Temperature and pressure changes within subglacial sediments caused by déglaciation favor a phase change from methane hydrate to methane gas. If released to the atmosphere, this gas could have contributed significantly to the increased atmospheric methane concentration during interglacial periods. Using a numerical reconstruction of sediment temperatures and ice-sheet loads, the total sediment volume in the Eurasian Arctic that is subject to this phase change is estimated to be 2.8 × 1014 m3.

Spatial representativeness and an understanding of controls on chemical species distribution are essential requirements of any significant ice core investigation. Snowpit studies provide an essential tool in this process. In preparation for the central Greenland deep drilling effort a series of snowpits was sampled in detail for oxygen isotopes, major anions, major cations, total acidity and radionuclides. The results of this sampling program are used to define: (1) the chemical composition of the snow in the region, (2) the input timing and spatial distribution of major chemical species, (3) the potential dependence of species concentration on accumulation rate, and (4) the signal characteristics identifiable in the region over the last few years.

Measurements at Wolverine Glacier, Alaska, from 1968 to 1988 indicate unsteady increases of air temperature and precipitation since the early 1970s. These increases were due almost entirely to changes in winter. Variations in annual temperature and precipitation at Wolverine Glacier and at Seward, a nearby climatological station at sea level, correlate positively with global temperature variations and are in general agreement with the changes at high latitudes predicted by five recent general atmospheric circulation models forced by anticipated rises of CO2.

A consequence of the air temperature and precipitation increases at Wolverine Glacier was a change to a generally positive mass balance after 1976. Although these observations in the coastal maritime climate of Alaska run against the common, oversimplified notion that in a warming climate glaciers will melt, causing sea level to rise, they are logical and easily understood when the sensitivity of the glacier to the seasonal distribution of the changes is considered. The observed seasonal changes at Wolverine Glacier also are in agreement with global climate models. Snow precipitation and glacier accumulation increased, but at the same time warming affected only these those temperatures below about −5°C, and melting was not altered. The extent of this response is not well known, but the process may be taking place in other important glacierized regions.

The dielectric stratigraphy of a 130 m ice core from Dolleman Island, Antarctic Peninsula, shows large variations in the dielectric relaxation process and in conductivity. A comparison with the chemical stratigraphy of the core demonstrates the decisive role played by both acids and salts in determining the electrical behaviour of natural ice. The dielectric response is sensitive both to the type of impurity and to its distribution within the ice fabric. The evidence supports other observations of the localization of sulphuric acid at three-grain boundaries: in contrast, the salt impurity appears to be largely incorporated into the ice lattice. The overriding importance of the dielectric profiling technique is that it is the only profiling tool so far developed that is sensitive to the presence of salt in polar ice cores.

The 550-year records of δ18O and dust concentrations from Siple Station, Antarctica suggest warmer and less dusty atmospheric conditions from 1600 to 1830 A.D. which encompasses much of the northern hemisphere Little Ice Age (LIA). Dust and δ18O data from South Pole Station indicate that the opposite conditions (e.g. cooler and more dusty) were prevalent there during the LIA. Meteorological data from 1945–85 show that the LIA temperature opposition between Amundsen-Scott and Siple, inferred from δ18O, is consistent with the present spatial distribution of surface temperature. There is some observational evidence suggesting that under present conditions stronger zonal westerlies produce a temperature pattern similar to that of the LIA. These regional differences demonstrate that a suite of spatially distributed, high resolution ice-core records will be necessary to characterize the LIA in Antarctica

Shallow corings and pit studies were carried out in 1987 on the Chongce Ice Cap in the West Kunlun Mountains which lie along the southern edge of the Taklamakan Desert. Stratigraphic observations, as well as measurements of insoluble particles, chemical constituents and liquid conductivity, on the core and pit samples allowed identification of annual layers, although visible dirt layers are not always annual signals. A vertical profile of stable isotope content showed a periodic variation near the surface, strongly attenuated with depth.

Ice sheets are always advancing or retreating, and these processes are not uniform along the ice edge. The result is that lines of maximum and minimum extent are envelopes of successive instantaneous ice edges. Generically, the envelopes can possess only two kinds of special points: crossings and cusps. These will often be arranged in characteristic patterns (unfoldings of higher mathematical singularities). The recognition of these features makes possible a novel approach to the interpretation of field evidence. It is shown, for example, how separate families of parallel erosional and depositional features seen at an interior site are related by coming together in lines of cusps at the envelopes, which may be marked by terminal moraines.

Landsat-5 Thematic Mapper (TM) and SPOT data collected two years apart from an identical area of Dronning (Queen) Maud Land, Antarctica, have been analyzed to detect variations in surface features that may signal climatic change, and to establish a technique that readily identifies such changes. We found that selective principal component analysis (Chavez and Kwarteng 1989), on band ratios of near-IR/green, highlights changes in blue ice areas. The formation and preservation of blue ice is poorly understood, but we suggest that it generally takes longer to increase a blue ice area than to decrease it, and that blue ice extent is most sensitive to changes in accumulation rate. The investigated blue ice area shows a decrease in extent over the two-year period caused by incursion of snow that probably resulted from an increase in accumulation rate. Comparison of two TM images collected 18 days apart shows that transitory snow drifts have little effect on blue ice extent.

