Determining Vertical Soil-Water Flux in Glaciated Terrains Using a Convective Heat Flux Model and Measured Transient Soil Thermal Properties

Abstract

Several workers have applied water flux models based on convective heat-transfer principles and measured temperature profiles to quantify vertical water movement through unsaturated soils. However, flux estimates can vary significantly due to uncertainty in soil thermal properties such as thermal diffusivity, thermal conductivity, and soil volumetric heat capacity and their relationship to changing degrees of saturation. In this study, soil temperature profile data is combined with in situ measurements of soil thermal properties to estimate both upward and downward water fluxes through soils developed in glacial parent materials. A regression analysis of 1.2-m discrete depth in situ volumetric heat capacity and degree-of-saturation data indicates that there is a good correlation for sites underlain by coarse-grained diamicton, outwash, and alluvium (R = 0.66 – 0.87), whereas sites underlain by fine-grained glacial diamicton show little to no correlation and are therefore deemed unfit for the modelling approach. The results of the convective heat model for the outwash and alluvium sites are compared to water fluxes previously simulated using the one-dimensional Richard’s Equation. Preliminary results at a midwestern U.S. glacial outwash site indicate that daily downward flux results compare well with the previously published estimates after eliminating extreme values caused by temperature signal noise. At the coarse-grained diamicton and alluvium sites having lower hydraulic conductivities, the fluxes generally are temporally correlated, but the thermal model tends to overestimate flux by an order of magnitude. This ongoing work aims to refine the convective flux model while optimizing vadose-zone monitoring networks such that real-time percolation estimates are possible.