Origin of Rare Gases in the Terrestrial Atmosphere and Planetary Comparisons;

We have previously presented a model for the distribution and transport of rare gases within the Earth [1,2,3,4] which explains the available observational data for mantle He, Ne, Ar, and Xe isotope compositions and provides specific prdctions regarding the rare gas isotopic compositions of the lower mantle, interactions between rare gas reservoirs, and mantle rare gas concentrations. Here we discuss the constraints on the acquisition and evolution of atmospheric rare gases that are derived from this model, and show that these constraints are compatible with planetary evolution processes put forth by others. Early losses of rare gases has occurred on both the Earth and Mars, and a late atmosphere was added to the Earth by accretion of gas-rich materials. Both atmospheres have suffered loss of rare gases, causing elemental and isotopic fractionation, as well as small additions by outgassing of the planetary interior. These processes are reflected in the relationship between the atmospheric and interior rare gases of both planets. The model for the Earth has two mantle reservoirs, an approximately closed system lower mantle and a highly depleted upper mantle that is open to interactions with the lower mantle by mass transport and atmosphere. It is assumed that rare gases in the upper mantle are derived from the mixing of three components: rare gases from the lower mantle, subducted rare gases, and ra&ogenic nuclides produced in situ over a residence time of -1.4Ga. The concentration of each rare gas isotope in the upper mantle is assumed to be in steady state. From the available data regarding the present composition of the upper mantle and the production of 4 ~ e , 2 1 ~ e , 4 0 ~ r , and 1 3 6 ~ e in the upper mantle through decay of 4'k, 2 3 8 ~ , and 2 3 2 ~ h , constraints are obtained on the composition of the lower mantle rare gas component. The lower mantle is assumed to have evolved as a closed system over -4.5Ga, and has been subject to isotopic shlfts from initial values due to decay of 4 k , 2 3 8 ~ , 2 3 2 ~ h , 1 2 9 ~ , and 244Pu. Concentrations of nomadiogenic rare gases in the lower mantle are calculated from the concentrations of parent elements and the constraints on the present isotopic compositions. For isotopes that are not presently generated in the mantle, isotopic variations are attributed to only mixing of atmospheric rare gases and lower mantle rare gases. For example, the upper mantle 1 2 9 ~ e / 1 3 0 ~ e ratio is greater than that of the atmosphere[5], and so the lower mantle ratio is constrained to be equal to, or greater than, the upper mantle ratio and more radogenic than atmospheric Xe. The model treats the atmosphere as a separate reservoir with rare gas isotope compositions that are distinct from those in the mantle and makes no initial assumptions regarding its orign. This differs from the approach of earlier models of mantle rare gas isotope evolution [6] which assumes that the atmosphere was formed by degassing of the upper mantle. In their model, rare gases presently in the upper mantle are then residual to this degassing, with isotopic compositions that reflect the integrated history of degassing and grow-in of radiogenic isotopes. Furthermore, the lower mantle is assumed to have isotopic compositions equal to those of the atmosphere. Other workers have argued that losses of atmospheric rare gases by hydrodynamic escape is the cause of the fractionation of nomadiogenic Xe isotope compositions [7,8] and possibly of Ne isotopes [9] relative to anticipated rare gas sources. Late addtions of gas-rich material have also been invoked [e.g.11,12]. These processes of open system behavior are incompatible with simple models of a closed system Earth. The present model has direct consequences for the evolution of the terrestrial atmosphere and its relation to rare gases retained in the lower mantle: 1. The atmosphere is derived from a rare gas reservoir with distinct radiogenic isotope characteristics and which is no longer represented in the Earth. Xe that has been preserved in the lower mantle since early Earth history has 129~e/13'%e and ' 3 6 ~ e / 1 3 0 ~ e ratios that is more radogenic than those of the atmosphere, and represents a reservoir with greater I/Xe and PuIXe ratios than the source of the atmosphere. 2. The atmosphere has suffered early losses of radiogenic 1 2 9 ~ e and 1 3 6 ~ e [13,14] that was necessarily accompanied by loss of nomadiogenic Xe. This has occurred over timescales of -108yrs. Lower mantle Xe has also suffered losses on a timescale similar to that of the atmosphere [3]. These loses may have been due to the effects of impact of a Mars-size body to form the moon [15,16]. Alternatively, t h ~ s may reflect accretional losses during late accretion of the Earth. The coincidence of the timescale of these losses with the young 'O'~b-~06pb age of the Earth and moon [17,18] suggests that Pb/U fractionation may have also been occurring during these events. 3. Subsequent to these losses, the nonradiogenic atmospheric rare gases were added by late accretion. The source of these gases is constrained to be more gas-rich than the lower mantle and CI chondntes, as these sources have isotope characteristics that are too radiogenic.