Skip to content

Re-evaluating phosphorus and potassium management for corn, soybeans and wheat in Ontario

Principal Investigator

Horst Bohner & David Hooker

Research Institution

Ontario Ministry of Agriculture, Food and Rural Affairs (OMAFRA) and University of Guelph (U of G)

Timeline

January 2012 – February 2018

Objectives

  • To initiate a corn-soybean-wheat rotation on four sites with relatively low soil test P and K levels.
  • To build-up areas within each site to moderate P, moderate K, and moderate P and K soil test levels (P>20 ppm, K>120 ppm). High rates of P and/or K fertilizer during the first few years after initiation were used to build these areas.  Maintenance P and K will be applied to keep soil tests adequate.  These built P and K soils will then be compared to soils with low background fertility.
  • To test various strategies (broadcast and starter) within each management approach (build and maintain vs. sufficiency) for economic yield responses in corn, soybean and wheat.
  • To test whether current dated OMAFRA P and K recommendations ensure economic yields in high yield environments.

Impacts

  • The most economically effective phosphorus (P) and potassium (K) fertility program (sufficiency vs. build-and-maintain approaches) will allow for increased economic sustainability by allowing farmers to be more efficient in their use of P and K fertilizer.

Scientific Summary

Current phosphorus (P) and potassium (K) recommendations for Ontario are based on data from the 1960-1970s. Doubling of phosphorus and potassium fertilizer prices over the past 10 years has led some producers to reduce P and K fertilizer application rates. Crop yields during this same time period increased resulting in greater demand for, and removal in grain, of P and K. Currently, corn yields exceeding 200 bu/ac, soybean yields exceeding 50 bu/ac and wheat yields exceeding 100 bu/ac are becoming increasingly common.  When these three crops are planted in rotation at yields of 200 bu/ac for corn, 50 bu/ac for soybeans and 100 bu/ac for wheat, the average yearly phosphorus and potassium removal is about 65 lb/ac for both P and K.  There is concern that current OMAFRA P and K recommendations do not adequately provide for yields that are much higher than those of 25 years ago. There is very little long term research in Ontario comparing grain yields on soils with lower P and/or K levels (where only maintenance P and K fertilizer rates were applied) against yields on soils where P and K levels were built to relatively high levels.

This proposed research investigated the best P and K fertility approach with today’s grain yields: current sufficiency approach versus “build and maintain” approach. The project established four long-term field trials to evaluate corn, soybean and wheat response. Yields were measured where only sufficiency P and K fertilizer rates were applied on low testing soils against where soil-test P and K levels were built to greater than 21 ppm P and 120 ppm K and then maintained. A three-year soybean-winter wheat-corn rotation was established with each crop present for each year of the trial. The goal is to maintain these trials for at least 10 years.

Results – Corn:

Corn yields responded to a starter blend of P and K in relatively low background fertility levels (i.e., P<20, K<120 ppm) by 35 bu/ac across the 21 site/years of this study (P<0.05). On plots with moderate soil test P and K values (i.e., P>20, K>120 ppm) corn responded to dry starters containing both P and K in a 2×2 band by 9 bu/ac (P<0.05).

As expected, corn response to starter fertilizers decreased as the soil test P and/or K increased.  Averaged across background fertility, liquid starter added 10 bu/ac, MAP alone added 13 bu/ac, potash alone added 15 bu/ac, and a blend of MAP and potash added 24 bu/ac compared to no starter fertilizer (P<0.05).

The highest corn yields were produced where soil test values were moderate in both P and K (i.e., P>20, K>120 ppm).  Averaged across all starter treatments, grain corn yields averaged 25 bu/ac higher in plots with moderate soil test values compared to low soil test values (i.e., P<20, K<120 ppm).  Of particular note, corn yields were 9 bu/ac higher in plots with moderate soil test P and K in the high starter treatment compared to relatively low soil test P and K across 21 site-years (P<0.05; 188 vs. 200 bu/ac); the starter rate approximated or exceeded the rate currently recommended given soil test P and K recommendations.  This finding supports the notion that starter fertilizer rates according to current OMAFRA recommendations may not produce the highest grain corn yields if background soil test levels are relatively low.

