Phosphorus cycling from residual pools underlying efficient P fertilization in rotation systems in tropical agriculture

Across the Brazilian Cerrado, the land area under soybean-maize (double cropping) and maize-brachiaria (intercropping) systems has been expanding. In this study, we evaluated the efficiency of P fertilization and the response of residual P compartments to i) soybean monoculture in conventional (CT) or no-tillage (NT), ii) soybean in rotation systems including the maize+brachiaria consortium; and iii) soybean in succession systems including soybean-maize double cropping. These factors were combined into eight treatments in an experiment conducted at Itiquira, Mato Grosso, Brazil. Our study was laid out following a completely randomized block design with four replicates. We determined the apparent efficiency of P fertilization using the balance method for 12 years, after which residual P was evaluated in soil samples submitted to a soil P fractionation scheme using CaCl₂, Mehlich-3, NaOH (inorganic-Pi and organic-Po) and HCl. P occluded was estimated as the difference between total soil P and the sum of the extractable P fractions. Our data showed apparent efficiency of P fertilization about 78.3 % under soybean-fallow (CT or NT) and about 93.0–94 % under the soybean-successions. Under the rotations, the apparent efficiency of P fertilizations was about 100–123.0 % coupled to some depletion of P-Mehlich-3 and Pi-NaOH. We found positive correlations between P-Mehlich-3 and Pi-NaOH, whereas both Po-NaOH and P-Mehlich-3 showed strong negative correlations with P occluded. Overall, the rotation systems evaluated in this research appear to benefit from the transfer of P between P-Mehlich-3 and Pi-NaOH pools, coupled to the formation of Po-NaOH limiting the accumulation of P occluded. These characteristics allow reconciling productive crop grain systems with efficient P fertilizations and recycling of P in soils with high residual P pools in tropical agroecosystems.