The difficulty of finding conceptually correct yet practically feasible unsaturated hydraulic conductivity curves.

In recent years, models for the unsaturated hydraulic conductivity have emerged that separate the soil hydraulic conductivity to constituting conductivities of water held in soil pores by capillary forces and of water adsorbed in films on the surface of soil grains. Some models include an equivalent hydraulic conductivity that accounts for diffusive movement of water vapor. The bulk soil hydraulic conductivity for a particular matric potential is the sum of these constituting conductivities.

The additivity attribute of the constituting conductivities relies on implicit assumptions regarding the configuration of the three domains: capillary water, water adsorbed in films, and the gas phase. These assumptions are not met in natural soils, and so alternatives to straight-forward addition are examined.

This examination unfortunately shows that any type of averaging of the constituting hydraulic conductivities leads to a non-monotonic hydraulic conductivity curve if the capillary water content reaches zero at a distinct matric potential above oven dryness. In fact, a correct expression for the hydraulic conductivity based on physically sound configurations of the domains can be shown to be realistically unattainable, leaving us without many alternatives for the additive model.

This being the case, an additive conductivity model and one alternative that is also unconditionally monotonic are introduced for a recently described parameterization of the soil water sigmoidal retention curve with a distinct air-entry value and a logarithmic dry branch terminating at a finite matric potential.