Regional nitrogen budgets of agricultural production systems in Austria constrained by natural boundary conditions

The impacts of climate change on nitrogen losses to the environment in Austria: A dual model analysis across spatial and temporal scales to support policy decisions

The amounts and pathways of reactive nitrogen (Nr) losses in Austria into the surface water, soil, and atmosphere were determined under four climate change scenarios for the period 2041-2070. Two nutrient models were used to undertake the analysis at two different scales. Firstly, a semi-empirical, conceptual model (MONERIS) was setup for Austria to calculate the overall annual Nr surpluses, categorise flows of Nr, and identify regional hotspots of Nr losses. Secondly, a physically based eco-hydrological model (SWAT) was setup in three agricultural catchments to determine the hydrological processes related to Nr transport and quantify the amounts transported by various pathways in cropland at a detailed spatial and temporal resolution. The agricultural N surplus calculations for Austria were revised and used as input data for both models. The MONERIS and SWAT simulated inorganic N loads transported into waterbodies are overall similar, with average differences for the subsurface inorganic N loads of ±3 kg ha-1 yr-1 and for surface inorganic N loads of +0.4 to -0.03 kg ha-1 yr-1. Crop level N losses under future climate scenarios was contingent upon the fertilizer type, the crop grown and its accumulated biomass, as well as the type of climate scenario (wet or dry). In the SWAT model, an examination of the sensitivity of the input data (climate data and parameter values) found the dominant contribution to the sensitivity of simulated monthly discharge was from the climate data (69 % to 98 %). For simulating N loads, the climate scenarios contributed 30 % to 89 % of the sensitivity. Simulating Nr flows under climate scenarios is policy relevant to assess critical areas of N losses and identify future N transport pathways. Using a dual-model approach saves on resources required to set up a complex, data intensive model at a large scale, and can focus on critical catchments in detail.