Impacts of Spatially-Varying Vertical Mixing in an Arctic Ocean Model
Impacts of Spatially-Varying Vertical Mixing in an Arctic Ocean Model
Abstract:
Numerical models are integral to our understanding of the rapidly changing and difficult-to-observe Arctic Ocean. In models used for a variety of applications, it is common practice to prescribe a single background vertical mixing rate uniformly throughout the model domain. However, recent observational analysis suggests that Arctic mixing rates are highly spatially variable, spanning up to three orders of magnitude between regions characterized by strong vs. weak mixing. In this work, we aim to better understand model sensitivity to spatial variation in the prescribed vertical mixing rate, with an emphasis on large-scale circulation, water mass composition, and Arctic Ocean exports to the subpolar seas. To do so, we perform a number of different experiments with a coarse-resolution model that vary the spatial map of the background vertical mixing rate using both idealized and observationally-informed distributions. We employ a regional configuration of the MITgcm, with 50 vertical levels and an average horizontal resolution of 36 km. The mixing values and distributions we prescribe are derived from a new map of diffusivity estimates that has been constructed by applying a finescale parametrization to historical ice-tethered profiler (ITP) data, and covers a substantial portion of the Arctic Ocean. We find that varying both the magnitude and spatial distribution of vertical mixing imparts significant changes to circulation pathways and water mass composition within the Arctic Ocean. These changes, in turn, modulate the exchange of heat and freshwater with the North Atlantic, and also modify the state of the Beaufort Gyre. Additionally, we observe that the distribution and exported flux of heat appears to be strongly influenced by the mixing rates prescribed on the continental shelves, while those of freshwater respond more strongly to the mixing prescription in the deep basins. A consideration of the implications of our findings for modelling studies and projections of future Arctic Ocean state is ongoing.