Mercury and the Moon are believed to have large reservoirs of ice within the permanently shadowed cratered regions near their poles. Given that Mercury's surface area is only twice that of the Moon, if all delivery/preservation processes were equal, we might expect the worlds to have comparable quantities of ice. They do not. The ice reservoir for Mercury is thought to be between 2e16 g to 1e18 g, while those for the Moon are only ~2e15 g, at least an order of magnitude lower.
Here we examine whether the delivery of water via asteroids, comets, or micrometeorites are plausible sources for the ice reservoirs found on Mercury and the Moon. Our assumptions for our modeling work are as follows. (i) Comets are composed of ~50% water ice. (ii) Primitive asteroids and micrometeorites (i.e., CM, CI, CR composition) are ~10% water, locked up in the form of hydrated silicates. (iii) The number of potential water-delivery impact events found on the Moon and Mercury are constrained by the number of Copernican and Eratosthenian craters ( 10 km. (iv) The impact probabilities, impact velocities, and physical properties of the impacting asteroids and comets hitting Mercury and the Moon for the last 3 Gyr are reasonably approximated by the NEO model discussed in Granvik et al. (2016; Nature). (v) The nature of the micrometeorite population is set by the modeling work of Nesvorny et al. (2009); they estimate that 1600 metric tons of MM hit the Moon per year, and 7 times this flux hits Mercury per year. Note that these rates are high enough to explain why ~2% of lunar regolith is CM-like and why most Antarctic micrometeorites are CM-CI-CR-like. (vi) High velocity impacts are energetic enough to drive off most water vapor after a large impact event; the fraction retained at given velocities is set by the estimates provided in Ong et al. (2010). (vii) The fraction of water retained by the Moon and Mercury from small, low velocity MM impacts is likely to be close to 100%. (viii) Only ~4% of all water delivered to Mercury and the Moon reaches a permanently shadowed region (e.g., Moores 2016).
If these assumptions are valid, our model results indicate that asteroids and comets are marginal players in the delivery of volatiles to Mercury and the Moon. On the other hand, MMs may have delivered >2e16 g and 2e15g to Mercury and the Moon, respectively, over the last 3 Gyr. These values are the right order to explain the observed abundances and the mismatch in ice abundance between the worlds. If more are needed, consider that additional volatiles could also be delivered if we were to relax our 3 Ga timeline (e.g., MMs deliver ~1e21 g of MMs to Mercury and 2e20 g to the Moon during the late heavy bombardment).