NASA flights over southern Greenland in 1991, 1992 and 1993, using a scanning laser altimeter with Global Positioning System (GPS) navigation, have demonstrated a capability to measure ice-surface elevations to an accuracy of 10-15 cm. Flights over Jakobshavns Isbræ revealed winter thickening by several meters between September 1991 and April 1992. By July 1993, surface elevations showed a small additional increase, possibly associated with the cold 1992 summer. Data collected over the ice sheet east of Jakobshavns Isbræ show negligible change over the same period; but further south, at latitude 65 N, the western part of the ice sheet appears to have thickened by up to 2 m between 1980 and 1993. It is clear that such measurements must be continued over many years, both to quantify the effects of inter-annual variability and to measure long-term trends. To this end, we plan to complete a first survey of all major drainage basins on the ice sheet in May-June 1994, and then to resurvey all flight lines at 5a intervals, with more frequent flights over selected routes.

Glacier variations in Iceland have been recorded systematically since the 1930s at 27 different glacier termini. The advance/retreat records of non-surging glaciers show a clear relationship to climate. A change in the climate typically leads to a response at the snout within a time period of 10 years. The records of surge-type and mixed-type glaciers show variations that are unrelated to climate. However, the maximum extension of surge-type glaciers at the end of surges and the minimum extension just before a surge appear to be influenced by long-term climate changes. A strong warming in the 1920s was a turning-point in the climate of Iceland which led to a rapid retreat of most glaciers in the country in the 1930s. The summer temperature fell gradually after 1940, with a notable drop in the mid-1960s. Since about 1970, more than half of the glaciers in Iceland have been advancing. In the western part of the country, the recovery is about one-quarter of the ground lost and in the southern, central and northern parts it is about one-half. In southeastern Iceland, some of the glaciers have been stationary for about 30 years while others have advanced slightly. Glacier snow-budget index computed from meteorological data indicates that the timing of the turning-point around 1970 coincides with a minimum in the cumulative net glacier mass balance.

The first European Remote Sensing Satellite (ERS-1) synthetic aperture radar (SAR) images offer opportunities for studying glacier surface properties and near-surface features. Analysis of back-scatter values from digital SAR data from 18 January, 7 June, 1 September and 25 October 1993 of Brúarjökull, an outlet glacier on the northeastern margin of the Vatnajökull ice cap, Iceland, that has a history of episodic surges, reveals several back-scatter boundaries that may relate to glacier facies and, inferentially, to mass balance. For example, a strong back-scatter boundary on the 18 January image of the snow-covered glacier, representing a back-scatter coefficient, σ°, difference of 4.34dB, appears to coincide with the position of the transient snow line at the end of the 1990–91 budget year. The boundary is visible on the 7 September 1991 Landsat thematic mapper (TM) image. The terminus is very difficult to define because of back-wasting from the last surge (1963–64) but is most easily delineated on the 1 September 1993 SAR and the 7 September 1991 TM images, in part due to the presence of ice-margin lakes.

Pine Island Glacier is one of the major ice streams draining West Antarctica. We calculated average velocities for both its grounded and floating parts by tracking crevasses and other patterns moving with the ice on two sequential images acquired in February and December 1992 by ERS-1 SAR (European Remote-Sensing Satellite, Synthetic Aperture Radar). Velocities in the fast-moving central parts of the glacier range from about 1.3 km a−1 on the grounded part to about 2.6 km a−1 on the floating part. Velocity increases rapidly just below the grounding line. For the floating part, velocities are approximately 0.3 km a−1 faster than previously established from Landsat images by other workers. This observation suggests that the discharge is larger and, therefore, the mass balance less positive than previously thought; however, our study also shows that current information is insufficient to allow for precise mass-balance calculations. Images of the glacier front show that a calving event released an iceberg about 5 km wide between February and December 1992. The position of the glacier front has not changed substantially between 1973 (Landsat image) and 1992 (ERS-1 images).

