Decadal Climate Predictability and Prediction

What: Forty-five researchers met to discuss strategies to better understand and predict decadal climate variability. When: 12–15 October 2009 Where: St. Michaels, Maryland T he importance of decadal climate variability (DCV) research is being increasingly recognized, including by the World Climate Research Program (WCRP) and the Intergovernmental Panel on Climate Change (IPCC). An improved understanding of DCV is very important because stakeholders and policymakers want to know the likely climate trajectory for the coming decades for applications to water resources, agriculture, energy, and infrastructure development. Responding to this demand, many climate modeling groups in the United States, Europe, Japan, and elsewhere are gearing up to assess the potential for decadal climate predictions. The magnitudes of regional DCV often exceed those associated with the trends resulting from anthropogenic changes. Therefore, differentiating between the two is also very important for planning, implementation, and national and international treaties. This brief meeting summary (presentations available at www.DecVar.org/auditorium.php) provides an overview of the consensus of a research community workshop1 on DCV held in the fall of 2009. The consensus of the participants was that decadal climate predictions may become useful to society some time in the future, but will be highly experimental in the near future. Ref lecting the increasing interest in decadal climate predictions since the previous DCV Workshop in May 2007, the main theme this time was “Decadal climate predictability and prediction: Where are we?”; the purpose was to describe and summarize the status of decadal climate nowcasting and forecasting, thus outlining a research strategy to secure the needed observations and methods to quantify predictability of DCV, and to identify unsolved science problems. Gaps in our knowledge of DCV were identified, as were areas of needed research. Recommendations in areas of predictability and prediction; observations, theory, and modeling; and societal impacts are summarized. Highest-priority