This study supports existing P and K fertility work that corn is highly responsive to starter fertilizers when soil test levels are relatively low; however, the new finding is that starter fertilizers cannot maximize yields with relatively low soil test levels. Results to date from this study also demonstrate that the sufficiency approach (highest rate of starter P and K) yielded significantly less corn (12 bu/ac; P<0.05) than soils that have been built to moderate levels of P and K.

Results – Soybean:

Soybean yields responded to a starter blend of P and K in low background fertility soils (P<20, K<120 ppm) by 5 bu/ac across the 21 site-years of this study (P<0.05). In plots with relatively moderate soil test values (P>20, K>120 ppm), soybean response to starter fertilizer was less than 2 bu/ac. Averaged across background fertility, liquid starter added 1 bu/ac, dry MAP added 3 bu/ac, potash added 1 bu/ac, and the MAP and potash blend added 4 bu/ac compared to no starter fertilizer (all responses were statistically significant at P=0.05).  On low background fertility soils liquid starter added 2 bu/ac, MAP added 3 bu/ac, K added 2 bu/ac, and a blend of P and K added 5 bu/ac compared to no starter fertilizer.

This study, as expected, supports existing work that soybeans respond to starter fertilizers when soil test levels are low; response to starter fertilizers becomes less as soil test values increase. This study has shown P nutrition is critical for the highest soybean yield; potash by itself was not sufficient to maximize soybean yields.

Similar to the new finding in corn, averaged across starter treatments, soybean yields were 7 bu/ac higher (P<0.05) in plots with relatively moderate soil test levels (60 bu/ac) compared to low background fertility (53 bu/ac).  This trend of higher yields with background soil test values occurred at all 4 locations, but was the greatest at Elora, where soybean yields were 15 bu/ac higher in plots with moderate soil test P (53 bu/ac) and K compared to low testing plots (38 bu/ac) averaged across starter treatments (P<0.05).

In the treatments where the starter rate approximated or exceeded the rate currently recommended by OMAFRA from soil test P and K, soybean yields were 4 bu/ac higher in plots with moderate soil test P and K in the high starter treatment compared to relatively low soil test P and K across 21 site-years (P<0.05; 60 vs. 56 bu/ac).  This significant finding indicates that starter fertilizer rates, according to current OMAFRA recommendations, may not produce the highest soybean yields if background soil test levels are relatively low.

Results – Wheat:

As expected, winter wheat yields responded to MAP placed in-furrow with low background soil test P and K levels (P<20, K<120 ppm) by 15 bu/ac across the 17 site/years of this study.  The addition of potash to the MAP did not increase yield. Averaged across background fertility, liquid starter added 5 bu/ac, MAP added 10 bu/ac, potash alone did not increase yield, and a blend of P and K together added 10 bu/ac compared to no starter fertilizer (P<0.05).

On soils with moderate soil test P and K (P>20, K>120), wheat showed less than 3 bu/ac response to any starter fertilizer compared to no starter. This study found that wheat is responsive to starter fertilizers especially when soil test levels are low.  It also demonstrates that winter wheat is highly responsive to starter P.

The highest wheat yields were produced in plots with moderate soil test values for P and K.  Across all starter treatments, wheat yields averaged 10 bu/ac higher in plots with moderate soil test P and K compared to plots with relatively low soil test P and K.  Most notably, where high rates of P and K were applied in the starter, wheat yields were 11 bu/ac higher with moderate soil test compared to low background fertility (P<0.05).

Results from this study demonstrate that the sufficiency approach (highest rate of starter P and K) yielded significantly less wheat (11 bu/ac) than soils that have been built to moderate levels of P and K.  This indicates that starter fertilizer rates, according to current OMAFRA recommendations, may not produce the highest wheat yields if background soil test levels are relatively low.

Results – Soil test responses to crop removal and added fertilizer P and K:

Soil test P and K was monitored at each of the four field locations.  Where P and K fertilizers were applied, soil test values increased faster than anticipated. The amount of fertilizer P and K needed to increase soil test values is critical for economical analysis of the build/maintain approach compared to the sufficiency approach.  It is likely that recent additions of fertilizer P and K may temporarily increase extractable P and K in soil tests, thus underestimating the amount of fertilizer P and K necessary to increase background soil test P and K.

External Funding Partners

Ontario Agri-Food Innovation Alliance

The project was funded in part by the Ontario Agri-Food Innovation Alliance, a collaboration between the government of Ontario and the University of Guelph.

Project Related Publications

None.