In spite of their importance to global climate and sea level, the mass balance of the Antarctic ice sheet and the dynamics of the coast of Antarctica are largely unknown. In 1990, the U.S. Geological Survey, in cooperation with the Scott Polar Research Institute. U.K., began a long-term coastal mapping project in Antarctica that is based on analysis of Landsat images and ancillary sources. The project has live objectives: (1) to determine coastline changes that have occurred between the mid-1970s and the late 1980s/early 1990s; (2) to establish an accurate base-line series of 24 1: 1 000 000 scale maps that define the glaciological characteristics of the coastline: (3) to determine velocities of outlet glaciers, ice streams and ice shelves: (4) to compile a comprehensive inventory of outlet glaciers and ice streams: and (5) to compile a 1: 5 000 000 scale map of Antarctica derived from the 24 maps. Analysis of images used in producing the first five of the 24 maps has shown that ice fronts, iceberg tongues and glacier tongues are the most dynamic and changeable features in the coastal regions of Antarctica. Seaward of the grounding line of outlet glaciers, ice streams and ice shelves, the floating margin is subject to frequent, large calving events and rapid flow. Although calving does occur along ice walls, the magnitude of their change on an annual to decadal basis is generally not discernible on Landsat images; therefore, ice walls can be used as relatively stable reference features for measuring other changes along the coast. Velocities of outlet glaciers, ice streams and ice shelves range from 0.1 to several kilometers per year.

Sequential visible and radar satellite imagery is used to document the fluctuations of 39 outlet glaciers on James Ross Island (JRI; 64° S, 58° W), Antarctic Peninsula. The monitoring time interval of 17.4 years extends from late 1975 until early 1993. The study is based on synergistic use of different sensors for monitoring glacier variations. The digitally enhanced Landsat TM images also allowed definition of boundaries between grounded and floating ice within Röhss Bay and detection of ice rumples, resulting in an improved measurement of the areal extent of JRI. In addition to glacier boundaries, the analyzed ERS-1 SAR images provide information on morphological properties of snow and ice. The satellite data show a general glacier retreat since 1975, resulting in clear decrease of overall glacier area on JRI. Of 39 glaciers measured, 33 glaciers retreated in the period 1975–93, 15 of which showed significant retreat in comparison to glacier size. This is in accordance with the pronounced atmospheric-warming trend recorded during the last two decades on the Antarctic Peninsula. The analysis of glacier areas on JRI thus provides a base line for detection of future changes.

Large retreats were revealed for most glaciers in Patagonia, South America, by analyzing satellite images and air photographs. For example, Glaciar O’Higgins retreated 13 km during 41 years from 1945 to 1986 and Glaciar Upsala retreated about 3 km between 1968 and 1990. During the 41 years former period, however, the southern tongue of Glaciar Pio XI advanced by up to 8.5 km and Glaciar Moreno remained almost in a steady state. Considerable ice-thinning rates, from 3.0 to 11 ma−1, were measured by surveying surface profiles in the ablation areas of Tyndall and Upsala glaciers, respectively, during the period from 1985 (or 1990) to 1993. The ice thickness of Glaciar Moreno, however, has changed very little.

Numerical experiments using a simple mass-balance model show that a 100 m rise in the equilibrium-line altitude due to climatic change would result in about a 200–350 m rise in the frontal altitude at Glaciar Upsala corresponding to a retreat of 5–8 km, while it would cause only about a 70–100 m rise at Glaciar Moreno. The large difference between these two neighbouring glaciers results from the difference in contributions of the calving amount to the total mass balance, as well as the difference in the altitudinal distributions of drainage areas.

