ABSTRACT

The problem of separating variations due to natural and anthropogenic forcing from those due to unforced internal dynamics during the twentieth century is addressed using IPCC climate simulations and observations. An unforced internal component that varies on multidecadal time scales is identified in the twentieth century IPCC climate simulations by a new statistical method that maximizes integral time scale. This component, called the Internal Multidecadal Pattern (IMP), is stochastic and hence does not contribute to trends on long time scales, but can contribute significantly to short-term trends.

The warming and cooling of the IMP matches that of the Atlantic Multidecadal Oscillation and is of sufficient amplitude to explain the acceleration in warming during 1977-2008 as compared to 1946-1977. The forced component is increasing at the same rate during these two periods. The amplitude and time scale of the IMP are such that its contribution to the trend dominates that of the forced component on time scales less than 16 years, implying that the lack of warming trend during the past ten years is not statistically significant. Furthermore, since the IMP varies naturally on multidecadal time scales, it is potentially predictable on decadal time scales, providing a scientific rationale for decadal predictions.

In the second part of the presentation, an attempt is made to optimize the most predictable land surface air temperature pattern at continental scales related to global SST. It is found that the optimized land surface air temperature pattern has predictability of only 3 - 5 years (for land precipitation only 1 - 2 years). Since this is an optimized predictable pattern it will be difficult to find additional predictability. This result raises some questions about the prospects for multidecadal prediction of unforced variability over land.