Acres of Answers

3 Errors to Watch For in Yield Data

| No Comments

Subscribe for Updates!

Flex leasing eBook

yield data

Yield Data Errors

This harvest, you know the yield data coming in from the combine isn’t perfect. There’s a lot of potential benefit, for sure—but the data has to be cleaned before you take it at face value.

Any sensor will record information with random error and with systematic error. Your yield monitor is made of a suite of sensors—a load cell that feels the weight of grain coming in, a moisture sensor so the weight of excess water can be subtracted from the calculation, and your speedometer so your traveling speed can be used with your header’s width to find the area you took that grain from. Of course, your GPS uses a sensor too, but decades of work have pared down GPS error to be pretty small for our agricultural needs.

3 things you should watch for when looking at your data

1. Each of these sensors has the capability of making “random” mistakes. Electrical hiccups can cause high and low outliers – but they’re easy to notice, and easy to throw away with a standard deviation filter. Some software even does this automatically for you, just be sure this is done with your data.

2. Each sensor can also record data with “systematic” errors. If you haven’t calibrated your yield monitor recently enough, it can shift your whole data set higher or lower than it should be. This is okay if you are just looking within a field (“What is the best part of the field? What is the worst part?”), but it won’t work for asking “Is this field doing better than another?”

In the current field, maybe reality is the pink data represented below, but your sensor showed the blue data. Or maybe you used a different combine on one field, so it shows you the pink data when the reality was blue.

yield data 2.png

You can only make a good comparison of fields after you have calibrated your yield data with your cart weights (or at least your field average).

3. There is another form of error that yield monitor suites are prone to, and it’s easy to clean up, but you will hurt yourself if you don’t account for it when making management zones based on past yield.

Your yield monitor figures out your yield by seeing how much weight of grain comes into the hopper, and saying “this is how big of an area we took that grain from.” When you’re at the end of a pass, your swath doesn’t change, but the weight of grain sure does—because you already harvested that!

yield data 3.png

Here is the same portion of the same field – but now the yield data has been “cleaned” of its “end pass error” before making the management zones. It’s really striking when you look at the whole field, but the farmer that this map came from asked only to show a small portion to protect his privacy.

yield data 4.png

Data cleaning can be a tedious task, but there is software to make it easier, and FamilyFarms Group offers one-on-one and group trainings for you and your trusted advisor to make sure you are only using information that will add value to your operation and help you make the best decisions.

– Allison Jonjak
GIS Technician, FamilyFarms Group

Needs assessment


Written By

Allison Jonjak

Allison Jonjak

Precision Ag Analyst |

Subscribe for Updates!

Start digging in now. Request a free needs assessment. START HERE >>