Past atmospheric pollution can be reconstructed from ice core trace element records retrieved from mountain glaciers. However, the current global temperature increase can result in post-depositional melt processes, significantly altering the originally stored information. Here, we present a comprehensive study on the behaviour of 35 trace elements (TEs) during meltwater percolation in a high-Alpine ice core segment from upper Grenzgletscher, Switzerland. Some TEs revealed significant concentration depletion, whereas others were well preserved depending on their water solubility and location at the grain scale. TEs present in insoluble minerals, typically enriched at grain boundaries, were found to be mostly preserved because their insolubility in water results in immobility with meltwater percolation. Water-soluble TEs revealed a variable meltwater-mobility. Whereas ultra-TEs tend to be preserved, likely due to incorporation into the ice lattice, abundant TEs are prone to relocation from grain-boundary regions. We propose that at Alpine sites, Ag, Al, Bi, Cu, Cs, Fe, Li, Mo, Pb, Rb, Sb, Th, Tl, U, V, W, Zr and the rare-earth elements may still be applicable as robust environmental proxies even if partial melting occurred, whereas Ba, Ca, Cd Co, Mg, Mn, Na, Ni, Sr and Zn are prone to significant depletion.

The Greenland ice sheet is experiencing dramatic melt that is likely to continue with rapid Arctic warming. However, the proportion of meltwater stored before reaching the global ocean remains difficult to quantify. We use NASA MODIS surface reflectance data to estimate river discharge from two West Greenland rivers – the Watson River near Kangerlussuaq and the Naujat Kuat River near Nuuk – over the summers of 2000–12. By comparison with in situ river discharge observations, ‘inundation–discharge’ relations were constructed for both rivers. MODIS-based total annual discharges agree well with total discharge estimated from in situ observations (86% of summer discharge in 2009 to 96% in 2011 at the Watson River, and 106% of total discharge in 2011 to 104% in 2012 at the Naujat Kuat River). We find, however, that a time-lapse camera, deployed at the Watson River in summer 2012, better captures the variations in observed discharge, benefiting from fewer data gaps due to clouds. The MODIS-derived estimates indicate that summer discharge has not significantly increased over the last decade, despite a strong warming trend. Also, meltwater runoff estimates derived from the regional climate model RACMO2/GR for the drainage basins are higher than our reconstructions of river discharge. These results provide indirect evidence for a considerable component of water storage within the glacio-hydrological system.

The geometry of the ice sheets during the Pliocene to early Pleistocene is not well constrained. Here we apply an ice-flow model in the study of the Greenland ice sheet (GIS) during three extreme intervals of this period constrained by geological observations and climate reconstructions. We study the extent of the GIS during the Mid-Pliocene Warmth (3.3–3.0 Ma), its advance across the continental shelf during the late Pliocene to early Pleistocene glaciations (3.0–2.4 Ma) as implied by offshore geological studies, and the transition from glacial to interglacial conditions around 2.4 Ma as deduced from the deposits of the Kap København Formation, North Greenland. Our experiments show that no coherent ice sheet is likely to have existed in Greenland during the Mid-Pliocene Warmth and that only local ice caps may have been present in the coastal mountains of East Greenland. Our results illustrate the variability of the GIS during the Pliocene to early Pleistocene and underline the importance of including independent estimates of the GIS in studies of climate during this period. We conclude that the GIS did not exist throughout the Pliocene to early Pleistocene, and that it melted during interglacials even during the late Pliocene climate deterioration.

The possibility of replacing or supplementing direct measurements of mass balance by estimates calculated from equilibrium-line altitude (ELA) measurements is investigated by statistical analyses of data from 31 glaciers. A linear relationship between mass balance and ELA in terms of two parameters, the effective balance gradient and the balanced-budget ELA, is tested. It is concluded that existing mass-balance series can be usefully extrapolated by using ELA data for additional years. However, accurate mass balance cannot be calculated for glaciers where no such measurements have been made because of the difficulties in prescribing the two model parameters with sufficient accuracy. For example, the effective balance gradient is of the order of 5 mm water/m so that errors of only a few decametres in the estimation of the balanced-budget ELA can have a great effect upon calculations of mass balance.

