Thermal Modeling of Permafrost Melt by Overlying Lava Flows with Applications to Flow-associated Outflow Channel Volumes in the Cerberus

Introduction: The Cerberus region of Mars has numerous geologically recent fluvial and volcanic features superimposed spatially, with some of them using the same flow channels and apparent vent structures [1, 2]. Lava-water interaction landforms such as psuedocraters suggest some interaction of emplacing lava flows with underlying ground ice or water (e.g. [1, 3]). This study investigates a related interaction type a region where the emplaced lava might have melted underlying ice in the regolith, as there are small outflow channel networks emerging from the flank flows of a lava shield over a portion of the Eastern Cerberus Rupes. Specifically, we use high-resolution Mars Orbiter Laser Altimeter (MOLA) topography to constrain channel and flow dimensions, and thus estimate the thermal pulse from the emplaced lava into the substrate and the resulting melting durations and refreezing intervals. These preliminary thermal models indicate that the observed flows could easily create thermal pulse(s) sufficient to melt enough ground ice to fill the observed fluvial small outflow channels. Depending on flow eruption timing and hydraulic recharge times, this system could easily have produced multiple thermal pulses and fluvial releases. This specific case suggests that regional small water releases from similar cases may be more common than suspected, and that there is a possibility for future fluvial releases if ground ices are currently present and future volcanic eruptions in this young region are possible. Data: We use both MOLA profiles for detailed measurements, as also regridded the MOLA data to construct a local crossover-corrected topography grid at 128 pixels/deg. longitude by 256 pixels/deg. latitude using the approach of Neumann et al. [4]. We use a suite of tools within the IDL-based Gridview program [5] to measure parameters such as channel depths and volumes and flow heights and volumes, which are input as constraints into the thermal models. Channel volumes. The channel volumes are measured with profile and gridded data. Since they are locally V-shaped, multiple cross-sectional areas are sequentially integrated from cross-section to crosssection to obtain a total volume. For the small channel system emerging from under the flows we estimate a total volume of 7.9 x10 m ± 62% Lava volumes. The lava volume is integrated from multiple profiles with the assumption of a planar surface (best-fit to the shield margin locations) prior to emplacement of the shield, with the embayed (older) high standing terrains removed from the shield volume calculation. . We estimate a lava volume of 7.66 x10 m ± 25% for the shield segment directly over and upslope of the channel network. Modeling Approach and Results To estimate the depth and intensity of the thermal pulse(s) over time due to the lava flow emplacement, we use a wellknown solution to the unsteady heat conduction equation [6]. which assumes that the lava is far enough from its source (here assumed to be the volcano summit) that it is not actively flowing. The thermal pulse as a function of depth T(z) is then described by STUDY AREA