Recent observations suggest that high-frequency forcing events have profound influence on the air-sea-ice interactions in the polar region. Studying these events with sea-ice models requires close examination of the model sensitivity that may arise from the high-frequency variability of the forcing. We show that the maximum layer thickness is dictated by the time-scale of the forcing variability and that the computation of the surface temperature develops enhanced sensitivity at high-frequency forcing. We resolve these constraints by developing an "adjustable-layering" thermodynamic formulation for ice and snow that re-computes the number of layers required each time-step to satisfy this maximum thickness, which preserves the total enthalpy and general internal thermal gradients. The conservation equations form a tri-diagonal system ideal for a fast and efficient implicit solution. Furthermore, we resolve the issue of the high sensitivity of the surface flux balance by solving the linearized version of the flux boundary condition simultaneously with the overall conservation system. In this paper we develop the analyses specifying the model requirements, describe the model system and test its algorithmic implementation.

Temperature measurements taken in young, landfast Antarctic sea ice show daily oscillations consistent with heating by solar radiation. We present and solve a heat conduction model for the temperature with a non-linear thermal capacity and a distributed source term for solar power absorption based on Monte Carlo scattering simulations of penetrating photons. We observe two characteristic modes for solar heating in sea ice, one dominated by travelling thermal waves or conduction in the upper half, and the other dominated by in-place solar heating in the lower half. We note that deep thermal responses to solar radiation are larger by a factor of ∼ 10 than predicted by scattering measurements, due possibly to the presence of algae and/or dissolved organic material.

All-wave albedo and spectral albedo data were collected over snow-covered pack-ice floes during summer 1999 (January and February) in the Ross Sea, Antarctica. Temporal variation of the all-wave albedo and spectral albedo was measured from the northern edge to the southern edge of the pack ice along three lines of longitude: 165° W, 150° Wand 135° W. Snow depth, snow-cover stratification, snow-temperature profiles, and grain-size and morphology were also documented. It was observed that daily-averaged albedos were 0.70−0.86 for cloudy conditions over the pack ice. Only two sets of daily-averaged albedo were collected for clear-sky conditions (0.788 and 0:825). Albedo was lower at the marginal edges of the pack ice than in the central pack ice. Albedo was higher over the southern part of the central pack ice than over the northern part. Clear- and cloudy-sky albedos measured on the same site indicate that the average increase in albedo due to clouds is 1.4% (maximum 4.0%). Spectral albedos of the pack-ice floes were similar under clear-sky and cloudy conditions. Both are mainly controlled by the snow grain-size, especially in the top snow layer. All-wave albedos derived from measured visible and near-infrared spectral albedos, with extrapolation to ultraviolet and shortwave infrared regions for both clear- and cloudy-sky conditions, agreed well with all-wave albedos from direct measurements.

The surface radiation budget was continuously measured in the sea-ice zone between 140° E (Terre Adélie) and 180° (McMurdo Sound) close to mid-summer, when the sea ice is disintegrating. These measurements were carried out during a cruise of the USCGC Polar Sea from Hobart, Tasmania, to McMurdo station, Antarctica, in 1998/99. Some of the findings are: the solar radiation is the major atmospheric energy source for the melting of ice. The sun was above the horizon for 24 h for most of the cruise. Due to a high amount of fractional cloudiness, the global radiation was somewhat reduced when compared to areas with lesser cloud cover Mean noon values were around 400 W m−2, while at midnight a value of 30 W m−2 was measured. Daily mean values of the net shortwave radiation varied widely, a function of the reflectivity of the surface, which is strongly dependent not only on the ice concentration, but also on the ice type (e.g. whether it is covered with snow, flooded, melting or dry). Detailed ice observations were carried out. Hourly values of the albedo varied from 6% (open water) to 84% (10/10 sea ice with a dry snow cover). The mean net longwave radiation was only modestly negative. The high amount of fractional cloud cover increased the longwave incoming radiation from the atmosphere. A mean value of −40 W m−2 was measured, which displayed only a very weak diurnal course. The sum of the short- and longwave radiation, the total radiation budget, showed the expected diurnal variation, with slightly negative values at night (for 6 h), and a mean maximum at solar noon of around 220 W m−2. A mean daily value of 98 W m−2 was calculated.

The physical, chemical and biological characteristics of surface and freeboard habitats in the summer pack ice in the eastern Ross Sea, Antarctica, were documented in a continuing effort to determine the factors controlling the distribution, production and succession of sea-ice biota. Three longitudinal transects from approximately 65° to 74° S in the western Ross Sea along 135°, 150° and 165° W were visited where samples of slush and slush interstitial water from surface and freeboard habitats as well as sea water were collected at every degree of latitude. Freeboard and surface habitats, found at all stations in the pack ice, contained a large range (five orders of magnitude) of microalgal biomass (measured as chlorophyll a concentrations) and nutrients ranging from below levels of detection to those of the surrounding sea water. The geophysical attributes of the freeboard habitat (i.e. a layer of semi-consolidated ice overlying a layer containing unconsolidated ice crystals and sea water) are consistent with previous descriptions of this environment. However, additional information is presented on the range of biomass concentrations as well as the small-scale distributions of the habitat and biota.