Synthetic-aperture radar interferometry data and airborne ice-sounding radar (ISR) data are employed to obtain modern estimates of the inland ice production from Nioghalvfjerdsbræ (NB) and Zachariae Isstrøm (ZI), the two largest glaciers draining the northeast sector of the Greenland ice sheet. Ice fluxes are measured at the grounding line (14.2 ±1 km3 ice a−1 for NB and 10.8 ±1 km3 ice a−1 for ZI) with an ice thickness deduced from ice-shelf hydrostatic equilibrium, and along an ISR profile collected upstream of the grounding line (14.3 ± 0.7 km3 ice a−1 for NB and 11.6 ± 0.6 km3 ice a−1 for ZI). Balance fluxes calculated from a map of snow accumulation and model predictions of surface melt are 11.9 ± 2 km3 ice a−1 for NB and 10.0 ± 2 km3 ice a−1 for ZI at the grounding line, and 12.2 and 10.3 km3 ice a−1, respectively, at the ISR line. The two glaciers therefore exhibit a negative mass balance equivalent to 14% of their balance flux, with a ±12% uncertainty. Independently, we detect a retreat of the grounding line of NB between 1992 and 1996 which is larger at the glacier center (920 ± 250 m) than on the sides (240 ± 50 m). The corresponding ice-thinning rates (2 ± 1 m a−1 at the glacier center and 0.6 ± 0.3 m a−1 on the sides) are too large to be accommodated by temporal changes in ablation or accumulation, and must be due to dynamic thinning.

The dynamics of a marine ice sheet's grounding lines determine the rate of ice discharge from the grounded part of ice sheet into surrounding oceans. In many locations in West Antarctica ice flows into ice shelves through ice streams experiencing low driving stress. However, existing simple theories of marine ice sheets are developed under the assumption of high basal and driving stress. Here we analyze the grounding line behavior of marine ice streams experiencing low basal shear and driving stress. We find that in this regime, the ice flux at the grounding line is a complex function of the geometry of the ice-stream bed, net accumulation rate and gradient of the net accumulation rate. Our analysis shows that the stability of distinct steady states is determined by the same parameters, suggesting a more complex (in)stability criterion than what is commonly referred to within the context of the ‘marine ice-sheet instability hypothesis’. We also determine characteristic timescales (e-folding time) of ice-sheet configurations perturbed from their steady states. These timescales can be used to determine whether particular configurations can be considered in isolation from other components of the climate system or whether their effects and feedbacks between the ice sheet and the rest of the climate system have to be taken into account.

Air bubbles in ice cores play an essential role in climate research, not only because they contain samples of the palaeoatmosphere, but also because their shape, size and distribution provide information about the past firn structure and the embedding of climate records into deep ice cores. In this context, we present profiles of average bubble size and bubble number for the entire EDML (Antarctica) core and the top 600 m of the EDC (Antarctica) core, and distributions of bubble sizes from selected depths. The data are generated with an image-processing framework which automatically extracts position, orientation, size and shape of an elliptical approximation of each bubble from thick-section micrographs, without user interaction. The presented software framework allows for registration of overlapping photomicrographs to yield accurate locations of bubble-like features. A comparison is made between the bubble parameterizations in the EDML and EDC cores and data published on the Vostok (Antarctica) ice core. The porosity at the firn/ice transition is inferred to lie between 8.62% and 10.48% for the EDC core and between 10.56% and 12.61 % for the EDML core.

The ice-sheet margin at Russell Glacier, West Greenland, advanced ∼7 m a−1 between 1968 and 1999. As the ice advanced over moraine ridges, small changes in position caused major changes in the routing of proglacial water and sediment. These included changes in the distribution of ice-marginal lakes, in the periodic drainage of ice-dammed lakes, in the routing and sediment content of meltwater draining into the proglacial zone, and in the release of sediment from the moraines by erosion and mass movements. Proglacial hydrology and sediment flux appear to be controlled not simply by glacier mass balance, but by evolving ice-marginal geomorphology, which must be accounted for in palaeoenvironmental interpretation of proglacial sediments.

