then the possibility exists that external forcing variations might determine the timing of events.
For example, the earth’s orbital parameters vary over time, affecting the distribution of solar energy delivered to the planet in time and location. The fastest of these variations is the precessional cycle, which takes about 22,000 years to complete and determines when the summer solstice occurs relative to the time of closest approach of the earth to the sun. In the tropics, that yields an 11,000 year half-precessional rhythm, and there are substantial changes in the insolation pattern over as little as 5,000 years. Those changes could have a major impact in the tropics through their effect on the latitudinal excursion of the ITCZ and its consequent effects on the Hadley circulation. The effect of precessional insolation changes on monsoonal circulations has been implicated in the moistening and drying of the Sahara, with strong feedbacks linked to land-surface processes, including changes in vegetation (Kutzbach et al., 1996; Claussen et al., 1999; Carrington et al., 2001). Precessional insolation changes possibly could have large effects on El Niño occurrence (Clement et al., 2000, 2001).
The effects and magnitude of fluctuations in solar output are less well constrained. There is an observable modulation of the sun’s brightness over the 11-year solar cycle, but this is too frequent an effect to alter climate much. On longer time scales, there are no direct observations of the fluctuation of solar output. Observations and proxies for solar activity going back centuries or more do indicate long-term fluctuations in activity, as measured by sunspot number or solar-wind effects; it is not known how much fluctuation in solar brightness goes with such fluctuations in activity. It has been suggested that the solar influence on climate can be mediated directly by the influence of cosmic-ray fluxes (modulated by the solar wind’s effect on the earth’s magnetic field) on cloud formation (Svensmark and FriisChristensen, 1997; Svensmark, 1998), but the magnitude or even sign of the effect has not yet been quantified. Solar influences have been suggested as the cause of the Little Ice Age (Broecker, 1999), and indeed changes in solar activity do seem to line up with some major climate fluctuations. Solar forcing has been tied to drought frequency and effects on Mayan civilization (Hodell et al., 2001).
In addition to the well-studied mechanisms detailed above, there might be other threshold phenomena in the climate system that are difficult to assess quantitatively. The possibility of catastrophic release of methane by