When it comes to on-farm research, the key things to remember are preparation, location, and repetition.
“The rigour you have to use goes up considerably if you want to turn it into research,” said Dan Heaney, research and development vice-president with Farmers Edge.
Long, narrow configurations work better than block configurations for check strips, Heaney said at canola Science-O-Rama earlier this month.
Also, good land yields the best results. Heaney said he has seen lots of trials shoved into the corners of fields or in the headlands, which create problems in the trial. Satellite imagery, yield maps from previous years, soil maps, and soil tests can help growers and agronomists pick the ideal location for a trial.
“Pick your spot and pre-assess the variability of where you will put your treatment,” said Heaney.
It’s easy to build the treatment right into the controller of the equipment, and growers and agronomists no longer have to flag a check strip.
“On all our variable-rate fields, we build our check into our treatment map, and it comes off automatically at cost and rate,” he said. “We can do the same thing with sprayers and just about every other piece of equipment.”
It’s important to make sure equipment is properly calibrated, and satellite imagery can be used to tell growers and agronomists the effects of post-treatments, he added.
And to get accurate results, you need to do a trial more than once.
“If you want to do research, you have to get more replication in,” said Heaney. “You can replicate by repeating a check in the field or a treatment in the field. That’s the traditional way.”
But another way to get more data is to replicate the same treatment on different fields or on different farms. Agronomists at Farmers Edge have been repeating the same experiments on fields from different clients to get more results.
“We can use the power of repetition to pull out what’s working and what’s not,” said Heaney.
He’s seen growers coming together to pool their data into groups. Grower data goes into an anonymous pool, and the numbers can be crunched to determine whether or not a treatment is effective.
One experiment can then be repeated on 20, 30, 40 or 100 fields across a region or across provinces, said Heaney.