Annualmeeting:2017 CSDMS meeting-071
The effects of changing boundary conditions on modelled heat and salt diffusion in subaquatic permafrost offshore of Muostakh Island, Siberia.
[[Image:|300px|right|link=File:]]Geophysical datasets, thermal modelling, and drilling data suggest that most Arctic shelves are underlain by submarine permafrost due to their exposure during the glacial low water stands. The degradation of subsea permafrost depends on the duration of inundation, warming rate, the coupling of the seabed to the atmosphere from bottom-fast ice, and brine injections into the seabed. The impact of brine injections on permafrost degradation is dependent on seawater salinity, which changes seasonally in response to salt rejection from sea ice formation and terrestrial freshwater inflows. The relative importance of the upper boundary conditions responsible for permafrost table degradation rates, however, remain poorly understood. This study evaluates the effects of changing upper boundary conditions on subaquatic permafrost thaw rates using CRYOGRID, a one-dimensional heat diffusion model, which was extended to include coupled dissolved salt diffusion. More specifically, the impacts of using a seasonally varying seabed temperature function compared to a mean annual seabed temperature for both freshwater and saline water bodies were assessed. For saline conditions, the effects of different salinity regimes at the seabed, including mean annual concentrations and seasonal variations. Daily observations of seabed temperature and electrical conductivity from 01-09-2008 to 31-08-2009 offshore of Muostakh Island in Siberia were used to set up the upper boundary conditions for the base case model runs. For saline water bodies, sensitivity analyses for mean annual salt concentrations and seabed sediment type were also performed. In all model runs, a steady-state heat conduction function was used to calculate the initial ground thermal regime prior to inundation. The initial state of permafrost was assumed to contain no salt and the ramp-up time from a terrestrial to a sub-aquatic upper boundary condition was one year for all simulations. Generally, it was found that using a mean annual seabed temperature overestimates subaquatic permafrost thaw for shallow freshwater by approximately 2 metres after 65 years of inundation. Seasonal variation of the seabed temperature led to seasonal freezing and thawing of the sea bed. However, for water bodies with high mean annual concentrations of salt (i.e. 420 moles NaCl/m3), it was found that the difference between using mean annual versus seasonally varying seabed temperatures was negligible. Dissolved salts below the seabed depress the pore water freezing point sufficiently to prevent ice formation in the near-surface sediment despite sub-zero winter temperatures. Given the current trend of freshening in the Arctic Ocean, we expect seasonal freezing of the seabed to be more common for newly submerged permafrost caused by coastal erosion, and thus potentially leading to slower permafrost table degradation rates.