By Lynn Betts
Progressive Farmer Contributing Editor
Farmers embrace all sorts of technology to enhance efficiencies and productivity. Yet, when it comes to pinpointing a field's fertilizer program, decisions can still largely hinge on a simple metal probe used to collect soil samples.
What has changed is the ability to better define areas with similar soil characteristics and the ability to feed that information into geographic information systems (GIS) that can deliver variable rates of fertilizer to selected areas of the field.
Of course, that data is only as good as the person collecting it. Tuning up your fertilizer program requires you to double down on your soil-sampling skills.
SPRING OR FALL?
When is the best time to pull samples? Some soil scientists point out readings for some nutrients may be more accurate in the spring, when soils are likely to have more moisture. Some farmers prefer spring soil tests so they have more time to analyze results ahead of fall fertilizer applications.
Still, many university recommendations stick with time-tested fall sampling -- and there's agreement that overall, spring/fall timing doesn't make much difference. The best time to sample is when you have time to do a good job.
"The biggest errors in soil testing come from the sampling process," said Shannon Gomes, an independent crop consultant in northeast Iowa for the past 20 years. Gomes has six years of soil-mapping experience and a master's degree in soil fertility from Iowa State University. "Quality information from proper sampling takes time and effort, but that's what you need for solid data to make good fertilizer decisions."
He urged farmers to talk with their advisers and ask in detail how they handle their soil-sampling work. See if they have verification of their sampling methods -- or even go along with them at times to observe.
Soil specialists recommend farmers soil-test at the same time of year, whichever season is chosen, after the same crop. Be sure to provide samples to soil depths that match the depth the lab is expecting. "That consistency will improve your ability to compare results over time," explained University of Wisconsin (UW) associate professor of soil science Carrie Laboski.
For nutrients like potassium and sulfur, however, your best bet may be taking samples in the spring. "Potassium soil-test readings are more variable in the fall, and if a sulfur soil test is used, results may be too low to represent what will be available to the plants in the spring," said John Kovar, a soil scientist with the National Laboratory for Agriculture and the Environment based in Ames, Iowa. "Potassium readings in the fall aren't as reliable because potassium leaches out of crop residues in the fall and over the winter.
"And from my experience, you can't trust fall testing for sulfur," he continued. "It's a little like nitrogen; it's a moving target as it mineralizes. A spring test before planting will give you the best idea of what's there before planting."
"That said," Kovar added, "we know the labs couldn't handle all the tests in the short window in the spring before planting. Fall soil testing is certainly better than no soil testing at all."
HOW OFTEN?
There isn't universal agreement on frequency of testing. Kovar noted the majority of samples come from grid soil sampling, and more frequent testing can be expensive. "The three- to four-year standard should work for most people unless there's a change in management, like the addition of cover crops or a change in rotations," he said.
There are exceptions. Most university soil specialists suggest fields with high fertility needs or problem areas should be soil-tested every two years or, in some instances, annually.
UW recommends sampling once every four years unless high-value vegetable crops or high-nutrient removal crops like alfalfa and corn silage are grown, Laboski said. Most guidelines stick with a three- to four-year frequency or once every crop rotation.
Gomes has clients who test every two years in a corn-soybean rotation, but most test every four years. In continuous corn, testing is done every three years.
Ryan Moellers grows corn and alfalfa for a 700-cow Holstein dairy herd near Cresco, Iowa. "We soil-test every three years and have learned how to adjust our rates accordingly," he said. "We've been growing some of our corn with injected manure as our sole source of nutrients, applied according to our soil tests based on phosphorus."
He points out some years he doesn't put any manure or commercial phosphorus or potassium on the field. "Soil testing and fertilizer determinations take time and knowledge—you get both from a good adviser [like Gomes]."
PULL REPRESENTATIVE SAMPLES
Gomes emphasizes it's crucial to collect enough core samples using a sampling pattern that gives results representative of the field.
"You can get pretty maps, but that's meaningless in telling you the true fertility of your field if the data is poor," he said. "If I pull one or two ears of corn and make a yield estimate based on that ear or two, how reliable would that be compared to taking 15 to 20 ears?" he asked.
"Soil sampling is the same idea -- more core samples mean more representative data," Gomes continued. "There are people who take as few as three cores for a 2 1/2- to 5-acre grid sample. The pretty map you get from that sampling has an error rate that could result in over- or under-fertilizing."
For example, say your optimum phosphorus range is 16 to 21 parts per million (ppm). Your field measures 18 ppm. With a plus or minus accuracy range of 4.5 ppm, because of fewer core samples or less intense sampling, you could have P levels as low as 13.5 ppm or as high as 22.5 ppm. That means the field could be at a low or high P level; you just don't know.
"I'd rather take 15 to 20 cores for a representative sample you can have confidence in," Gomes said. "You need to know the numbers behind the colors."
Growers' adoption of precision ag practices has increased the popularity of smart or directed soil sampling. These usually fall into two categories: grids or zones.
Dallas County, Iowa, no-till farmer Raynold Johnson has tested soils on 2.5-acre grids since the practice was first offered by the West Central Cooperative, Woodward, Iowa. He gets soil tests and fertilizer recommendations, chemical and fertilizer products and applications, cover crops and advice from the cooperative. "It all starts with a good soil test," Johnson said. "It's basic -- you have to know where you are before you can have any confidence in knowing what you need to apply."
Mike Schrum, an agronomy field marketer for West Central, said he's seen more people move to grid sampling on 2.5-acre grids. But the grid system can have drawbacks.
"It's a customer choice, but my advice is to reduce the error rate or the variability in samples as much as you can," he said.
That means using a smaller grid, unless you have really high levels of P and K present, he said. "On my own land," Schrum continued, "I've done some 1.1-acre grid sampling. It costs about twice as much as our standard 2.5-acre grids, so not many people want to do it, but it does give you more accuracy with variable-rate applications."
Wisconsin's Laboski agreed. "Grid sampling is not very useful if the grids are too large (for example, 5 acres) because variability in the field is not adequately mapped."
Research in Wisconsin indicates fields that had very high or excessively high P and K levels the last time they were tested should be sampled in an unaligned pattern on a 300-foot grid, taking 10 to 20 cores per sample. That's one sample every 2 to 2.5 acres. If either P or K levels were in the high category or lower the last time they were tested, UW recommends using a 200-foot grid with a sample from every acre.
Gomes isn't a proponent of grids, either. "In the Prairie Pothole area, for instance, an Okoboji soil could have a pH of 6.5," he said. "But Canisteo or Harp soils that surround Okoboji soils in a ring, in that same grid, could have a pH as high as 8.4. Phosphorus needs change with pH, so that's a sampling error with impact."
Instead, Gomes prefers zone samples, taking plenty of cores, basing zones on electrical conductivity rather than soil survey or grid lines. "The resolution of the soil survey is 2 acres; any soil less than that was called an inclusion, and it could have entirely different soil properties," Gomes explained. "With electrical conductivity and GPS, we can get 100 times more accurate than that, delineating soils down to 900 square feet or even 300 square feet with more accurate soil boundary lines."
Regardless of the approach you use, it boils down to basics. "Poor sampling doesn't provide any useful information for whole-field or variable-rate fertilizer management," Laboski said.
**
Editor's note:
This is the fifth in DTN/The Progressive Farmer's series on Finesse Field Fertility, helping farmers fine-tune how to feed their crops and get the most value from every input and practice.
(ES/BAS)
© Copyright 2016 DTN/The Progressive Farmer. All rights reserved.