The Coffee Talk Journal · Soil
What Your Soil Will Actually Hand Over
Counting nitrate tells you what is in the soil. Edgar Hammermeister of Western Ag on why supply is the better question, how roots change the answer, and what to check before you pay for variable rate.
Swap spring wheat for oats and the soil hands over 60 pounds of nitrogen instead of 24, and 29 pounds of phosphate instead of 12. That is what the model showed when Edgar Hammermeister ran it on Coffee Talk on September 24, 2026. The soil stayed the same. The roots changed.
Edgar has been with Western Ag since 1998. The company was Western Ag Labs until 2013, when it became Western Ag Professional Agronomy, a consulting company with its own way of analyzing soil. He owns the franchise for southeast Saskatchewan and Manitoba, and that franchise owns the lab that processes all of Western Ag's samples. So take this as one company's method, explained by a man who runs it. The idea under it is worth having either way.
Contents versus supply
Edgar's complaint about conventional soil tests is that they focus only on nitrate and leave out nitrogen mineralization. He cited older research showing that an organic matter measurement alone doesn't predict how much nitrogen a soil will mineralize.
The traditional lab method for mineralization, Stanford and Smith, takes 24 weeks. Western Ag's answer is the plant root simulator, or PRS, probe: a resin membrane developed at the University of Saskatchewan. In the lab the soil is moistened and warmed, and the probes sit in it for 24 hours. What they pick up is extracted and fed into a crop simulation model called Crop Caster.
Why the roots change the answer
Crop Caster starts with the soil itself: a water storage index based on clay content, soil density to account for compaction and hard pans, pH and salinity. Then it adds the season: stored moisture, rainfall and irrigation against long-term Environment Canada averages, plus temperature and growing degree days. What comes out are fuel gauges for nitrogen, phosphate, potassium and sulfur, and those gauges move with the rain and with the crop.
- Oats
- Spring wheat
Edgar Hammermeister, Coffee Talk, September 24, 2026
See the numbers
| Oats | Spring wheat | |
|---|---|---|
| Nitrogen | 60 lb | 24 lb |
| Phosphate | 29 lb | 12 lb |
The model holds rooting characteristics for 27 crops, and oats get there on an extensive, aggressive root system. The model's VAM button works on the same logic: it simulates mycorrhizal fungi expanding root surface area, which helps with phosphate, potassium and micronutrients, particularly for organic farmers or when phosphate is expensive.
When Dean York asked whether biological products and soil primers change soil supply power, Edgar reached for the oats again. If a product really does enhance root development, the crop explores more soil and can take up more. Whether probe readings pick that up, he planned to ask Western Ag's research side.
None of this means oats everywhere. Asked about disease, Edgar stuck with a four-year rotation: canola one year in four, and the cereal years alternating among different crops.
Run the decision before you spend
Once the soil is in the model, the questions turn to money: maximum yield against grain bin yield, protein targets, local blend costs. In his September 2026 demonstration, $9 wheat showed a net return of $192, tailored malt barley at $6 showed $227, and $4.50 oats showed an advantage as well. Those were one scenario on one day. The habit is what lasts: compare net returns across crops, and run canola at 6, 7 and 8 inches of rain to see the risk and reward before your capital is on the line.
Checking the model against the combine
A model you can't check is only an opinion, so Western Ag backcasts: it runs the model in reverse against what really happened. Edgar showed a southeast Saskatchewan wheat field. With actual rainfall, 11 inches of useful rain, and the heat stress added in, the model predicted 72 bushels, which lined up with what the farm saw.
Edgar Hammermeister, Coffee Talk, September 24, 2026
See the numbers
| Bushels | |
|---|---|
| Model, after the season | 72 bu |
| Expectation | 65 bu or better |
| Yield map, field average | 56 bu |
The yield map averaged 56 bushels against that expectation of 65 or better. Edgar put the gap down to how the combine header ran through the turns and a 5-inch rain that drowned out part of the crop. That is the honest edge of any soil test. It was close where the soil was sampled, and it had nothing to say about a storm or a header.
The backcast is only as good as the rain data, from whatever manual or digital gauge the farmer has. Gauges can under-measure, Edgar allowed, but they give a workable baseline.
What the test and model can and can't tell you
It can help with
- What this soil can supply to a given crop
- How that supply shifts with rain and heat
- Which crop and blend pencil out at today's prices
It can't see
- How much rain actually comes
- A 5-inch storm that drowns out a crop
- What happens at the combine
Less phosphate than the old rules said
A farmer on the call asked the obvious question for high fertilizer and low grain prices: how far can you cut phosphorus and potassium before it costs you yield? Edgar cited recent Prairie research from the University of Saskatchewan and Manitoba Agriculture showing crops are more efficient than the old International Plant Nutrition Institute rules of thumb suggested, at roughly two-thirds of a pound of phosphate per bushel of canola. The old rules overestimated, and nutrients built up in the soil.
Edgar said a yield penalty typically takes a couple of years to show, and that it depends heavily on whether springs are warm and moist or cold and dry. He also said soil phosphate below 15 parts per million needs added nutrient.
The variable rate question
This was where Edgar was bluntest. He is concerned about political promotion and government programs for variable rate, because the research doesn't uniformly show an economic benefit for the farmer footing the bill. He wants objective, field-by-field assessments instead of broad mandates. A farmer on the call added that nearly all seeding tools sold in Western Canada over the past eight years can do variable rate, yet utilization stays low.
Part of the reason is in the slope. Edgar said soil assessments often show less overall fertility variability than people expect. A low spot can supply an extra 30 pounds of nitrogen through mineralization compared with the mid-slope, and that flattens yields out.
If you do go to zones, Western Ag samples each one individually, usually 3 to 5 in a field, as if it were a separate field and not a composite. Dean asked about SWAT maps, which combine electromagnetic measurements with topography and water modelling. Edgar said biomass and yield maps need hard scrutiny over multiple years before they're used for variable rate.
Then there's the iron. A seeder 60 to 80 feet wide can cover several zones at once. Edgar's practical middle ground is to treat the real problem spots, like sand lenses short on potassium and sulfur, and run a simpler average blend across the rest.
Five questions for your next soil test
- 1Is it counting nitrate, or looking at supply?
Ask how mineralization is handled.
- 2Which crop was it run for?
Same soil, different roots, different number.
- 3How much rain was assumed?
Ask to see more than one rainfall scenario.
- 4Has it been checked against my yields?
Backcast with your own rain gauge records.
- 5Were the zones sampled separately?
And does the plan pay on this field, with your bill?
Worth keeping
- Contents and supply are different numbers.
- Change the crop or the rain, and the supply changes.
- A protein target is an economic decision.
- Variable rate has to pay on your field, not on paper.