Understanding the response of present and former ice caps to climatic fluctuations depends on an effective means of modelling mass balance. Ablation can be reasonably well approximated from surface energy-balance calculations, but changes in snowfall, the primary mass-gain mechanism, are poorly understood. This is a particularly important limitation in those mid-latitude areas where the location of precipitation belts changes from glacial to interglacial conditions. This paper develops a method of modelling the spatial and temporal variability of snowfall in Patagonia. Snowfall is predicted from a vertically integrated moisture-balance equation representing winter, equinoctial and summer seasons. Inputs to the model are sea-surface temperatures and the geostrophic wind at 700 mbar. The dominant process affecting precipitation is the advection of moisture in an air column whose ability to hold moisture is dependent on temperature and altitude. Relative humidity controls evaporation and precipitation, both of which increase with higher wind speeds. Snowfall is calculated from seasonal precipitation and the annual temperature regime, while ablation is estimated from degree-day calculations. The model successfully simulates present-day precipitation and mass-balance patterns in Patagonia. We also run the model for full-glacial conditions with and without an ice cap, revealing the extreme variability of mass balance at different stages of a glacial cycle and the relative importance of altitude/mass-balance feed-back in ice-cap growth. Since the model is driven by wind and temperature fields it has the potential to use the results of GCMs as input.

The calving rates and calving styles of temperate glaciers that calve into fresh water are distinctively different from those of temperate tide-water glaciers. These contrasts are important for interpreting and predicting the response of ice masses to climate change. Glaciar Upsala is a large calving outlet of Hielo Patagónico Sur (southern Patagonia ice field). Its twentieth-century retreat has been climate-driven but significantly modulated by calving dynamics and by the transition from melting to calving at its eastern terminus. Here, the onset of rapid calving in the early 1980s initiated retreat at ≤440 m a−1. The 1992–93 calving rate (vc) is estimated to be 60 m a−1 in a mean water depth (hw) of 67 m. A vc/hw relationship for fresh water based on 14 sites around the world, including seven deep-water sites, confirms both the linear dependency of vc on hw and the contrast between calving rates in tide water and fresh water. As yet, no physical explanation for this contrast exists, but differences in subaqueous melt rates, longitudinal strain rates and crevassing may provide a partial explanation.

The glacier inventory in Chile covers all the arid north of the country (18° to 33° S) and the central part down to the lake district (33° to 41° 30′ S) except for the Río Maule and Río Itata basins (35° to 37° S). The most glaciated region of Chile is Patagonia, characterized by a wet climate, but glacier data are lacking over vast areas of this region. A total area of 5515 km2 has been inventoried in Chile; this does not include Hielo Patagónico Sur (southern Patagonia ice field), by far the largest glaciated area in Chile with an area of 13000km2, part of which is in Argentina. Including the uninventoried giaciers of Maule, Itata and Patagonia, the total glaciated area in Chile is estimated to be 201715 km2.

Information on glacier variations in the north of Chile is very scarce. In the central and south-central parts, and especially in Patagonia where glacier variations have been most studied, there is ample evidence for general glacier retreat from a neoglacial maximum, although a few glacier advances have been reported. The generalized retreat might be a result of tropospheric warming observed in the second half of the 20th century.

A numerical model is used to study the spatial and temporal variability of ice thickness in the Arctic. The model is run to cyclo-stationary equilibrium, forced with daily varying geostrophic winds and monthly varying surface air temperatures from 1951 to 1990. Decadal-average winter ice-thickness fields exhibit a trend of increasing ice thickness in the East Siberian Sea and somewhat thicker ice overall in the 1980s. Inter-annual variability was largest in the Beaufort and East Siberian Seas and generally lower in the Laptev Sea and central Arctic. Spatial correlation patterns show that ice thickness at a point is correlated with ice thickness over a surrounding area of roughly 4 × 105 km2 with a correlation coefficient of 0.7. Temporal auto-correlation functions indicate that the total ice volume has a correlation time-scale of roughly 7 years, while the volume of ice between 2 and 5 m thick and the volume of ice less than 1 m thick have correlation times of roughly 2 years and 2 months, respectively.

The Resolute Bay area in the central Canadian Arctic Islands is in a region of maximum variability in 50 kPa heights in spring. This is the northern part of an oscillation in large-scale atmospheric circulation called the Baffin Island West Atlantic Oscillation (BWA). The historical record of temperature at Resolute Bay, Northwest Territories, over the 1948–93 period is compared with circulation patterns for extremely warm and cold spring seasons, using a composite technique and a simple index of the intensity of the BWA. The extremely cold spring of 1992, the second coldest of the 46a record, was probably caused by reduced solar radiational forcing due to the increased dust veil caused by Mount Pinatubo.

