Eyeball to Eyeball with Precision Farming

Dennis Demmel

"Precision farming" refers to a complex analysis of crop factors measured with high-tech instruments such as computerized yield monitors and GPS (global positioning satellite technology). The goal of this management approach is to analyze yield information so that inputs may be variably applied to optimize net returns. So far, precision farming has focused on N-P-K fertility levels, though some applications have looked at variable seeding and irrigation rates.

Those of us in sustainable agriculture have been monitoring our yields by the "eyeball" or "tractor seat" method for some time. I, for example, do most of the harvesting by combine on my farm and have an idea where yields are generally lower than other areas. This is the benefit of farms where the manager is still the equipment operator: we have a good idea of what’s going on in the field.

However, I was intrigued by reports of extreme yield variations being found by farmers with yield monitors. A farmer in York, Nebraska, found corn yields varying from 100 bushels to 205 bushels in a field averaging 180 bushels/acre. Similarly, a soybean field with an average yield of 50 bushels/acre ranged in yields from 10 to 76 bushels/acre. These extremes made me even more curious.

I looked into the requirements to get into precision farming. Just a yield monitor mounted on a combine to indicate yields-on-the-go costs $3,500 and up. A monitor with a complete GPS unit to record the monitor’s yields on maps can cost up to $8,000. "Grid sampling" for soil tests is very expensive on 5-acre increments, for example. Variable rate application equipment to apply different rates of fertilizer or seed can add thousands more to equipment costs. Soil sampling and variable application equipment can cost over $7 per acre, according to one source.

Bob Nielsen, a Purdue University agronomist, advises farmers to study yields for several years before deciding to use precision farming technology. Nielsen said, "You probably need four to five year’s worth of data before you can make that decision."

Even with the extensive soil sampling, many farmers and consultants find no correlation between yields and fertility. Yields seem to vary more with soil type, compaction, insects, and weed populations. Gary Hergert, director of the University of Nebraska’s North Central Research and Extension Center in North Platte, said, "There are not real big economic gains from the fertility side" for precision or site-specific farming.

A consultant, Don Larson of Iowa, who has worked closely with GPS technology for years, reported, "We could not show any linkage between fertility levels and crop production levels. ... Three factors show a correlation between soils and yield: 1. organic matter: not necessarily the amount but its biological activity; 2. water-holding capacity and the proper amount of moisture; 3. calcium levels and pH. These are the only three issues where we’ve found a correlation between computerized soil maps and yield maps from GPS yield monitors."

Another factor to consider is the topography of the field, manifested as varying elevations in the field that correspond with different soil types and organic matter levels. Vern Hofman, extension ag engineer at North Dakota State University, commented, "We feel because of lack of reliability and because of high cost, grid sampling doesn’t fit. That’s why we’re taking soil samples based on elevation in the field." Dave Franzen, extension soils specialist at NDSU, bases soil sampling on topography. "It was far less expensive than grid testing," he said. "With the topographic map, you can define the major boundaries and only test appropriate zones."

A Kansas farmer found that slight elevations in his fields had 4 inches less topsoil than lower elevations. Soil tests across one field were very similar although grain sorghum yields varied from 60 to 170 bushels per acre on the nearly flat field. The farmer felt that extra topsoil in low areas gave higher yields.

On our own farm we have found similar results, although we have greater extremes in elevations. Our heavier soils have slight elevations where the topsoil is fairly shallow. These soils have 1% less organic matter content than the "average" soil of the farm. These soils also seem to hold less moisture for crops and tend to harden easier. A subsoiler or ripper pulls much harder through these elevations and it is not uncommon to shift down two gears when ripping through these areas.

Weigh wagon analysis in the early 90s showed 10-12 bushels/acre reductions in yields for soybeans and 30 bushels/acre reductions in corn grown on the poorer soils under irrigation. We think that the current yield variations are less than 10 years ago, since we have been spot-applying compost for this long.

On our farm, we feel that applying compost to the poorer, more elevated soils is one of the most valuable things we can do to improve organic matter levels and yields. In earlier years we implemented "full field" compost spreading, but now we recognize that "precision" applications to poorer soils may give the best return to labor and other costs. We apply compost to poorer soils of irrigated fields to help improve water-holding capacity. Since fixed costs of irrigation equipment, land, taxes, and basic field operations are similar for all acres, humus improvements that increase yields on poorer soils should give good returns to compost application costs if yield increases are large enough. Humus improvements from compost should also provide longer term soil improvements than short-term N-P-K applications.

Deep subsoil tillage or "ripping" is another aspect of our "site-specific" strategy. The poorer, more elevated soils generally are harder, partly due to lower organic matter levels. We generally "spot" rip or subsoil these areas. We have also spot-treated poorer soils with humates and microbe inoculants to improve humus levels, and we have spot-treated with extra nitrogen applications during regular cultivation. We have seen yield improvements during recent years, particularly in corn and soybean yields in our irrigated corn-soybeans-wheat-sunflower-legume rotation.

Other simple, low-tech management choices can achieve goals similar to precision farming. We have noticed that sunflowers added to our dryland rotation have helped improve soil structure and increase wheat yields. There is less slabbing in heavier soils and fewer grassy weeds.

An old idea of the 70s has gained renewed interest and may bring high-tech and low-tech closer together. Remote sensing photography using infra-red light may be another low-cost method of identifying problem spots within fields. Conventional aerial photos can be used to correlate color variations with field conditions. Studies have shown a close correlation between color patterns in aerial photos and soil organic matter levels. I would guess that the colors would also correlate closely with elevation variations, where higher elevations would have lower organic matter levels. Hergert from North Platte cautions that "ground truthing" is not always available for infra-red remote sensing; it is not always known what the colors mean.

The cost effectiveness of high-tech precision or site-specific farming methods seems to depend on the farm in question. "Even the economics of precision agriculture are site-specific," stated Jess Lowerberg-DeBoer, Purdue ag economist, a recognized expert on the economics of precision farming. Some large farms might be able to justify the costs of the technology and accompanying soil testing. On smaller farms with less variability, the system using costly technology may not pay its way. Hergert has proposed "a new paradigm in sampling soils" that may be more appropriate technology for the family-sized, sustainable farmer (see sidebar at right).

If you decide to invest in some of the new precision farming technology, the best place to start would be with a yield monitor. It would indicate any major variations in yields, useful information even without the mapping and recording equipment. Calibration of the yield monitor’s sensor is critical and it should be re-set several times throughout harvest. Combine speed, grain moisture, hills, and slopes can create errors of up to 20%. Test plots where yield variations may be small should be analyzed by conventional weigh wagons. Such weigh wagons can be used to compare yields of poorer soils to yields of "average" soils. This approach will take more time, but much can be learned at a lower cost.

Hergert summed many of these issues up when he said, "How much do you want in accuracy and how much can you afford?" He further cautioned, "Once you know the problem, can you afford to fix it?" Some areas are easier to change than others. Should we be using the "tractor seat and common sense" or high technology?

Dennis Demmel farms near Ogallala and serves on the NSAS Board of Directors. This story was edited by Jane Sooby.


Farm Journal, Jan. 1998.

Soybean Digest, Oct. 1997

High Plains Journal, Dec. 16, 1996

Pro Farmer News, Feb. 17, 1996

Nebraska Farmer, Jan. 1998

Farm Industry News, special issue, 1996

Nebraska Sustainable Agriculture Society: Home        Features