The following provides a list of topics and some key questions. We hope that the workshop participants will address these and other related questions in their presentations, posters, and discussions.

1. Observed decadal variability: What is the observed decadal variability in the climate system (observations and syntheses)? What are the observed climate impacts? What do the paleoclimate records show? What are the observed signals in the ocean? Are the present ocean observations adequate for decadal variability studies? What new observations are needed?

2. Predictability and state of the ocean models: Is there any predictability in the climate system? What are the sources of such predictability? What are the roles of natural and forced variability? Are the ocean models up to the task? What are the CORE-II hindcast simulations and ocean syntheses showing? How sensitive are small/regional changes to the ocean initial state? Are all the participating models in CORE-II reproducing the observed and synthesized variability robustly? How sensitive are the model results to model resolution? Are the ocean models robust in their internal variability? How is the internal variability affected by ocean model parameterization choices?

3. Physical Mechanisms: What are the sources of decadal variability? What determines the propagation and decay of decadal anomalies? What are the physical mechanisms in the ocean for decadal variability? How robust are these mechanisms? How are CORE-II experiments and ocean data assimilation improving our understanding of oceanic decadal variability mechanisms?

4. Initial conditions, predictions, and verification: What initialization techniques are used in the community? Are they robust in their outcomes across different models? What fields should be carefully initialized in the ocean? Can the CORE-II hindcast simulations be successfully used to initialize the ocean state for prediction experiments? Are they as useful as ocean syntheses? Is any one of the initialization approaches clearly superior to the others? What are the common verification techniques?

5. CMIP5 decadal prediction experiments: What do the early results show? Do the models agree? Is there any predictability? Are the ocean models capturing observed variability during the hindcast period? Do the hindcast experiments show that it is meaningful to run projection / prediction experiments?