The meltwaters draining from two glaciers in the Italian Alps contain metabolically active yeasts isolable by culture-based laboratory procedures. The average number of culturable yeast cells in the meltwaters was 10–20 colony-forming units (CFU) L–1, whereas supraglacial stream waters originating from overlying glacier ice contained <1 CFUL–1. Yeast cell number increased as the suspended-sediment content of the water samples increased. Basidiomycetous yeasts represent >80% of isolated strains (Cryptococcus spp. and Rhodotorula spp. were 33.3% and 17.8% of total strains, respectively). Culturable yeasts were psychrotolerant, predominantly obligate aerobes and able to degrade organic macromolecules (e.g. starch, esters, lipids, proteins). To the authors’ knowledge, this is the first study to report the presence of culturable yeasts in meltwaters originating from glaciers. On the basis of these results, it is reasonable to suppose that the viable yeasts observed in meltwaters derived predominantly from the subglacial zone and that they originated from the subglacial microbial community. Their metabolic abilities could contribute to the microbial activity occurring in subglacial environments.

The Lena river is one of the four largest rivers flowing into the Arctic Ocean. It has a complicated hydrological system that is affected not only by the amount of precipitation but also by the timing of ice break-up. To determine the mechanisms of runoff formation, a numerical analysis based on modeling was carried out for the period 1986–2000. The results show that (1) the timing of flood rise and peak can be modeled at Tabaga, which represents the upper and central portions of the Lena river; (2) river-freeze processes delay the spring, snowmelt-dominated flood by about 23 days; and (3) the difference between the break-up dates at Tabaga and Kirensk has ranged from several days to >2 months, and a maximum value of 69 days was recorded in spring 1998.

We carried out snow-pit observations at Nagaoka, Niigata prefecture, Japan, where the snow layers were at the melting point. It was observed that the water content in the snowpack was nearly constant at approximately 10%, and the coarsening rate of snow particles was about 0.4×10–3mm3 h–1, which was in the range between the rate for dry snow and that for snow soaked in water. The isotope change of snow particles by melting and freezing in a closed system under isothermal conditions at 0˚C was modeled. The temporal change in isotope concentration was calculated for wet snow layers, based on the fractionation between snow particles and liquid water in between the particles, in association with the coarsening of snow particles. The results compared well with field observations. These results suggest that the isotope concentration of the pore water that flows downward from the surface contributed significantly to the isotope change of snow particles.

This paper explores changing ice–water interactions during jökulhlaups from Grænalón, a 5 × 108m3 subaerial lake dammed by Skeiðarárjökull, Iceland. Unstable drainage of Grænalón since the early 20th century has resulted in 45 jökulhlaups whose hydrologic character has varied enormously. Geomorphic observations and geophysical measurements from the inlet and outlet zones of the subglacial floodwater tract constrained the hydromechanical factors governing ice–water interactions at Grænalón. To date, three distinct drainage regimes have occurred in response to the changing surface elevation of Grænalón. Shifts from one drainage regime to another involved pronounced changes in jökulhlaup magnitude, timing and cyclicity. Present hydraulic conditions for lake drainage differ from the classical view of a pressure-coupled lake draining directly beneath an ice dam. Instead, low-amplitude drawdown occurs at regular, frequent intervals when hydrostatic pressure in a shallow, rock–ice trench enables water to flow beneath a sagging ice barrier. Floodwater exits Skeiðarárjökull in a supercooled state due to rapid hydraulic displacement from an overdeepened subglacial basin.