Creep tests were performed in torsion and torsion–compression on polycrystalline ice at temperatures near the melting point. Syntectonic recrystallization occurs at strains of the order of 2–3%, leading to a rapid increase in strain-rate. It is shown that the increase of creep-rate during tertiary creep arises from the development of fabrics favouring the glide on basal planes but also from the softening processes associated with recrystallization. The c-axis fabric of recrystallized ice developed in simple shear consists of two-maxima, one at the pole of the permanent shear plane and the other between the normal of the second plane of maximum shearing stress and the principal direction of compression. In torsion–compression, a three- or four-maximum fabric is formed according to the intensity of different components of the stress tensor. The maxima are clustered around the principal direction of compression. Processes of fabric formation are discussed. The experimentally developed fabrics are probably produced by the strain-induced recrystallization, for which the driving force is provided by differences in stored plastic strain energy. However the degree of preferred orientation of ice c-axes must be a function of the total strain when syntectonic recrystallization becomes less important. In this case, fabrics are principally formed by plastic flow and a steady state is obtained for very high strains.

New experiments reveal that the flexural strength of fresh water, columnar-grained ice loaded normal to the columns may be increased by about a factor of two upon reversed cyclic loading at a frequency of ~0.1 Hz at stress amplitudes of 1.3–2.6 MPa. The effect is explained in terms of a reduction in deformation-induced stress concentration through the activation of grain boundary sliding, which is evident through boundary decohesion.

Variations in accumulation rate As(t) and temperature Ts(t) at the surface of firn cause changes in the rate of firn compaction (FC) and surface height H(t) that do not involve changes in mass, and therefore need to be accounted for in deriving mass changes from measured H(t). As the effects of changes in As(t) and Ts(t) propagate into the firn, the FC rate is affected with a highly variable and complex response time. The H(t) during measurement periods depend on the history of As(t) and Ts(t) prior to the measurements. Consequently, knowledge of firn response times to climate perturbations is important to estimate the required length of the time series of As(t) and Ts(t) used in FC models. We use our numerical FC model, which is time-dependent on both temperature and accumulation rate, to examine the response times of both H(t) and the rates of change dH(t)/dt to variations in As(t) and Ts(t) using sample perturbations and climate data for selected sites in Antarctica. Our results show that the response times for dH(t)/dt, which are of particular interest, are much shorter than the responses of H(t). Typical response times of dH(t)/dt are from several years to <20 years. The response times are faster in warmer and higher-accumulation areas such as Byrd Station, West Antarctica (4 years), and slower in colder and lower-accumulation areas such as Vostok, East Antarctica (18 years). The response times to temperature are much faster (0.9 year at Byrd and 2.2 years at Vostok), but the corresponding height changes persist much longer. The associated variations in firn density are significantly preserved in the density–depth profiles. For typical fluctuations of surface weather, the Ts(t) from satellite observations since 1982 and As(t) from meteorological data since 1979 are essentially of sufficient length to correct for FC height changes for measurements beginning in 1992.

Snow properties relevant to the fracture processes involved in dry-snow slab avalanche release include weak layer specific fracture energy, slab elastic modulus and density. Various techniques exist to determine these snow mechanical properties, but it is presently unclear how values determined with different methods compare. In the laboratory, the 3-D microstructure of cm-sized snow samples is reconstructed by micro-computed tomography (μCT) so that density and elastic modulus can be computed. In the field, fracture energy and modulus are estimated based on particle tracking velocimetry (PTV) of the displacement field observed during propagation saw tests. Snow stratigraphy is measured with the snow micro-penetrometer (SMP) in either, field or laboratory. We compared SMP-derived properties to corresponding μCT- and PTV-derived values. Values of snow density related well to μCT results and so were SMP-derived elastic moduli related to PTV-derived values. By taking into account snow anisotropy a good relation between SMP- and μCT-derived moduli resulted suggesting the SMP-derived modulus characterizes the components of the modulus perpendicular to the axis of penetration. SMP- and PTV-derived values of fracture energy were correlated. The SMP can provide a bridge between scales and techniques, yet further improvements in signal interpretation are still needed.

