Interpretation of Soil Phosphorus Tests for Corn in Soils with Varying pH and Calcium Carbonate Content

There is uncertainty about appropriate soil-test extractants and interpretations in regions with soils varying in pH and calcium carbonate content. The objective of this research was to compare extractable P using the Bray-P₁ (B1), Olsen, and Mehlich-3 (M3) extractants and to obtain soil-test interpretations for corn (Zea mays L.) in soils varying in pH and calcium carbonate equivalent (CCE). The extractants were compared by correlating the amounts of P extracted from 350 soil samples and by calibrations based on 61 field response trials. Correlations between P extracted by the B1 extractant and the Olsen or M3 extractants improved markedly when calcareous soils were excluded from the analyses. The correlation between the Olsen and M3 extractants was high and independent of soil pH. Neither soil pH, CCE, soil type, nor organic matter content completely explained the low correlations between P extracted by the B1 and the Olsen or M3 methods in high pH soils. Critical concentrations across all soils defined by the Cate-Nelson and linear-plateau models were 7 and 9 ppm for the Olsen extractant and 14 and 24 ppm for the M3 extractant. When highly calcareous soils were excluded, the critical concentrations for the B1 extractant were 11 and 20 ppm, respectively. The Olsen and M3 extractants are more reliable than the B1 for estimating plant-available P of Iowa soils over a broad range of soil pH and CCE. Research QuestionThere is uncertainty about appropriate soil tests for P and interpretations in regions with soils varying in pH and calcium carbonate content. The Bray-P₁ (B1) and Olsen (or sodium bicarbonate) P extractants are widely used in the North Central Region, however, the B1 extractant often underestimates plant-available P in calcareous soils. The uncertainty about an appropriate extractant has induced many routine soil testing labs to use both the B1 and Olsen extractants. This reduces lab efficiency at a time when increased environmental awareness and adoption of grid soil-sampling techniques are increasing the demand for soil testing. The use of a multielement extractant for soils with a broad pH range is an attractive possibility. The Mehlich-3 (M3) extractant was developed for routine soil testing of P, K, and other nutrients. This test, however, has not been thoroughly evaluated in calcareous soils. The objective of this research was to compare P extracted by the B1, Olsen, and M3 extractants and to obtain soil-test interpretations for corn in Iowa soils varying in pH and calcium carbonate content. Literature SummaryThe B1 and Olsen soil tests for P are widely used in the North Central Region. Research has shown that the B1 extractant is reliable on neutral or acid soils but often underestimates available P on calcareous soils. The Olsen extractant, on the other hand, has traditionally been considered a reliable extractant for calcareous soils as well as in many acid soils. The M3 extractant is an attractive procedure for routine soil testing because it seems reliable for P, K, and many secondary nutrients and micronutrients. Research comparing P extracted by the M3 and B1 extractants suggests that the amounts of P extracted by the M3 extractant are similar to or slightly higher than amounts extracted by the B1 extractant. The ratio of P extracted by B1 and M3 extractants, however, often is different depending on the soil type. Moreover, some studies suggest that the M3 extractant extracts more P from calcareous soils than does the B1 extractant. There is little information on soil test interpretations of the M3 extractant in soils of the central and western Corn Belt, especially on areas having both calcareous and noncalcareous soils. Published field calibrations for soils of other regions probably do not apply because of major differences in soil types. Study DescriptionThe B1, Olsen, and M3 soil tests for P were compared by using two approaches. In one approach, the amounts of P extracted were compared by analyses of 350 soil samples varying in soil pH and calcium carbonate equivalent. The soil samples were selected to include many soil series and to encompass a wide range of soil pH and calcium carbonate equivalent (CCE). Ranges of soil test values for these samples were pH 5.1 to 8.2, Olsen P 2.5 to 42 ppm, M3 P 3 to 91 ppm, B1 P 1 to 96 ppm, and CCE 0.5 to 17%. In the other approach, long-term and short-term P-response studies with corn (61 site-yr) were conducted on farmers' fields and experimental research centers from 1989 to 1995. Ranges of soil test values were pH 5.1 to 8.2, Olsen P 2.5 to 42 ppm, M3 P 3 to 80 ppm, B1 P 1 to 59 ppm, and CCE 0.5 to 4.1%. Briefly, field treatments included a nonfertilized control and fertilization rates that varied from 10 to 65 lb P/acre and there were three or four replications depending on the trial. Critical concentrations of soil tests were estimated by fitting the Cate-Nelson statistical procedure and the linear-plateau segmented model to relationships between relative corn yields and extractable P. Applied QuestionsWhat were the relationships between soil P extracted by the different extractants? The correlations between P extracted by the three extractants were strongly dependent on soil pH and CCE of the soils. The amounts of P extracted by the B1 and those extracted by the Olsen or the M3 extractants were very poorly related when all soil samples were included in the analyses but were highly correlated when soils with pH equal to or greater than 7.4 or soils with more than 4% calcium carbonate equivalent were not included in the analyses. These results suggest that over all soils analyzed (including calcareous soils), the M3 extractant is more comparable with the Olsen extractant than with the B1 extractant. What would be the critical concentrations for these extractants? The critical concentration ranges determined by using the Cate-Nelson and linear-plateau models across all soils were 7 to 9 ppm for the Olsen extractant, 14 to 24 ppm for the M3 extractant, and 11 to 15 ppm for the B1 extractant. Which soil test would be better across calcareous and noncalcareous soils? The Olsen and the M3 soil P extractants are more reliable diagnostic tools than the B1 extractant for estimating available P on Iowa soils having different soil pH and CCE. The B1 extractant often underestimated plant-available P in soils with pH equal to or greater than 7.4. RecommendationThe Olsen and the M3 extractants are more reliable than the B1 extractant for estimating plant-available P of soils varying in soil pH and calcium carbonate content. The Olsen or the M3 extractants would be better choices than the B1 extractant if a single soil-test for P is used for routine analysis across many Iowa soils.