Interpreting results of the Mehlich-3 ICP Soil Phosphorus Test

Encyclopedia Article

Iowa State University (ISU) has provided soil-test interpretations for the Bray-1, Olsen, and Mehlich-3 (M3) soil phosphorus (P) tests since 1996. These tests use a colorimetric analysis method to measure the P extracted from the soil. This method is based on the intensity of a blue color that develops after adding appropriate chemicals to the soil extract. The M3 and Bray-1 P tests measure approximately the same amount of P in acid or neutral soils. However, the Bray-1 test measures less P than it should from some high-pH (calcareous) soils and is not recommended for soils with soil pH 7.4 or higher. If only one test is to be used across all Iowa soils, the Olsen and M3 P tests produce the most reliable results. Some laboratories use the Bray-1 test for samples with acid to neutral pH and the Olsen test for high-pH samples, which is acceptable.

Since November 2002, Iowa State University also provides soil-test interpretations for a new version of the M3 P test. This new version uses an inductively coupled plasma (ICP) instrument to measure the P extracted from the soil. Soil test laboratories are adopting the new method rapidly because the ICP instrument can be used to measure amounts of other nutrients in the same soil extract. However, using the ICP method results in higher P test values. The reason is that soil extracts for all P tests have the orthophosphate P form and also small amounts of other P forms (such as other inorganic and simple organic forms). The ICP instrument measures all P forms in the sample.

Iowa research shows that use of the M3 extractant with an ICP determination of extracted P usually measures between 10 and 15 ppm more P than the original test does. The variation is large, however, and a single transformation factor to get results for one test from the other would give unreliable results. Thus, use of the M3 extractant combined with the new ICP analytical method should be considered as a different test (the Mehlich-3 ICP test), and interpretations should be based on field calibrations with crop yield response.

Field calibration research conducted during recent years provides the basis for soil-test interpretations. The graphs in Fig. 1 show, as an example, the relationship between relative corn yield response to P fertilization and soil P measured with both M3 tests. The M3-ICP test values needed for optimum yield are higher, and not all soils reflected the same change along the soil-test P axes of the graphs. The data indicate no clear advantage of one or the other test in predicting crop response to P fertilization.

Figure 1. Relationship between relative corn yield response to P fertilization and soil P measured with two versions of the Mehlich-3 soil P test.


Complete soil-test interpretations for both M3 tests were published in the November 2002 revised publication A General Guide for Crop Nutrient and Limestone Recommendations in Iowa, (PM 1688), which is available at your Iowa State University Extension county office. Table 1 was copied from that publication, and shows soil-test P interpretation classes for the Bray, Olsen, traditional M3, and M3-ICP tests. Because of results discussed previously, interpretations for the M3 colorimetric test are similar to those used for Bray-1 test and are the same to those recommended since 1999. The amounts of P fertilizer recommended for the soil-test interpretation class are the same for all soil P tests, and are not shown here (see PM 1688).

The differences between the two M3 P tests discussed above do not apply for other nutrients that can be measured in M3 soil extracts, such as K, calcium (Ca), and magnesium (Mg) among others. Using an ICP instrument to measure these nutrients produces similar results to the traditional methods (flame photometry and atomic absorption spectrometry). Moreover, the M3 extracts amounts of K that are similar to those extracted by the commonly used ammonium-acetate test across all Iowa soils. Thus, interpretations for K using the ammonium-acetate and M3 tests shown in Table 1 and in publication PM 1688 are similar, and no distinction is made between ICP and the other determination methods.

Table 1. Interpretation of soil test values for phosphorus (P) measured by the Bray-P1, Olsen, Mehlich-3, Mehlich-3 ICP tests and for potassium (K) measured by the ammonium acetate and Mehlich-3 tests (6 to 7-inch deep soil cores).*

Relative Level Phosphorus Potassium
Wheat, Alfalfa All other crops

Subsoil P

All crops

Subsoil K

Low High Low High
Bray-1 or Mehlich-3 Ammonium acetate or Mehlich-3
Very Low 0-15 0-8 0-5 0-90 0-70
Low 16-20 9-15 6-10 91-130 71-110
Optimum 21-25 15-20 11-15 131-170 111-150
High 26-30 21-30 16-20 171-200 151-180
Very High 31+ 31+ 21+ 201+ 181+
Very Low 0-10 0-5 0-3  
Low 11-14 6-10 4-7  
Optimum 15-17 11-14 8-11  
High 18-20 15-20 12-15  
Very High 21+ 21+ 16+  
  Mehlich-3 ICP  
Very Low 0-20 0-15 0-10  
Low 21-30 16-25 11-20  
Optimum 31-40 26-35 21-30  
High 41-50 36-45 31-40  
Very High 51+ 46+ 41+

*Relative levels and soil-test values in this table are exactly the same as those in Table 1 of A General Guide for Crop Nutrient and Limestone Recommendations in Iowa (PM 1688) for crops listed in that publication.

This article originally appeared on pages 178-179 of the IC-490(24) -- November 17, 2003 issue.

ICM News Archive