The horizontal and vertical motions of eight points on the surface of Highway Glacier, Baffin Island are reported. The average horizontal speed below the lowest tributary glacier is 56 metres/year. The major part of this motion arises from the glacier sliding on its bed. The mean shear stress on the bed is practically constant and equal to about 0.9 bars, but it is much smaller beneath the retreating tongue. A measure of the retreat in the lower part of the glacier is obtained from the difference between the ice discharge and the net ablation.

Unlike in most other regions, Karakoram glaciers are either stable or advancing, a phenomenon known as the Karakoram anomaly. Despite studies of glacier surges and the derivation of surface velocity maps, the spatiotemporal variability of glacier dynamics still remains poorly understood, particularly in the Eastern Karakoram Range. We use Advanced Land Observing Satellite/the Phased Array type L-band Synthetic Aperture Radar (ALOS/PALSAR)-1/2 data from 2007 to 2011 and 2014 to 2015 to examine detailed surface velocity patterns of the Siachen, Baltoro, Kundos, Singkhu and Gasherbrum Glaciers. The first three glaciers show considerable velocity variability (20–350 m a−1), with clear seasonal patterns. Although all glaciers, except for Baltoro, flow slowest in 2015, the velocity structures are individual and vary in space and time. In Gasherbrum Glacier, peak surge-phase velocities are seasonally modulated, with maxima in summers 2006 and 2007, suggesting surface melt plays an important role in maintaining the active phase. Given the relatively close proximity of these glaciers, we assume that surface melt timing and rates are comparable. We therefore argue that the observed spatiotemporal and interannual velocity patterns are determined by local and internal mechanisms, including englacial and subglacial hydrology, thermal processes and tributary configuration of each individual glacier.

Dissolved and particulate sodium, magnesium and calcium are analyzed in ice cores to determine past changes in sea ice extent, terrestrial dust variability and atmospheric aerosol transport efficiency. They are also used to date ice cores if annual layers are visible. Multiple methods have been developed to analyze these important compounds in ice cores. Continuous flow analysis (CFA) is implemented with instruments that sample the meltstream continuously. In this study, CFA with ICP-MS (inductively coupled-plasma mass spectrometry) and fast ion chromatography (FIC) methods are compared for analysis of sodium and magnesium. ICP-MS, FIC and fluorescence methods are compared for analysis of calcium. Respective analysis of a 10 m section of the Antarctic WACSWAIN Skytrain Ice Rise ice core shows that all of the methods result in similar levels of the compounds. The ICP-MS method is the most suitable for analysis of the Skytrain ice core due to its superior precision (relative standard deviation: 1.6% for Na, 1.3% for Mg and 1.2% for Ca) and sampling frequency compared to the FIC method. The fluorescence detection method may be preferred for calcium analysis due to its higher depth resolution (1.4 cm) relative to the ICP-MS and FIC methods (~4 cm).

Deposition of snow from precipitation and wind events creates layering within seasonal snowpacks. The thickness and horizontal continuity of layers within seasonal snowpacks can be highly variable, due to snow blowing around topography and vegetation, and this has important implications for hydrology, remote sensing and avalanche forecasting. In this paper, we present practical field and post-processing protocols for recording lateral variations in snow stratigraphy using near-infrared (NIR) photography. A Fuji S9100 digital camera, modified to be sensitive to NIR wavelengths, was mounted on a rail system that allowed for rapid imaging of a 10 m long snow trench excavated on the north side of Toolik Lake, Alaska (68°38′ N, 149°36′ W). Post-processing of the images included removal of lens distortion and vignetting. A tape measure running along the base of the trench provided known locations (control points) that permitted scaling and georeferencing. Snow layer heights estimated from the NIR images compared well with manual stratigraphic measurements made at 0.2 m intervals along the trench (n = 357, R2 = 0.97). Considerably greater stratigraphic detail was captured by the NIR images than in the manually recorded profiles. NIR imaging of snow trenches using the described protocols is an efficient tool for quantifying continuous microscale variations in snow layers and associated properties.

