Abstract. Earth appears to have been warm during its early history despite the fainmess of the young Sun. Greenhouse warming by gaseous CO2 and H20 by itself is in conflict with constraints on atmospheric CO2 levels derived from paleosols for early Earth. Here we explore whether greenhouse warming by methane could have been important. We find that a CH 4 mixing ratio of 10 -4 (100 ppmv) or more in Earth' s early atmosphere would provide agreement with the paleosol data from 2.8 Ga. Such a CH 4 concentration could have been readily maintained by methanogenic bacteria, which are thought to have been an important component of the biota at that time. Elimination of the methane component of the greenhouse by oxidation of the atmosphere at about 2.3 -2.4 Ga could have triggered the Earth's first widespread glaciation.

Abstract. The late Archean atmosphere was probably rich in biologically generated CH4 and may well have contained a hydrocarbon haze layer similar to that observed today on Satum's moon, Titan. Here we present a detailed model of the photochemistry of haze formation in the early atmosphere, and we examine the effects of such a haze layer on climate and ultraviolet radiation. We show that the thickness of the haze layer was limited by a negative feedback loop: A haze optical depth of more than •0.5 in the visible would have produced a strong "antigreenhouse effect," thereby cooling the surface and slowing the rate at which CH 4 was produced. Given this climatic constraint on its visible optical depth, the amount of UV shielding provided by the haze can be estimated from knowledge of the optical properties and size distribution of the haze particles. Contrary to previous studies [Sagan and Chyba, 1997], we find that when the finite size of the particles is taken into account, the amount of UV shielding provided by the haze is small. Thus NH3 should have been rapidly photolyzed and should not have been sufficiently abundant to augment the atmospheric greenhouse effect. We also examine the question of whether photosynthetically generated 02 could have accumulated beneath the haze layer. For the model parameters considered here, the answer is "no": The upper limit on ground level 02 concentrations is • 10 -6 atm, and a more realistic estimate for pO2 during the late Archean is 10 -8 atm. The stability of both 02 and NH3 is sensitive to the size distribution and optical properties of the haze particles, neither of which is well known. Further theoretical and laboratory work is needed to address these uncertainties.

No abstract

Methane is a key atmospheric constituent in origin of life theories that rely on an atmospheric source of reduced organic compounds. A CO2/N2-dominated primitive atmosphere (Holland, 1984), or even a CO~/N2/CO-dominated atmosphere (Kasting, 1990) would have allowed the formation of only minimal anaounts of HCN from lightning or photochemistry (Stribling and Miller, 1987). The presence of even a few parts per million of CI-L, however, would have allowed HCN to be synthesized by a mechanism in which the by-products of methane photolysis recombine with N atoms streaming down from the ionosphere (Zahnle, 1986). HCN is an essential starting material for the prebiotic synthesis of both amino acids and RNA.The amount of methane that would have been present in the early atmosphere depends on the balance between its volcanic source and its photochemical sink. The present-day volcanic source of methane is small, but finite: about 1% of the carbon found in midocean ridge hydrothermal vent fluids is CH 4 (Welhan, 1988). (Surface volcanism releases virtually no methane.) The early mantle, however, was probably more reduced than today. Data from 3.3 b.y. old diamond inclusions indicate that the mantle oxygen fugacity was at least 2 log units below the present value (Kasting et al., 1993). A new analysis of mantle siderophile abundances (Walter and Thibeaux, 1995) supports the idea that the primitive mantle was reduced. Thus, it is possible that the CHJCO2 ratio in hydrothermal vent fluids was >1 and that the vents provided a substantial flux of methane to the primitive oceans and atmosphere.We have performed new photochemical model calculations to investigate the fate of methane in the primitive atmosphere. Our calculations indicate that most of the methane would be oxidized, instead of forming hydrocarbon polymers such as those thought to form Titan's haze. Hence, the chance of forming a hydrocarbon smog layer that would have protected the surface from solar UV radiation seems rather remote. We find, however, that the photochemical lifetime of 219

No abstract