Repeat photographs and field survey reveal the mechanism of short-term ice-cliff evolution at Maud Glacier, a temperate lake-calving glacier in New Zealand. Calving is cyclic, each cycle involving four stages: (1) waterline melting and collapse of the roof of a sub-horizontal notch at the cliff foot; (2) calving of ice flakes from the cliff face leading to a growing overhang from the waterline upwards, and crack propagation from the glacier surface; (3) large but infrequent calving of slabs in response to the developing overhang, returning the cliff to an “initial” vertical profile; (4) rare subaqueous calving ofa submerged ice foot. Results indicate that the rate-controlling process is the speed of waterline melting, and that calving rate is independent of water depth (at least at time-scales of weeks to months). Slowly calving lake-terminating glaciers have mass balances more negative than land-terminating glaciers, but nevertheless advance and retreat in response to mass-balance driven changes in ice velocity.

Bakaninbreen, Svalbard, started to surge during 1985–86, and developed a surge front up to 60 m high. Associated with down-glacier propagation of this surge front was the formation of shear zones and thrust faults, some of which revealed basally derived debris at the glacier surface. Hot water drilling and sampling of basal material showed the glacier bed to be soft sediment more than 1 m thick. A high-resolution ground-penetrating radar (GPR) survey at 100 MHz was conducted along three 500 m lines parallel to glacier flow on the surge front. The aims were to investigate the internal geometry of the thrust features, and the processes of entrainment of basal debris into bulk glacier ice.

A strong linear reflector was seen on the survey, but it is about 15–20 m above the bed as identified from drilling depths. It probably represents the upper interface of a layer of debris-rich basal ice. Several extensive englacial reflectors were interpreted as debris-laden emergent thrust features, varying in thickness from 0.1 to 1.1 m. These features were mapped at the glacier surface, and drilling and sediment sampling verified the interpretation. Other englacial features included regions of incipient thrusting at the basal reflector, and an extensive region of scattering up to 30 m above the basal reflector that we interpret as folds, or blind thrusts that terminate englacially. Our results clearly demonstrate the potential of GPR for mapping internal glacial structure, and suggest that thrusting is an important process by which sediment is incorporated into glacier ice in the highly compressive region at the surge front.

Mass-balance measurements carried out since 1963 on six glaciers in southern Norway indicate that glaciers to the west of the main water divide have increased their mass, whereas glaciers further east have decreased. However, when small glaciers within the same mountain area are studied, the picture is more complicated. Different exposures, the terrain forms on the upwind side and area distribution of the glaciers are important factors with regard to the effect of the wind-driven snow.

The three glaciers studied in this paper are located within a 10 km radius in the Aurland Mountains, but differ in size, exposure and dynamic condition. Volume changes were calculated from two repeated mappings of two glaciers. Svartavassbreen increased in volume in all areas; Storskavlen decreased over most of its area and increased in some parts. Vargebreen has lost nearly all its ice mass in the eastern part, but has changed little in the western part. It is concluded that the volume changes found on these glaciers mainly reflect changes in local snow accumulation. Analysis of air-pressure data from the period 1952–94 is presented and supports the hypothesis that there has been a change to stronger westerly winds from the mid-1960s onwards.

It is recommended that glaciers chosen for long-term studies of volume changes should represent different exposures. The topography of the surrounding terrain is also important.

This paper presents an approach to long-term glacier monitoring. Mathematical surface descriptors, such as altitude, slope and curvature (surface form) are used to classify and quantify glacier surface developments. The analysis is based on photogrammetically derived grid-based digital elevation models over a period of decades. This paper outlines the concept and applies it to five valley glaciers in Spitsbergen, Svalbard, which differ with respect to size, thermal regime and dynamics. The results reflect differences between the glaciers investigated which are attributable to glacier dynamics, in particular concerning the glacier’s possible surge behaviour during a period with retreat and mass losses.

Patterns of retreat from maximum Little Ice Age (LIA) to present limits are studied at 20 glaciers in the relatively humid northwestern front ranges and arid inner areas of the Tien Shan, Central Asia. The depression of equilibrium-line altitudes has been calculated using several approaches. Data on changes of elevation ranges, glacier length and area are used to compare the patterns of glacier fluctuation. It is found that the large LIA glaciers in the warm and humid northwestern frontal ranges were 1.5–1.9 times larger in area than the modern glaciers; and the LIA glaciers in cold and arid inner parts of the Tien Shan were only 1.03–1.07 times larger. The changes in terminus-to-headwall elevation ranges are about 1.3–1.6 and 1.02–1.10, respectively. The largest LIA glaciers were 1.4–1.9 times longer than modern glaciers in “humid” ranges and only 1.02–1.12 times longer in “arid” areas. The maximum equilibrium-line depressions are approximately 100–200 m in “humid”areas and 20–50 m in “arid”areas. These results suggest that the glaciers in the “humid” areas are likely to be more variable than those in “arid” areas. The differences may be explained either by differences in the sensitivity of glaciers to climate change or by variability of climate signals from one area to another.

