Modelling water fluxes in a soil profile of a degraded peatland site

Reducing greenhouse gas emissions from degraded, agriculturally used peatlands is a vital contribution to meeting the German climate protection targets by 2050. It requires an understanding their closely coupled hydraulic and geochemical processes, which is the basis of sustainable land and water management on these sites. Hydraulic modelling of the fluxes in a soil profile was done to characterise the hydraulic state of a degraded peatland site with spatially and temporally high resolution. The basis of the model were measurements from groundwater lysimeters in a site with three horizons in Spreewald wetland, Germany. The model was implemented in the one-dimensional hydraulic modelling software Hydrus-1D. The first step was the determination of initial soil hydraulic properties using soil physical properties and the ROSETTA tool based on pedotransfer functions, which is integrated in Hydrus-1D. Two model variants were set up that differed in their lower boundary condition – either the measured pressure head (representing groundwater level) or the measured flux at the lower boundary of the lysimeter. The modelled volumetric water contents, pressure heads and groundwater table (variant 1) or fluxes at lower boundary (variant 2) were validated with the measured lysimeter data. In a second step, the soil hydraulic parameters were inversely optimised based on measured time series data, for both variants. To further improve the model results, dual porosity type flow was implemented in the upper two horizons. The different steps were able to continuously improve the model. The choice of lower boundary condition had an effect on the quality of the model results: The use of groundwater table as lower boundary condition improved the modelled volumetric water contents and pressure heads, but yielded deviating fluxes at the lower boundary in comparison to the measurements. The application of flux as a lower boundary condition produced deviations in the modelled groundwater table, the water contents and the pressure heads, especially after heavy rainfall events. The integration of preferential flow (dual porosity) into the model improved the vadose zone pressure head and water content results significantly.