Understanding the relative roles of Atlantic vs. Pacific coupled dynamics in initialized decadal climate predictions
Grant
Overview
Affiliation
View All
Overview
abstract
This collaborative effort between NCAR, Texas A&M University, and the University of Washington/JISAO will advance decadal prediction science by deepening our understanding of the regionally-dependent coupled dynamics at work in the Atlantic and Pacific in initialized decadal prediction ensembles using full-complexity coupled general circulation models. The motivation for this proposal is that current state-of-the-art decadal prediction systems, such as the CESM Decadal Prediction Large Ensemble (DPLE), show areas of tantalizing promise, but the coupled mechanisms explaining the variations in skill across different regions and fields remain poorly understood. We will analyze the DPLE (in conjunction with its uninitialized counterpart--the CESM Large Ensemble) with the aim of developing improved mechanistic understanding of the origins of skill in interannual-to-decadal predictions of sea surface temperature, sea-ice, atmospheric circulation and moisture transport, and high-impact near-term climate change over populated regions. Our overarching goal is to contribute to improved decadal predictions by elucidating the relative roles of the Atlantic and Pacific Oceans in giving rise to predictable climate signals; investigating the ramifications of model drift; improving our understanding of initialization shock; identifying the origins of low signal-to-noise in key climate fields; and developing process-level understanding of the factors governing skill for a selection of high-impact regional climate variations including tropical cyclones, precipitation over land, and Arctic sea-ice. We will synthesize our findings to develop some specific recommendations for new systems that will be in development during the lifetime of this proposal.