Changes in the taxonomic composition of diatoms in fast ice as well as in the underlying platelet layer were followed from June to December 1995 in Atka Bay Antarctica. Four communities were clearly distinguished: (1) an interior community dominated by flagellates in spring; (2) a bottom community dominated by a small form of Fragilariopsis cylindrus; (3) a platelet-ice layer dominated by Amphiprora Kufferathii and Thalassionema sp. growing attached to the ice platelets; and (4) an interstitial community dominated by Chaetoceros neglectus growing in the water between the platelets. Species composition distribution and succession in these communities were related to changes in silicate concentration, pore space and ice-formation processes and the ability of some cells to grow attached to ice platelets. The presence of the platelet-layer communities may have implications for paleoglaciology

It has been hypothesized that there are significant dissolved organic matter (DOM) pools in sea-ice systems, although measurements of DOM in sea ice have only rarely been made. The significance of DOM for ice-based productivity and carbon turnover therefore remains highly speculative. DOM within sea ice from the Amundsen and Bellingshausen Seas, Antarctica, in 1994 and the Weddell Sea, Antarctica, in 1992 and 1997 was investigated. Measurements were made on melted sea-ice sections in 1994 and 1997 and in sea-ice brines in 1992. Dissolved organic carbon (DOC) and dissolved organic nitrogen (DON) concentrations in melted ice cores were up to 1.8 and 0.78 mM, respectively, or 30 and 8 times higher than those in surface water concentrations, respectively. However, when concentrations within the brine channel/pore space were calculated from estimated brine volumes, actual concentrations of DOC in brines were up to 23.3 mM and DON up to 2.2 mM, although mean values were 1.8 and 0.15 mM, respectively. There were higher concentrations of DOM in warm, porous summer second-year sea ice compared with colder autumn first-year ice, consistent with the different biological activity supported within the various ice types. However, in general there was poor correlation between DOC and DON with algal biomass and numbers of bacteria within the ice. The mean DOC/DON ratio was 11, although again values were highly variable, ranging from 3 to highly carbon-enriched samples of 95. Measurements made on a limited dataset showed that carbohydrates constitute on average 35% of the DOC pool, with highly variable contributions of 1−99%.

Microbial populations and activity within sea ice have been well described based on bulk measurements from melted sea-ice samples. However, melting destroys the micro-environments within the ice matrix and does not allow for examination of microbial populations at a spatial scale relevant to the organism. Here, we describe the development of a new method allowing for microscopic observations of bacteria localized within the three-dimensional network of brine inclusions in sea ice under in situ conditions. Conventional bacterial staining procedures, using the DNA-specific fluorescent stain DAPI, epifluorescence microscopy and image analysis, were adapted to examine bacteria and their associations with various surfaces within microtomed sections of sea ice at temperatures from −2° to −15°C. The utility and sensitivity of the method were demonstrated by analyzing artificial sea-ice preparations of decimal dilutions of a known bacterial culture. When applied to natural, particle-rich sea ice, the method allowed distinction between bacteria and particles at high magnification. At lower magnifications, observations of bacteria could be combined with those of other organisms and with morphology and particle content of the pore space. The method described here may ultimately aid in discerning constraints on microbial life at extremely low temperatures.

Sea-ice and surface-water samples collected in January-April 1999 in coastal areas in the northern Baltic Sea were analyzed for particle, nutrient and trace-element concentrations and salinity. Stratigraphic analyses of ice cores were also carried out. Bulk nutrient and trace-element concentrations in sea ice fluctuated widely. Nutrient concentrations in sea ice normalized to sea-water salinities showed that sea ice had, almost without exception, an excess of nutrients compared to underlying waters. For phosphorus and phosphate this can be explained by particle incorporation and snow-ice formation, whereas for nitrogen and the sum of nitrite and nitrate snow-ice formation and other mechanisms are important. The levels of Al, Cu, Fe and Ni in the ice were similar to those observed in underlying waters. Pb was observed in detectable concentrations in the ice only. This indicates that sea ice contributes lead to underlying waters during melting, and in some degree also affects the other elements. Furthermore, the observations indicate that incorporation of lead into the ice cover is governed by different processes than for the other elements studied.

