Effectiveness of using pedo-transfer functions to quantify the spatial variability of soil water retention characteristics.

Accurate knowledge of soil hydraulic properties is of crucial importance for reliable applications of recently developed distributed models to environmental studies and land-use planning. To provide such information in a cost-effective way, indirect estimation of water transport parameters from easily measurable or already available soil data using pedo-transfer functions (PTFs) is becoming increasingly popular. However, distributed hydrological modeling requires that soil hydraulic characterization also takes account of the description of spatial variability. The objective of this study was to evaluate some published PTFs in the light of their ability to quantify the spatial structure and variability of soil water retention adequately. Four PTFs were tested: two provided only values of water content at specific pressure potentials (PTF Group A), whereas the remaining two estimated the parameters of closed-form relations describing the water retention function (PTF Group B). Measured data for testing were obtained from undisturbed soil samples taken from the top layer of different soils along a 5 km transect with constant spacing of 50 m. Overall, summary statistics and sample distributions of the PTF-estimated retention characteristics at selected pressure potentials are close to those of the retention variables used as reference for comparisons. The largest discrepancies are related to the use of PTFs pertaining to Group A. Although the quality of kriged interpolations based on soil property data obtained by simplified methodologies still gives cause for concern, results show that the structure of spatial variability exhibited by the variables considered along the study transect is described well enough when using PTFs for determining soil water retention characteristics.