Funções de pedotransferência para solos do sul do Uruguai

Abstract

The continuous search for information that will assist agricultural practices is today relevant, being the primary use of methodologies to facilitate the determination of relevant parameters in the agricultural productivity. Alternative methods for determining the physical and hydro soils can auxiliary in the monitoring of agricultural activity. Among these, the use stands of pedotransfer functions (FPTs), among other properties, provide an indirect measure of water retention of soil, reducing cost, time and manpower. This study aims to use multiple linear regression models to determine the FPTs for the soils of southern Uruguay to the practical use at laboratories in soil physics, to obtain the curve of retention. For the realization of the work, used 251 samples of superficial, intermediate and inferior soils of southern Uruguay, available in the database of physical analysis of the INIA (Instituto Nacional de Investigación Agropecuaria). This database consists of information of soil texture (percentage sand, silt and clay), bulk density, porosity of the soil, besides the information of water content at potentials of 0, -3, -30, -100, -500 and -1500 kPa. The organization of the database was organized randomly, which is subdivided into two banks, one with 60% of the data (development of FPTs) and the other with 40% of the data (validation of FPTs). The data used to develop the FPTs were included in the Statistical Analysis System (SAS), in its method of determining equations of multiple linear regression (stepwise). Being generated FPTs to estimate water retention at potentials of -30, -100, -500 and -1500 kPa for the superficial layers (0-20 cm), intermediate (20-40 cm) and bottom (40-60 cm ) of the soil profile. The equations presented in its constitution datas textural and structural soil and observed that the statistical indicators mean error and root mean square error squared values were near zero, indicating good accuracy in the determination of residual moisture through the FPTs. In general can say that the attribute structure, water content remaining in the soil after application potential of -3 kPa, was significant in all the FPTs developed. We conclude that the FPTs developed to estimate the water content remaining in the soil at potentials of -30, - 100, -500 and -1500 kPa were good, could be used as an alternative in the construction of water retention curve in soils that have characteristics that fall under the requirements of the FPTs developed for the soils of southern Uruguay.