It is well established that during sea-ice formation, crystals aggregate into a solid matrix, and dissolved sea-water constituents, including inorganic nutrients, are rejected from the ice matrix. However, the behaviour of dissolved organic matter (DOM) during ice formation and growth has not been studied to date. DOM is the primary energetic substrate for microbial heterotrophic activity in sea water and sea ice, and therefore it is at the base of the trophic fluxes within the microbial food web. The aim of our study was to compare the behaviour of DOM and inorganic nutrients during formation and growth of sea ice. Experiments were conducted in a large indoor ice-tank facility (Hamburg Ship Model Basin, Germany) at −15°C. Three 1 m3 tanks, to which synthetic sea water, nutrients and dissolved organic compounds (diatom-extracted DOM) had been added, were sampled over a period of 5 days during sea-ice formation. Samples were collected throughout the experiment from water underlying the ice, and at the end from the ice as well. Brine was obtained from the ice by centrifuging ice cores. Inorganic nutrients (nitrate and phosphate) were substantially enriched in brine in comparison to water and ice phases, consistent with the processes of ice formation and brine rejection. Dissolved organic carbon (DOC) was also enriched in brine but was more variable and enriched in comparison to a dilution line. No difference in bacteria numbers was observed between water, ice and brine. No bacteria growth was measured, and this therefore had no influence on the measurable DOC levels. We conclude that the incorporation of dissolved organic compounds in newly forming ice is conservative. However, since the proportions of DOC in the brine were partially higher than those of the inorganic nutrients, concentrating effects of DOC in brine might be different compared to salts.

We review the various numerical methods that have been developed for calculating the reflection and transmission of ocean waves at a land-fast ice boundary, including recent developments. While an integral form of the solution, found by the Wiener-Hopf technique, has been known for many years, direct numerical computation of this exact solution has been thought to be prohibitively difficult. Instead, several numerical "matching" procedures have been developed, including some that are only approximate, along with asymptotic solutions based on the integral form. Recently it has been discovered that direct calculation of the integral form is feasible, actually requiring less computation than the matching methods. We outline the actual computations required and contrast each method, and provide examples of computation from the integral form.

By deriving the appropriate Green’s function, a model is developed that allows the interaction of normally-incident, ice-coupled waves with any number of cracks to be studied analytically. For a single crack a simple formula for the reflection and transmission coefficients, R and T, emerges that yields identical results to the computationally intensive work of Barrett and Squire (1996) but is much easier to apply. A crack is found to behave as a steep low-pass filter, allowing long waves through while inhibiting shorter waves, although there is also some fine structure to the response curve. The introduction of more cracks is straightforward. While in that case a formula for R and T is also possible in principle, it is easier to express the result as the solution of a simple matrix equation of order 2N, where N is the number of cracks. It is found that perfect transmission (|R| = 0) occurs at a set of discrete periods, hereinafter called a comb, for N > 1 and that the comb becomes finer as period decreases. For both periodically distributed cracks and ones that are randomly spaced, the gross shape of the response curve remains similar. The results suggest that it is improbable that waves travelling through the Arctic basin can be used as a remote-sensing agent to determine mean ice thickness. The Green’s function technique employed in this paper furnishes solutions to other problems of interest.

It is well known that an incoming ocean swell produces a strain field in a land-fast ice sheet. The attenuation and spectral content of this strain field can be calculated and has been measured. The response of the sea ice to this type of cyclic forcing has also been measured, and in particular we are able to estimate the number of cycles to failure for sea ice loaded at constant amplitude. In this paper we consider the response of the land-fast ice sheet or vast floe to a measured ice-coupled wave field of variable amplitude. We use the Palmgren-Miner cumulative damage law and stress-lifetime curves taken from field experiments to predict the lifetime of the sea-ice sheet as a function of significant wave height and sea-ice brine fraction. Calculations are performed to account for the swell entering a land-fast sea-ice sheet at arbitrary angle, and the influence of c-axis alignment and the presence of pre-existing cracks are discussed.

We present a method for measuring the characteristic length of sea ice based on fitting to a recently found solution for the flexural response of a floating ice sheet subject to localized periodic loading. Unlike previous techniques, the method enables localized measurements at single frequencies of geophysical interest, and since the measurements may be synchronously demodulated, gives excellent rejection of unwanted measurement signal (e.g. from ocean swell). The loading mechanism is described and we discuss how the effective characteristic length may be determined using a range of localized measurements. The method is used to determine the characteristic length of the sea ice in McMurdo Sound, Antarctica.

An experiment to investigate wave-induced floe response and floe-floe interactions was conducted in the Ross Sea marginal ice zone during austral summer 1999. Three types of sensors were used: global positioning system (GPS) receivers; triaxial accelerometers; and compasses. The accelerometer data reveal consistent bands of energy centred at about 0.1 and 1.35 Hz, the latter an unexplained but common feature of such experiments. The GPS data also contain energy near 0.1 Hz, which may suggest that GPS receivers can detect the ocean-wave-induced lateral motion of ice floes.

