We’ve use a lot of ink these days talking about precision agriculture. At the recent Agri-Trend conference in Saskatoon, I heard several presentations on the topic and was heartened by what I heard. One presentation pointed out that precision agriculture covers a very broad front and can involve drainage (a swear word in many circles), spot herbicide or pesticide application, and many other aspects of crop production.
A presentation on variable rate fungicide application, based on a real time air photo, was impressive. The applicator put fungicide on the areas of rank growth where the disease would be worst. To me, the big bang from this practice was not only the money saved on product, but also the much reduced amount of chemical that enters the environment. I think that is a winner. There is no real way to objectively compare the outcome with a full field rate, but if chemical use is reduced and there is no disease, at least it looks like a winner.
In most cases when we talk about precision farming, we think variable rate fertilization within one field unit. That topic takes me back 43 years. My first job in soil fertility was related to variable fertilizer requirements within a field unit. In the spring of 1965, my job as a soil surveyor was to set up “strip test” fertilizer experiments. The objective was nitrogen soil test correlation, and at the same time we hoped to identify which soils in the field responded the most to the nitrogen. The experiments worked for the first objective but failed in the second. At that time there was no practical way to vary rates within a field anyway so the idea was dropped.
In the 1970s our soil survey people co-operated with Ag Canada in Swift Current to conduct simple replicated phosphorus rate experiments on various soils within a field. We found that eroded knolls responded well to added P fertilizer. Many longer-term experiments also showed that single large applications of P on those bald knolls would kick out more yield for years to come.
But the big variable in fertilizer is always nitrogen. That is where the serious money is spent and that is what makes the trucker run at harvest time. And we have known for years about the huge interaction of N fertilizer response with moisture — available water at seeding plus growing-season rain. The more water we have, the more N we use and the more bins we need.
Some versions of variable rate N talk about putting more N in the more productive parts of the field and less N in the less productive parts. This makes sense if the difference in productivity is due to moisture, but it does not make sense if the difference is due to topsoil depth and variation in natural productivity.
If we have a quarter section that has a mix of sandy soils that only hold three inches of water when at “field capacity” and heavy clay soils that hold eight inches of water at “field capacity,” it is simple. Apply lots more N to the heavy clay and lots less N to the sand.
Elsewhere in this issue is an article based on a talk an Aussie (Roger Lawes) gave at the Agri-Trend conference. Australians base N rates on soil moisture differences and that will work. Lawes also talked about EM38 maps. I am a huge believer in EM38, so much so that I invested good money to own one. But we will leave that for a future article.
Back to the moisture question. The sand and clay example I gave only works if both soils are full (at field capacity). In our situation that is not often the case. Just because a soil holds eight inches of available water does not mean that the water is there. A five-gallon pail of water with three gallons in it still gives us three gallons! The Stubble Soil Moisture map tells us if the pail (soil) is full or not.
In the 1980s colleagues of mine at U of S laid out experiments to determine if we should put more N on the hilltops, which are known to be deficient in N but lack moisture, or put more N in the lower slope positions which have more moisture and more yield potential. But those same lower slope positions also have much more organic matter and much more capability to provide N throughout the growing season. So that project came to no firm conclusion either.
What we have now is technology in spades to be able to mix and match fertilizer types and rates across a quarter section as we please. And we can make pretty yield maps from the combine seat And we can measure yield on the go from the combine seat. But we still lack the proper experiment to tell whether the variable fertilizer rate prescription is providing more profit at the end of the year or not. Many articles talk about a profit of tens of dollars per acre for variable over single rate, but they provide no measured facts to back up the conclusion. Comparisons with an adjacent quarter section or from a previous year are not suitable.
The real truth of the matter is that there is no single way that will work for a wide variety of circumstances. What we have now are many young and technology-savvy farmers and agronomists who recognize that they have to develop what is right for their particular situation.
continued on next page