Evidence for a Rapid Turnover of Argon in the Lunar Exosphere

Abstract

We have developed a numerical model of the Moon's argon exosphere. The results from our simulations are tested using measurements from the Lunar Atmosphere and Dust Environment Explorer (LADEE) and the Lunar Atmosphere Composition Experiment (LACE). We find that the local time of the near-sunrise peak in exospheric density provides a strong constraint on the nature of the surface interactions of the argon molecules. The time of this peak is the same over both the maria and highlands, and can be reproduced by simple surface interaction models with a single desorption energy of 28 kJ mol^-1. The density at all local times of day is also reproduced to within a factor of 2 with the inclusion of a `squirrelling' process, by which particles build up a subsurface population during the night that reappears the following day. We demonstrate that the persistent enhancement in argon density over the western maria in the LADEE dataset cannot be explained by locally changing the surface interactions over the maria, because this inevitably leads to a decrement at other local times of day, as well as shifting the position of the sunrise peak away from that in the highlands. The only possible explanation for this overdensity appears to be that it is driven by a source coincident with the potassium overabundance in the western maria, and that the average lifetime of argon molecules in the exosphere is very brief (~1.4 lunar days, 41 days). This implies high source and loss rates of 1.1*10^22 argon atoms s^-1 , which could require permanent cold traps comparable with the area of permanently shadowed regions, or a more highly localised source. We demonstrate that the long-term variation in the global argon density seen by LADEE could be explained by the presence of large seasonal cold traps, the required sizes of which depend on the area of the permanent traps.