Forecasts of climate variability at the decadal, seasonal, and monthly time scales will be discussed in this session. This will include presentations and discussions of predictions for a time horizon of up to ten years from dynamical ensemble forecast systems and statistical models. Further topics include forecast quality assessment, physical mechanisms for skill and predictability, role of anthropogenic climate change, implementation of seamless forecasting systems, ensemble forecast generation and initialization, approaches to address model uncertainty, downscaling and calibration, and demonstrations of end-user value for climate risk applications and climate change adaptation using the recently developed climate services framework. A special focus will be put on results from the CMIP5 decadal prediction experiments (and projects Natclim, SPECS and Euporias). We also encourage contributions on the analysis of predictability, downscaling and calibration for long time scales in simple dynamical systems (as proof-of-concepts), and the use of advanced analysis techniques to clarify the long term climate predictability (such as information theory, bayesian approaches, networks, weather regime dynamics..).