Sammendrag
Numerical weather prediction (NWP) combines physical models and observations to monitor and predict the evolution of the Earth system. In the Arctic, there are specific challenges related to process understanding, modelling and observations. MET Norway has developed and implemented a short-term, high-resolution weather prediction system - AROME-Arctic - for the delivery of reliable and accurate weather forecasts and warnings.
Based on AROME-Arctic, the Copernicus Arctic Regional Reanalysis (CARRA) combines past observations with NWP models to provide a comprehensive description and consistent time series of the observed climate as it has evolved during recent decades. AROME-Arctic performs relatively well in terms of accuracy. A set of common weaknesses across forecast systems are identified. To advance our understanding and model representation of key processes, the Polar Prediction Project coordinates a process-based model evaluation project based on observations at selected Arctic supersites, including MOSAiC - the largest polar expedition in history. Impact-based forecasting combines a forecast of a weather or climate hazard and an assessment of possible impacts. Experiments show that the reliability in the weather predictions benefits from better uncertainty estimation at the smaller spatial scales, including the sea ice, snow on sea ice and sea surface temperature. At present, satellite observations are not used optimally for weather prediction and climate monitoring in the Arctic, particularly in seasons and areas with snow and sea ice. We show the relative impact of different observations on forecast accuracy, disentangling the benefits of observations on local forecast accuracy. There is a tremendous opportunity and role for physical modeling and data assimilation (including reanalysis) in making observations into usable products, and investments in observing systems must be accompanied by a continued investment in NWP and high performance computing.
Vis fullstendig beskrivelse
