solar luminosity increased, the global temperature of the modern Earth would be too hot for nearly all forms of present-day life (Kump et al., 2000). The suggested solution to this problem is once again a silicate weathering-CO2 negative feedback. Higher temperatures and larger continents would have increased continental silicate weathering rates, thereby decreasing the CO2 greenhouse effect and cooling the Earth. Thus, many regard the silicate weathering-CO2 negative feedback as a thermostat that prevented permanent freezing of the early Earth and, later, prevented permanent temperatures too hot for life.

It should be emphasized, however, that much of what has been said above is based more on models and inference from evidence of warmth on early Earth than on conclusive proxy evidence. In addition, some have argued that methane may have been an important constituent of the early atmosphere (Kasting and Catling, 2003; Rye et al., 1995; Catling et al., 2001; Hessler et al., 2004) and that production of volcanic CO2 may have been an important source of the early Earth’s higher CO2 greenhouse effect (Ruddiman, 2001). More speculative solutions of the Faint Young Sun Paradox suggest that the early Sun might have been hotter than previously thought (Wuchterl and Klessen, 2001) or that a decreased cosmic ray flux, resulting from an early Sun’s stronger solar wind, may have reduced cloud cover and raised global temperatures (Shaviv, 2003).

Geological evidence also suggests that atmospheric CO2 changed dramatically on timescales of a few to tens of millions of years during the Phanerozoic (Figure 3-2). These changes are on the order of 5 to 10 times present atmospheric level. The record suggests that for at least two-thirds of the last 400 My (million years), levels of atmospheric CO2 were much higher than at present. It appears that these oscillations in atmospheric CO2 were linked to recurring changes from greenhouse to icehouse climate states.

The cause of the large atmospheric CO2 changes during the last 400 My is hotly debated. One view, known as the Berner-Lasaga-Garrels (BLAG) hypothesis, proposes that atmospheric CO2 changed in response to changes in seafloor spreading rates (Berner et al., 1983). Higher spreading rates increase volcanic activity at both divergent and convergent plate boundaries, thereby increasing the rate of release of volcanic CO2 from the large rock reservoir of carbon. Plate motion reconstructions for the last 100 My (the limit for which this can be done) suggest that at about 100 Ma (million years ago), spreading rates were 50 percent faster than today; however, during the last 15 My, CO2 levels fell at the same time that spreading rates increased, calling into question any simple relation between spreading rates and CO2 (Ruddiman, 2001).

A second view is that plate tectonic-driven uplifts of large plateaus, formation of mountain ranges, and amalgamation of supercontinents (which appear to be associated with low relative sea level) cause large



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement