Acessibilidade / Reportar erro

Use of buffer methods to estimate the potential acidity of Mato Grosso soils

Uso de métodos tampões para estimar a acidez potencial de solos de Mato Grosso

ABSTRACT

Mato Grosso, with a total area of 903357 km², does not have an official methodology for estimating soil potential acidity (H + Al), and determination of H + Al using the standard method is onerous and time consuming. The objective of this study was to compare estimated values of H + Al determined using the standard calcium acetate method with those obtained using three buffer methods, namely, the Shoemaker, McLean, and Pratt (SMP) buffer, Sikora buffer, and Santa Maria buffer (SMB) methods, with samples of the main classes of cultivated soils in the state of Mato Grosso. A total of 196 soil samples were collected from the arable layer (0-20 cm) in agropastoral and adjacent native systems. Statistical models were obtained and compared with models used by laboratories that are hypothetically inadequate because there is no calibration for soils in the state. After laboratory analyses, the paired H + Al and equilibrium pH values corresponding to the equilibrium of each buffer solution (SMP buffer, SMB, and Sikora buffer) were subjected to nonlinear regression analysis (P < 0.05). The SMB method, which does not release pollutant residues into the environment, was better than the Sikora and SMP methods for replacing the standard method used in state laboratories for soil analysis, that is, H + Al (cm3 c dm-3) = 51.189 -25.70 ln(pHSMB) (R2 = 0.88, P < 0.0001). Thus, if laboratories use uncalibrated equations to estimate soil potential acidity, the recommended limestone correction will be underestimated or overestimated, which may compromise crop productivity in Mato Grosso.

Index terms:
Limestone; SMP buffer; Sikora buffer; Santa Maria buffer

Editora da Universidade Federal de Lavras Editora da UFLA, Caixa Postal 3037 - 37200-900 - Lavras - MG - Brasil, Telefone: 35 3829-1115 - Lavras - MG - Brazil
E-mail: revista.ca.editora@ufla.br