Best management practices for harvest data

As with other agronomic measurements, yield data is only helpful if it actually reflects what’s going on in the field, and if the individual(s) analyzing it know what they are doing.

Peter Gredig, a grain farmer from London Ontario, technology consultant, and ag-tech enthusiast, discusses best management practices for getting the best bang-for-buck from yield data post harvest.

Calibrate, and calibrate again

Calibrating your combine’s yield monitor is a simple – and likely obvious – step. Still, it can get missed in the tumult and frenzied activity that often characterizes harvest season.

Even if yield monitors are appropriately set for the crop at hand, though, Gredig says it’s important to check whether those same settings are working in other crops. Different corn hybrids, for example, may have different test weights which throw off the previously accurate monitor.

t’s not a ‘do it once and its good for rest of harvest’ type of thing

“You have to be aware of anomalies too, like if you’re going up a hill as opposed to down a hill. The data can be skewed…Look for fliers and odd patterns. It’s not a ‘do it once and its good for rest of harvest’ type of thing.”

Don’t be afraid to ask for help

Looking for fliers and anomalies relates directly to data cleaning – that is, running through the numbers after the fact and removing anything which is clearly too high or too low. “Scrubbing” is the term Gredig usually applies. He encourages growers to seek professional help, particularly if they are strapped for time, not overly interested, or otherwise don’t have a good understanding of what they’re looking for.

“For most farmers of a certain scale, collecting is easy. It’s about what you’re going to do with it, what software you run it through, and what partner you have to help,” says Gredig.

“What software and partnership you use is almost bigger than the hardware itself. Before you collect data in a serious way, have a really good conversation with [a service partner] that’s data savvy. The farm supplier who is usually the best partner otherwise, they may not be there yet. You need more expertise.”

A tale of two camps

Getting a feel for what you’re trying to accomplish through data accumulation is important, though perhaps a more difficult to grasp part of the process. Gredig classifies growers as falling into one of two general camps.

Gredig himself falls into the first camp – those taking a directed approach. This translates to a simpler strategy of comparing different factors in the field.

I learn more directing the data

“It’s basically on-farm trials at field scale. Doing on-off fungicide treatments, for example. For me, the biggest value in yield data is when I set my field up in such a way that I’m measuring treatment A versus B, on or off application, three versus single pass with cultivator, that kind of thing. I learn more directing the data.” Gredig says.

Collecting data passively

Passively collecting data, and trying to layer data sets to create management zones over time, is the second camp. Gredig’s experience in the ag-tech space shows many growers find this approach useful and fruitful. However, it can get more complicated and time consuming – something which reinforces the value of partnering with a person or company that really digs data analysis.

In his own case, Gredig says every time production zones have been generated for his fields, they exactly match the topographical record – regardless of what practices he does or does not employ. It’s one of the reasons he opts for a more targeted strategy.

The production zone approach can also be complicated by the addition of in-field comparisons (e.g. hybrid variety differences).

“A downside where you try to get five years of data, normalize it, add layers on it, create production zones – you actually work against yourself at bit. You can create variability with on-off product treatments, for example.”

Whatever strategy a person chooses, remembering yield data is not a tell-all solution is critical. “There’s the more complex long view, with variable rate and production zones, then there’s the ‘does it make sense to apply this product?’ view. Where lots of people get hung up is assuming there will be magic coming out of the box.”