Within each spring season, large-scale synoptic conditions provided an important control on the timing of critical stages leading to full snow melt on the sea-ice surface. The first stage is the change from dominance of the region by wintertime continental polar air to the frequent invasion of maritime polar air masses, resulting in increased cloud cover and reduced frequency of Arctic inversions. The second stage is the transition to full melt on the ice with a sharp drop in surface albedo. During the spring on 1993, the first stage was triggered by a major storm which moved through the area on 9–10 May 1993. The second snow-melt stage was triggered by a stationary upper ridge which remained over the area for a 2 week period from 15 May to early June and produced clear to scattered-cloud conditions and high downward solar radiation.

Surface spectral-albedo data collected over snow-covered first-year and multi-year sea ice under diffuse sky conditions during the springtime transition are examined. Of specific interest is the relationship between changes in the visible and near-infrared albedo of sea ice and concurrent changes in the geophysical characteristics of the ice volume. With the onset of melt conditions, visible and near-infrared sea-ice albedo decreased due to physical changes within the snow and ice volumes. Visible albedo was found to be sensitive to changes occurring throughout the sea-ice volume, while the near-infrared albedo appeared most influenced by near-surface conditions.

For the Last Glacial Maximum, (LGM; 21 000 BP), simulations using atmospheric general-circulation models (AGCMs) are very sensitive to the prescribed boundary conditions. Most of the recent numerical experiments have used the CLIMAP (1981) data set for ice-sheet topography, sea-ice extent and sea surface temperatures (SSTs). To demonstrate the impact of ice-sheet reconstruction on the LGM climate, we performed two simulations: one using CLIMAP (1981) ice-sheet topography, the other using the new reconstruction provided by Peltier. We show that, although the geographical structure of the annually averaged temperature is not modified, there are important seasonal and regional impacts on the temperature distribution. In a second step, to analyze the effects of cooler SSTs and sea-ice extent, we performed a simulation using CLIMAP (1981) for the ice-sheet topography, but with present SSTs. We find that the cooling due to ice sheets for the LGM climate is one-third of the global annually averaged cooling, and dial the southward shift of the North Atlantic low in winter is not due to sea-ice extent, but is an orographic effect due to the Laurenride ice sheet. This set of sensitivity experiments allows us also to discriminate between thermal and orographic forcings and to show the impact of the ice-sheet topography and cooler SSTs on the pattern of planetary waves during the LGM climate.

Recently reported decreases in the summertime ice coverage of the Bellingshausen Sea over the period 1988–91 have been mentioned in the popular press with some loss of the larger spatial and temporal context in which they occurred. Experience over the past two decades with the continually lengthening satellite record has frequently revealed prominent increases or decreases in regional or even hemispheric sea-ice coverage which last for a several-year period and are then reversed. Also, in almost any several-year period in the short satellite record available, at least one region could be highlighted as having prominent ice-cover increases and at least one as having prominent ice-cover decreases. In the 1988–91 period of low ice coverage in the Bellingshausen Sea, mid-winter sea-ice extents noticeably increased in the Baffin Bay/Davis Strait region. Placed in the longer temporal context of ice extents since the late 1970s, the ice-extent increases from 1988 to 1991 in Baffin Bay/Davis Strait conform much better with an interpretation of cyclically varying wintertime ice extents than with one of a long-term upward trend. The time series for the Bellingshausen Sea also shows marked multi-year fluctuations and, although in the 1989–91 period the summertime coverage was unusually low, the wintertime ice coverage noticeably increased.

A vertically integrated, zonally averaged energy-balance climate model coupled to a two-dimensional ocean model with prescribed overturning pattern is employed to assess the seasonally and latitudinally varying response of the climate system to changes in radiative forcing. Since the sensitivity of the climate system depends on its actual state, considerable attention is given to the correct simulation of the important features of the present-day climate (such as surface air temperature, sea-ice and snow amount and meridional energy transport). The climate variability induced by the various elements of the albedo-temperature (e.g. sea-ice and snow) feed-back is quantified. It appears that the variability caused by sea-ice variations is approximately twice as large as for snow variations.

