PARADIGM - Prediction and Regional Downscaling Models
What are the goals of the project?
The main goal of the project is to improve regional predictions of climate both on land and in the ocean. For this we are developing a framework to dynamically downscale climate predictions that we will evaluate and improve in terms of regionally focused analyses of predictive potential. The Norwegian Climate Prediction Model (NorCPM) will be used to drive the regional models and its predictive skill with regard to meteorological indices and ocean ecosystem indicators is being evaluated. The time scale of the predictions is from a season to decades.
What are the main results so far?
The main focus in the project so far has been to develop, improve, and evaluate NorCPM using dynamical covariance, which turns out to give promising results in subsurface hydrography in the North Atlantic down to at least 1000 m, especially for temperature. A version also assimilating sea ice has been developed, is now running, and will be evaluated in the near future.
The downscaling activity in the project has been divided into an ocean and atmospheric part. The atmospheric part focuses on the quantification of errors and uncertainties due to nesting, resolution jumps, and spin-up, and is crucial in order to evaluate the reliability of the regional forecasts.
The oceanic part is slightly changed from the project proposal due to issues related to the ROMS model in reproducing realistic deep water characteristics. Two regional ocean models (NEMO and ROMS Agrif) have been set up on the same Atlantic-Arctic grid as proposed in the project description. They are now running, and will be evaluated shortly to decide which model to be used in future climate prediction downscalings. Work on identification of specific precursors for the regional climate and ecosystem productivity in the Barents Sea has started.
What are the future plans?
The future plans include further development and evaluation of the NorCPM, specifically the usefulness of sea ice assimilation, including the predictive skill of meteorological indices and and ocean ecosystem indicators. Based on the evaluation of the regional ocean models, the NorCPM will be downscaled for the ocean and the results will be evaluated with focus on the Nordic and Barents Seas, and the northwestern European climate. Also, the performance of other CMIP5 decadal prediction systems will be investigated with respect to the identified key precursors, primarily in the Barents Sea.
Who is involved?
NERSC: Francois Counillon, Laurent Bertino, Yiguo Wang, Madlen Kimmritz
UiB: Noel Keenlyside, Corinna Schrum, Tor Eldevik, Marius Årthun,
Co-leader Stefan Sobolowski,
Uni Research: Muralidhar Adakudlu
IMR: Bjørn Ådlandsvik, Trond Kristiansen, Paul Budgell, Øystein Skagseth
Duration: 2015 - 2017