The phenomenon known as pacific decadal variability (PDV) exerts a profound influence on the climate of the U.S. And its territories, yet prediction efforts using initialized forecasts with the…

The phenomenon known as pacific decadal variability (PDV) exerts a profound influence on the climate of the U.S. And its territories, yet prediction efforts using initialized forecasts with the current generation of earth system models show limited skill. Further, the influence of external radiative forcings on recent and future PDV evolution remains unclear. We hypothesize that many of the outstanding questions regarding potential predictability of PDV could be answered using high resolution (HR) models that resolve key small-scale ocean and atmospheric processes. Our ideas are motivated by recent HR simulations performed with the community earth system model (cesm) showing: 1) enhanced PDV predictive skill relative to the standard lower resolution (LR) version and 2) an externally forced PDV that explains observed changes. As HR models become available, understanding the mechanisms underpinning their skill in predicting forced and unforced PDV is critical, particularly since IT is not fully understood if these mechanisms arise from increasing oceanic or atmospheric resolution. Our preliminary analysis of the HR-cesm reveals several physical processes arising from both increased atmosphere and ocean resolution. These mechanisms are not included in current mechanistic explanations of PDV and are poorly simulated by the LR-cesm. To further understand the salient PDV processes that depend on resolution, we propose a systematic exploration of mechanisms involved in forced and unforced PDV in a hierarchy of existing simulations and proposed new experiments with the HR-cesm. We leverage the unique features of the cesm data and modeling infrastructure, in particular: 1) availability of ensembles under historical and future emissions scenarios configured with different resolutions (HR, LR and a hybrid LR-ocean/HR-atmosphere); and 2) ability to perform simulations with components of reduced complexity to isolate mechanisms. We expect that our results will advance understanding of the causes of past decadal trends in pacific/north american climate and their potential predictability, improve future projections of PDV over the coming decades, and reveal shortcomings of standard-resolution climate models.