Effective hydraulic properties of layered soils at the lysimeter scale determined by inverse modelling

Abstract

The transferability of soil hydraulic properties measured in the laboratory to the field or catchment scale is a key problem in soil hydrology. To narrow the gap between laboratory and field scale we investigated soil water movement at the lysimeter scale. Specific questions of interest are the existence and uniqueness of effective hydraulic properties for lysimeters that are internally heterogeneous. To answer these questions, synthetic measurements of water contents, pressure heads, and fluxes across the system boundaries were generated by forward modelling of water flow, based on the Richards equation, for a variety of homogeneous and layered soils and under varying types of boundary conditions (multistep outflow, evaporation, transient atmospheric). We evaluated the measurements by inverse modelling assuming a homogeneous system. The globally convergent shuffled complex evolution algorithm was applied to avoid parameter estimation problems caused by a rough topology of the objective function. The hydraulic properties of homogeneous soils were always uniquely identified, not only for the classic multistep outflow and evaporation experiments but also under natural atmospheric boundary conditions. Moreover, for the weakly heterogeneous layered soils, the soil water dynamics could be well described with effective homogeneous properties. For highly heterogeneous, layered soils, it was still possible to match the boundary fluxes for all types of experiments. Internal system states could not be matched, however, and the properties identified depended on the type of experiment. For these soils no unique effective soil hydraulic properties exist. We found that for strongly layered soils the simultaneous determination of the hydraulic properties of multiple soil layers by inverse modelling is a practicable alternative to the description with effective parameters.

Does textural heterogeneity matter? Quantifying transformation of hydrological signals in soils

“…A large number of scientific papers deals with finding effective soil hydraulic properties (e.g. Bayer et al, 2005;Durner et al, 2008) and averaging vadose zone variables (e.g. De Lannoy et al, 2007;Schlüter et al, 2012a;Vogel et al, 2010).…”

Seasonal groundwater contribution to crop-water use assessed with lysimeter observations and model simulations

“…Capillary rise from groundwater and salt accumulation in the soils are difficult parameters to measure in the field. Lysimeters offer an option for simultaneously measuring these and other components of the water budget in a field-like situation Hermsmeyer et al, 2002;Soppe and Ayars, 2003;Kelleners et al, 2005;Durner et al, 2008). However, it is still difficult for the limited observations that are usually undertaken within limited time to quantify the whole spectrum of groundwater contribution to crop-water use under the influences of inter-seasonal rainfall variability.…”

Estimating hydraulic conductivity of internal drainage for layered soils in situ

“…The expression is modified by directly substituting analytical functions for K and reformulating data analysis in terms of the gravity-drainage optimisation code UNIGRA (Sisson and van Genuchten, 1992). This expression was extended to layered soils by scaling water content to assume equivalent homogeneous soil profiles (Shouse et al, 1991;Durner et al, 2008). By so doing, the K functions for the different layers are linearized into a single effective property, and the effect of spatial variability is minimized.…”