What is the Role of the Sensible Heat Flux on the Surface Heat Budget of Multi-Year Sea Ice? (INVITED)

Covariance measurements during the SHEBA project by the Atmospheric Surface Flux Group (ASFG) indicate that the mean (approximate 95% confidence) annual turbulent sensible heat flux at the main tower site was -2.1 (0.2) Wm^-2 (negative downward, i.e. surface heating). Monthly means ranged from a minimum of -8.0 Wm^-2 (0.3) in February to a maximum of 2.4 (0.3) Wm^-2 in August. Standard deviations of hourly averages ranged from about 11 Wm^-2 in winter to 6 Wm^-2 in summer. Although these values seem small compared to typical radiative fluxes of 100's of Wm^-2, the sensible heat flux over thick ice is nonetheless an important parameter for several reasons. The radiative fluxes tend to cancel in the mean. For example, the SHEBA annual mean net surface radiation value of only 2.6 (1) Wm^-2 is comparable in magnitude to the sensible heat flux mean. Hourly fluctuations in sensible heat flux often partially negate surface cooling or heating by radiation; in April over 30% of the hourly surface net radiation was counteracted by surface sensible heat fluxes. The sensible heat flux is a key parameter in terms of lower atmospheric dynamics and stability because it represents the primary means of heat transfer between the surface and the air just above. Sensible heat flux "events" can be divided into two types: advective and radiative. During advective events, temperature changes in the air above the surface lead temperature changes of the surface proper. During radiative events, the surface temperature leads. Each of these types of events are associated with time scales that will be quantified using the SHEBA results. Differences in sensible heat flux values between the 3 m and 8 m tower levels, implies a divergence of heat by turbulence during the winter and a convergence of heat during summer, suggesting that winter heating and summer cooling from leads does not entirely reach the upper tower levels. The key parameter related to sensible heat flux representations in numerical models is the value of the sensible heat transfer coefficient, C_H. A preliminary estimate of the annual median 10-meter CH during SHEBA for near-neutral conditions is 1.4 (0.3) x 10^-3. The large confidence interval is due to uncertainties in the surface temperature measurements and questions about the validity of the covariance method during highly-stable situations. Current work is underway to improve this estimate and to quantify stability effects. These results will benefit modeling efforts by providing quantification of sensible heat flux parameterizations, as well as demonstrating the types of heat flux behaviors that should be reproduced in numerical simulations.