Acid fallout from volcanic eruptions is well documented in the Greenland and Antarctic ice sheets (Hammer and others, 1980; Hammer, 1984; Legrand and Delmas, 1987). However, to date, no volcanic ash (tephra) layers have been positively identified in association with any of the high electrical conductivity layers that mark the volcanic acid deposition. In this paper we report the results of a study of the chemical composition of insoluble microparticles filtered from five intervals of a core from the South Pole. These five intervals were identified by Kirchner (1988) as being due to volcanic fallout, on the basis of electrical conductivity and sulfuric acid measurements.

The major element composition of tiny (<5 µm) glass shards found in these layers was determined and compared with analyses of volcanic ash from known eruptions or from volcanic sources suspected of having produced the fallout. Glass shards from volcanic eruptions of both local (Antarctic and sub-Antarctic) and of global (Indonesian/South American) importance have been identified in this study.

Records from the expeditions of Cook, Bellingshausen, Wilkes, and Ross in the late 18th and early 19th centuries have been examined for the information they provide on locations of the Southern Ocean sea-ice edge during the period of the late Little Ice Age in much of the Northern Hemisphere. When these locations are compared with satellite-derived ice edge locations in the mid 1970s, there is a suggestion of particularly heavy ice covers in the eastern Weddell Sea in December 1772, in the Amundsen Sea in March 1839, and perhaps, on the basis of an isolated observation, in a portion of the western Weddell Sea in January 1820. However, overall no strong Little Ice Age signal is found for the sea ice of the Southern Ocean. Many of the observations from the four expeditions indicate sea-ice edge locations that lie within the range of ice edge locations at the same time of year in the mid 1970s, and a few of the observations suggest a less extensive ice cover than in the 1970s.

Comparison of monthly averaged sea-ice distributions in the Sea of Okhotsk with atmospheric pressure data during the four winters having passive-microwave sea-ice coverage from the Nimbus 5 satellite, 1973–76, revealed a strong apparent relationship between the extent of the sea-ice cover and the influence of the Siberian High atmospheric pressure system. Examination of data for the years 1978–86, having passive-microwave coverage from the Nimbus 7 satellite, reveals that the strong correspondence found for 1973–76 between Okhotsk sea-ice extents and the Siberian High was not maintained in the 1978–86 period. A weaker correspondence continued, however, between the sea ice and the combined Siberian High/Aleutian Low system. A Siberian High/Aleutian Low index was created, and the correlation coefficient between that index and sea-ice extents in the midwinter month of February is 0.97 for the 1973–76 period and 0.52 for the 1978–86 period. Primary reasons for the lack of a consistently strong monthly averaged ice/atmosphere correspondence are: the various oceanographic influences on the sea-ice cover, the failure of monthly averages to reflect fully the important shorter-term interactions between the ice and the atmosphere, and the fact that ice conditions in one month are influenced by ice conditions in previous months.

The lower latitude extension of polar water during Quaternary glaciations has two opposing effects on the mass balances of adjacent maritime ice sheets. Cooler air temperatures reduce ablation and increase the fraction of precipitation falling as snow, but also lead to reduced atmospheric moisture content and reduced precipitation. The effects of these contrasting processes on ice-sheet initiation are investigated using a coupled ice-sheet-atmospheric moisture model of the Loch Lomond ice sheet in Scotland (10 000 a B.P.). There is a delicate balance between the degree of cooling required to initiate ice accumulation and that which restricts the flow of moisture over the area. At the regional scale of Scotland, the combined effects of orographic enhancement and the inherent feedback between the rising ice-surface elevation and increasing mass balance are dominant, and ice-sheet growth accelerates. Comparison of the results from models using different wind directions suggests that south-westerly winds were prevalent during the Loch Lomond glaciation in contrast to the dominant westerly winds of the present day. The modelling experiments demonstrate the sensitivity and complexity of the links between ocean surface cooling and ice-sheet growth, particularly at the early stages of a glacial cycle.

Identification of present-day climate setting and alpine glacier-balance gradients indicates that the balance gradient of alpine glaciers is primarily determined by climatic conditions. Determination of balance gradients for specific climatic settings on present-day ice sheets provides an analog for determining the mass balance on paleo and future ice sheets.

Sea ice covering the polar oceans is only a thin veneer whose areal extent can undergo large and rapid variations in response to relatively small changes in thermal forcing. Positive feedback between variations in ice extent and global albedo has the potential to amplify small changes in climate. Particularly difficult to model is the summer decay and retreat of the ice pack which is strongly influenced by shortwave radiation entering the upper ocean through leads (Iw). Most models assume that all of this energy is expended in lateral melting at floe edges. In reality, only a portion of Iw contributes directly to lateral melting, with the remainder going to bottom ablation and warming of the water. This partitioning of Iw affects not only the magnitude, but also the character of the predicted ice decay, reducing the change in ice concentration and enhancing the thinning of the ice and the storage of heat in the water. In this paper we present an analytical model which includes many of these processes and is stable regardless of time step, making it suitable for use in climate simulations.