Ionic analyses of snow samples collected from snow pits and cores, and samples of meltwater collected from the only river draining the Norwegian glacier Austre Okstindbreen during the 1995 summer, showed that the input/output relations of sodium and chloride were nearly identical. Sampling at the pits revealed that some 72% of the Na+ and Cl– ions which accumulated in winter drained from the snowpack in early June with the first 12% of the meltwater. Concentrations and loads of the two ions in the first meltwater draining from the glacier in late May were relatively high. There then followed a slight decrease in both the concentrations and the loads of the two ions in samples of the river water, and more-or-less steady levels were present for a long period. In late July, concentrations decreased. Loads decreased in early August, indicating that most of the winter-accumulated ions had drained from the catchment. Input and output calculations for Na+ and C1– indicate a total drainage time for the first meltwater from the uppermost part of the accumulation area of 6–7 weeks, equivalent to an average drainage velocity of 100 md–1.

A dark line appears on a recent satellite image of McMurdo Ice Shelf, Antarctica. It is parallel to the calving front. Initial thoughts were that the line marks an opening crevasse associated with an impending major calving event. The feature was studied by means of a strain and surface-elevation grid that was surveyed twice, 25 d apart, using global positioning system (GPS) techniques. Results show that the dark line is not due to an opening crevasse. The feature is probably the surface expression of firn collapse over sea water soaking horizontally into the ice shelf.

The difficulty of using traditional direct mass-balance methods on large glaciers and ice caps requires the development of new methods. The objective of this paper is to evaluate whether a precise global positioning system (GPS) survey of geometry changes can be used to detect and monitor mass-balance changes. A joint project between Norway, Poland and Russia has investigated three high-altitude, large ice masses on Spitsbergen (Lomonosovfonna, Amundsenisen and Kongsvegen). The GPS survey was conducted using two methods.

1. (1) Static GPS survey of stakes drilled into the ice was used for flow-velocity and emergence-velocity measurements.

2. (2) Kinematic GPS survey was used to measure longitudinal profiles of surface elevation between the stakes.

The accuracy of the measurements has proved to be better than 5 cm in horizontal position and 10 cm in height. Comparisons show that the GPS survey is applicable and gives sufficient accuracy to replace traditional methods and is thus especially useful on large glaciers where traditional surveying by electronic distance meter is impossible.

We have analyzed records of glacier-front variations, mass-balance reconstructions, temperature and precipitation data of Alpine stations, and found that the difference between summer and winter temperature in connection with winter precipitation is a useful indicator of glacier activity. Application of this parameter to the records of six stations since 1860 indicated the advance around 1920 had been preceded by a decade with frequent positive mass balances, while the period 1928–64 was characterized by increased climatic continentality and strong, uniform glacier retreats.

A transverse bedrock ridge, or riegel, in the bedrock topography under Storglaciären, northern Sweden, induces a drastic change in flow regime. The part of the glacier located downstream of the riegel exhibits velocity variations correlated to water-pressure variations under this part of the glacier. Similar velocity variations are also observed up-glacier from the riegel despite a lack of significant water-pressure variations there. Field data and a numerical model suggest that the variations in surface velocity observed on the glacier upstream of the riegel are a result of pulling from down-glacier through longitudinal coupling across the riegel.

A model has been formulated to determine the stability regimes for water flow in a Subglacial conduit draining from a reservoir. The physics of the water flow is described with a set of differential equations expressing conservation of mass, momentum and energy. Non-steady flow of water in the conduit is considered, the conduit being simultaneously enlarged by frictional heating and compressed by plastic deformation in response to the pressure difference across the tunnel wall. With the aid of simplifying assumptions, a mathematical model has been constructed from two time-dependent, non-linear, ordinary differential equations, which describe the time evolution of the conduit cross-sectional area and the water depth in the reservoir. The model has been used to study the influence of conduit area and reservoir levels on the stability of the water flow for various glacier and ice-sheet configurations. The region of the parameter space where the system can achieve equilibrium has been identified. However, in the majority of cases the equilibrium is unstable, and an initial perturbation from equilibrium may lead to a catastrophic outburst of water which empties the reservoir.

