Water retention in Australian soils. 2.* Prediction using particle size, bulk density, and other properties

Abstract

Different approaches were investigated for estimating the parameters in the Campbell soil water characteristic (SWC) equation from soil attributes such as particle size distribution (PSD), bulk density, and organic matter content. Predicted soil water characteristics were compared with measured values for soils of the wheatbelt of south-eastern Australia. A method of prediction is proposed incorporating an empirical relationship for estimating the slope of the SWC from the slope of the cumulative PSD. A power-law form is assumed for both the SWC and PSD functions. One measured SWC point is then used to locate and thus define the SWC curve. When SWC points predicted with this 'one-point' method were compared with measured values, the mean absolute value of the difference between each measured and predicted SWC point was 0.016 m³/m³ for the Geeves data and 0.027 m³/m³ for the Forrest data. Eight sets of predictive equations, previously developed using multiple regression analysis, were also evaluated. Whilst the equations predicted the slope of the SWC curves reasonably well, predictions of the air entry potential were poor. Although less accurate, the equations developed by multiple regression are less demanding in data requirement compared with alternative SWC prediction methods.

The one-point method gave better predictions than the multiple regression approach but was less accurate than the 'two-point' method proposed in the first paper in this series. The one-point method should be considered where PSD data and 1 measured SWC point are available. In most other circumstances it will be more accurate and cost-effective to measure 2 SWC points to define the soil water characteristic function (the two-point method).

* Part I, Aust. J. Soil Res., 1996, 34, 195–212.