Thermal regime and sublimation rates of subsurface ice in Antarctica and on Mars under current and past environmental conditions: Insights from modeling and ground data

Subsurface ice is common in regions where mean annual temperatures are below the freezing point of water. Once present, this ice may persist for long periods of time even under unfavorable environmental conditions. Subsurface ice is fundamentally important as a source of water for plants and microbes, a proxy for paleoclimatic information, and a major component controlling the periglacial landscape. This dissertation examines the stability and persistence of ground ice both in the Dry Valleys of Antarctica, a small part of the continent that is not covered in ice, and at the equatorial region on Mars. This research was initially motivated by claims that the ice buried under several decimeters of dry regolith in Beacon Valley is more than 1 Ma old and perhaps over 8.1 Ma, thereby making it the oldest known ice on Earth. My study uses numerical modeling to determine the short-term and long-term rates of sublimation of ground ice; the latter bears directly on the longevity of subsurface ice. The modeling utilizes extensive soil and environmental data from a well characterized site in Beacon Valley. One component of the model is to reconstruct subsurface temperatures based on measured surface temperatures. It is validated using ground temperature measurements. Both thermal and sublimation components of the model are then coupled, along with environmental data from the Curiosity Rover, to numerically simulate the sublimation of ground ice in Gale Crater, and to consider the potential for ground ice to have persisted there for long periods of time.;The contemporary sublimation rate of ground ice in Beacon Valley, Antarctica is modeled using a vapor diffusion model constrained by 12 years of climate and soil temperature data, and field data of episodic snow cover and snowmelt events that have not been represented in previous models of the ground ice sublimation. The model is extended to reconstruct the sublimation history over the last 200 ka using paleotemperatures estimated from ice core data from nearby Taylor Dome, and a relationship between atmospheric temperature and humidity derived from our meteorological records. This study provides a realistic estimate of the long-term sublimation history in Beacon Valley. Sublimation occurs throughout the modeled period; however, the rates are slow enough that the residual buried ice is likely older than 1 Ma.;I also studied the subsurface thermal regime and soil thermal properties in Beacon Valley and, using temperature dependent thermal properties, numerically solved the heat diffusion equation using the finite volume method. The modeled temperatures approximate closely the measured temperatures at all depths with average differences ranging from 0.01°C to 0.03°C. The latent heat from documented episodic snowmelt events or modeled changes in ice content due to condensation or sublimation have negligible effects on the temperature. This study validates that our model is applicable to ground conditions in the cold, dry environments in general, including other regions in Antarctica, high mountain ranges in the central Asian and South America, and on Mars.;The heat and vapor diffusion models developed for Beacon Valley were then applied to Gale Crater, Mars to study the ground temperature and stability of ice that formed potentially during the last high obliquity phase of Mars. This study of subsurface heat and vapor transport is made use of one year of extensive ground based measurements by Curiosity?s Rover Environmental Monitoring Station (REMS). As in Beacon Valley, ground ice is currently unstable along the Curiosity traverse; furthermore, also analogous to Beacon Valley, Gale Crater may have relict ground ice from as recently as ~0.5 Ma during the last high obliquity period. Similarly, modeling the sublimation rates, adjusted for Mars conditions, shows that water vapor loss deceases increasingly with depth as overlying dry regolith thickens. My study suggests ground ice that may have formed ~0.5 Ma ago could persist within meters of the surface; this finding may be important for future mission considerations. Overall, my studies advance our understanding of the thermal regime and long-term stability of subsurface ice in frigid, hyperarid environments.