Reversals in the ice/air surface slope are important in geomorphic and glaciological contexts, thus motivating consideration of the conditions under which they form. Surface slope reversals are seen in numerous places, such as ice rumples on ice shelves, as surficial lakes, and at the down-glacier end of Vostok lake, Antarctica. Such slope reversals can reduce or reverse the subglacial hydrological gradient, thereby rerouting subglacial water transport and possibly leading to the creation of subglacial lakes. Supraglacial lakes produced by slope reversals in ablation zones may aid in driving water-filled cracks that allow surface water access to the bed. Surface slope reversals, in the absence of a concomitant reversal in ice flow, indicate a local violation of the so-called ‘shallow-ice’ approximation, and in this circumstance the longitudinal deviatoric stress becomes critical in the stress equilibrium. Using a simple numerical model, we have explored the conditions under which surface slope reversals form for certain simple scenarios. The results indicate that ice which initially possesses a normal slope will tend toward a reversed slope if the ice is thinned, the bed is strengthened or the downstream buttressing is increased.

The nature of microscopic particulates in meteoric and accreted ice from the Vostok (Antarctica) ice core is assessed in conjunction with existing ice-core data to investigate the mechanism by which particulates are incorporated into refrozen lake water. Melted ice samples from a range of ice-core depths were filtered through 0.2 μm polycarbonate membranes, and secondary electron images were collected at ×500 magnification using a scanning electron microscope. Image analysis software was used to characterize the size and shape of particulates. Similar distributions of major-axis lengths, surface areas and shape factors (aspect ratio and compactness) for particulates in all accreted ice samples suggest that a single process may be responsible for incorporating the vast majority of particulates for all depths. Calculation of Stokes settling velocities for particulates of various sizes implies that 98% of particulates observed could ‘float’ to the ice–water interface with upward water velocities of 0.0003 ms–1 where they could be incorporated by growing ice crystals, or by rising frazil ice crystals. The presence of particulates that are expected to sink in the water column (2%) and the uneven distribution of particulates in the ice core further implies that periodic perturbations to the lake’s circulation, involving increased velocities, may have occurred in the past.

Linear trends in 1 April snow water equivalent (SWE) in California, USA, are determined from a dense network of observations over the period 1950–2002. These trends are compared to concurrent time series of precipitation and temperature and Pacific Ocean climate indices. We find increased winter temperatures have accompanied both decreasing and increasing SWE trends, resulting in a weak overall negative trend in spring SWE. The spatial distribution of SWE trend is dependent on both latitude and elevation. Increases in both precipitation and temperature have led to increases in SWE at high elevations in southern Sierra Nevada and decreases at lower elevations. Increased temperature and decreased precipitation are associated with SWE loss in northern California. Trends in SWE can partially be attributed to shifts in Pacific Ocean climate indices.

Large variations in both snow water equivalent (SWE) and snow slope stability are known to exist in the alpine snowpack, caused by wind, topographic and microclimatic effects. This variability makes extrapolation of point measurements of snowpack properties difficult and prone to error, but these types of measurements are used to estimate SWE and stability across entire mountain ranges. Radar technology provides a promising alternative to point measurements, because large areas can be covered quickly and non-intrusively. There is great potential for obtaining information on a large spatial scale from airborne applications. Frequency-modulated continuous wave (FMCW) radar measurements were made from the ground in several different alpine snowpacks, along with manual and in situ electrical measurements. The surface and ground reflections from the radar data, combined with an average density estimate, can provide a useful estimate of SWE. In addition, the locations of internal reflections are highly correlated with both visually identified layers and measured changes in in situ dielectric properties.

Snowmelt is measured over a 34 day summer period at 2440ma.s.l. on Tasman Glacier (>500m above the equilibrium-line altitude) using a tipping-bucket lysimeter and an array of ten ablation stakes. A degree-day factor for snowmelt is calculated using a linear relationship between combined measured melt and the number of degree-days. The slope of the regression line for these data points provides an estimate of the degree-day factor for use in runoff models. Average snowmelt is 17.8 mmd–1, but varies between 0 and 78 mmd–1. Melt occurs in a series of distinct cyclical events or pulses, each of which lasts 5–8 days. These correspond to the eastward passage of anticyclones, then troughs over the Southern Alps. When all days with northwest airflow across the Southern Alps are excluded, the melt factor is 3.4 mm˚C–1 d–1. Northwest days belong to a different population with a much higher average melt factor of 9.1 mm˚C–1 d–1, but more measurements are required to better understand key processes.

