Submitted to: International Soil Science Congress Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 11/25/2001
Publication Date: 8/14/2002
Technical Abstract: Soil hydraulic pedotransfer functions (PTFs) transfer simple-to-measure soil survey information into soil hydraulic characteristics, that are otherwise costly and cumbersome to measure. Soil texture is among the key inputs for most current PTFs. Most PTFs are developed using texture data compatible to the USDA/FAO particle size system.However, many laboratories use different particle size systems, and that often hinders the developmen and/or application of PTFs. The objective was to develop a reliable method to interpolate between particle size systems. A total of about 118.000 measured soil particle-size distribution (PSD) curves were used from the NRCS, BIS (The Netherlands) and HYPRES databases to test two methods for the interpolation of the PSD curve. The 'similarity procedure' does not rely on mathematical interpolation but involves searching in a sufficiently large external reference data set for a number of soils that have a particle-size distribution similar to the PSD of the soil in question. Similarity between soil PSDs was quantified based on correspondence of mass fractions at common particle sizes. Secondly, neural network (NN) models combined with the bootstrap method were developed to predict intermediate points on the PSD curve from measured PSD points. The accuracy of each method was tested on independent data. Both methods outperformed the log-linear interpolation and showed comparable prediction accuracy. The prediction error of each method varied by texture and by the particle diameter (size) that was predicted on each PSD curve. The proposed new techniques offer a solution for expanding applicability of pedotransfer functions in various regions of the world.