This paper discusses the current procedures used by the Environmental Agency of the Republic of Slovenia (EARS) for forecasting streamflows and extreme events. It focuses on the decision-making process, and how new emerging science and technology is being used to improve the quality of water resource management and protection of life and property in floods and droughts. With Slovenia’s entry into NATO and the European Union in 2004, it is seeking to enhance its ability to fully use new science and technology to improve hydrometeorological forecasts. Slovenia receives between 800 and 3000 mm of annual precipitation from the Hungarian plain to the Julian Alps respectively. This range of spatial variability creates unique challenges for water managers and hydrometeorological forecasters especially under recent changing climatic conditions, when extremes of drought and floods appear to occur more frequently.

Examples of the extreme drought of 2003 and very wet 2004 water years will be presented showing the dramatic range of annual variability, and how the hydrologic and atmospheric forecasters are able to provide information of immediate value to the water managers on the Sava, Soca, Drava, and other major river systems in Slovenia. EARS produces daily forecasts of discharges of all major Slovene rivers. One of its main purposes is early warning of floods. EARS decided to use the HBV model developed by SHMI. In 2003 and 2004 preliminary forecasts and calibration of the model were performed. There are still many imperfections, mainly in the data input side. This paper will discuss how the LDAS – LIS products may improve the streamflow forecasts. The limited area numerical model ALADIN has been used operationally in weather forecasting practice for eight years in EARS. The performance is generally good, excellent in some fields (e.g. wind forecasts) and quite poor in case of precipitation. Implementation of detailed microphysics and nonhydrostatic dynamics is planned in forthcoming years in the ALADIN/AROME community; it is expected that precipitation forecasts scores will improve. Results from this model improvement will be evaluated using the RiverWare modeling framework.

The RiverWare modeling framework used for decision-making will be presented showing how advanced scientific methods may be used to improve water management within Slovenia. RiverWare, developed by the Center for Advanced Decision Support for Water and Environmental Systems (CADSWES), University of Colorado in partnership with the US Bureau of Reclamation, Tennesee Valley Authority and US Army Corps of Engineers, is a state-of-technology information system that enables water managers to fully describe the hydrologic and hydraulic characteristics of a river system including dams, diversion structures, and hydropower plants, to better manage the distribution of water. It is driven by streamflow forecasts at key inflow points, thus it is an ideal test bed for evaluating impacts of scientific enhancements to hydrologic forecasts.

Plans to improve hydrometeorological analysis and forecasts using new products from the multi-agency Land Data Assimilation System (LDAS) and Land Information Systems (LIS) developed at NASA, NOAA, and with the academic community will be discussed. Demonstrations of these emerging technologies will be conducted in 2004-2005 to evaluate the accuracy of snow water equivalent estimates from satellite and radar systems, and land surface models of the snowpack evolution using high resolution (1 km) grids and surface observations at key sites including Mount Triglav Observatory (2864 m msl), Slovenia’s highest mountain where a meteorological observatory is manned throughout the year. Future presentations will provide documentation of the success of this newly emerging technology in managing Slovenia’s water resources and improving flood and drought responses.