Meltwater retention by freezing is a highly climate-sensitive term in the mass budget since the cold content is directly controlled by winter climate, which is expected to change most in an anthropogenic-driven climate change. Meltwater released at the surface percolates into dry snow in a pattern with alternating horizontal and vertical water-flow directions, where the processes of pore refreezing (RF) (vertical flow) and superimposed ice (SI) formation (horizontal flow) occur. The flow cannot be forecasted and quantified when water first enters cold, dry snow. However, because the two processes are driven by different physical mechanisms, their potential magnitude can be estimated, which has been done in a transect at 66° N. Results show that SI declines from west to east and inversely correlates with accumulation. From the total retention capacity, theoretical lowest-runoff lines were determined at ∼1400ma.s.l. in the west and ∼1600ma.s.l. in the east. Since the SI potential is high in most places and the warming from SI formation predominately occurs near to the surface, it is argued that winter cooling effectively recharges the cold content of the snow/firn/ice pack, preventing the development of isothermal conditions and subsequent runoff. However, SI formation declines over time, so an extension of the melting season could result in deeper percolation beyond the SI layer.

Ice-dammed Hidden Creek Lake, Alaska, USA, outbursts annually in about 2–3 days. As the lake fills, a wedge of water penetrates beneath the glacier, and the surface of this ‘ice dam’ rises; the surface then falls as the lake drains. Detailed optical surveying of the glacier near the lake allows characterization of ice-dam deformation. Surface uplift rate is close to the rate of lake-level rise within about 400 m of the lake, then decreases by 90% over about 100 m. Such a steep gradient in uplift rate cannot be explained in terms of ice-dam flexure. Moreover, survey targets spanning the zone of steep uplift gradient move relative to one another in a nearly reversible fashion as the lake fills and drains. Evidently, the zone of steep uplift gradient is a fault zone, with the faults penetrating the entire thickness of the ice dam. Fault motion is in a reverse sense as the lake fills, but in a normal sense as the lake drains. As the overall fault pattern is the same from year to year, even though ice is lost by calving, the faults must be regularly regenerated, probably by linkage of surface and bottom crevasses as ice is advected toward the lake basin.

The microstructural evolution of brine inclusions in granular and columnar sea ice has been investigated through magnetic resonance imaging (MRI) for temperatures between –28 and –3˚C. Thin-section and salinity measurements were completed on core samples obtained from winter sea ice near Barrow, Alaska, USA. Subsamples of granular (2–5cm depth in core) and columnar sea ice (20–23 cm depth) were investigated with morphological spin-echo and diffusion-weighted imaging in a Bruker 4.7T MRI system operating at field gradients of 200 mTm–1 at temperatures of approximately –28, –15, –6 and –3˚C. Average linear pore dimensions range from 0.2 to 1 mm and increase with bulk liquid volume fraction as temperatures rise from –15 to –3˚C. Granular ice pores are significantly larger than columnar ice pores and exhibit a higher degree of connectivity. No evidence is found of strongly non-linear increases in pore connectivity based on the MRI data. This might be explained by shortcomings in resolution, sensitivity and lack of truly three-dimensional data, differences between laboratory and field conditions or the absence of a percolation transition. Pore connectivity increases between –6 and –3˚C. Pore-number densities average at 1.4±1.2mm–2. The pore-number density distribution as a function of cross-sectional area conforms with power-law and lognormal distributions previously identified, although significant variations occur as a function of ice type and temperature. At low temperatures (< –26˚C), pore sizes were estimated from 1H self-diffusivity measurements, with self-diffusivity lower by up to an order of magnitude than in the free liquid. Analysis of diffusional length scales suggests characteristic pore dimensions of <1 μm at < –26˚C.

