An investigation of the stably-stratified atmospheric boundary layer over the Arctic Ocean during stable, clear-sky, winter conditions

The stably stratified atmospheric boundary layer (ABL) occurring over the Arctic Ocean snow/ice pack during clear-sky, winter conditions is examined using data from the Surface Heat Budget of the Arctic Ocean (SHEBA) field experiment, a large-eddy simulation (LES) model and a single-column model (SCM).; Surface heat budgets computed during four predominantly clear-sky periods of the SHEBA experiment indicate that the fluxes of sensible heat into the surface comprise significant portions of the budgets during clear-sky conditions. The cooling of the ABL implied by these fluxes is found to be predominantly balanced by entrainment of heat from above, which provides locally large heating rates at the ABL top due to the interaction of weak subsiding motions with the strong temperature gradients surmounting the ABL. The importance of this vertical transfer of heat to the snow/ice pack heat and mass budgets are discussed.; The ABLs occurring during the periods examined above are investigated further using LES of two case studies taken from the SHEBA winter. The LES data provide insight into the dynamic and thermodynamic processes occurring within these ABLs and the importance of the large-scale subsidence rate, S. Simulations conducted with S = 0 produced marked departures from the observations, while those forced with S = 0.001 and 0.002 ms-1 significantly improved agreement among several ABL parameters.; The ability to adequately represent the effects of the ABL processes investigated above in numerical weather prediction (NWP) models is examined using SCM simulations conducted with four representative NWP ABL parameterization schemes forced with the same data as the LES. None of the ABL schemes were able to consistently match the LES data in many key respects. Simulations conducted with S = 0 produced significant departures in several important ABL parameters from the LES. Simulations conducted with S = 0.002 ms-1 compounded these errors by incorrectly prescribing the evolutions of several key variables under the effects of subsidence. The nature of the errors produced by each of the schemes were investigated further using sensitivity experiments, and some suggestions for improvement are presented.