ABSTRACT The optimal soil pH for most annual crops in Brazil varies between 5.7 and 6.0. Numerous methods have been developed for estimating lime requirement (LR), but they vary widely in their predictions and fail to raise pH to desired values for optimum crop production in the highly weathered soils of Brazil. The objectives of this study were to (i) compare seven traditional methods for estimating LR in Brazilian soils; (ii) assess the effects of LR predicted by these methods on soil-acidity related properties, and (iii) determine if these methods are predicting LR to attain target pH values of 5.8 and 6.0, which are within the pH range recommended to optimize crop yields. The traditional LR methods evaluated in this study are based on the following criteria: exchangeable acidity (EA), base saturation (BSAT), exchangeable acidity along with Ca2+ and Mg2+ as proposed by the 4th (MG4A) and 5th (MG5A) Approximations to the Minas Gerais State, SMP soil-buffer pH (SMP), potential acidity (PA), and soil pH along with organic matter (pHOM). These methods were compared with the standard incubation method using correlation-regression analysis and, alternatively, the identity test designed for assessing equivalence between methods. Representative agricultural soils (n = 22) were incubated for 60 days with incremental amounts of lime determined by the tested methods. On average, LR predictions differed among methods, and increased in the following order: EA < BSAT ≈ MG5A ≤ MG4A ≈ SMP ≤ PA < pHOM. Suitable changes in soil pH, exchangeable acidity, potential acidity, base saturation, and Ca2+ and Mg2+ were achieved upon application of LR estimated by all methods except the EA and pHMO, which resulted in undesirable soil acidity characteristics. All methods evaluated in this study were unable to predict LR for attaining target pH values of 5.8 and 6.0 as revealed by the identity test, even though they were moderate to strongly correlated with the standard incubation method as indicated by the correlation-regression analysis. Further research should focus on the development of reliable methods for predicting LR to attain desired pH values and consequently maximize crop production on Brazilian soils.

ABSTRACT Increasingly rational use of phosphate fertilizers by agriculture is important, especially in Brazil, due to its importance for global food security and its high dependence on phosphate fertilizers for crop production. Thus, correlation and calibration researches with soil available phosphorus (P) extractants to improve the recommendation and the use of P are extremely important. Our objectives were (i) to determine soil P recovery rates and critical levels by the extractants Mehlich-1 (M1), Mehlich-3 (M3), and Ion Exchange Resin (RTI); (ii) to adjust predictive models of P recovery rates and critical levels for each extractant, as a function of soil properties related to the soil P buffer capacity, and (iii) to adjust the an interpretation table of soil P availability by M3 taking into account soil P buffer capacity indices. A pot experiment under greenhouse condition was conducted, in which the treatments were generated by the combination of a factorial 12 × 6, being 12 soil samples collected at a layer of 0.00-0.20 m, and six doses of P. The experimental units were constituted by plastic pots with capacity for 2.0 dm3 of soil, in which four hybrid corn plants (Zea mays L.) were cultivated for 45 d. The experimental design was a randomized complete block with four replicates. After cultivation, soil samples were collected from each experimental unit, and the determination of soil available P was performed by the extractants. The remaining P and the clay content were the soil properties that best correlated with the P recovery rates and P critical levels in the soil by the M1 and M3. The IER did not show sensitivity to soil P buffer capacity. The M3 had a discontinuous loss of extraction capacity with increasing soil P buffer capacity. Therefore, it is recommended the use of M3 as P extractant in soils with different characteristics, but a measure of soil P buffer capacity such as remaining P or clay content have to be used.