The Swedish state-of-the-art seasonal hydrological forecasting service - An example of successful science-policy-practice integration

Abstract

The Swedish Meteorological and Hydrological Institute (SMHI) occupies a distinctive position at the intersection of science, policy, and practice. The institute is uniquely equipped to directly translate state-of-the-art research into its operational services. Seasonal hydro-meteorological forecasts have been used in decision support tools for various water-related sectors that require long-term planning. Several seasonal hydrological forecasting systems have been employed, with streamflow climatology, ensemble streamflow prediction (ESP), and dynamical GCM-based streamflow prediction (DSP) being among the most common systems. Each system has its advantages and disadvantages, however, DSP is overall considered the most accurate, particularly during the first lead weeks (about up to 6). Currently, SMHI provides seasonal hydrological forecasts based on climatology, which can be openly accessed (https://vattenwebb.smhi.se/ in Swedish). Building on its existing seasonal forecasting system, SMHI recently operationalized dynamical seasonal forecasts for Sweden, utilizing bias-adjusted ECMWF SEAS5 meteorological forecasts. With this work, we showcase the implementation of cutting-edge meteorological and hydrological research into a useful service for long-term decision-making. We present our workflow, address potential (scientific and technical) obstacles, and outline strategies for the accessible and reliable distribution of seasonal hydrological forecasts to the general public.