Groundwater recharge and capillary rise in a clayey till catchment

Quantification of recharge rates and patterns is essential for sustainable groundwater abstraction. On northern mid- to high latitudes, the common weather is humid and a common geological unit is clay till. Mapping of the tortuous flow through clay till has an unrealistically high cost and models integrating the processes in the surface-groundwater contact have only been developed over the past decade. One traditionally disregarded process is the capillary rise from groundwater in response to surface drying. Diversion via the artificial subsurface drainage can be presented by a linear reservoir, and this contribution dominates the primary calibration target for hydrological models: streamflow. The discharge from deep aquifers to streams is approximately linear and can accordingly be modeled by a number of linear reservoirs. The tempting conclusion is that groundwater recharge is equally linear, so the capillary rise should not be present at all. Moreover, many have speculated whether removal of the artificial subsurface drainage would enhance the aquifer replenishment and thereby the level of sustainable groundwater abstraction. A literature survey of methods for determination of groundwater recharge clarified that usually only the percolation at some level above the watertable is considered, and the subsequent capillary rise is neglected. Monitoring the very watertable by a number of wells captures the lag phase of recharge, its spatial distribution, and allows for model interpretation of also the capillary rise. Though the number of wells was limited to eleven, they covered twothird of the 16 km2 study catchment in terms of terrain level. Adopting the variable source concept, they sufficed for a monthly distinction between the dry and wet state in six terrain intervals. From the wet stat area, the uniformly distributed subsurface drainage diverts recharge to the stream. A sequential parameterization of existing modeling tools was unfolded with the assistance of electrical sounding, a minor equilibrium model for determination of the proper boundary condition between the unsaturated zone and groundwater, and tools for mathematical inversion. This study finds that watertable monitoring render calibration of a groundwater model’s surface contact superfluous apart from drain time; vertical conductivity of weathered soil, and snow storage. We have found a first guess of the drain time in 1D by streamflow separation, and calibrated in 3D by MIKE SHE modeling. Preferential conduits are merged into the soil by 2D modeling, and snow storage parameterised by comparison of an automatic and a manual precipitation gauge. The capillary rise accounts for an uphill increase in the long-term net recharge from 200 to 260 mm/y, implying that local topography is mirrored into the aquifer circulation faster than found by regional modeling without this feedback process. Perhaps the capillary rise ties annual evapotranspiration to the North Atlantic Oscillation better than thought of for Denmark, implying that hydrological modeling of clayey catchments can establish area integrated values, which may contribute to the debate on the oscillation’s spatial coverage. Moreover, subsurface drainage reduces evaporation from waterlogged areas by an increase in streamflow that is one magnitude larger than the 2 mm/y decrease in aquifer replenishment.