The densification of water-saturated firn, which had formed just above the firn-ice transition in the wet-snow zone of temperate glaciers, was investigated by compression tests under pressures ranging from 0.036 to 0.173 MPa, with special reference to the relationship between densification rate, time and pressure. At each test, the logarithm of the densification rate was proportional to the logarithm of the time, and its proportionality constant increased exponentially with increasing pressure. The time necessary for ice formation in the firn aquifer was calculated using the empirical formula obtained from the tests. Consequently, the necessary time decreased exponentially as the pressure increased, which shows that the transformation from firn in ice can be completed within the period when the firn aquifer exists, if the overburden pressure acting on the water-saturated firn is above 0.12–0.14 MPa. This critical value of pressure was in good agreement with the overburden pressure obtained from depth–density curves of temperate glaciers. It was concluded that the depth of firn–ice transition was self-balanced by the overburden pressure to result in the concentration between 20 and 30 m.

The formation of superimposed ice at the surface of high-Arctic glaciers is an important control on glacier mass balance, but one which is usually modelled in only a schematic fashion. A method is developed to predict the relationship between the thickness of superimposed ice formed and the mean annual air temperature (which approximates the ice temperature at 14 m depth). This relationship is used to investigate the dependence of the proportion of snowpack water equivalent which forms superimposed ice on changes in mean annual temperature and patterns of snow accumulation.

Increased temperatures are likely to reduce the extent of the zone of superimposed-ice accumulation and the thickness of superimposed ice formed. This will have a negative effect on glacier mass balance. This is true even if warming occurs only in the winter months, since near-surface ice temperatures will respond to such warming. For John Evans Glacier, Ellesmere Island, Nunavut, Canada (79°40’ N, 74°00’ W), a 1°C rise in mean annual air temperature due solely to winter warming is predicted to reduce the specific mass balance of the glacier by 0.008 m a–1 as a result of decreased superimposed-ice formation. Although such a response is small in comparison to the changes which might result from summer warming, it is nonetheless significant given the very low specific mass balance of many high-Arctic glaciers.

A satellite radar altimeter can be used to monitor surface elevation change over polar ice sheets. Thirty-five months of Geosat Exact Repeat Mission (ERM) data from November 1986 to September 1989 over a section of East Antarctica (69–72.1 ∘S, 80–140∘ E) have been used in this study. A model that considers both surface and volume scattering was used to retrack the altimeter waveforms. Surface elevations for each month after the first three were compared to the average elevations for the first 3 months through a crossover method. The averaged crossover elevation difference changed with time in a way that suggests a yearly cycle in surface elevation. The average amplitude of the cycle is about 0.6 m. We have been unable to find any satisfactory explanation for the observed changes, in terms of either sources of error or contributors to real surface-height changes. We strongly suspect that orbit error plays a major role in producing the variations, although we know of no quantitatively satisfactory source of a quasi-seasonal variation in orbit error. Other possible contributors include a real seasonal variation in accumulation rate, seasonal changes in the delay of the radar signal as it propagates through the atmosphere, unmodeled variations in the depth of penetration of the radar pulse into the firn, changes in the thickness of the ice and the firn zone in response to seasonal variations in pressure and temperature, and the inverted barometer effect. Even though we do not know the cause of the variations, the results show the importance of comparing elevations at the same time of year for observations that are not continuous, while at the same time showing that even annually spaced measurements may not be free of substantial errors associated with interannual variability. The quasi-periodic variations obscure any evidence of a moderate secular change in surface height, if there is one, but a dramatic lowering at rates approaching 1 ma–1, such as are known elsewhere in Antarctica, can definitely be ruled out.

A one-dimensional ice flow model was adapted to reconstruct fluctuations of Hintereisferner, Ötztal Alps, since the last postglacial maximum in the mid-19th century. Both front positions and longitudinal ice thickness profiles were considered.

As forcing function the specific mass balance was used. The model was calibrated with the period 1953–91, because since 1953 the mass balance has been determined directly. For the period before 1953, as a first step so-called dendro-mass balances (derived from tree rings) were used. Then the mass balance was also parameterized as a function of temperature and precipitation from adjacent climate stations. With both forcings Hintereisferner could be reconstructed back to about 1850 with an accuracy that lies within the accuracy of the model (200 m grid-point distance).

