The difficult-to-measure unsaturated hydraulic properties generally can be theoretically estimated only as rough approximations, for example by methods developed by Brooks and Corey (1966), Mualem (1976), van Genuchten (1980), and Arya and Paris (1981). These can extend the utilization of hydraulic property information from sparse or indirectly related data sources. Such methods either (1) predict hydraulic properties from easy-to-measure properties, for example hydraulic conductivity functions from particle-size distributions, or (2) predict a full range of the desired properties from a small number of measured data. One example of the latter would be to compute a complete water retention curve, over a wide moisture range, from two or three measured points on the curve. One development of this type is the water retention relation of Rossi and Nimmo (1994), described in connection with low water contents, which covers the entire range of water contents from saturation to oven-dryness in a realistic way. Other developments by our project, described below, relate to simple ways of representing the changes in hydraulic properties that occur with temperature, and with the direction of change (either drying or wetting).
For systematically predicting and correcting temperature effects on soil-water retention, the gain-factor model of Nimmo and Miller (1986) has become widely used. The gain-factor model is a practical quantification of temperature dependence of the soil water system, expressed in terms of the factor by which this temperature dependence exceeds that of the surface tension of pure water.
Unsaturated hydraulic properties are hysteretic, that is, their measured values are quantitatively different for soil becoming drier than for soil when the soil becoming wetter. Hysteresis is a serious practical problem when soils that are subject to change by both wetting and drying influences, a common occurence in the field. In principle, a huge amount of property data would be necessary to characterize the unsaturated properties at many possible moisture histories. Various hysteresis models, including that published by Nimmo (1992), address this problem by enabling prediction of hydraulic properties that have not been measured, as a function of other, related data. These models serve both as a practical advance, decreasing the data requirement for working with unsaturated-zone hysteresis, and as a needed tools for structural studies, in that it gives parameter values directly related to pore size and shape. The model of Nimmo (1992) has served as a stepping stone to other advances, for example in relating water retention to other properties (Nimmo, 1997b; Nimmo, 1998).
Another development of this type is our analysis using and extending the Arya-Paris (1981) model for the prediction of retention curves from particle-size and porosity data. Deason (1997) and Nimmo and others (2001) have developed means of predicting approximate unsaturated hydraulic properties from data that are qualitative and related only indirectly, for example the verbal descriptions of subsurface materials in driller's logs.
| National Research Program | USGS | Water Resources | UZ Flow Homepage |