Estimation of Saturated Volumetric Moisture of Brazilian Soils for Sustainable Water Management in Agriculture

Abstract

Pedotransfer functions are used to estimate soil hydraulic characteristics, which are laborious and involve high costs, especially in a country of continental dimensions like Brazil. Since the determination of saturated volumetric humidity helps with the sustainability of the water used in agriculture, avoiding waste, the present work aspired to develop a new pedrotransfer function to predict said property for Brazilian soils using HYBRAS data. A total of 1053 soil samples from the North, Northeast, Midwest, Southeast and South of the country were considered, for which the database informs the hydraulic parameters of each sample collected. Using the Multiple Linear Regression method, the function was developed from the value of the density measured in loco and the fractions of sand and clay, having been validated through the comparison of the results with those obtained by models available in the literature. When compared to existent pedotransfer functions, the new function obtained a sufficient behavior, presenting values close to the ideal ones for the following parameters: Mean Square Error, Model Efficiency, Residual Mass Coefficient and Deviation Ratio. Thus, it can be applied to the different Brazilian soils present in the database used.