Open access publications in the SLU publication database

Abstract

This thesis examines water discharge and NO₃-N loads from drained agricultural land in southern Sweden by modelling at field and watershed scale. In the first stage of the work, the ability of DRAINMOD to simulate outflow in subsurface drains and that of DRAINMOD-N II to simulate NO₃-N loads in these drains was evaluated in field experiments. In addition, the ROSETTA pedotransfer model was used to estimate soil hydraulic properties required by DRAINMOD. In the second stage, DRAINMOD was integrated with Arc Hydro in a GIS framework (Arc Hydro-DRAINMOD) to simulate the hydrological response of an artificially drained watershed. DRAINMOD-N II and a temperature-dependent NO₃-N removal equation were also included in Arc Hydro-DRAINMOD to predict NO₃-N loading. Arc Hydro-DRAINMOD used a distributed modelling approach to aggregate the results of field-scale simulations, where the Arc Hydro data model described the drainage patterns in the watershed and connected the model simulations from fields through the stream network to the watershed outlet. GLUE methodology was applied to estimate uncertainties in the framework inputs. At field scale, monthly values of drain outflows simulated by DRAINMOD and NO₃-N loads simulated by DRAINMOD-N II showed good agreement with observed values. Good agreement was also found between observed and DRAINMOD-simulated drainage rates when ROSETTA-estimated Ks values were used as inputs in DRAINMOD. At watershed scale, temporal trend and magnitude of monthly measured discharge and NO₃-N loads were well predicted by Arc Hydro-DRAINMOD, which included uncertainty estimation using GLUE methodology. Sensitivity analysis showed that NO₃-N loads from the stream baseflow and N removal in the stream network processes had the most sensitive parameters. These results demonstrate the potential of DRAINMOD/DRAINMOD-N II and Arc Hydro-DRAINMOD for simulating hydrological and N processes in drained agricultural land at field and watershed scale. These models can contribute to improve water use efficiency in watersheds and to evaluate best management practices for preventing surface water and groundwater pollution.