Hypothesized causes of decade-to-century-scale climate variability: Climate model results

There are at least five hypothesized causes of decadal to century-scale climate variability, both 'natural' and anthropogenically-induced: (1) inherent ('random') variability in the atmosphere (i.e. no external forcing); (2) inherent or forced variability in the ocean dynamical system (e.g. North Atlantic Deep Water fluctuations); (3) solar variability (e.g. the Maunder Minimum); (4) variability in volcanic aerosol loading of the atmosphere (e.g. Tambora); and (5) atmospheric trace gas variability (e.g. CO₂ methane). Modeling experiments conducted for each of these potential mechanisms show that they have different signatures in time and space which may allow for discrimination in the climate record. The effects of variations in ocean thermohaline circulation are likely to be largest in those regions adjacent to and downwind of areas of deepwater formation. Decreased insolation and large low latitude volcanic eruptions affect all latitudes, including changes in Hadley cell-dominate low-latitude systems. In contrast, trace-gas induced warming is likely to be amplified at high latitudes, although still containing a significant low latitude effect. Each of the hypothesized causes may have been important, and it is likely that several may have acted in concert to produce the observed climate record. During the Holocene, natural global mean decade-to-century-scale variability is likely to have been limited to 0.5 to 1.0°C, and to have been non-uniformly distributed over the globe. These natural sources of climate variability are thus unlikely to counter future greenhouse warming in any long-term way.