This is the first in a series of three articles on precision agriculture.
Precision agriculture has been with us now for almost 20 years. There is a journal (Precision Agriculture) devoted to just that topic that is now in Volume 12, and there have been at least that many world conferences on the subject.
I am a firm believer that farming individual soils rather than the whole landscape is the way to go. Some of my very first research was about soil fertility differences within a single field.
Much work was done on “Farming within a field” in the 1960s and 70s but it was pie in the sky as there was no technology to make it happen. At the University of Saskatchewan there was work done on variable rate nitrogen in the 80s and early 90s. But this research reconfirmed that water and the interaction with nitrogen is the big issue, not nitrogen alone, so it was dropped again.
I have watched the development without being directly involved. I see many great things, but some not so great. Almost the only regret I have about officially retiring early is not being mixed up in the middle of the exciting business of farming soils instead of fields. Fancy technology and creative mathematics are at the forefront and basic soil science and agronomy have been left behind. Don’t get me wrong — just because I am too old to handle the fancy technology does not mean I do not respect it. There are huge opportunities.
Variable rate fertilization
At the January Saskatchewan Soil Conservation Association conference in Saskatoon I heard a talk analyzing the yield results and economics of variable rate N applications in Alberta. The study included data from a number of sites. The bottom line was positive for some and negative for others but the overall conclusion was that it is not worth the hassle. That data, and other examples, left me thinking that a new approach to variable rate is needed.
In the whole scheme of precision agriculture there are some runaway winners. Sprayers that shut off boom sections to avoid overlap win on all fronts. Fungicide applications based on an air photo a day or two ahead of spraying and a map to target the dense vegetation seems to me to be a winner.
Another runaway winner is to map saline areas with Veris or EM38 and reduce or eliminate fertilizer altogether in those areas. Saline soils are highly fertile and often very high in organic matter. Adding more fertilizer makes them worse, not better.
Research from Australia
In today’s world I can sit at the 27-inch computer screen in my home office and in a few hours scan through the entire 12 volumes of the Precision Agriculture journal and I have just now done exactly that.
In Volume 1, Number 1 (1999) a paper from Australia looked at Variable Rate N for wheat in an environment where moisture is important.
They concluded that varying the nitrogen rate had little benefit in an environment where nitrogen response is dependent on soil moisture at seeding plus growing season rain. That is precisely why we changed nitrogen recommendations in Saskatchewan in 1971 to give three rates — one each for dry, average and wet years.
In Volume 12 (2011), a paper based on uniform and variable nitrogn rates at eight locations over three years with wheat in Oklahoma showed no advantage to all the fancy technology.
I noticed that most of the articles in the journl used fancy technology and mathematics to “deal” with soil variation. Much of the fancy math assumes that soils are a strictly random event. For sure, soils can be hugely variable within one field.
But, when Mother Nature took the soil material left behind by glaciers and their meltwaters, she did not make our soils in a completely random fashion. One of the biggest soil forming factors is the climate — particularly the amount of water that has percolated through the soil over 10,000 years and the native vegetation the soil and water regime would support.
Within one field the major soil forming factor is the topography (lay of the land) which rearranges rainfall and snow melt.
In the photo of the field with rolling topography just east of Saskatoon, the land is Weyburn loam. The “Weyburn” part means it is in the dark brown soil zone. Soils in that land unit are developed from glacial till — material unceremoniously dumped directly from a glacier complete with lots of clay but also a bit of sand and silt and stones.
Within this field of Weyburn loam there are several types of soil. And these differences occur in a repeatable pattern over the field.
To deal with the soil types in this kind of land, a contour map with one meter intervals would be an imperative first step. Veris, EM38 and other options make that easy. Veris or EM38 will also provide electrical conductivity (EC) numbers that can be interpreted in various ways. EC is primarily to measure soil salinity but it can also be used for other purposes. I claim to have drug an EM38 over more acres than most and point out to folks new to the business:
“The EM38 always tells the truth — it’s up to the operator to figure out what that truth is.”
It has been a great thrill for me to see Veris and EM38 units mapping individual quarter sections, and I see great improvements to precision agriculture as private agrologists figure out applications.
When we farm the land we do one thing over the whole quarter section of Weyburn loam. When we farm the soil, we treat different areas differently.
The second photo shows Regina heavy clay on nearly level topography near Milden. Even level land can have quite different soils within one field. It takes little elevation difference to result in runoff of rain or snowmelt, and accumulation in shallow sloughs can change the soil dramatically. But the variation here is less than the Weyburn loam, and the approach to precision agriculture would be very different.
So, you see, land and soil are not completely random events and a fancy mathematical formula that pretends that they are is a very crude instrument.
In almost all the scientific papers I have read or heard about precision agriculture, the land or soil type is seldom if ever mentioned. I fail to see how we can make the most out of farming soils instead of land without using the wealth of information already available in map form.
The next article in this series on precision farming will deal with soil classification and soil maps in the three Prairie provinces. Stay tuned. †