To improve our understanding of the deformation properties of cold-based polar glaciers, we examine here some of the factors leading to the localization of strain within the amber ice facies. We present a crystallographic case study of amber ice (a fine-grained bubbly ice containing a relatively high impurity content) sampled at the base of two Antarctic glaciers. The crystal fabrics and textures of amber ice were computed by application of a recently developed automated method. To date, it was tedious and awkward to determine amber ice facies accurately because of the sub-millimetric crystal size and relatively high debris content of this facies. The authomatic analytical method applied in this study allows not only for improving analytical accuracy in this task but also for considerably reducing the time of analysis. Our investigations reveal highly homogeneous crystallographic properties for the studied amber ice. The ice crystals are mainly polygonal, equant and sub-millimetric, and show a strong lattice-preferred orientation. These properties, beside the relatively high impurity content, are likely to exert a major control on strain enhancement in amber ice when this facies is present at the base of cold glaciers.

Glaciers are permeated by liquid veins that line the triple junctions between ice grains and act as conduits for material transport. The density difference between water and ice produces a hydraulic gradient that drives liquid flow down towards the glacier bed. We investigate how variations in this transport rate are manifest in the lower regions of a glacier that floats on a subglacial reservoir. When the temperature rises towards the glacier bed, the associated permeability increase leads to more rapid fluid transport; internal melting supplies the changing flow. A reduction in hydraulic gradient results from surface energy effects, and causes a decreasing transport rate near the base; hence, freezing occurs within the polycrystalline ice. For representative values of the heat flux, soluble impurity loading and grain-size, the downward flux peaks at approximately 1–100mma–1 several tens of centimeters above the glacier bed. Beneath this level, the effects of surface energy control the fluid transport and cause the flow to reverse directions, pulling liquid upwards into the polycrystalline ice so that the flow does not introduce glacial material into the underlying reservoir.

The Canadian Lake Ice Model (CLIMo), a one-dimensional, thermodynamic model with unsteady heat conduction and penetrating solar radiation, is used to simulate ice growth and decay on shallow ponds in and near Fairbanks, central Alaska, USA. Simulations are compared with observations of ice thickness and composition (snow ice, congelation ice), freeze-up, break-up and duration. Simulations run using US National Weather Service meteorological data as input variables do not agree well with ice-thickness measurements. The simulations improve significantly when the model is run with more representative, locally measured data for air temperature and depth of snow on the ice. The causes of some discrepancies between simulations and observations are discussed and some suggestions for model improvements are made.

The 2001/02 austral summer was the warmest summer on record in Taylor Valley, Antarctica, (∼78° S) since continuous records of temperature began in 1985. The highest stream-flows ever recorded in the Onyx River, Wright Valley, were also recorded that year (the record goes back to the 1969/70 austral summer). In early January 2002, a groundwater seep was observed flowing in the southwest portion of Taylor Valley. This flow has been named ‘Wormherder Creek’ (WHC) and represents an unusual event, probably occurring on a decadal time-scale. The physical characteristics of this feature suggest that it may have flowed at other times in the past. Other groundwater seeps, emanating from the north-facing slope of Taylor Valley, were also observed. Little work has been done previously on these very ephemeral seeps, and the source of water is unknown. These features, resembling recently described features on Mars, represent the melting of subsurface ice. The Martian features have been interpreted as groundwater seeps. In this paper we compare the chemistry of the WHC groundwater seep to that of the surrounding streams that flow every austral summer. The total dissolved solids content of WHC was ∼6 times greater than that of some nearby streams. The Na : Cl and SO4 : Cl ratios of the seep waters are higher than those of the streams, but the Mg : Cl and HCO3 : Cl ratios are lower, indicating different sources of solutes to the seeps compared to the streams. The enrichment of Na and SO4 relative to Cl may suggest significant dissolution of mirabilite within the previously unwetted soil. The proposed occurrence of abundant mirabilite in higher-elevation soils of the dry valley region agrees with geochemical models developed, but not tested, in the late 1970s. The geochemical data demonstrate that these seeps could be important in ‘rinsing’ the soils by dissolving and redistributing the long-term accumulation of salts, and perhaps improving habitat suitability for soil biota. The H4SiO4 concentration is 2–3 times greater in WHC than in the surrounding streams, indicating a large silicate-weathering component in the seep waters.

