Assessing model fidelity and sources of bias for predictions and projections

Timothy DelSole, Professor, Atmospheric, Oceanic and Earth Sciences (AOES) and Martha Buckley, Research Associate Professor, AOES, are working to improve NOAA's climate prediction capability by identifying and correcting errors in the Geophysical Fluid Dynamics Laboratory's Seamless System for Prediction and Earth System Research (SPEAR) forecast system. 

They are doing this by comparing predictable variability in observations and models using a rigorous statistical method that overcomes major limitations of previous methods and provides a foundation for evaluating and testing prediction systems on multi-annual to multi-decadal timescales.  

They are using this method to assess how well modes of variability are represented in SPEAR, Coupled Model Intercomparison Project Phase 6 (CMIP6), and large ensembles.  

DelSole and Buckley received $473,318 from NOAA for this research. Funding began in Sept. 2023 and will end in late Aug. 2026