Glacier-dammed lakes can yield subglacial outburst floods (jökulhlaups) repeatedly. Predicting flood timing is crucial for hazard mitigation, but incomplete understanding of flood-initiation physics makes this challenging. Here we examine the predictability of the timing of jökulhlaups from Merzbacher Lake, Kyrgyzstan, using five flood-date prediction models of varying complexity. The simplest model, which offers a benchmark against which the other models are compared, assumes that floods occur on the same date each year. The other four models predict flood dates using a flood-initiation threshold approach and incorporate weather forcing (approximated by the output of two climate reanalyses) behind the meltwater input to the lake; the most complex of these models accounts for a moving subglacial water divide beneath the glacier that dams the lake. Each model is optimized against recorded flood dates to maximize its prediction ability. In terms of their flood prediction ability, our two best models are those that assume a variable outburst threshold governed by the rate of meltwater input to the lake and the rate of lake-level rise. They excel over the simplest and most complex models and correctly predict flood dates to within ±20 days 57.4% of the time. We also quantify the impact of weather uncertainty on prediction success. Our findings can inform practical flood-forecasting schemes and future investigations of flood-initiation physics.

We report sea-ice temperature and bulk salinity measurements as well as textural analysis from 33 first-year drift- and fast-ice stations sampled between November 2007 and June 2008 in the southern Beaufort Sea–Amundsen Gulf, Canadian Arctic, during the International Polar Year Circumpolar Flaw Lead (IPY-CFL) system study. We use this significant dataset to investigate the halo-thermodynamic evolution of sea ice from growth to melt. A strong desalination phase is observed over a small time window in the spring. Using calculated proxies of sea-ice permeability (brine volume fraction) and of the intensity of brine convection (Rayleigh number) we demonstrate that this phase corresponds to full-depth gravity drainage initiated by a restored connectivity of the brine network with warming in the spring. Most stations had a textural sequence typical of Arctic first-year ice, with granular ice overlying columnar ice. Unusual textural features were observed sporadically: sandwiched granular ice, platelet ice and draped platelet ice. We suggest that turbulence in leads and double diffusion in strong brine plumes following the refreeze of cracks are plausible mechanisms for the formation of these textures.

As part of the Ross Ice Shelf Geophysical and Glaciological Survey (RIGGS), ice velocities were measured on the Ross Ice Shelf (RIS) during 1973–78. Comparisons of these with velocity estimates at the same locations derived from RADARSAT synthetic aperture radar (SAR) measurements in 1997 and 2009 show velocity reduction in the southeast quadrant of the ice shelf by almost 200 m a−1, with deceleration rates increasing with time. Large areas of ice shelf in this region are lightly grounded, forming an ‘ice plain’ that increases local buttressing of the ice streams. ICESat measurements show this ice plain to be thickening. The observed decrease in ice-shelf velocities implies a total reduction in the mass of ice flowing into the RIS from the West Antarctic ice sheet (WAIS) by ~23 Gt a−1, shifting the mass balance of the WAIS drainage basin from strongly negative in the 1970s to strongly positive in 2009. The resulting decrease in ice advection should lead to ice-shelf thinning further seaward of the ice plain. This thinning would reduce the lateral drag and back-stress of the shelf ice, further contributing to thinning through an increase in spreading rate. ICESat measurements show recent thinning of most of the freely floating ice shelf.

We explore the small-scale spatial and temporal transferability of model parameters between two points in the ablation zone of tropical Glaciar Shallap, Cordillera Blanca, Peru (9°S, −77° W; ~4800 m a.s.l.) in order to provide a robust assessment of the performance of a process-based glacier mass-balance model. Relative surface height change is calculated at hourly time-steps, and cumulative values are compared to surface height measurements made at irregular intervals (14–64 days) over the course of two continuous hydrological years (August 2006–August 2008). Best-performing parameter combinations were determined for each point from the outcome of 1000 model simulations for which parameters were varied randomly within a defined range. With these parameter combinations measurements for a specific location and time-span are well reproduced. Transferring the parameter combination as optimized for one location to the other location in the ablation zone increases the errors of modeled cumulative mass balance by 5–1326 mm ice eq.a−1. Transferring the parameter combinations as optimized for one year to the other year increases the modeled errors in cumulative mass balance by 18–3179 mm ice eq.a−1. Model errors generally increase during periods with frequent snowfall and snow cover. This could reflect either the inherent difficulty of modeling complex snow processes, or the inability of the model to correctly capture the pattern of albedo evolution at this site. The magnitude of errors associated with parameter transfer in space and time highlights the need for improving model performance for robust climatological and/or hydrological analyses on tropical glaciers.