A marginal ice-contact lake at Miage glacier, Mont Blanc, Italian Alps, has been studied to reconstruct changes in lake area. Historical sources, comprising sketches, maps, photographs and scientific surveys, have been supplemented by recent field surveys. These include surveys of glacier surface velocity (which varied along the glacier tongue from 70 ma–1 in the upper part to about 6 ma–1 close to the snout, consistent with data in the literature, showing that velocity rates have remained constant during the last 40 years), volumetric ice-cliff loss (–92 000±180m3 in 2002–03), lake temperature and bathymetry, and qualitative observation of calving events, crevassing, and meltwater production. Results indicate that the lake has been stable for the last half-century following a period of enlargement due to ice-marginal retreat. The lake hydrology is complex, with possible reversals of englacial water flow causing infrequent emptying episodes. The debris cover on the glacier and ice-cliff surfaces seems to have played an important role in the ice-cliff evolution and the calving phenomena; calving is driven by undercutting at the water-line aided by the opening of water- and debris-filled crevasses in the glacier surface.

The disintegration of Larsen A and B ice shelves in 1995 and 2002, respectively, was preceded by intense surface melting during the summer of ice-shelf collapse and previous summers. To understand the transition of the ice-shelf surface from dry to wet conditions, we developed a one-dimensional model, describing the mass, heat and force balances of water and firn within the ice-shelf surface layer. The model is run using atmospheric data from an automatic weather station on Larsen C ice shelf (World Meteorological Organization station ‘Larsen Ice Site’) located south of Larsen A and B. The model’s derived melting rate is greater than melting predicted by the positive degree-day (PDD) approach, common in studies of ablating ice sheets, such as Greenland. The model shows that the years of ice-shelf break-up (1995 and 2002) are distinguished from previous years by local maxima in the number of melting days. When the PDD approach is considered, however, a maximum in the number of positive degree-days appears in the 2002 break-up year, but not in 1995.

Sub- and proglacial bed conditions influence advance and retreat of an ice sheet. The existence and distribution of frozen ground is of major importance for better understanding of ice-flow dynamics and landform formation. The southern margin of the Laurentide ice sheet (LIS) was dominated by the presence of relatively thin ice lobes that seem to have been very sensitive to external and internal physical conditions. Their extent and dynamics were highly influenced by the interaction of subglacial and proglacial conditions. A three-dimensional thermomechanical ice-sheet model was coupled with a model for the thermal regime in the upper Earth crust. The model has been applied to the LIS in order to investigate the spatial distribution of thermal conditions at the bed. The evolution of the whole LIS was modeled for the last glacial cycle, with primary attention on correct reconstruction of the southern margin. Our results show extensive temporal and spatial frozen ground conditions. Only a slow degradation of permafrost under the ice was found. We conclude that there are significant interactions between the ice sheet and the underlying frozen ground and that these influence both ice dynamics and landform development.

The Ragnhild glaciers are three enhanced-flow features situated between the Sør Rondane and Yamato Mountains in eastern Dronning Maud Land, Antarctica. We investigate the glaciological mechanisms controlling their existence and behavior, using a three-dimensional numerical thermomechanical ice-sheet model including higher-order stress gradients. This model is further extended with a steady-state model of subglacial water flow, based on the hydraulic potential gradient. Both static and dynamic simulations are capable of reproducing the enhanced ice-flow features. Although basal topography is responsible for the existence of the flow pattern, thermomechanical effects and basal sliding seem to locally soften and lubricate the ice in the main trunks. Lateral drag is a contributing factor in balancing the driving stress, as shear margins can be traced over a distance of hundreds of kilometers along west Ragnhild glacier. Different basal sliding scenarios show that central Ragnhild glacier stagnates as west Ragnhild glacier accelerates and progressively drains the whole catchment area by ice and water piracy.