Without more data, precision ag is all promise

Precision agriculture. It sounds high tech and scientific. Unfortunately, our ability to apply exact amounts of inputs to a specific chunk of soil greatly exceeds our ability to know what the correct amounts actually are.


Knowing the thousand kernel weight of your seed along with its germination and vigour is useful information as you set your seeding rate, but exactly how many plants per square foot or square metre should you be targeting?


Most of the advice provides a wide range for optimum plant count. Depending on the weather, the soil moisture, the slope, the soil texture and a plethora of other possible factors, 13 plants per square foot may or may not be better for a particular crop than just 10. That’s why a lot of producers just seed X pounds per acre because that’s what has worked in the past. 


The big variable, especially on small seeded crops, is survival rate. Will 55 percent of the seeds actually emerge or will it be 70 per cent? It makes a big difference.


The tendency in precision farming is to seed fewer pounds per acre. Rather than five pounds of canola, producers are cutting back to four or even three because their seed placement and therefore survival percentage is assumed to be higher.


But with hybrid varieties, canola seed has tended to become larger. There are fewer seeds per pound – one of the reasons why the thousand kernel weight is useful information. Unfortunately, there’s debate over whether a larger seed also means a more vigorous seedling and a higher survivability rate. 


Without that knowledge, the whole exercise isn’t really very precise. 


It has long been the gospel to recommend soil tests on every field. In fact, for variable rate applications, there’s often extensive soil sampling so the fertilizer prescription can be varied across the landscape. 


This is heresy for an agrologist to say, but soil tests are not the precise tools that some people make them out to be. 


It depends upon whether representative samples are gathered in the first place, the nutrient assays that are used and the yield response that is assumed. 


Some professional services will have you using a bunch of micronutrients. Others will advise that only the macros will give you a bang for your buck. And no one with any precision can predict the weather, which is a vital variable. 


GPS guidance has been a miracle for preventing overlap and putting what we want exactly where we want it. But there isn’t enough basic research to update yield response curves that in many cases were developed with older crop varieties in much different farming systems than what we practice today.


Do we really need full herbicide rates in a year with excellent growing conditions? Probably not, but most of us will apply full rates just in case. 


The lentil field I had just sprayed with Solo was hit with a heavy shower shortly after I finished a tank full. The herbicide book says, “Do not spray if there is forecast of rain during or soon after application as it may reduce control.” 


Other than the random shower hitting the precise field I had just sprayed, there appears to be little precision in what can be expected for reduced weed control.


Fungicides are worse. Since they’re largely preventive, you seldom know for sure how much they truly accomplished. 


Overall, precision agriculture may be a worthy goal, but we’re not there yet.

Kevin Hursh is an agricultural journalist, consultant and farmer. He can be reached by e-mail at kevin@hursh.ca.