There are many challenges associated with the development of a credible estimate of the evolution of the climate system with detail down to regional spatial scales. First, a means of informing the climate system model of the current state of the atmosphere and, more importantly, the ocean, land, and sea ice states must be developed for the initialization phase. Second, a means of producing a representative ensemble of depictions of the future of the climate system over a period of 10 to 30 years must be developed for the decadal prediction and its uncertainty quantification. Third, a method to add regional value to this ensemble must be developed. The proposed research will address these three issues, as well as test the merits of alternative approaches, by leveraging several new and promising efforts at NCAR, Sandia, and SUNY Stony Brook. The three main science objectives of this research are:

1. To understand the role of low-frequency modes of climate variability (primarily rooted in the ocean) in affecting atmospheric variations that impact the surface precipitation and temperature climate anomalies and weather patterns over land. These variations include hurricanes, mid-latitude winter storms, extreme temperature and precipitation events, and persistent regimes of drought, floods,heat waves and cold waves
2. To understand and quantify the limitations and uncertainties of coarse resolution (1degree) coupled decadal predictions through a comparison of the results and mechanisms obtained in global high resolution decadal projections in both historical prediction mode and for predictions of the coming decades. High resolution refers to integrations using 1/4 degree resolution in the atmosphere and 1/10 degree resolution in the ocean
3. To quantify the efficacy of regional refined resolution comparing predictions using either local mesh refinement in the global atmospheric model or two-way nesting and comparing the results with the uniformly high resolution global model predictions. Importantly, the local refinement/nesting will not only cover land regions of interest such as the continental U.S. but, in addition, regions of importance for coupling the atmosphere and ocean.