The propagation of ocean waves in the marginal ice zone (MIZ) is investigated with the aim of determining whether the loading and scattering of waves by ice floes is significant. Measurements made using instrumented ice floes in the MIZ north of the Ross Sea, Antarctica, during June 1998 are used to determine the frequency-wavelength relationship for propagating ocean waves in that region. This measured-dispersion equation is related to the effective large-scale properties of the MIZ that occur in models for wave propagation and scattering. We present the measured wave speeds to enable estimation of the parameters in these models.

From many field observations, it has become well known that pancake ice is ubiquitous in a wave-dominated sea. These strikingly uniform circular floes are consistently found in the Antarctic seas during the ice-formation season. Their presence has also been reported in the Bering Sea, the Greenland Sea and in polynyas and leads within pack ice. In one estimate made by P. Wadhams (personal communication, 1991) the areal coverage of pancake ice in the Southern Ocean marginal ice zone is around 6 × 106 km2. Pancake ice forms through a combination of thermodynamic growth and mechanical thickening, caused by rafting of floes that is driven by wave motion This complex growth process is much faster than pure thermodynamic growth and hence may be the main factor responsible for ice-edge advance in marginal ice zones. In this paper we simulate wave/ice-floe dynamics by combining a new three-dimensional discrete element model with wave hydrodynamics. In one set of simulations we calculate ice-drift speed as a function of wave amplitude and ice collisional dynamics. In a second set of simulations we place a non-reflecting vertical barrier meant to simulate the fast-ice edge. The simulation results show the rate of increase of the ice impact force on the barrier, and the accompanying thickening of the rafted region in front of the barrier, as a function of wave amplitude. Results from this study will help quantify the formation process for pancake-ice covers in a wave field.

It is well known that waves jostle frazil-ice crystals together, providing opportunities for them to compact into floes. The shape of these consolidated floes is nearly circular, hence the name: pancakes. From field and laboratory observations, the size of these pancakes seems to depend on the wavelength. Further aggregation of these individual pancakes produces the seasonal ice cover. Pancake-ice covers prevail in the marginal ice zone of the Southern Ocean and along the edges of many polar and subpolar seas. This theoretical paper describes conceptually the formation process of a pancake-ice cover in a wave field. The mechanisms that affect the evolution of the entire ice cover are discussed. Governing equations for the time-rate change of the floe diameter, its thickness and the thickness of the ice-cover are derived based on these mechanisms. Two criteria are proposed to determine the maximum floe size under a given wave condition, one based on bending and the other on stretching. Field and laboratory observations to date are discussed in view of this theory.

During August 1999, detailed data were collected in the Mertz Glacier polynya along the coast of Antarctica on the drift of newly forming ice tracked with drifting buoys and the ice thickness in the vicinity of the buoys over time. Using these measurements, we estimate the ice-growth rate and the processes which are important in the early stages of ice formation. We find that although there is rapid frazil formation in the open-water areas near the coast and Mertz Glacier Tongue, this frazil ice can take several days to consolidate. A period of warmer weather, when temperatures reached as high as 0°C, delayed the consolidation of the frazil for > 4.5 days. The undeformed new-ice growth rate averaged about 4 cm d-1 for the first 5 days of formation. Ridging and rafting doubled the total growth rate to an average of 8 cm d-1. Blowing and falling snow was incorporated into the surface of the newly forming ice, with 16 of 22 ice cores having some snow in the top few centimeters.

A detailed analysis of hydrographic data from a period of 20 years (1980−99) has shown that the persistent presence of a flaw polynya influences mesoscale hydrography of the Laptev Sea, Russian Arctic. Based on these data, the interannual variability of surface water salinity within the polynya has been estimated. As the salinity increase in the surface water layer is mainly caused by the formation of new ice within the polynya, the average ice-production rate of the polynya was calculated. The results indicate an ice production of 3−4 m per season. A further aim of this study was to calculate the probability that the convective mixing in the polynya penetrates to the sea-floor. It is demonstrated that the probability is maximal in the flaw-polynya area, but does not exceed 20% in the eastern and 70% in the western part of the polynya as a result of strong vertical density stratification from river runoff, especially in the eastern Laptev Sea. Additional studies of water circulation in the marginal zone of the flaw polynya were carried out during field observations in April-May 1999. On the basis of conductivity-temperature-depth and current measurements we deduce that high current velocities (62 cm s-1) recorded in surface waters near the fast-ice edge are caused by a convectively driven circulation system under the polynya. Our measurements indicate that these high-velocity currents are part of a cellular circulation, which results from the rejection of brine during intensive ice formation in the polynya. The observed azimuthal alignment of the crystalline structure of sea ice is also, most probably, the consequence of this quasi-stationary cellular circulation.