Recent modeling work by Ganopolski and Rahmstorf (2001, 2002) and Alley and Rahmstorf (2002) suggests that the North Atlantic branch of the global thermohaline circulation possesses two potential modes of operation during glacial times: a cold stable mode and a warm marginally unstable mode, the latter of which typically lasts for but a few hundred years.  The cold stable mode is characterized by deep-water formation south of Iceland; while the warm unstable mode is characterized by deep-water formation in the Nordic Seas and shares many characteristics with the circulatory mode of the current interglacial, although it is not quite as strong.

All else being equal, the cold stable mode of the ocean's thermohaline circulation would be expected to persist throughout an entire glacial period.  However, as Ganopolski, Rahmstorf and Alley (GRA) note, a weak real-world forcing with a periodicity on the order of 1500 years produces small cyclical variations in freshwater input to high northern latitudes at approximately the same periodicity; and these perturbations, when in the declining phase, often, but not always, initiate a transition to the warm unstable mode of thermohaline circulation, which includes a shift in the location of deep-water formation from south of Iceland to the Nordic Seas.  This new mode of circulation (warm unstable, which is accompanied by rapidly warming air temperatures) then persists for a few hundred years before reverting back (because of its inherent instability) to the cold stable mode of circulation (and its accompanying colder air temperatures).

An interesting aspect of this recurring rapid-warming-followed-by-slower-cooling scenario is that the cyclical perturbation that leads to the change in the ocean's mode of thermohaline circulation is directly responsible for only a small fraction of the change in deep-water formation that is required to trigger the rapid warming events.  By applying the concept of stochastic resonance to the problem, however, Ganopolski and Rahmstorf (2002) demonstrate that it is the background noise in the climate system that "triggers the events and thus amplifies the weak cycle into major climatic shifts with global reverberations."

These several observations, some empirical and some theoretical, suggest a number of important things.  First, very weak forcing factors may well have the potential to produce large changes in earth's climate under certain circumstances; and one such forcing factor that presents itself to our minds within this context is solar variability.  This possibility has also presented itself to GRA.  Ganopolski and Rahmstorf (2001), for example, state that the low-amplitude cycle in freshwater forcing responsible for the large-amplitude cyclical changes in glacial climate could be "ultimately due to solar variability," while Alley and Rahmstorf (2002) say that "a possible cause could be a weak periodic variation in the output of the sun."  In fact, Bond et al. (2001) have actually committed themselves to this conclusion, particularly as it applies to the Holocene, for which period of time they have assembled a vast array of compelling evidence that essentially proves the sun-climate connection.

An interesting thing about the Holocene, however, is the fact that Ganopolski and Rahmstorf (2002) report that - in their model, at least - its climate "is not susceptible to regime switches by stochastic resonance with plausible parameter choices and even unrealistically large noise amplitudes, and neither is it in conceptual models."  Also, as they correctly report - and of even more significance, since the observation is based on real-world data - "there is no evidence for regime switches during the Holocene."

This, thus, is the other important lesson to be learned from these several studies: Holocene climate - both in theory and point of fact - is not susceptible to catastrophic changes.  Indeed, the Holocene is only known to have experienced much more modest climatic oscillations of the Medieval Warm Period-to-Little Ice Age-to-Modern Warm Period type, which are serious enough when in the cooling mode, but actually welcome when in the warming mode.

In conclusion, we are about as convinced as we can possibly be that predictions of catastrophic CO2-induced global warming are totally out of sync with reality.  Also, there is no question in our minds but what the historical increase in global temperature over the past two centuries, which has recently been characterized by Esper et al. (2002), is solar-induced and represents a return to climatic conditions akin to those of the Medieval Warm Period.  We welcome this modest climatic transition; and we welcome the contemporaneous increase in atmospheric CO2 concentration, which poses no threat of additional warming, but holds out instead the promise of enhanced biological activity.

Reference
Alley, R.B.S. and Rahmstorf, S.  2002.  Stochastic resonance in glacial climate.  EOS, Transactions, American Geophysical Union 83: 129, 135.

Bond, G., Kromer, B., Beer, J., Muscheler, R., Evans, M.N., Showers, W., Hoffmann, S., Lotti-Bond, R., Hajdas, I. and Bonani, G.  2001.  Persistent solar influence on North Atlantic climate during the Holocene.  Science 294: 2130-2136.

Esper, J., Cook, E.R. and Schweingruber, F.H.  2002.  Low-frequency signals in long tree-ring chronologies for reconstructing past temperature variability.  Science 295: 2250-2253.

Ganopolski A. and Rahmstorf, S.  2001.  Rapid changes of glacial climate simulated in a coupled climate model.  Nature 409: 153-158.

Ganopolski, A. and Rahmstorf, S.  2002.  Abrupt glacial climate changes due to stochastic resonance.  Physical Review Letters 88: 038501.