If one extrapolates back to 4,6 Ga, using Equations (1) and (2) and holding the planetary albedo constant, one finds that 7e should have been lower than today by about 22 K. If atmospheric CO* had remained constant, T, should have dropped by an even larger amount, —33 K, because the amount of water vapor in the atmosphere would have declined as the surface cooled off. The predicted changes in Tr and 7J with time are shown by the dashed curves in Figure 13.1. The surface temperature calculations were performed using a one-dimensional radiative-convective climate model (Kasting and Ackerman, 1986) using a fixed tropospheric relative humidity distribution and neglecting any kind of cloud feedback. The results are similar to those found by Sagan and Mullen (1972), who used a somewhat simpler climate model. In both calculations, Ts drops below the freezing point of water prior to —2,0 Ga, imply ing a totally ice-covered Earth, Geologists know, however, that this is not what actually happened on the early Earth, inasmuch as there is evidence for life hack to at least 3.5 Ga (Schopf, 1993) and evidence for water-lain sedimentary rocks back to almost 4.0 Ga (Sagan and Mullen, 1972; Schopf, 1983), The conflict between the geologic record and the predictions of such simple climate models has come to be called the "faint young Sun paradox."

The faint young Sun paradox is not a paradox at all, of course, if either Earth's albedo was lower or its greenhouse effect was higher in the past. A greatly lower albedo is possible, in principle, if cloud cover was substantially reduced at that time. This has been suggested several times (Henderson-Sellers, 1979; Rossow ct aL, 1982), but it has yet to be shown to be very likely. General circulation model (GCM) studies with a swamp ocean predicted a 20% decrease in cloudiness as a consequence of the Earth's faster rotation {Jenkins, 1993; Jenkins et al., 1993), but a more recent study by the same author (Jenkins, 1996) indicates that this effect goes away if sea surface temperatures are specified. A much colder Earth might conceivably have had fewer clouds, but any reduction in planetary albedo would likely have been offset by an increase in highly reflective sea ice. So cloud cover changes remain an improbable solution to the faint young Sun problem.

Fortunately, there is another, more intuitive way of solving this problem, namely, by postulating that the atmospheric greenhouse effect was higher in the past. Several different infrared-active gases might, in principle, have been involved. Sagan and Mullen themselves (1972) suggested ammonia, NH*. Subsequent investigations (Kuhn and xAtreya, 1979; Kasting, 1982) have shown, though, that ammonia would have been rapidly photolyzed to N!2 and H>, unless it was protected by a UY shield (Sagan and Chyba, 1997). I shall return to the question of possible UV screens in a moment, For now, let us simply note that ammonia is less likely to have contributed substantially to the greenhouse effect on the early Earth than are several other gases.

13,2 CO2 and the Carbonate-Silicate Cycle

One greenhouse gas that was almost certainly abundant on the early Earth is carbon dioxide, CO2. CO2 is the second most important greenhouse gas in the present atmosphere (after H2O) and, as mentioned at the outset, can provide more than enough warming to solve the faint young Sun problem. (Note that water vapor, by itself, cannot do so because it is always close to its saturation point. Hence, H2O acts as a feedback mechanism, rather than as a forcing mechanism, for climate change.) The idea that high CO2 concentrations could have compensated for low solar luminosity was first suggested by Hart (1978), although it remained for Owen et aE (1979) to demonstrate this with a detailed radiad ve-convective climate model. Subsequent calculations by Kuhn and Kasting (1983), Kasting et aE (1984), and Kiehl and Dickinson (1987) have confirmed this conclusion and have demonstrated reasonable agreement between the different climate models. COj could, in fact, have overcompensated for the faint young Sun and produced a very warm early Earth. Walker (1985) suggested that the C02 partial pressure on an ocean-covered early Earth could have been as high as 10 bars for the first several hundred million years of the planet's history. According to Kasting and Ackerman (1986), this would have produced a mean surface temperature of 80-90 °C. Although this sounds extremely warm, there is no evidence to prove that this could not have been the case. Note that the oceans would not have been anywhere near boiling at this temperature because the total atmospheric pressure was considerably higher than 1 bar. Indeed, the oceans would not have boiled until surface temperature reached the critical point (374°C) because the atmosphere-ocean system works much like a pressure cooker.

It is not only the large potential for greenhouse warming by CO2 that makes this gas a good candidate for solving the faint young Sun problem, There are good, independent reasons for expecting that CO? levels should have been high on the early Earth. On long (> year) time scales, atmospheric COj concentrations are controlled primarily by the carbonate-silicate cycle (Figure 13.2). CO¿ dissolves in rainwater, forming carbonic acid (H2CO3). Carbonic acid dissolves silicate rocks on the continents in a process called silicate weathering. The by-products of silicate weathering,

0 0

Post a comment