Was There an Early Habitability Window for Earth's Moon?

Our moon is uninhabitable and lifeless today. It has no significant atmosphere, no liquid water on its surface, no magnetosphere to protect its surface from solar wind and cosmic radiation, no polymeric chemistry, and it is subject to large diurnal temperature variations (e.g., Vaniman et al., 1991; Schulze-Makuch and Irwin, 2008). Thus, associating our Moon with habitability seems outrageous, and certainly it would have been just a decade ago. However, results from recent space missions, as well as sensitive analyses of lunar rock and soil samples, have indicated that the Moon is not as dry as previously thought (e.g., Anand, 2010; Hauri et al., 2017). In addition to the probable occurrence of water ice in permanently shadowed polar craters (e.g., Feldman et al., 1998; Baker et al., 2005; Lawrence, 2017), spectroscopic studies also indicate the presence of hydrated surface materials at high, but not permanently shadowed, latitudes (Clark, 2009; Pieters et al., 2009; Li and Milliken, 2017), with evidence for temporal variations over the course of a lunar day (Sunshine et al., 2009). In addition, recent studies of the products of lunar volcanism indicate that the lunar interior also contains more water than was once appreciated and that the lunar mantle may even be as comparably water-rich as Earth’s upper mantle (see Hauri et al., 2017, for a review). The existence of indigenous sources of water implies that the Moon may not always have been as dead and dry as it is today. Insofar as water is required for habitability (e.g., Kasting et al., 1993; although it is not the sole criterion, see Schulze-Makuch et al., 2011), we can speculatively identify two possible windows for lunar habitability. These may have occurred immediately following the accretion of the Moon and some hundreds of millions of years later following outgassing associated with lunar volcanic activity. Current understanding is that the Moon originated from a gigantic impact 4.5 billion years ago (e.g., Stevenson and Halliday, 2014). The extent to which volatiles were preserved in the Moon-forming debris disk produced by this impact is model-dependent, but impediments to the diffusion of water molecules in a silicate-dominated vapor are expected to result in some water retention in the disk and therefore in the newly formed Moon (Nakajima and Stevenson, 2014; Hauri et al., 2017). The evidence for water concentrations of several hundred parts per million in the mantle source regions of lunar basalts (Hauri et al., 2017; Lin et al., 2017) indicates either that volatiles were indeed preserved during the formation of the Moon or that they were added shortly afterwards by impacting asteroids (e.g., Barnes et al., 2016). Following accretion, the Moon is expected to have been largely molten, with its silicate components existing in the form of a lunar magma ocean (LMO). Such magma oceans are expected to outgas volatiles, leading to the formation of significant transient atmospheres (Elkins-Tanton, 2008). Indeed, Lin et al. (2017) have invoked degassing from the LMO to reconcile the relatively low abundance of water in the post-LMO lunar mantle (at most a few hundred parts per million) with their predictions of much higher values (possibly >1000 ppm) prior to LMO crystallization. On the other hand, some authors have argued that the LMO would have been initially dry following accretion, with the current mantle volatile budget having been added by a subsequent ‘‘late veneer’’ of asteroidal volatiles (e.g., Barnes et al., 2016; Hauri et al., 2017). However, in either case, it appears that significant quantities of water were present in the final stages of LMO evolution. Here, we merely note that outgassing 500 ppm water during the LMO phase (which would be required to bring the higher original values predicted by Lin et al. [2017] into agreement with current estimates) could in principle result in a surface water layer of an order of 1 km thickness. Of course, this would be a very optimistic estimate for the depth of any early lunar oceans—water would only be stable at the surface if protected by a sufficiently dense atmosphere, and significant losses would be expected owing to impact erosion (e.g., Melosh and Vickery, 1989)—but it illustrates how much water might potentially have been available. Needham and Kring (2017) have suggested a second phase of outgassing, and associated peak in lunar atmospheric pressure, as a result of mare basalt eruptions *3.5 billion years ago. Gases derived from lava outpourings may have built up an atmosphere of about 10 mbar, which is above the triple-point pressure of water and about 1.5 times the present atmospheric pressure on Mars (and about 3 times as massive as the current martian atmosphere, given the difference in surface gravities). For comparison with the discussion above, Needham and Kring’s estimated outgassing of water (*10 kg) would equate to a global layer having an average depth of *3 mm.