The reliability of soil analysis depends on the correlation of nutrient concentration in the soil obtained by a given method with the plant nutrient uptake. The objective of this study was to evaluate the correlations between phosphorus content estimated by different methods and phosphorus quantity absorbed by rice plants in some flooded soils from Rio Grande do Sul state (RS), Brazil. Soil samples from 16 locations in RS, largely different in soil chemical and physical properties, were collected from the top layer (0-20 cm). Each soil was treated with three P levels: P0 (without P); P1 (½ P2) and P2 (P dose sufficient to obtain 0.2 mg L-1 of P in soil solution, based on the maximum phosphorus adsorption capacity of each soil). The pots used to cultivate rice plants in a greenhouse were arranged in a completely randomized design, with three replications. Rice shoots were cut 45 days after emergence to evaluate P uptake. The soil phosphorus concentration was evaluated by the following methods: Mehlich-1, Mehlich-2, Mehlich-3, anion exchange resin in sheets and spheres (AER), Bray-1, Bray-2, Olsen, Morgan, Calcium Lactate, Texas, Oxalate, EDTA, H2SO4, and NaOH. The soil P concentration by the distinct extractors were compared with the P absorbed by plants through linear correlation analysis. Among these methods, H2SO4, Texas, Calcium Lactate, EDTA, Olsen and Resin were the best, but their efficacy was similar to that of Mehlich-1.
calibration; correlation; methods; soil analysis
