GIS-based modelling of solute fluxes at the catchment scale: a case study of the agricultural contribution to the riverine nitrogen loading in the Vejle Fjord catchment, Denmark

A model that predicts nitrogen loading is developed and applied on an agricultural catchment draining to a Danish Fjord. Increases in nitrogen loadings of the Danish coastal waters have been recorded and the transformation in the Danish agricultural sector is established to be the main cause. The development in Danish agriculture in the last 40 years has resulted in significant changes in the agricultural production structure which are of major significance to the water quality. Therefore, there is a need for methods which can relate the consequences of changes in the agricultural production structure to the nitrogen loadings of coastal waters. In this study riverine nitrogen loads are modelled by combining spatially distributed data for: (1) non-point source nitrogen leaching; (2) point source nitrogen emissions; (3) soils; and (4) elevation data of the terrain and of the ground water head, with a nitrate decay process. The decay process is expressed as a first-order decay with respect to distance, having two different decay constants according to different flow velocities. N-leaching was partitioned into a component with high velocity (mainly tile drainage) and a component with low velocity (mainly baseflow) based on soil type/flow-path correlations. The nitrogen loading model was implemented into GIS to calculate nitrogen loads for sub-catchments in the Vejle Fjord catchment, Denmark. The two nitrate decay-coefficients were estimated by a trial and error method. Discrepancies between model calculated and measured N-loads vary from -16 to 17%. A source apportionment of the agricultural contribution to the riverine nitrogen loading revealed that livestock production is responsible for about 60% of the total non-point source loading while part lime farming contributes with approximately 23% of the total non-point source load. The remainder of the N-loadings originates from plant production and background sources. (C) 1998 Elsevier Science B.V. All rights reserved.