Avaliação da qualidade física de solos irrigados do Rio Grande do Sul e de funções de pedotransferência para a estimativa da retenção de água

Abstract

Knowledge of the physical and hydric characteristics of the soil is essential for proper management, aimed at conciliate good yields and soil quality. In agriculture, crop productivity tends to be maximized and environmental degradation, and soil tends to a minimum, while the physical quality of the soils is within a range of values considered ideal. There is no simple way to evaluate the physical quality of the soil, and, creating indexes based on some soil properties (density, porosity, retention curve, water infiltration, etc.) may assist in this evaluation. However, many of these indicators interact with each other, hindering its use in a generalized way in the evaluation of soil quality. Many of the indicators used are directly or indirectly related to the storage and availability of water, aeration, soil bulk density, macroporosity, organic carbon, etc., these parameter settings are determined in the laboratory, which is getting expensive and impractical for large-scale application. An alternative to the direct measurement is the use of pedotransfer functions (PFT's), which can estimate previously measured parameters, as well as predicting these same parameters for other locations. The objectives of this study were: (a) develop consistent indexes to evaluate physical properties of irrigated soils of the Rio Grande do Sul State; (b) generate and validate PFT's to estimate and water retention; (c) test the literature PFT's accuracy to estimate the water retention at field capacity (FC) and permanent wilting point (PWP) for database soils. A set of 3288 soil samples from three deeps (surface, intermediate and lower), belonging to the Irriga System database was used to classify the physical quality of the soil by macroporosity and soil clay content. The data set was fractionated into two sets for calibration and validation of FPT's through statistical indicators. Of the total samples evaluated, 39% had some indication of soil compaction. PFT's generated demonstrated efficacy in estimating the water content of the soil in the potential of -1, -6, -33 and -100 kPa, from the texture, bulk density and particle porosity and, most FTP's of literature tested did not show good capacity to estimate the water content in FC and PWP. Featured for FPT proposed by Tomasella and Hodnett (1998), in which it was obtained some ability to predict the water retention in sandy soils in potential CC and PMP.