Resumo O objetivo deste trabalho foi avaliar a acurácia das regressões que definem os limites do intervalo hídrico ótimo e a validade do uso deste indicador ao se considerar a variabilidade espacial de lavouras sob sistema plantio direto. Foram coletadas 100 amostras de solo com estrutura preservada, em áreas com dimensões de 10x10 m, em Latossolo Vermelho distrófico, Nitossolo Vermelho distroférrico e Latossolo Vermelho aluminoférrico, manejados sob sistema plantio direto. Determinaram-se porosidade de aeração (0,10 m3 m-3), capacidade de campo (potencial de -0,006 MPa), ponto de murcha permanente (potencial de -1,5 MPa) e resistência mecânica do solo à penetração crítica (2 MPa). As regressões que descreveram a capacidade de campo tiveram os menores coeficientes de determinação, enquanto as que descreveram a porosidade de aeração, o ponto de murcha permanente e a resistência mecânica do solo à penetração apresentaram os maiores valores. As regressões que definem o limite da capacidade de campo não são acuradas em descrever a variação na umidade volumétrica do solo por meio da densidade do solo, e os solos avaliados apresentam elevada variabilidade espacial para o intervalo hídrico ótimo.

Abstract The objective of this work was to evaluate the regressions that define the least limiting water range limits and the validity of using this indicator considering the spatial variability of crops under a no-tillage system. One hundred undisturbed soil samples were collected in areas with dimensions of 10x10 m, in two different Oxisols and an Ultisol managed under no-tillage. Air-filled porosity (0.10 m3 m-3), field capacity (potential of -0.006 MPa), permanent wilting point (potential of -1.5 MPa), and soil mechanical resistance (2 MPa) were determined. The regressions that described field capacity had the lowest coefficients of determination, whereas those that described air-filled porosity, permanent wilting point, and soil mechanical resistance to penetration showed the highest values. Regression adjustments that define field capacity limits are not accurate in explaining variation on volumetric soil water content through bulk density, and the evaluated soils show high spatial variability regarding the least limiting water range.

ABSTRACT Inceptsoils have high aluminum contents, and amendments are required to ensure a satisfactory crop development. Liming is efficient in neutralizing Al3+, but when applied to the topsoil its action is restricted to the surface layers, and sometimes lime incorporation into the soil is recommendable. However, tillage may negatively alter physical soil properties. Thus, gypsum could be an alternative to increase Ca2+ levels and reduce Al saturation in deeper layers, without requiring tillage. An experiment was initiated in 2010, to assess chemical and physical soil properties. Soil samples were collected in February 2013. A random block design with sub-subplots and two replications was used. Lime was applied to the surface of the main plot (0, 8 and 16 Mg ha-1) and incorporated into the soil by plowing and light harrowing. Gypsum was applied to the subplot (0 and 6 Mg ha-1), and the sub-subplots corresponded to the sampled layers (thickness of 0.05 m, to a depth of 0.25 m). The following parameters were evaluated: pH in water, base saturation, Al saturation, Ca2+, Mg2+, soil organic matter, water retention, and least limiting water range (LLWR). Lime incorporated into the soil reduced Al3+ and increased pH and Ca2+ levels, especially in the deeper soil layers, although pH levels did not exceed 5.1 and base saturation was lower than 60 %, still inadequate for crops. The gypsum rate of 6 Mg ha-1 did not decrease Al saturation, but reduced Mg2+ levels in top soil layers. Soil water retention and LLWR were not affected by plowing and harrowing within a period of three years. Gypsum rates below 6 Mg ha-1 were not effective in reducing Al saturation and increasing Ca2+ levels. Incorporated lime is more effective in correcting the acidity of the soil profile and, after three years, soil water retention and LLWR were similar under both liming methods.