ARS | Publication request: SCALING IN SOIL STRUCTURE AND PEDOTRANSFER FUNCTIONS

Submitted to: Soil Structure Water and Solute Transport
Publication Type: Abstract Only
Publication Acceptance Date: September 30, 2001
Publication Date: October 8, 2001

Technical Abstract: Soil hydraulic properties have to be estimated to carry out large-scale projects in agronomy, hydrology, and remote sensing. Estimations are made using pedotransfer functions (PTFs), i.e., regression equations to relate hydraulic parameters to basic soil properties available from soil survey. Presently, pedotransfer functions are developed with input data taken from laboratory soil analyses of 5 cm laboratory samples. Soil information from scales other than the laboratory, i. g., from maps and field soil descriptions, has to be used in large area estimates. The objective of this work was to present examples of using soil structural parameters along with texture to relate hydraulic and basic soil properties across scales. Observations of differences in "field-lab" water retention are presented to show that fractal scaling of the bulk density can explain these differences. PTFs built from the laboratory water retention data can overestimate available water content, and underestimate saturated hydrauli conductivity and sorptivity values. At coarser scales, augmenting textural class with data on grade of soil textural units provided a significant improvement in PTF accuracy. As scales become yet coarser, only the predominant textural class is often available for estimation of soil hydraulic properties. To incorporate structure of soil cover in PTFs, we defined topotextural groups that have an advantage of lesser demand to the accuracy of soil texture. Soil structural properties provide an important information to improve estimation of soil hydraulic properties at various scales, and to define scale-related corrections for such estimates.