What does an agronomist or crop consultant do when researchers at a major land grant university claim existing soil tests for potassium are so faulty as to have little if any value? As a full-service ag retailer, how do you respond to their argument that “an extensive survey of more than 2,100 yield response trials confirmed that KCl (potassium chloride) fertilization is unlikely to increase crop yield.” With the research only published this past October, the entire issue may still be news to you, but it is a subject likely to come up with customers. To date, little discussion has been had ... publicly.

“There has been a noticeable silence,” said Richard Mulvaney, co-author of “The Potassium Paradox: Implications for Soil Fertility, Crop Production and Human Health.” “The public deserves a scientific discussion of the data.”

Why the silence? It is not for lack of author credentials, depth of material, nor for lack of review. Lead author Saeed Khan, Mulvaney and Timothy Ellsworth are all professors in the Department of Natural Resources and Environmental Sciences, University of Illinois, Urbana-Champaign. The paper cites 500 peer-reviewed, university-produced research papers on the subject of KCl. First submitted to the Soil Science Society of America Journal in 2012, peer reviewers recommended against publishing without giving any science-based reasons. The authors then turned to Renewable Agriculture and Food Systems, a less well-known journal, but one willing to present the controversial findings.

“The Potassium Paradox” revisited research Khan carried out for his Ph.D. in the late 1980s. It included multi-sample, bi-weekly soil sampling for K in corn/soybean rotation plots, including one that had received no K fertilizer for 15 years. Sampling was carried out year round from March 1986 to March 1990. Soil K testing was performed on the field-moist samples before and after air-drying, common practice for commercial labs.

Variations from day to day throughout the year and throughout the four years were dramatic, as were the variations between test results. However, both tests showed steadily increasing K levels in the plot where no K was applied.

“I told my advisor that soil K testing does not work,” recalled Khan.

Due to Khan’s return to work in Pakistan and health issues of his advisor, the research was never published. Nearly 20 years passed until Khan worked with Mulvaney and Ellsworth on several projects related to soil nitrogen management. Discussions led them to review his earlier research and the published literature concerning KCl fertilization. The three researchers agreed on the importance of a comprehensive review of the scientific literature and were careful to focus their efforts on peer-reviewed and university publications as the most objective resource available. The process took five years.

“Farmers have been led to believe that the soil K test was not perfect, but was the best we had and was rated up to 80 percent effective in the Illinois Agronomy Handbook,” said Mulvaney.

“In fact, it doesn’t work, and it never will. How do you manage what can’t be measured?”



Paul Fixen, senior vice president, International Plant Nutrition Institute (IPNI), doesn’t agree with some of the paper’s conclusions; however, he does agree that many of the questions raised are valid. Although the authors may not be hearing much official response, Fixen said the paper was a hot topic of discussion at a recent annual meeting of the American Society of Agronomy. He questions reference selection and interpretation by the researchers. However, he said he had no argument with much of the paper.

“There have been gaps in the science of assessing soil K plant availability and predicting needs,” admitted Fixen.

“Those questions have been on the radar screen of the scientific community for the past few years, and concerns have been intensifying for a number of reasons.”

Fixen cited the increasing cost of K as an input, as well as the impact of long-term reduced tillage and resulting stratification from what some call “nutrient uplift,” biocycling from the root profi le to the surface. Adding to the uncertainty and difficulty in tracking K and predicting response is that these changes may be affecting basic soil mineralogy.

“There are two dominant types of clays in Midwest soils: smectite, an expanding clay, and illite, a non-expanding clay that traps K between its layers,” explained Fixen. “What mineralogists are seeing is if you create a higher K environment, over time you may convert smectite clay to an illite-like clay. Soil test extracting solutions don’t detect K trapped by illite clays.”

He noted that so-called fixed or nonexchageable K is neither fixed permanently nor very nonexchangable. The pools of fixed and exchangeable K are more variable within a season than previously thought. This suggests that nutrient uplift may create a big reserve that slips back and forth between exchangeable and nonexchangeable pools.

“There is a lot of stuff we don’t understand very well,” said Fixen.



Perhaps even more controversial and of greater concern for fullservice ag retailers selling KCl is the statement mentioned above on yield response. The 2,100 yield response trials, including 774 grain production trials in North America, showed KCl as 93 percent ineffective at increasing yield with yield reduction more likely.

“There were cases where soybean yields were significantly decreased with KCl applications, and one reason is that K depresses calcium and reduces uptake,” reported Mulvaney. “The chloride in KCl competes with nitrate uptake. You don’t want anything competing with N.”

The authors are quick to say that sandy or shallow soils and certain crops, such as alfalfa hay, sometimes benefit from K applications. However, they recommend using potassium sulfate (K2SO4) instead of KCl, noting that unlike chloride, sulfur does not depress nitrate uptake. With most soils and crops, and in particular grain production, the researchers argued that frequent K applications are not needed.

Fixen argued in turn that acting on the authors’ position could be economically damaging to farmers. He noted that an IPNI soil test summary of 700,000 samples from Iowa in 2010 showed 70 percent of samples testing below Iowa State University’s (ISU) new critical level of 200 ppm K and 49 percent testing below 160 ppm (by dry test). He added that an ISU data set showed that the soil test does work, though it has limitations, some of which are removed with the fi eld moist soil procedure.

“How does one draw the conclusion they draw when one of the most extensive K data sets in North America showed this kind of response?” asked Fixen, who noted the annual removal of K in harvested grain. “The K has to come from somewhere.”

Khan credited K mobility and abundance in the soil. He compared crop removal (in the case of grain) to removing a cup of salt from the ocean and expecting salinity to change. As K uplifts to the surface, the mass of K in the soil profile is essentially untouched, he argued.

He pointed to the famed Morrow plots in continuous corn for 137 years. “In 1955, a K test was 216 pounds per acre in a plot with no K ever added,” he said. “In 2005, it was 360.”

 “Potassium Paradox” Questions





Where Fixen and the University of Illinois researchers do agree is the need for on-farm trials. Mulvaney added that ag retailers, agronomists and crop consultants have a vital role to play here.

“Help farmers establish split trials on their farms,” he said.

“Now is the time for farmers to utilize precision technology to check for a yield response to K. It can only improve management of K.”

Noting that it can take months for scientific review and reaction to an article, Fixen advised against throwing the baby out with the bathwater. “Our old soil tests are serving us pretty well, but as Ronald Reagan said, ‘trust, but verify,’” he said. “There is no better way of verifying than maintaining comparison strips using precision ag technology. If the strips show up, you know you have a problem. If they don’t, you aren’t shorting the crop. Replicate to eliminate other problems like drainage. It’s an easy thing to do.”