C51C-01 INVITED 08:00h
The Phase Behavior of Sea-Ice
The basic tenets of bulk and interfacial thermodynamics, equilibrium and nonequilibrium statistical mechanics, and transport theory are standard tools in many areas of physical science and engineering, and are essential for any attempt at explaining the phase behavior of sea-ice and the crystal growth phenomenon that characterize it on a macroscopic scale. Because the effective (homogenized) properties of sea-ice underlie its large scale dynamic and thermodynamic properties a rigorous predictive theory is warranted. In this talk, I describe where these properties come from, how we can study them in a laboratory setting, and how these properties manifest themselves in the polar regions. Due to the fact that at least some aspects of nearly all of the phase change problems we observe in sea ice are found in a host of other systems, I will emphasize the connections between floating ice and other materials of geophysical and technical interest.
C51C-02 INVITED 08:15h
Thermal Conductivity of Sea ice: Measurement and Modelling Parameterisation
The thermal conductivity is a fundamental parameter in sea ice modelling, but due to measurement difficulties is usually represented by simplified parameterisations or a constant value. We report new results from our program to measure the thermal conductivity of sea ice ($k_{si}$), new insight into previous results, and a comparison of our results with the common parameterisations used in sea ice modelling. The measurement is based on two different experiments. The first is based on temperatures recorded with thermal arrays in first year sea ice in McMurdo Sound, Antarctica and off Point Barrow, Alaska. From time-series analysis of these temperatures we have determined the depth- and temperature- dependence of ksi. Previous such analysis has suggested an unexpected decrease in the conductivity of first year ice of 25-50% over the top 50cm. We also report on a second, direct method for measuring the thermal conductivity of small sea ice cores in which we observe no difference in the conductivity of surface (0-10cm) and sub-surface (45-55cm) ice. Furthermore, we show that analytical artefacts might explain the apparent near-surface reduction in the array measurements, as well as other observed features. From our measurements we resolve the temperature dependence of $k_{si}$(T), and compare it with theoretical models and the 1964 parameterisation of Untersteiner which is commonly used in sea ice models. We discuss that scheme and possible amendments.
C51C-03 08:30h
Role of Heat Flux Through Snow in Antarctic Sea Ice Growth
The heat flux through snow on Antarctic sea ice was calculated to examine its role in controlling sea ice thickness and ice types. A statistical determination of the heat flow through the snow was calculated using over four thousand point temperature measurements of snow and sea ice from four cruises in the Ross, Amundsen, and Bellingshausen Seas. Calculated heat flux through the ice was used to determine an effective thermal conductivity for the snow. Two approaches were taken: (1) calculation base on the temperature profiles at individual snow pit and ice core sites, and (2) calculation based on point temperature measurements across ice floes. The effective thermal conductivity determined from the first method was 0.29 W m$^{-1}$ K$^{-1}$, and the second method was 0.30 W m$^{-1}$ K$^{-1}$, in close agreement with values typically used in large-scale sea ice models, but approximately twice the measured values for the same cruises. There was also significant variation between cruises, indicating a possible dependence on snow properties that is not captured by the simple snow parameterizations used in current large-scale sea-ice models. Analysis shows some dependence of effective thermal conductivity on snow depth. This may result from snow property variations but may also indicate a variation in heat flow mechanisms that depend on the heterogeneity of the snow cover. Unlike at the SHEBA site in the Arctic, the analysis showed no evidence for enhanced heat flow through ridges. We suggest that the discrepancy between measured and effective thermal conductivity may be due to the presence of icy layers within the snow cover, and brine at the base of the snow, both of which can enhance the heat flow. As these properties are controlled by environmental factors, significant variability in vertical heat transfer can result, as evidenced by the variability among cruises. This suggests that to accurately predict the response of Antarctic sea ice to environmental change may require more sophisticated treatment of processes within the snow cover. Finally, using a sophisticated one-dimensional thermodynamic model of sea ice, we investigate how variations in these snow processes affect the thickness of the underlying sea ice and snow ice production.
C51C-04 08:45h
A model of melt pond evolution on sea ice
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 2 stream radiation model is used with a parameterisation of summer-time evolution of optical properties. The heat transport within the sea ice is modelled using the equations describing a mushy layer. The model is tested by comparison with SHEBA data and previous modelling. The maximum melt pond depth is found to be highly sensitive to optical parameters and drainage.