X-ray microtomography has become an essential tool for investigating the mechanical and physical properties of snow, which are tied to its microstructure. To allow a quantitative characterization of the microstructure, the grayscale X-ray attenuation coefficient image has to be segmented into a binary ice/pore image. This step, called binary segmentation, is crucial and affects all subsequent analysis and modeling. Common segmentation methods are based on thresholding. In practice, these methods present some drawbacks and often require time-consuming manual post-processing. Here we present a binary segmentation algorithm based on the minimization of a segmentation energy. This energy is composed of a data fidelity term and a regularization term penalizing large interface area, which is of particular interest for snow where sintering naturally tends to reduce the surface energy. The accuracy of the method is demonstrated on a synthetic image. The method is then successfully applied on microtomographic images of snow and compared to the threshold-based segmentation. The main advantage of the presented approach is that it benefits from local spatial information. Moreover, the effective resolution of the segmented image is clearly defined and can be chosen a priori.

Ground-penetrating radar (GPR) surveys at a center frequency of 500 MHz were used to determine winter (2007/08) and net annual (2005–07) snow water equivalent (SWE) patterns across the upper parts of Belcher Glacier, Devon Ice Cap, Nunavut, Canada. The GPR measurements were validated against snow depths determined from avalanche probe measurements, and converted to SWE values using densities measured with a down-borehole neutron density probe and in shallow snow pits. Distinct internal reflection horizons (IRHs) in the GPR record were formed during warm summers in 2007 and 2005, and a large rain event in summer 2006 which caused ice to accumulate above the 2005 melt surface. Elevation provides the dominant control on winter SWE distribution across the basin, with surface topography (e.g. gullies) also being locally important. Based on the location where IRHs intersected the ice-cap surface, the basin-wide firn line occurred at an altitude of 1260–1300 m over the period 2005–08. Net mass balance across the accumulation area of Belcher Glacier averaged 0.24 m w.e. a−1 over the period 2005–07, mainly dependent on altitude. This is a little higher than most previous estimates for the period since the 1960s, although the differences lie within error limits.

Short-term ice-dynamical processes at Greenland’s Jakobshavn and Kangerdlugssuaq glaciers were studied using a 3 day time series of synthetic aperture radar data acquired during the 2011 European Remote-sensing Satellite-2 (ERS-2) 3 day repeat campaign together with modelled meteorological parameters. The time series spans the period March–July 2011 and captures the first ~30% of the summer melting season. In both study areas, we observe velocity fluctuations at the lower ~10 km of the glacier. At Jakobshavn Isbræ, where our dataset covers the first part of the seasonal calving-front retreat, we identify ten calving episodes, with a mean calving-front area loss of 1.29 ± 0.4 km2. Significant glacier speed-up was observed in the near-terminus area following all calving episodes. We identify changes in calving-front geometry as the dominant control on velocity fluctuations on both glaciers, apart from a <15% early-summer speed-up at Kangerdlugssuaq Glacier during a period of calving-front advance, which we attribute to enhanced surface melt-induced basal lubrication. Our 3 day velocity maps show new spatial characteristics of the ice melange flow variability in the Jakobshavn and Kangerdlugssuaq fjord systems, which are primarily controlled by calving-front dynamics and fjord geometry.

Winter balance is an important metric for assessing the change on glaciers and ice caps, yet measuring it using ground-based techniques can be challenging. We use the European Space Agency prototype Airborne SAR/Interferometric Radar Altimeter System (ASIRAS) to extract snow depths from the received altimeter waveforms over Austfonna ice cap, Svalbard. Additionally, we attempt to distinguish the long-term firn area from other glacier facies. We validate our results using snow depth and glacier facies characterizations determined from ground-based radar profiles, snow pits and a multi-look satellite synthetic aperture radar image. We show that the depth of the winter snowpack can be extracted from the altimeter data over most of the accumulation zone, comprising wet snow zone and a superimposed ice zone. The method struggles at lower elevations where internal reflections within the winter snowpack are strong and the winter snow depth is less than ~1 m. We use the abruptness of the reflection from the last summer surface (LSS) to attempt to distinguish glacier facies. While there is a general correlation between LSS abruptness and glacier facies, we do not find a relationship that warrants a distinct classification based on ASIRAS waveforms alone.

