U of M Extension Home : U of M Home

Minnesota Crop News > 2001-2008 Archives

Minnesota crop producers are always striving for higher yields. For various reasons, some have a perception that high soil test values are needed before it’s possible to achieve very high yields. With this perception fixed in mind, fertilizer applications (immobile nutrients) are planned so that soil test values will rise into a high or very high range. If there is success in raising soil test values to these elevated levels, the plan then shifts to keeping them at these levels by using fertilizer applications that supply the amounts of immobile nutrients that the crop removes. This approach to fertilization has frequently been referred to as the build and maintenance concept. Briefly stated, this approach to fertilization is designed to first build soil test values to some predetermined value, then, maintain at that level by using crop removal values.

A second concept for recommendations, adjusts the recommended rate of fertilizer for the measured relative level of that nutrient in the soil. There is no plan to create high or very high levels of immobile nutrients. This approach to recommendations has been labeled as the “sufficiency” concept.

So, there are obvious questions. Which of the two contrasting approaches leads to the higher yields and higher soil test values? Are there economic advantages for either approach? Long term research studies are needed to answer these questions.

The data needed to provide answers can be found in the results of a study designed to compare fertilizer recommendations from soil testing laboratories that used contrasting recommendation philosophies. These comparisons were made in Nebraska and Minnesota.

To reduce confusion, this discussion will focus on phosphate fertilizer application and changes in soil test phosphorus measured by the Bray and Kurtz #1 procedure. Data from three locations are summarized in Figures 1, 2, and 3. Data in Figure 1 were taken from an irrigated site in eastern Nebraska. Results from a non-irrigated field in northeastern Nebraska are summarized in Figure 2. Results from a non-irrigated site in southern Minnesota are provided in Figure 3. The data from Nebraska are for continuous corn grown for a period of 12 years. At Waseca, in southern Minnesota, corn was grown in years 1, 2, 4, 6, and 8. Data for soybeans grown at this site are not shown in this discussion.

For each site, information from a laboratory which used some variation of the “build and maintenance” concept (laboratory A) is compared to information from laboratories that use the “sufficiency” concept (University of Nebraska, University of Minnesota). It should be noted that laboratory A was not the same private laboratory in Nebraska and Minnesota.

For the irrigated site in eastern Nebraska, yield over the 12 year period was not significantly affected by the phosphate recommendations (Figure 1) The phosphate recommendations from laboratory A were associated with a total production of 1860 bu. per acre in 12 years. This is in contrast to a total of 1805 bu. per acre for 12 years following recommendations from the University of Nebraska. This yield difference was not statistically significant.
The initial soil test for P at this site was 12 ppm. The expected yield or yield goal was 170 bu. per acre. Both laboratories used nearly identical nitrogen recommendations.

Comparing laboratories, there was a substantial difference in the amount of phosphate fertilizer that was recommended. This was most notable in the first 3 years. With high amounts of phosphate applied (135 to 145 lb P₂O₅ per acre), soil test P increased to over 40 ppm. At the end of the study, phosphate recommendations from laboratory A were associated with soil test P values of about 40 ppm. With lower amounts of phosphate applied (University of Nebraska recommendations), soil test P was about equal to the initial value. All phosphate was broadcast and incorporated before planting.

Figure 1. Corn yield and soil test P as affected by contrasting phosphate recommendations from two soil testing laboratories.

There was a cost for building the soil test for P. Using a cost of $.25/lb. P₂O₅, total phosphate cost for the recommendations from laboratory A was $225.63 per acre. This compares to a total of $80.00 per acre for the phosphate recommendations from the University of Nebraska. This added expense of $145.63 did not produce additional yield. So, what, then, is the value of the higher soil test levels for P?

Because of major fluctuations in soil moisture, the data from the non-irrigated site in northeastern Nebraska are more erratic. Nevertheless, the results and conclusions are similar to those from the irrigated site (Figure 2). Drought in years 1, 2, and 3 seriously damaged yield.

For this site, the yield goal (expected yield) was 90 bu. per acre. Similar nitrogen recommendations were used by both laboratories.

Figure 2. Corn yield and soil test P as affected by contrasting phosphate recommendations.

Comparing recommendations from laboratory A and the University of Nebraska, there was no significant difference in yield. For 12 years, total yield was 1033 bu. per acre (laboratory A) and 1045 bu. per acre (University of Nebraska).

In the initial years, phosphate recommendations from laboratory A were substantially higher and soil test values for P increased. Initial values were in the range of 12 ppm to 15 ppm and increased to about 30 ppm when the higher rates of phosphate were applied. In subsequent years, phosphate fertilizer recommendations from both laboratories were nearly the same.
Throughout the study, higher soil test P values were associated with phosphate recommendations from laboratory A. For 12 years of continuous corn, phosphate recommendations from laboratory A cost a total of $103.75/acre while cost of phosphate recommendations from the University of Nebraska were a total of $30.00 per acre. In general, phosphate recommendation from laboratory A produced higher soil test values for P. If there was no difference in yield, it’s reasonable to question the real value of the higher soil test values for P.

It should be noted that soil test P values in the initial years increased even though no phosphate fertilizer was applied. This is attributed to mineralization of soil organic matter and little or no P removal because of severe drought. Ignoring the dry years, phosphate recommendations from the University of Nebraska produced no substantial changes in soil test P over the 12 years.
Since a corn/soybean rotation was followed after year 2, less information for corn is available from the Waseca site when compared to the Nebraska locations (Figure 3). However, corn yields resulting from the recommendations of laboratory A and the University of Minnesota were identical. Yield differences are not evident in Figure 3.

Figure 3. Corn yield and soil test P as affected by contrasting phosphate recommendations from two soil testing laboratories.

The expected corn yield for the first year at this site was 180 bu. per acre and was lowered to 160 bu. per acre in subsequent years. Nitrogen recommendations from both laboratory A and the University of Minnesota were nearly the same.

Except for year #5, the amount of phosphate fertilizer recommended for corn production by both laboratories was nearly the same. There were, however, substantial differences in the amount of phosphate recommended for 3 years of soybean production (175 lb. P₂O₅/acre for laboratory A; 80 lb P₂O₅/acre for University of Minnesota, data not shown). As a result, soil test values for P were higher in corn years 3, 4, and 5 when laboratory A is compared to the University of Minnesota (Figure 3).

As was the case for the two sites in Nebraska, money was spent for additional phosphate for the purpose of increasing soil test values for P. There was a measured increase in soil test P. There was, however, no increase in corn yield. Likewise soybean yields did not increase with added phosphate (data not shown). Again, a plan to purchase additional phosphate fertilizer for the purpose of increasing soil test P can be questioned.

Nearly everyone will agree that there is no single factor or management practice that is responsible for high corn yields. High yields are a consequence of using good agronomy combined with careful attention to management details.

Research from several universities conducted over the years has clearly shown that high corn yield cannot be achieved if phosphate fertilizer is not applied when soil test P values are medium or below. At the same time, results from studies cited in this discussion lead to the conclusion that high soil test values for P are not a prerequisite for high yield. The “crop sufficiency” approach to fertilizer recommendations can produce yields equal to those resulting from the use of the “build and maintenance” approach. In other words, if there is a known level of phosphorus in the soil and a realistic yield goal or expected yield, it is possible to apply phosphate fertilizer to achieve high yields without adding additional amounts for the purpose of building soil test levels. The economic return to the concept of building soil test values for P is questionable.