Decision support system for evaluation of available surface water resources for use in Slovenia – development of the basis

According to the Water Law of the Republic of Slovenia anyone who wants to use water resources in addition to the general use (e.g. drinking, swimming and other recreational uses) needs to acquire a water right. Approving the water rights is in principal based on two conditions: (i) the discharge downstream from the withdrawal should be at least equal to ecologically acceptable flow and (ii) the withdrawal of the water should not influence the natural conditions or other already present uses and needs of water resources. Through the development of the system’s basis we have focused on the first condition, the hydrologically available water resources.

The information about the amount of the water in a river or a stream is provided by the discharge measurements, performed at locations of the water gauging stations by the Slovenian Environmental agency (ARSO). However, when granting a water permit, the location of certain water withdrawal can be anywhere along the watercourse. Therefore, we have tested seven advanced statistical models to connect characteristic discharges (mean and mean minimal discharge for a selected 30-year period) at measured points with attribute data describing the properties of the corresponding catchment. The 49 attribute values were gathered through analysis of spatial data in GIS environment and provided information about precipitation, temperature, geological structure, land use and water use in the area. According to the comparison of models performance we have selected the neural networks. They were used to estimate characteristic discharges in 340 selected points on the water courses all over the country. These values provide the basis for calculation of the ecologically acceptable flow and the amount of discharge available for use under certain terms. In addition, for the selected points (340 points) also the natural (reference) discharge was estimated as a discharge which would be observed without any water use present upstream. Simulations of natural discharge situation were performed for various scenarios and multiple selected test cases. The estimated differences between natural and measured discharge for selected points was than generalized for all the other points using the hierarchical clustering approach. So far the basic information about the amount of water available for use on watercourses with basins larger than 10 km² was estimated. However, the project is ongoing with the focus on improving the models, including the complex interactions between surface waters and groundwaters, and taking into account the vulnerability of the natural environment and ecosystem services they provide as well as sectoral needs.