A model of melt pond evolution on sea ice

Abstract

[1] A one-dimensional, thermodynamic, and radiative model of a melt pond on sea ice is presented that explicitly treats the melt pond as an extra phase. A two-stream radiation model, which allows albedo to be determined from bulk optical properties, and a parameterization of the summertime evolution of optical properties, is used. Heat transport within the sea ice is described using an equation describing heat transport in a mushy layer of a binary alloy (salt water). The model is tested by comparison of numerical simulations with SHEBA data and previous modeling. The presence of melt ponds on the sea ice surface is demonstrated to have a significant effect on the heat and mass balance. Sensitivity tests indicate that the maximum melt pond depth is highly sensitive to optical parameters and drainage.

The refreezing of melt ponds on Arctic sea ice

“…where k i 5 1.16 (1.91 2 8.66 3 10 23 h 1 2.97 3 10 25 h 2 ) W (m K) 21 is the conductivity of pure ice [Schwerdtfeger, 1963;Sakazume and Seki, 1978], k a 5 0.03 W (m K) 21 is the conductivity of air [Weeks and Ackley, 1986], and we have assumed a constant V a 5 0.025 as the fractional volume of air in sea ice [Timco and Frederking, 1996].…”

“…where F LW and F SW are the downward fluxes of longwave and shortwave radiation, e 5 0.99 is the emissivity of sea ice, r 5 5.67 3 10 28 J (K 4 m 2 s) 21 is the Stefan-Boltzmann constant, a is the albedo of sea ice and the lid which are assumed to be identical, I 0 is the fraction of solar radiation penetrating the surface layer (considered to be of negligible thickness), and F sen T 5 T L S ocean ð Þ s and F lat are the sensible and latent heat fluxes [Ebert and Curry, 1993;Maykut and Untersteiner, 1971;Taylor and Feltham, 2004;Bailey et al, 2010].…”

“…While the albedo of ponded sea ice does decrease with pond depth [Taylor and Feltham, 2004], the dominant impact on the area-averaged albedo comes from increasing pond area [Scott and Feltham, 2010]. In the last decade, the relative fraction of FYI has increased in the Arctic, with a 40% reduction in MYI volume over the period 2005-2008[Kwok et al, 2009, and therefore, study of the physics of melt ponds and their proper representation in GCMs has become increasingly important.…”

“…Previous modeling studies of melt ponds have focused on their formation and summertime evolution [Taylor and Feltham, 2004;Flocco et al, 2010Flocco et al, , 2012L€ uthje et al, 2006;Skylingstad, 2009]. In autumn, the ponds refreeze at their upper surface to form an ice lid, which insulates them from the atmosphere and traps pond water between the sea ice and the ice lid.…”

“…The pond water is a store of latent heat that is released during refreezing, and until the pond freezes completely there can be minimal ice growth at the base of the sea ice. A first attempt at modeling the refreezing of melt ponds was described in Taylor and Feltham [2004], although this ignored the role of pond salinity. A simpler version of this approach, essentially treating the ice lid growth as a classic Stefan phase change problem, was incorporated into a climate sea ice model by Flocco et al [2010Flocco et al [ , 2012.…”