Dynamics of residual phosphorus forms under different tillage systems in a Brazilian Oxisol

•The dynamics of residual P forms in soil after 17 years of fertilization was studied.•After suspending fertilization, the available inorganic P fractions supplied this nutrient to the plants.•The no-till system presented higher capacity to replace labile P forms in the soil.•Under the no-till the soil superficial layers supplied residual P to the plants. Phosphorus (P) accumulation in soil, mainly from decades of agricultural use with frequent additions of phosphate fertilizers, results in a large reserve of this nutrient that can be exploited by farmers. However, the dynamics of soil P forms under the residual effect of phosphate fertilization is still little known. In an experiment located at Embrapa Cerrados, in Planaltina, DF, Brazil, corn or soybeans were grown during the summer for 17 years, receiving 35 kg of P ha−1 year−1. The same area was then cultivated with corn for 4 more years without the supply of this nutrient, in order to characterize the behavior of P fractions accumulated in the soil and identify those that contribute most to crop nutrition when phosphate fertilization is interrupted. Changes in total P (Pt), organic P (Po), inorganic P (Pi), and labile P extracted by the Bray-1, Mehlich-1 and Olsen methods (organic and inorganic) and organic carbon (C) were evaluated by means of samples obtained before and after cultivation without P fertilization in different soil layers subject to no-till (NT) or conventional tillage (CT), previously fertilized with reactive rock phosphate (RRP) or triple superphosphate (TSP) which were applied by broadcast spreading or band-applied to the planting row. The superficial layers were most responsible for crop nutrition in NT, especially in the mode of broadcast application, while in CT the relative contribution of each layer was less stratified. There was a drastic reduction in the stock of almost all P fractions, except for Po which showed an increase, notably in soil cultivated under CT (22% in the 0–10 cm layer), although the C content was reduced. This suggests that the little available inorganic fractions, partially replacing the labile fractions, were responsible for the supply of P to crops. This replacement was more effective in NT, since it could sustain greater P offtake by crops, despite the similar reductions in labile P fractions observed in both systems.