The role of climate variability in recent warming, and the ability of climate models to simulate this variability realistically, is difficult to assess given that the observed record is short, incomplete and contains the response to external drivers. The answers to the research questions i) to iv) also result in new estimates of the amount of the observed temperature, precipitation, and sea level pressure variability that originates from internal climate dynamics as it allows us to identify and remove, with uncertainties, the influence of external forcings from the observations. These estimates, based on a longer time-horizon, can provide a much improved estimate of the role of forced and internal variability in interdecadal climate fluctuations.
Coupled atmosphere-ocean climate model simulations that are selected to be close to the observed sea surface temperatures (SST) are used, applying a new method (Goosse et al., 2012) for the first time to a coupled model and to this period in order to understand the mechanisms underlying long-term observed variability, particularly the role of the ocean (see, e.g. Delworth and Knutson, 2000).