Development and analysis of a dual-permeability model for subsurface stormflow and solute transport in a forested hillslope

Preferential flowpaths related to the soil structure have a decisive influence on subsurface flow and transport processes in forest soils in the boreal region. In dual-permeability models, the soil pore space is divided into a slow flow matrix domain and a preferential flow domain that are connected by a transfer term. The main objective of this study was to develop an advanced, physics-based, three-dimensional dual-permeability model that captures the main mechanisms behind a subsurface stormflow and solute transport event in a forested hillslope. A variety of experimental data were collected to describe the soil properties, flowpaths and processes in Finnish conditions, and to support the model development, parameterisation and analysis. The collected dataset consists of soil analyses, field measurements and tracer experiments. The model developed was used to simulate tracer migration in the study slope during an artificially generated stormflow event. As many parameters as possible were a priori fixed to reduce problems with equifinality and model identifiability. Parameterisation of the soil matrix was fixed relying on the fact that fast subsurface stormflow is related to the preferential flow routes and is not sensitive to the slow matrix processes. Similarly, water retention properties were a priori fixed in the preferential flow domain. With respect to capturing the dynamics of the tracer plume, the model was sensitive to the remaining parameters, which include the saturated hydraulic conductivity of the preferential flow domain, the fractioning of the total porosity to soil matrix and preferential routes, and the transfer coefficient between the two pore domains. Identification of these parameters was based on the tracer experiment data. The parameterisation routine used proved to be applicable to the development of the dual-permeability model, and this study emphasized the importance of the tracer dataset for the development and analysis of the model. The parallel and coupled simulation of the soil matrix and the preferential flow domain with the dual-permeability model was found to be essential in reproducing the observed, dynamic changes in the moisture conditions, flow velocities and tracer concentrations during the different stages of the event. The dual-permeability model was able to capture the main features of the observed stormflow and solute transport event. A further development and evaluation of the model call for field data in dry conditions and advances in the measurement and parameterisation of the water retention properties and porosity values.