Assessing Organic Phosphorus Contributions for Predicting Soybean Response to Fertilization

Abstract

The use of organic P fraction in soil test P for soybean [Glycine max (L.) Merr.] under no-till could improve the accuracy to predict crops response to P fertilization. The research objectives were to assess the contribution and accuracy of P indices that include organic-P fractions to predict soybean yield response to P fertilization in comparison with Bray soil test P (BP). The study included P fertilization experiments conducted in the Pampas Region of Argentina during three growing seasons. We selected sites considering Cate and Nelson quadrants (I, II, III), established by a critical BP concentration of 9 mg P kg−1 and a relative soybean yield (RY) of 85%. Bray P, bioavailable (BioP), particulate (POM-P), and mineral-associated (MA-P) soil P fractions were determined in soil samples from the 0- to 5-cm and 0- to 20-cm depths. Bray P varied from 4.8 to 31 mg P kg−1 and BioP from 8 to 29 mg P kg−1 at the 0- to 20-cm depth. Phosphorus content in POM ranged from 19 to 171 mg P kg−1, and MA-P ranged from 208 to 446 mg P kg−1. Bioavailable P and MA-P correlated with RY, whereas POM-P did not. Bioavailable P performed similarly to BP test to predict soybean yield response to P fertilization. Phosphorus content in MA fraction reduced Cate–Nelson classification errors, improving the accuracy to predict soybean yield response to applied P in comparison with BP test. Sampling at the 0- to 5-cm depth did not improve soil test P performance compared with the 0- to 20-cm depth.