Submitted to: Soil Science Society of America Journal
Publication Type: Peer reviewed journal
Publication Acceptance Date: 12/11/2002
Publication Date: 8/16/2003
Citation: Tomasella, J., Pachepsky, Y.A., Crestana, S., Rawls, W.J. 2003. Comparision of two approximation techniques to develop pedotransfer functions for water retention of Brazilian soils. Soil Science Society of America Journal. 67:1085-1092. Interpretive Summary: The need in soil hydraulic parameters on a regional scale, where detailed measurements are impractical, has driven development of a large number of pedotransfer functions (PTF) to estimate soil hydraulic properties from soil basic information available from soil surveys. Soil hydraulic properties are characterized by dependencies of soil ability to retain and transmit water on soil suction. These dependencies can be reported either as tables or as equations. Large database of soil hydraulic properties in Brazil allowed us to set an objective of this work as comparing accuracy of PTF for tabulated water retention and PTF for water retention equation, both for further use in modern hydrologic models. Flexible neuron-net-type algorithms were used to develop PTF of both types. Results show that estimating tabulated water retention provides substantially better results for Brazilian soils. This work presents first comprehensive PTF for water retention of tropical soils that can be used in many hydrologic applications in tropical regions.
Technical Abstract: The need in soil hydraulic parameters on a regional scale, where detailed measurements are impractical, has driven development of a large number of pedotransfer functions (PTF) to estimate soil hydraulic properties from soil basic information available from soil surveys. Soil water retention is measured at selected soil water potentials. An analytical expression of the eretention curve, however, is needed for applications in environmental modeling. To-date, two approaches has been used for obtaining such analytical expression with PTFs: i) soil basic information is used to estimate soil water retention at specific soil water potentials; and then an analytical expression of the retention curve is fitted to the estimated soil moisture; ii) soil basic information is used for direct estimation of coefficients in the analytical expression of water retention curves. Our objective was to compare the two techniques using Brazilian soil data. The database consisted of 839 samples representing main five geographical regions of Brazil. The Group Method of Data Handling has been used in developing PTFs both for coefficients in the van Genuchten equation and for water contents at -6, -10, -33, -100, and -1500 kPa. Coarse sand, fine sand, silt, clay, and organic matter content along with bulk density were used as the PTF inputs. Results show that the approach based on the estimation of water contents at specific water potentials with following fitting the van Genuchten equation to the estimates provides substantially better results in Brazilian soils.