spikelets for an average kernel per spikelet number.  Small errors in this number result in unrealistically high yield estimates; consequently, using a fixed number that accurately reflects long term yield trends is best.  Usually 2.3 gives the most accurate results, or 2.1 when the crop has been stressed.
4.  Finally, determine the drill row width.   When seed is planted in a band or paired row the band width plus the distance between bands is used.
5.  Account for variability within a field. Counts should be repeated not less than eight or ten times per field and the average used. Small errors in counting the components that contribute to yield can result in large errors in the yield estimate.

Example:
Next, simply plug the numbers into the formula.  For example, if you counted 70 heads in three feet of row, 12 spikelets per head, and your row spacing is seven inches, the yield estimate would be (70 x 12 x 2.3 x 0.142) ¸ 7 = 39 bu/A.

This process can be used to estimate yield of wheat and durum as early as late boot stage.  At this stage, early aborted spikelets can be determined; however, some spikelets will abort as the season progresses.

As I said before, estimating yield using this process is obviously subjective.  The correlation between an estimate and the final yield is related to how close head and spike counts are to the real numbers and crop development during the remainder of the growing season.  Further, this estimation does not take into account the potential for environmental stress, such as hail, drought, early frost, and pests to impact yield.  Therefore, the farther away from harvest yield is estimated, the greater the error in the estimation. 

The value in estimating yield might come in estimating what impact a disease like scab might have on yield.  For example, if a scab infection destroys on average one spikelet per head in our example above, that would reduce the yield from 39 to 36 bu/A.  Remember, this is only an estimation. The actual impact will be different depending on timing of infection and other factors present in the field being evaluated. Still, now you have some general guidelines to go by.  Have fun counting—I hope you're pleased with the numbers this year.