Effects of terrain slope on long-term and seasonal water balances in clayey, subsurface drained agricultural fields in high latitude conditions

• 3D FLUSH supports year-round water balance simulation in undulating field area.
• Most outflow from the clay fields occurred through subsurface pathways.
• Slope increased groundwater outflow and decreased drain discharge.
• Runoff events triggered mostly by saturation excess mechanism.
• Frost hydrological impacts in spring may be occasionally high but typically low.

Water outflow pathways affect environmental loads from agricultural fields, but the pathways and effects of terrain topography on their proportions and on drainage design are not known in detail. In this study, a long-term hydrological dataset and 3D FLUSH model were applied to a field-scale assessment of multi-yearly and seasonal water balance in two adjacent clayey subsurface drained agricultural fields with different slopes (1% and 5%). The model was calibrated and then run with an hourly time step of input data throughout five studied years, and it was able to reproduce the measured water balance components. The results suggested that macropore flow had an essential role in the field-scale hydrological processes in clayey agricultural fields. The model provided a quantification how terrain slope increased the amount of groundwater outflow and correspondingly decreased the amount of drain discharge. The implication is that the hydrological effects of topography of the field and surrounding areas should be taken into account when optimizing drainage intensities. Though most of the groundwater outflow occurred outside the growing periods, sustained groundwater outflow occurred throughout all seasons. A correspondence was observed between the near-saturated surface soil conditions and tillage layer runoff (TLR) events, which suggests that TLR events in high-latitude clayey fields are mainly triggered by saturation excess mechanism. During two springs (once in both field section) TLR was clearly higher than during other seasons, which was likely induced by soil frost. However, the model without computational schemes for frost-induced changes on soil hydraulic properties satisfactorily reproduced drain discharge during most spring periods, and the amount of measured TLR was low during most springs, which indicates that typically frost-induced changes on TLR generation may be small. However, inaccuracies in the quantification of TLR induced by snowmelt formed uncertainty to the estimate of the water balance components during snowmelt.