Asynchronous coupling schemes between ice sheet and atmospheric forcing models are evaluated for use in long-term ice-age simulations. In these schemes the ice sheet and atmospheric forcing are run together for short synchronous periods (Ts), alternating with longer asynchronous periods (TA) during which the ice sheet is run with atmospheric information extrapolated from the previous synchronous period(s). Two simple ice-sheet models are used that predict ice thickness as a function of latitude, and the atmosphere is represented by a prescribed pattern of net annual accumulation minus ablation. The pattern is shifted vertically to represent long-term orbital variations, stochastic inter-annual weather variability and ice-sheet albedo feedback.

Several asynchronous schemes are evaluated by comparing results with those from fully synchronous runs. The best overall results are obtained using a scheme in which the forcing during each asynchronous period is linearly extrapolated from its means in the previous two synchronous periods. Differences from the synchronous results are caused primarily by poor sampling of the stochastic forcing component, which exaggerates the stochastic ice-sheet fluctuations. We examine how these errors depend on Ts and TA, and outline implications for GCM ice-age simulations.

A method of determining the temporally varying “state” of the ice cover (the concentrations of three surface types: open water, first-year ice, and multi-year ice) is presented. The methodology is that of Kalman smoothing: a physical model and a measurement model are used to blend satellite passive microwave data and buoy data to give an optimal estimate of the ice state. The estimates are optimal only to the degree that model parameter values are known and assumptions about variances are met. Uncertainty about these values and assumptions, and lack of independent data with which to compare results,leaves self-consistency as the most important test of results. A four-year record (1979-82) of the estimated Arctic Ocean ice balance is presented and shown to be self-consistent. Results are discussed in terms of the Arctic multi-year ice balance, which may be an important factor in the interaction ofocean, sea ice and climate because of its relationship to the minimum summer ice extent. The estimated area of multi-year ice decreases each year, but the decrease is small and insignificant based on four years of results. Furthermore, the observed decrease may be due to instrument drift or changes in the multi-year ice signature.

We have investigated two 30 m cores at two different spots in the most heavy snow-accumulation regions on Mizuho Plateau, East Antarctica. Marked seasonal variations periodically appear in oxygen isotope records of the cores. We analyzed one core with no trace of snow melting and found it had a complete record showing a yearly change of annual net snow accumulations from 1920 through 1980. The analysis shows that a yearly variation of annual net accumulation (N.A.) has some relations with that of the annual maximum value (δmax) of δ18O and that of the annual amplitude (Δδ) of the δ18O-change in an annual snow layer. Power spectral analyses with respect to the variation of N.A., δmax and Δδ also indicate that there is commonly a predominant periodicity of about five years.

Mass balance observations on Urumqi River No. 1 Glacier and Tuyuksu Glacier in Tianshan show a large deficit during the 1980s as compared to the decades before. The Qinghai Lake in northeastern Qinghai-Xizang (Tibet) Plateau and the Issyk-kul Lake in Soviet Tianshan have been continuously shrinking during the past few centuries. Since the maximum of the Little Ice Age which occurred mainly in the 18th century, glaciers have decreased in area by about 44% in the Urumqi valley. These data and other evidence from glaciers, lakes, etc. clearly indicate that the climatic warming and drying tendency grows stronger in this century and will possibly persist to the early decades of the next century. The growing greenhouse effect due to the increase of CO2 and other related gases will enhance this tendency in the near future. However, should the high temperature period of the early and middle Holocene reappear, the climate would become humid.

We present a simple parameterization of the effect of open leads in a general circulation model of the atmosphere. We consider only the case where the sea ice distribution is prescribed (i.e., not interactive) and the fraction of open water in the ice is also prescribed and set at the same value at all points in the Southern Hemisphere and a different value in the Northern Hemisphere. We approximate the distribution of sea ice over a model “grid box” as a part of the box being covered by solid ice of uniform thickness and the complement of the box consisting of open water at a fixed -1.8 C. Because of the nonlinearity in the flux computations, separate calculations are performed over the solid sea ice and over the open leads. The net fluxes conveyed to the atmosphere over the grid box are determined by performing the appropriate area-weighted average over the two surface types.

We report on an experiment designed to assess the sensitivity of the modelled climate to the imposition of a 50% concentration in the winter Antarctic sea ice. Significant warming of up to 6°C takes place in the vicinity of and above the Antarctic sea ice and is associated with significant changes in the zonal wind structure. Pressure reductions are simulated over the sea ice, being particularly marked in the Weddell Sea region, and an anomalous east-west aligned ridge is simulated at about 60°S. Very large changes in the sensible heat flux (in excess of 200 Wm−2) are simulated near the coast of Antarctica.