My first work on precision agriculture was actually in the 1960s and colleagues had been doing work on the idea before that. Back then it was all about identifying specific soil profile types in a field and trying to determine if their variable properties could be managed differently. Some things were clear:
1. Leached (white but not salty) sloughs were very high in available phosphorus but they still responded to a dab of seed-placed phosphorus — largely because of the cold, wet conditions in the spring. The bottom line was to leave out those soils when taking a field composite sample.
2. Soil nitrate nitrogen was low on poor, low organic matter (OM) soils like eroded knolls and much higher in lower areas with very deep topsoil and high OM. So, the answer should be simple: pork lots of nitrogen on the knolls and back off for the rich soils in low lying areas.
But, in this case, the confounding factor is moisture. The knolls are poor because they are dry and the low areas produce well because they may have more moisture. Again, the answer is simple: in a wet year pork lots of nitrogen on the knolls and leave it off the low areas that may often be drowned out.
The kicker is that we do not know if or when it is going to rain. We can easily determine fall soil moisture and use that in planning, but in-season rain is a complete crap shoot.
At that time there was no technology to effect soil-specific applications within a field so the idea was dropped.
Now the technology is there in spades, but most are using it with no or little attention to the soil types involved.
A scientific and peer reviewed journal with the title Precision Agriculture started in 1999 now has 18 years of results. It is available with a few mouse click so I have checked it all out. Most of the papers deal with various fancy technologies, but few deal with agronomics.
When a precision ag method is being planned, the first order of business should be to look at the soil map and determine what input is to be varied over a field and what soil profiles are to be separated for treatment. The examples to come in this piece are all from Saskatchewan but the principles apply to all three Prairie provinces.
Saskatchewan: The first soil survey
The first soil survey was really a long series of individual area surveys that took place from the 1920s to the 1950s. When the ag area was complete it was unified into a single concept and presented in two reports with maps.
Soil Survey Report #12 covered all of Saskatchewan from the U.S. border to the top of Township 48 (i.e. Prince Albert). The map scale was one inch = eight miles. Not a lot of detail but a lot of good information.
The ag area north of township 48 was published as Soil Survey Report #13. These maps were at a scale of one inch = three miles. The #13 map of the P.A./Carrot River area does a great job of telling us the location of very severe potassium deficiencies — just look at the area mapped as Carrot River Association.
The basic unit of mapping was the soil association and texture. Examples are Weyburn loam, which is a soil formed from glacial droppings in the dark brown soil zone and Blaine Lake loam which is a soil formed from lacustrine (lake) deposits in the black soil zone.
The framework provided by Soil Maps #12 and #13 was the basis of all agronomic work for 50 years. With that information it was possible to be much more specific with agronomic recommendations.
The second soils survey
The second Saskatchewan soil survey was also a series of surveys carried out from the 1950s to the 1990s. They were published in three separate formats:
1. Large topographic map sheets with coloured maps that were easy to read and useful on the wall. The scale was one inch = two miles. Areas covered were Weyburn, Willowbunch, Regina, Rosetown, Swift Current, Saskatoon and Hudson Bay.
2. Atlases of individual RMs or groups of RMs with very good information on geology as well as soils and with several interpretive maps such as ag capability, irrigation suitability, wind and water erosion risk. They are very good but the expense of such a fancy document meant that few were published.
3. Black and white maps of individual RMs. These maps and reports have good information but are hard to read and not at all inviting to use.
Because the second soil survey had never been made into a unified piece, it has not been fully utilized as an agronomy framework.
The information is all available online — there are PDFs of each map and report. Go to the federal government’s website, select your language, then “soil survey reports and printed maps.” At the bottom of the page click “Show availability of scanned soil surveys (KML)” and a map will pop up that you can use to find the area you want.
The future: SM SIS
I am happy to report that by the time you read this the SK SIS (Saskatchewan Soil Information System) will likely be launched and available online.
That website will provide online, seamless access to all of the second soil survey information. An area of interest can be located by zooming in on the presentation map or by a specific legal location. A coloured map will appear, and if you slide back a transparency bar a recent colour air photo appears. I am having great fun using it to check out the “Land For Sale” column in the Western Producer.
The SK SIS is the work of Dr. Angela Bedard-Haughn, the current head of the Soil Science Department at the University of Saskatchewan. Thanks Angela.
I must confess that my idea was that the entire province should be converted to large, coloured printed maps first and then proceed to digital. But, I was wrong. SK SIS was the way to go and we look forward to continuous updates that can occur in both technology and information accuracy.
The bottom line with respect to precision ag is that the starting point should be to take a peek at the soil map for your area and then decide what approach is appropriate for that area. One size does not fit all.