Unlocking the full strength of your data can turn small changes into bigger returns, and smart, localized insights can boost a farm’s bottom line, Brunel Sabourin of Antara Agronomy says.
Sabourin was speaking to an audience of farmers and agronomists at St. Jean Farm Days, an agricultural event at St. Jean Baptiste, Man., about 60 km south of Winnipeg.
“A lot of promises have made around big data,” Sabourin says. “Huge venture capital dollars were invested into big data, and we were going to revolutionize farming, but it’s fallen short of that expectation.”
Read Also
Good news, bad news for fungicides meant to fight stem rot in canola
A report shows overall insensitivity of sclerotinia to three fungicide groups hasn’t changed in a big way between 2010 and 2024 — but shows some sclerotinia populations have been discovered with elevated insensitivity to all three.
A major reason for this, he says, is that collecting good-quality data remains challenging. Sabourin highlights how variability in fields, and across farms, complicates data analysis. He pointed to the huge variability in terms of rainfall that can exist within a field from one end to the other — not to mention the differences from a farm 20 miles away, or in a neighbouring province.
The result, he says, is a lack of reliable comparisons due to insufficient quality of data.
Sabourin’s company addresses this by starting with small, localized data before scaling to larger datasets. By focusing on small geographies and comparing data at the field and farm levels, farmers can achieve more accurate insights before considering regional trends.
Antara also pools data from its clients to derive what they call “area insights,” covering a 20- to 25-mile radius from their home base.
“We want to be able to compare apples to apples as best we can,” he adds.
Soil testing
Soil testing is one of the simplest and most effective tools for gathering on-farm data.
“I’m a firm believer in soil sampling and doing annual testing, to establish trend lines,” Sabourin says.
For example, he says, when he looks at differences in something like phosphate, a non-mobile nutrient that wouldn’t be expected to change much year over year, he will sometimes track huge differences these numbers. Identifying those trends can reveal important insights.
“We’re growing some heavier phosphate-using crops like corn and soybeans. We’re starting to see on a lot of the fields a trending decline in capacity levels,” Sabourin says. “We’re still well above our critical levels, but it is a trend that we need to keep watching.”
By keeping a spreadsheet that tracks soil samples, farmers can use this information to improve decision-making.
Keeping a nutrient balance sheet that tracks the crops grown, their yields and the fertilizer applied from year to year is also low-hanging fruit in terms of data collection.
Sabourin references an example where one customer who had been growing soybeans for several years, without applying any phosphate, was shocked at how depleted his soils had become. After 10 years, he had mined 170 pounds of phosphate out of his field.
Because it was identified, he’s now able to take corrective action to build it back up.
Sabourin also sees untapped potential in yield monitoring, and noted many farmers have yield monitors on their combines but don’t use them. By analyzing yield variability within a field, farmers can identify areas that need attention.
Driving up yield, driving down costs
The next question farmers must ask themselves is how they can increase yield or reduce expenses.
“This is where the strategy comes into play about budgeting nutrients across the rotation with something like phosphate,” Sabourin says. “I tell my growers to look at what their removal is over five to 10 years, and then advise them, that’s the amount that they should be applying every year.”
However, because you can’t apply all the phosphate at once with canola, farmers must apply it in smaller doses, either in separate applications or in a different year. Sabourin noted many farmers opt to apply extra phosphate in advance.
“They’ll double-fertilize ahead of corn because they’re going to follow it up with canola,” he explains. “They’ll be putting down a percentage of their phosphate ahead of the corn that’s really meant for the canola crop.”
Other data sources
LIDAR elevation maps provide another affordable source of field-level data farmers can access through the Aerial Photography Library portal on the Manitoba Conservation website.
“You can go in there and look through the archives of satellite or aerial photography going back to the 1930s,” Sabourin says. “Often you can lay this map over top of the regular map, and you see why the farmer would have separated things out in different paddocks and different areas.”
Weather stations are another useful tool, but can require an investment of $1,500 to $3,000.
“They hook up to the internet, and you can have all of the data live on your phone,” Sabourin says.
But the ranges for these weather stations are limited. In many cases a farm might need more than one of the units to cover their entire farm. Those expenses can add up.
However, with Antara’s pooled network, the company ran 15 stations this year and all the participants could see weather from all the different stations, spreading that cost over multiple farms.
“We wanted to have a weather station within three miles of all the fields in our system,” he says.
Of course all of this requires a good system of record-keeping. A logbook or spreadsheet that tracks data such as seeding depth, placement, plant counts and weed or insect pressure is invaluable for understanding field performance.
“How it came out of the ground, what was the seed bed like, residue management — these are very useful to come back at harvest time to say ‘Why did this field yield more or less than that field?’”