Recent acceleration and thinning of Thwaites Glacier, West Antarctica, motivates investigation of the controls upon, and stability of, its present ice-flow pattern. Its eastern shear margin separates Thwaites Glacier from slower-flowing ice and the southern tributaries of Pine Island Glacier. Troughs in Thwaites Glacier’s bed topography bound nearly all of its tributaries, except along this eastern shear margin, which has no clear relationship with regional bed topography along most of its length. Here we use airborne ice-penetrating radar data from the Airborne Geophysical Survey of the Amundsen Sea Embayment, Antarctica (AGASEA) to investigate the nature of the bed across this margin. Radar data reveal slightly higher and rougher bed topography on the slower-flowing side of the margin, along with lower bed reflectivity. However, the change in bed reflectivity across the margin is partially explained by a change in bed roughness. From these observations, we infer that the position of the eastern shear margin is not strongly controlled by local bed topography or other bed properties. Given the potential for future increases in ice flux farther downstream, the eastern shear margin may be vulnerable to migration. However, there is no evidence that this margin is migrating presently, despite ongoing changes farther downstream.

We present a high-resolution Gravity Recovery and Climate Experiment (GRACE) mascon solution for Gulf of Alaska (GOA) glaciers and compare this with in situ glaciological, climate and other remote-sensing observations. Our GRACE solution yields a GOA glacier mass balance of −65 ± 11 Gt a−1 for the period December 2003 to December 2010, with summer balances driving the interannual variability. Between October/November 2003 and October 2009 we obtain a mass balance of −61 ± 11 Gt a−1 from GRACE, which compares well with −65 ± 12 Gt a−1 from ICESat based on hypsometric extrapolation of glacier elevation changes. We find that mean summer (June–August) air temperatures derived from both ground and lower-troposphere temperature records were good predictors of GRACE-derived summer mass balances, capturing 59% and 72% of the summer balance variability respectively. Large mass losses during 2009 were likely due to low early melt season surface albedos, measured by the Moderate Resolution Imaging Spectroradiometer (MODIS) and likely associated with the 31 March 2009 eruption of Mount Redoubt, southwestern Alaska. GRACE data compared well with in situ measurements at Wolverine Glacier (maritime Alaska), but poorly with those at Gulkana Glacier (interior Alaska). We conclude that, although GOA mass estimates from GRACE are robust over the entire domain, further constraints on subregional and seasonal estimates are necessary to improve fidelity to ground observations.

Many glaciological and hydrological studies require outlines of individual glaciers rather than total ice cover. Here we develop a new semiautomatic algorithm that uses a digital elevation model (DEM) and outlines of glacier complexes to calculate the extents of individual glaciers. The algorithm first applies hydrological modeling tools to a modified DEM to calculate flowsheds. It then merges flowsheds that belong to individual glaciers using a distance-based approach, whose required empirical parameters are derived from the Juneau Icefield area in Alaska. In this region, 2% of ~1300 glaciers were misclassified. The algorithm was validated on >25 000 km2 of ice in other regions in Alaska and on >40 000 km2 of ice in Arctic Canada, resulting in ~2% and ~3% misclassified glaciers, respectively. Results indicate that the algorithm is robust provided the DEM and the outlines are of good quality.

The refreezing rate of a borehole drilled through a 252 m thick region of the Ross Ice Shelf, Antarctica, is determined using oceanographic measurements over two periods of a day. We first use a method based on the conservation of salt in the supercooled salt water of the borehole. This is compared to a model using a numerical solution of the heat equation to find the temperature distribution in the host ice, allowing ice growth to be calculated from the balance of heat fluxes at the ice/water interface. This second method broadly confirms the refreezing rates deduced from salinity measurements, giving confidence in the generalization of this simple heat-flux model to predict refreezing rates of other boreholes. Predictions from both are subject to uncertainty due to the poorly defined value of the solid fraction of ice that freezes in a supercooled volume of sea water. This is taken to be 0.5 ± 0.1 throughout this study. The predicted rates are also strongly dependent on the initial and boundary conditions chosen, but results show the initial diameter of 600 mm decreases at a rate of ~3–5 mm h−1 in an ice shelf with a minimum temperature of −22°C.

We study the approximations underlying the macroscopic theory of glacier response to climate, and illustrate some basic properties of glacier response using two simple examples, a block on an inclined plane and a block with a more realistic terminal region. The properties include nonlinearity and the usefulness of linear approximations, sensitivity to bed slope, timescale, stability, characteristic elevations of the equilibrium line at which the properties of the response change, the limit of fast response in a changing climate, the minimum sustainable size, increased sensitivity to climate change as a glacier retreats, and finally the similarity in the responses of simple glaciers with the same product of length and square of bed slope.