An energy-balance model is used to calculate mass balance and equilibrium-line altitudes (ELAs) on two northwest Spitsbergen glaciers, Austre Brøggerbreen and Midre Lovénbreen, whose mass balances are at present negative, and for which greater than 20 year records of mass-balance data are available. The model takes meteorological data, ice-mass area distribution with altitude, and solar radiation as inputs. Modelling uses mean daily meteorological data from a nearby weather station, adjusted for altitude. Average net balances modelled for 1980–89 using models tuned to the decade’s average were –0.44 and –0.47 m w.e. for Lovénbreen and Brøggerbreen, respectively, compared with the measured averages of –0.27 and –0.36 m. Sensitivity tests on glacier response to greenhouse warming predict a net balance change of –0.61 m year–1 per °C temperature rise relative to today, and a rise in ELA of 90 m °C–1. Modelling of Little lee Age conditions in Spitsbergen suggests that a 0.6°C cooling or a precipitation increase of 23% would yield zero net mass balance for Lovénbreen and that further cooling would increase net balance by 0.30 m year–1 °C–1. Set in the context of similar modelling of southern Norwegian, Alpine and Greenland ice masses, these results support the suggestion that glaciers with a maritime influence (i.e. higher accumulation) are most sensitive to climate change, implying a gradient towards decreasing sensitivity as accumulation decreases eastward and with altitude in Svalbard.

A grid-based glacier melt-and-discharge model was applied to Storglaciären, a small valley glacier (3 km2) in northern Sweden, for the melt seasons of 1993 and 1994. The energy available for melt was estimated from a surface energy-balance model using meteorological data collected by automatic weather stations on the glacier. Net radiation and the turbulent heat fluxes were calculated hourly for every grid point of a 30 m resolution digital terrain model, using the measurements of temperature, humidity, wind speed and radiative fluxes on the glacier. Two different bulk approaches were used to calculate the turbulent fluxes and compared with respect to their impact on discharge simulations. Discharge of Storglaciären was simulated from calculated meltwater production and precipitation by three parallel linear reservoirs corresponding to the different storage properties of firn, snow and ice. The performance of the model was validated by comparing simulated discharge to measured discharge at the glacier snout. Depending on which parameterization of the turbulent fluxes was used, the timing and magnitude of simulated discharge was in good agreement with observed discharge, or simulated discharge was considerably underestimated in one year.

A time-dependent model of glacier flow was used to predict the response of Storglaciären, a small valley glacier in northern Sweden, to different warming scenarios by imposing two possible climatic forcings: one in which temperature alone increases (T model), and one in which both temperature and precipitation increase (TP model). A range of possible changes in temperature and/or precipitation was related to changes in glacier mass balance through a multiple linear correlation of mean specific net balance with mean summer temperature and mean specific winter balance. The T model was run with mass-balance perturbations in the form of linear increases from the recent (1980–89) mean summer temperature of 1, 2 and 4°C over the next 100 years. Perturbations for the TP model also used linear increases in precipitation of 10, 20 and 50% over current mean winter values in addition to increases in temperature.

Results of the modeling suggest that initial changes in the glacier‘s profile due to increases in temperature, or in both temperature and precipitation, are of comparable magnitude to those that might be expected as the glacier completes its response under the existing climate. Changes in the glacier‘s surface profile and terminus position that can, with some certainty, be attributed to climatic warming may only become apparent several decades after warming has begun.

We use a three-dimensional thermomechanical ice-sheet model, previously tested on the Greenland ice sheet, to reconstruct Last Glacial Maximum (LGM) ice sheets. We compare the effects on the results of the ice-sheet model of three different accumulation parameterization schemes. In the first and second schemes, LGM precipitation is computed from the present precipitation, taking and not taking into account moisture transport. In the third scheme, LGM precipitation and surface temperatures are computed using outputs of an atmospheric global circulation model (AGCM), treated in anomaly mode.

Results are compared to the last reconstruction of the Northern Hemisphere ice sheets (Peltier, 1994), computed using global rebound rates in a visco-elastic model of the Earth’s crust. The first two accumulation parameterizations do not give satisfactory reconstructions of the LGM ice sheets, since they are unable to compute realistic LGM climatic conditions. The third method gives very satisfactory results, which leads us to conclude that the best way to obtain realistic LGM climatic conditions is to use AGCM outputs.

A large amount of glacier mass-balance data has been gathered during the last few decades. One of the ultimate goals with these data is to reveal changes in climate, recorded by the sensitive climatic gauges that glaciers are. How can we use the collected mass-balance data to retrieve knowledge of changes in climate and in the atmospheric circulation? One method is to examine the extremes of the summer and winter balance years from a glacier or a glaciated area, and from these reconstruct the mean synoptic situation for those years. We focus here on the long 50 year continuous mass-balance record from Storglaciären, northern Scandinavia, and analyze the extreme events in the data set. The results show that high net balance is favoured by stronger westerly air flow (more maritime influence) which increases accumulation in winter but reduces ablation in summer.