An idealized planetary flat-bottom geostrophic ice–ocean model is constructed with boundaries at latitudes 5° and 65° N and longitudes 50° W and 10° E in order to approximate the North Atlantic. The model is driven by fixed zonally averaged wind, surface air temperatures and surface ocean salinity. A dynamic thermodynamic sea-ice model is coupled to the ocean model. Only the thermodynamic insulating effects of the sea ice are considered, and no salt fluxes due to melting and freezing are included. Four equilibrium simulations of about 5000 years each are performed: two with interactive sea ice with and without ice dynamics, and two control simulations with either a fixed or no ice cover.

In the two simulations including interactive sea ice, characteristic oscillations in the ice thickness and ocean temperature are found to occur. The oscillations are smaller when sea-ice dynamics are included. The dominant oscillation occurs at about a 5 year period, with the key feature being that the presence of sea ice tends to insulate the ocean and hence allows an oceanic warming. This warming in turn eventually causes a melt-back of the ice and a subsequent cool-down of the ocean. Oscillations at longer periods of about 20 years in the thermohaline circulation are also observed. These longer-period oscillations are particularly pronounced in the northward surface water transport.

Based on presented field data, it is shown that snow contributes roughly 8% to the total mass of ice in the Weddell Sea. Snow depth averages 0.16 m on first-year ice (average thickness 0.75 m) and 0.53 m on second-year ice (average thickness 1.70 m). Due to snow loading, sea ice is depressed below water level and flooded by sea water. As a result of flooding, snow ice forms through congelation of sea water and brine in a matrix of meteoric ice (i.e. snow). Sea-ice growth has been simulated with a one-dimensional model, treating the evolution of salinity, porosity and thermal properties of the ice. Simulations demonstrate that in the presence of a snow cover, ice growth is significantly reduced. Brine volumes increase by a factor of 1.5–2, affecting properties such as ice strength. Snow-ice formation depends on the evolution of freeboard and ice permeability. Effects of accumulation-rate changes have been assessed, for the Weddell Sea with a large-scale sea-ice model accounting for snow-ice formation. Results for different scenarios are presented and compared with field data and one-dimensional simulations. The role of snow in modulating the response of Antarctic sea ice to climate change is discussed.

There are no systematic measurements of seasonal snow in the Southern Alps, New Zealand, so little information is available as to its past variability. To rectify this, a conceptual model is developed that calculates seasonal snow deposition, ablation and accumulation. The model is based on daily temperature and precipitation data from long-established climate stations about the Southern Alps. Output is given as daily specific net balance of snow at five elevation bands from 1000 to 2200 m and as total water stored as seasonal snow over several major river catchments. Model output is in general agreement when tested against the few historical observations of snow and is tuned to the long-term water balance. A chronology of seasonal snow is reconstructed from 1931 to 1993. Area-averaged annual maxima average 366 mm. They show no trend, but large inter-annual variability from less than 200 to over 650 mm w.e. Seasonal snow can peak at any time between September and January.

Long-term ice records (1823-1994) from six sites in different parts of the Laurentian Great Lakes region were used to show the type and general timing of climatic changes throughout the region. The general timing of both freeze-up and ice loss varies and is driven by local air temperatures, adjacent water bodies and mixing, and site morphometry. Grand Traverse Bay and Buffalo Harbor represent deeper-water environments affected by mixing of off-shore waters; Chequamegon Bay, Menominee, Lake Mendota, and Toronto Harbor represent relatively shallow-water, protected environments. Freeze-up dates gradually became later and ice-loss dates gradually earlier from the start of records to the 1890s in both environments, marking the end of the “Little lce Age”. After this, freeze-up dates remained relatively constant, suggesting little change in early-winter air temperatures during the 20th century. Ice-loss dates at Grand Traverse Bay and Baffalo Harbor but not at the other sites became earlier during the 1940s and 1970s and became later during the 1960s. The global warming of the 1980s was marked by a trend toward earlier ice-loss dates in both environments.