Recent studies of Himalayan glacier recession indicate that there is wide variability in terminus retreat rate and mass balance in the different sectors of the mountain range, primarily linked to the topography and climate of the region. Variable retreat rates of glacier termini and inadequate supporting field data (e.g. mass balance, ice thickness, velocity, etc.) in the Himalayan glaciers make it difficult to develop a coherent picture of climate change impacts. In this study, the results of a detailed mapping campaign and ground-based measurements of ablation rate, terminus retreat and ice loss are reported for the period 2003–10. In addition, background information from an old glacier map (Survey of India, 1962) was compiled and terminus recession measurements were carried out from 1990 field photographs of Chorabari Glacier, central Himalaya. Our ablation stake network results suggest that the influence of debris cover is significant for Chorabari Glacier mass balance and terminus retreat. The terminus survey finds that the glacier is retreating, but at a lower rate than many other non-debriscovered glaciers in the region. The recession and ablation data (particularly in the upper ablation area at higher altitudes) suggest that the ice volume loss of the glaciers is of greater magnitude than the slow terminus retreat and, if the process continues, the lowermost part of the glacier may reduce to a quasi-stationary position while significant ice loss continues.

We investigated the large-scale (10–1000 m) and small-scale (mm–cm) variations in size, number and arrangement of air bubbles in the EPICA Dronning Maud Land (EDML) (Antarctica) ice core, down to the end of the bubble/hydrate transition (BHT) zone. On the large scale, the bubble number density shows a general correlation with the palaeo-temperature proxy, δ18O, and the dust concentration, which means that in Holocene ice there are fewer bubbles than in glacial ice. Small-scale variations in bubble number and size were identified and compared. Above the BHT zone there exists a strong anticorrelation between bubble number density and mean bubble size. In glacial ice, layers of high number density and small bubble size are linked with layers with high impurity content, identified as cloudy bands. Therefore, we regard impurities as a controlling factor for the formation and distribution of bubbles in glacial ice. The anticorrelation inverts in the middle of the BHT zone. In the lower part of the BHT zone, bubble-free layers exist that are also associated with cloudy bands. The high contrast in bubble number density in glacial ice, induced by the impurities, indicates a much more pronounced layering in glacial firn than in modern firn.

Estimating regional-scale rates of glacier mass change from repeat airborne and satellite altimetry requires assumptions about the density of the material being added to, or removed from, the glacier surface. This may vary in both space and time, especially in a period of warming climate such as has occurred in Arctic Canada since 2005. Here we compare firn properties measured on 20 shallow cores recovered from elevations of 1400–1900 m a.s.l. on Devon Ice Cap in 2012 with equivalent measurements from 34 cores recovered from the same locations between 2004 and 2011. The average density of the top 2.5 m of the firn column has increased by 13–80%. As a result, measured rates of density increase in the top 1.25 mw.e. of the firn layer have likely thinned this layer at rates of 0.021–0.168 m a−1 over the period 2004–12. These thinning rates are significant relative to local rates of surface height change measured by repeat airborne laser altimetry between 2004 and 2012 (−0.077 to −0.252 m a−1) and indicate that, during periods of climate warming or cooling, firn densification rate changes must be taken into account when estimating mass change rates from measured height changes.

Studies addressing the response of glaciers to climate change have so far analyzed the effect of long-term trends in a particular set of meteorological variables only, implicitly assuming an unaltered climatic variability. Here a framework for distinguishing between year-to-year, month-to-month and day-to-day variability is proposed. Synthetically generated temperature and precipitation time series following the same long-term trend but with altered variability are then used to force an ice-dynamics model set up for Rhonegletscher, Swiss Alps. In the case of temperature, variations in the day-to-day variability are shown to have a larger effect than changes at the yearly scale, while in the case of precipitation, variability changes are assessed as having negligible impact. A first set of scenarios is used to show that compared to reference, doubling the temperature variability can reduce glacier ice volume by up to 64% within half a decade. A second set derived from the results of the European ENSEMBLES project, however, shows that such changes are expected to remain below 8% even for extreme scenarios. Although the latter results relativize the importance of the effect in the near future, the analyses indicate that at least caution is required when assuming ‘unchanged variability’.