C51C-05 09:00h
Transmission of Solar Radiation Through Summer Melt Ponds
During the summer melt cycle, the sea ice cover exhibits large changes in the amount of solar radiation backscattered to the atmosphere, absorbed, and transmitted to the ocean. These changes in radiative partitioning are, in part, driven by changes in the optical properties of the surface. Relatively high albedo areas of bare ice and remnant snow and relatively low albedo areas of open water and surface ponding create a spatially inhomogeneous surface with significant optical property variability on meter to decameter scales. Understanding how the complex surface partitions radiation requires understanding how individual surface types partition radiation. Measurements of spectral albedo, transmittance, and extinction were made at bare and ponded multiyear sea ice sites during the SHEBA summer. A radiative transfer model is used to simulate the optical property observations. Modeled and observed spectral albedos were compared and used to infer vertical profiles of scattering coefficients for each site. For bare multiyear ice, scattering coefficients were generally between 1 and 10 cm-1 (assuming g = 0.94) for the uppermost 10 -30 cm and between 0.25 and 0.9 cm-1 within the ice interior. For ponded multiyear ice, scattering coefficients were approximately uniform with depth and generally between 0.1 and 0.2 cm-1. The inferred inherent optical properties are then used in the radiative transfer model to predict spectral transmittance and extinction. For bare ice, the good agreement between observed and modeled transmittance suggests that the optical property observations are consistent. For ponded ice, the transmittances predicted by the model were significantly larger than those observed. Although the inferred model has some uncertainty, it is highly likely that the measurements are biased by the horizontal inhomogeneity of the ice cover. Monte Carlo radiative transfer simulations performed on an azimuthally symmetric domain indicate that transmittance measurements made under the center of ponds with diameter less than approximately four times the ice thickness can be strongly influenced by the optical properties of the surrounding ice. When the surrounding ice has higher extinction, transmittance measurements made at specific locations beneath ponds may underestimate the amount of shortwave energy reaching the ocean. The radiative transfer model is used to estimate and correct these biases in the measurements.
C51C-06 09:15h
Sea Ice Records Spatial and Temporal Variations in Arctic Surface Waters
The surface hydrography of the Arctic Ocean interior can be reconstructed from oxygen isotopic analysis of drifting Arctic sea ice, coupled with back trajectories of ice drift and an ice growth model. The results compare well with values obtained by traditional oceanographic methods and known water mass distributions. Elevated oxygen-18 in ice core profiles reflects growth from Atlantic Water in Barents Sea ice and from the Pacific inflow in floes that that transited the Chukchi Sea. Sea ice grown on the Siberian shelves and while drifting in the Beaufort Gyre has depleted oxygen-18 signatures reflecting river runoff. Ice grown in the Transpolar Drift has an intermediate Atlantic/Beaufort Gyre composition. Future spatial and temporal variations in the surface waters of the ice-bound Arctic Ocean interior, could be estimated by analysis of sea ice floes sampled at exit locations, for example, Fram Strait, and the Barents and Beaufort seas.
C51C-07 09:30h
A Mechanism for the High Rate of Sea-Ice Thinning in the Arctic Ocean
Submarine measurements of sea ice draft show that the ice has thinned in some parts of the Arctic Ocean at a remarkably high rate over the past few decades. The spatial pattern indicates that the thinning was a strong function of ice thickness, with the greatest thinning occurring where the ice was thickest initially. A similar relationship between sea ice thinning and the initial thickness is reproduced individually by three global climate models in response to increased levels of carbon dioxide in the models' atmosphere. All three models have weak trends in their surface winds and one model lacks ice dynamics altogether, implying that trends in the ice circulation are not necessary to produce a relatively high rate of thinning over the central Arctic or a thickness change that increases with the initial thickness. I will present a general theory to describe the thinning of sea ice subject to climate perturbations. Results from numerical models and theory alike indicate that the leading component of the thickness dependence of the thinning is likely due to the basic thermodynamics of sea ice. When perturbed, sea ice returns to its equilibrium thickness by adjusting its growth rate. The growth--thickness relationship is stabilizing and hence can be reckoned a negative feedback. The feedback is stronger for thinner ice, so thinner ice recovers more quickly from perturbations than thicker ice. I will discuss the implications of this feedback for recent observed changes in the ice thickness, including the reduced frequency of multiyear and ridged ice and increased frequency of firstyear ice.
C51C-08 09:45h
Sea ice properties and processes in a warming Arctic
The Arctic sea ice cover is changing. Over the past few decades there has been a marked decrease in the areal extent and thickness of multiyear ice. Open water fraction in summer has been increasing, with corresponding increases in first-year ice fraction during the remainder of the year. Such changes have a profound impact on air-sea-ice interactions in the Arctic. For example, there is increased input of solar heat to the ocean that is producing a positive feedback on ice formation and decay, as well as affecting biological activity in the ice and ocean. Recent observations in the Chukchi and Beaufort Seas help to illustrate the magnitude of the environmental changes. Work during the SHEBA Program has shown substantial thinning of second-year and multiyear ice in this region, with almost half of the total summer melt occurring at the base of the ice due to heat transfer from the ocean. Surface hydrography and tracer studies indicate that solar heating of the mixed layer, enhanced by a more mobile ice cover with larger fractions of leads and open water, plays the major role in ocean-ice heat transfer. Future sea ice research in the Arctic needs to focus strongly on the consequences of these and other ongoing changes. To do so will require interdisciplinary studies that combine field observations and large-scale modeling efforts. Such studies and a more fundamental understanding of the underlying processes are needed to assess potential changes in western Arctic shelf ecosystems and the related impacts on coastal communities.