Lava-loading of Ice Sheets in a Late Noachian “icy Highlands” Mars: Predictions for Meltwater Generation, Groundwater Recharge, and Resulting Landforms

Introduction: Evidence suggests that cold and dry conditions have persisted on Mars since the mid-to-late Hesperian period [1-3]. While it is likely that the climate conditions of Mars differed during the earlier Noachian period [e.g. 2], the exact nature of the martian climate during this earliest period of martian history remains unclear. Geological evidence has been interpreted by some previous investigators to suggest that the early Mars climate may have supported at least episodic “warm and wet” conditions [e.g. 4]. However, difficulties in reproducing a “warm and wet” early Mars climate with atmospheric General Circulation Models (GCM) [5], due to a faint young sun [6,7], have led others to propose that the early Mars climate may have instead been predominately “cold and icy” [e.g. 8]. Recent global climate modeling efforts predict generally “cold and icy” conditions, where a Late Noachian (LN) water cycle, combined with slight increases in the atmospheric pressure, results in atmosphere-surface thermal coupling, adiabatically cooling high standing areas [9,10]. As a result, water is transported to high elevation areas where cold temperatures preserve deposited ice, forming regional ice sheets that characterize the Late Noachian “icy highlands” (LNIH) Mars climate scenario [10]. However, the GCM predictions of “cold and icy” conditions give rise to a difficult paradox: How could the abundant features indicating at least transient “warm and wet” conditions for early Mars history have formed if the climate was predominantly “cold and icy”? Geological evidence unequivocably shows significant LN fluvial and lacustrine activity. Therefore, climate models which predict “cold and icy” early Mars conditions must fail to include some important processes that would allow for melting. Several studies have investigated the possibility for ice sheet basal melting [11,12] to have allowed for liquid water to be generated in spite of “cold and icy” conditions. However, broadscale ice-sheet basal melting has been shown to occur only at very significant ice thicknesses, geothermal heat fluxes, or through a combination of these factors [11,12]. Additional work has shown that the elevated geothermal heat flux values needed to induce basal melting were likely to have been achieved during the LN on highly localized scales in association with active volcanic features [12]. However, because of the limited scale, the amount of water that can be produced by this “heat-pipe drain pipe” melting is not significant [12]. Are there any other mechanisms for meltwater generation in the context of the LNIH climate scenario? Here we investigate the implications of supraglacial emplacement and loading of lava flows atop LNIH ice sheets for: (1) top-down and bottom-up ice sheet melting, (2) meltwater generation, transport, and fate (2) cryosphere thickness reduction, and ice-saturated cryosphere melting, and (3) groundwater aquifer recharge. We divide our assessment to reflect the time-line of lava flow emplacement and loading by considering: (1) Primary supraglacial lava flow emplacement (those lava flows that are emplaced directly atop the ice sheet), and (2) Long-term loading of lava flows atop the ice sheet. Water Inventory: We assume that the growth of the LNIH regional ice sheets is a supply-limited process (since there is no mechanism to return ice to the lowlands), constrained by a supply 5X the currently observed near-surface/polar water ice budget on Mars (current inventory ~30 m GEL) [13]. Distribution of this water ice reservoir across Mars, above a predicted equilibrium line altitude of +1 km [11], results in average ice sheet thicknesses of ~700 m [11]. Primary and Subsequent Lava Flows: Intuitively, the short-term thermal interaction between a supraglacial lava flow and the underlying ice is dominated by the transfer of heat from the cooling lava to the ice below and the atmosphere above. To determine the amount of heat that is transferred to the underlying ice we implement an analytical solution of the 1-D heat conduction equation [e.g. 14]. Under nominal LN climate conditions, lava flows that are emplaced across the top of the ice sheets will not encounter a surface comprised of solid ice, but rather snow and firn [15], resulting in more rapid melting. Analysis of the heat transfer from lava flows of varying thickness indicates that thinner lava flows contribute a much higher heat flux to the ice, but over a much shorter period of time, melting less ice (Table 1). After the emplacement of the first flow, subsequent flows will contribute much less heat to the underlying ice sheet because of the intervening solidified lava. Superposed lava flows will undergo subsidence equal to the total thickness of ice melted, with the potential for fracturing, and degradation of the lava flow morphology. 1 m 10 m 20 m 10 min 8,200 (27) 900 (3) 500 (1.5) 1 hr 7,000 (22) 900 (3) 450 (1.5)