Initiation and growth of martian ice lenses

Water ice in the upper meters of the martian regolith is a major volatile reservoir. Although the geographic extent, burial depth, and thermal stability of this shallow ice are well understood, its origin, history, and stratigraphy are not. Over the past decade, a growing body of observational evidence has indicated that shallow ground ice exceeds the pore volume of its host soil over large regions of both martian hemispheres. This is confounding, given that (1) the physical theory that accurately predicts the location of ground ice also assumes that ice should be pore-filling in the upper meter of regolith, and (2) the Phoenix spacecraft uncovered far more pore-filling ice than excess ice at its landing site in the northern hemisphere. The development of ice lenses by low-temperature in situ segregation - analogous to the processes that generate frost heave on Earth - has been hypothesized to explain shallow excess ice on Mars. We have developed a numerical model of ice lens initiation and growth in the martian environment, and used it to test this hypothesis for the first time. We carried out a large suite of numerical simulations in order to place quantitative constraints on the timing and location of ice lens initiation, and on the magnitude of ice lens growth in a variety of host soils. We find that ice lens initiation is a ubiquitous process in the martian high latitudes, but the ultimate magnitude of lens growth, or frost heave, is sensitive to the properties of the host soil. Depending on the specific properties of martian soils, in situ segregation may be a very slow process sufficient to explain the excess ice observed in the Dodo-Goldilocks trench at the Phoenix landing site, but without regionally significant effects. Alternatively, if clay-sized particles or perchlorate salts are present, in situ segregation may be a vigorous process that has significantly affected the stratigraphy of ground ice in the upper meter of regolith throughout the high latitudes.