gas and aerosol fluxes. Although understanding of how land-use change acts as a climate forcing is not as developed as that for greenhouse gas forcing, a recent review of research on this topic concluded that land-use change during the twentieth century had well-defined effects on regional climate (Kabat et al., 2004). In some cases, these regional effects are estimated to be as large as those due to a doubling of CO₂. In addition, land-use change may have possible significant effects on the global climate through teleconnections. Continued conversion of the landscape is expected to have a comparable effect in the next century.

The climate forcing of future landscape change on the surface energy and water budget, and how this effect is transmitted through the atmosphere, has been little studied. Most previous work on future landscapes has concentrated on how the landscape responds to future climate change simulated in models (e.g., NAST, 2001) and on surface greenhouse gas sources or sinks, rather than on how the landscape itself alters climate through its heat and moisture fluxes (e.g., Snyder et al., 2004).

Abrupt climate changes take place when “the climate system is forced to cross some threshold, triggering a transition to a new state at a rate determined by the climate system itself and faster than the cause” (NRC, 2002). Paleoclimate records indicate that abrupt changes have occurred frequently in the past and at rates high enough that human or natural systems may have had difficulty adapting. However, efforts to develop model-based scenarios for future abrupt climate changes are limited by our lack of understanding of the mechanisms that underlie such changes (see NRC, 2002, for a full discussion of this issue).

One example is a striking series of abrupt changes in temperature identified in Greenland ice core records from the last glaciation. Known as Dansgaard/Oeschger (D/O) cycles, these temperature shifts recur on millennial time scales. They have been attributed to threshold jumps in thermohaline circulation (THC), presumably triggered by sudden discharges of icebergs and freshwater from ice sheets (Broecker, 1997). Temperatures above the ice cap may have dropped by as much as 8°C, and large parts of western Europe may have cooled by a few degrees. The recent discovery of correlatives of D/O cycles in tropical latitudes, however, has raised an alternative possibility that the underlying mechanism was ocean-atmosphere interaction in low latitudes (Clement et al., 2000). Some have even argued that El Niño cycles of long duration or shifts in the ITCZ (intertropical convergence zone) might change water vapor export to high latitudes. Such changes could have amplified or, through impacts on THC, even caused the D/O variations (Peterson et al., 2000).