This is a renewal grant to continue an investigation of climate variability as simulated by a low resolution atmospheric general circulation model (GCM) coupled to interactive upper oceans formulations. Both interannual and intra-annual aspects of the model climate will be examined. The PIs will continue the analysis of variability within two already completed 100 year seasonal integrations one performed with the atmospheric model coupled to a 50m deep upper ocean and the other performed with a prescribed annual cycle of observed ocean conditions. New experimental initiatives with more sophisticated upper ocean and sea ice models are also proposed. These involve 100 year seasonal integrations of the atmospheric model coupled to a prescribed variable depth upper ocean component, a prognostic mixed layer ocean formulation, a tropical-dynamical upper ocean model, and a improved sea ice parameterization. The PIs will focus on characterizing the interannual variability of (1) ocean surface temperature in tropical and extratropical latitudes of the Pacific and Atlantic Ocean and (2) sea ice area and thickness in polar latitudes. They will aim to identify cause/effect relationships between anomalies in different parts of the atmosphere-ice-ocean system. The studies of intra-annual variability will focus on bimodality in planetary scale atmospheric circulation features of middle latitudes and on transition shifts in the seasonal structures of atmospheric circulation and energetic. Intercomparison of the different experiments will identify the relative importance of individual ocean-atmospheric processes that produce climate variability. These processes include ocean-atmosphere energy exchange,mechanisms involving mixed layer and sea ice development, and internal ocean dynamics. This research is important because it seeks to improve understanding of how the ocean-ice-atmosphere system interacts to cause climate variability.