Current capabilities and future directions of SPARE:WATER – a Site-sPecific Agricultural water Requirement and footprint Estimator

Sebastian Multsch and Lutz Breuer

Research Centre for BioSystems, Land Use and Nutrition (IFZ), Institute for Landscape Ecology and Resources Management (ILR), Justus Liebig University Giessen, Giessen, Germany, Sebastian.Multsch@umwelt.uni-giessen.de, Lutz.Breuer@umwelt.uni-giessen.de

The agriculture sector is responsible for large fraction of global water consumption. A proper management of irrigation agriculture could spare water resources. The agricultural water footprint addresses the quantification of water consumption in agriculture. By considering site-specific properties when calculating the crop water footprint, this methodology can be used to support decision making in the agricultural sector on local to regional scale.

Hence, we developed the spatial decision support system SPARE:WATER that allows us to quantify green, blue and grey water footprints on regional scale. SPARE:WATER is programmed in VB.NET, with geographic information system functionality implemented by the MapWinGIS library. Water requirement and water footprints are assessed on a grid-basis and can then be aggregated for spatial entities such as political boundaries, catchments or irrigation districts.

Improving the irrigation management and crop production is of major concern to save water resources, especially for countries in arid regions such as Saudi Arabia. SPARE:WATER has been used to investigate the effect of alternative production and irrigation scenarios on the water consumption of Saudi Arabia and its provinces. The prediction of water consumption and water footprints with simulations models comes along with uncertainties (e.g. structural and stochastic). These uncertainties have to be quantified if the model predictions are used to inform stakeholders such as authorities, scientists or decision makers. Therefore, in a second case study we conducted a comprehensive study on the uncertainty of estimating irrigation requirement for wheat in the Murray-Darling Basin (Australia). Uncertainty has been assessed with an ensemble of different potential evapotranspiration methods and crop parameter sets. Finally, we applied the Reliable Ensemble Averaging technique to reduce the predictive uncertainty of the model ensemble.

Estimating the water footprint in such diverse regions may require an adaption of the underlying simulation model, because the dominating biophysical processes will vary between the regions. E.g. capillary rise from groundwater is not important in Saudi Arabia but is a crucial process in irrigation agriculture in Central Asia. Hence, we aim to replace the static environmental model currently implemented in SPARE:WATER by flexible modelling frameworks such as the Plant growth Modeling Framework (PMF) and the Catchment Modeling Framework (CMF), which provide an interface written in the scripting language Python. Both frameworks represent building sets to set up individual plant and hydrological model setups, e.g. to compare different representation of biophysical processes or adapt the simulation model to site-specific conditions. The third part of our presentation will present our recent work in this domain.