In this study, an attempt is made to simulate the historical length variations of Hintereisferner (Austria). Key elements in the model are: (1) reconstruction of the local climate (Vent); (2) multiple linear regression analysis between measured climate and mass-balance variations; (3) reconstruction of the mass balance with (1) and (2); (4) a one-dimensional numerical ice-flow model forced by the mass-balance history. Progress relative to the previous, similar studies can be summarized as two points: (1) the mass balance is reconstructed more carefully, and (2) sensitivity experiments show that uncertainties and assumptions in the flow model hardly affect the response of the glacier to climate fluctuations with a time-scale of the order of (or larger than) the response time of the glacier (around 100 year).

In the regression equation, mass-balance fluctuations are determined by the temperature during the ablation season and the annual precipitation. Cloudiness has no effect. The local temperature can be reconstructed as far back as 1755 with the temperature series from Basel, but the reconstruction of precipitation is unreliable prior to 1900. Mainly due to this point, a simulation starting in 1755 failed completely. However, in a run starting in 1894, the total retreat since then (2.2 km) was reproduced within 10%.

During the Japanese-Swedish Antarctic traverse expedition of 2007/08, we measured the surface snow density at 46 locations along the 2800 km long route from Syowa station to Wasa station in East Antarctica. The mean snow density for the upper 1 (or 0.5) m layer varied from 333 to 439 kg m-3 over a region spanning an elevation range of 365-3800 ma.s.l. The density variations were associated with the elevation of the sampling sites; the density decreased as the elevation increased, moving from the coastal region inland. However, the density was relatively insensitive to the change in elevation along the ridge on the Antarctic plateau between Dome F and Kohnen stations. Because surface wind is weak in this region, irrespective of elevation, the wind speed was suggested to play a key role in the near-surface densification. The results of multiple regression performed on the density using meteorological variables were significantly improved by the inclusion of wind speed as a predictor. The regression analysis yielded a linear dependence between the density and the wind speed, with a coefficient of 13.5 kg m-3 (m s-1)-1. This relationship is nearly three times stronger than a value previously computed from a dataset available in Antarctica. Our data indicate that the wind speed is more important to estimates of the surface snow density in Antarctica than has been previously assumed.

A flowline model, coupled with a surface mass-balance model forced by climate data from Bergen, was used to simulate future frontal changes of Briksdalsbreen, a western outlet glacier from Jostedalsbreen, western Norway, under various future climate scenarios. The model was used to calculate the time-lag of frontal response to a sudden and short change in the mass balance. According to the model, the front has a time-lag for maximum advance rate of 4–5 years, in close agreement with previous studies. The response time for Briksdalsbreen was calculated by running the model for 200 years with different mass-balance perturbations. For mass-balance perturbations of +0.3 and +0.6 m w.e. the model yields response times of 52 and 60 years, respectively. We ran the model from 1963 to 2007 with measured mass-balance data, and from 2007 to 2085 using calculated mass balances from 12 different climate scenarios. The model predicts retreat up the steep valley from the lake inlet, with a total frontal retreat of 2.5–5.0 km by 2085. A spectacular icefall, one of the main tourist attractions in western Norway, may thus disappear and the glacier may become a plateau glacier that will gradually melt down.

Satellite remote sensing has come to dominate the measurement of glacier and ice-sheet change. Three independent methods now exist for assessing ice-sheet mass balance and we focus on progress in two: satellite altimetry (ICESat) and gravimetry (GRACE). With improved spatial and temporal sampling, and synergy with ice flow measurements, both the mechanisms and causes changing mass balance can be investigated. We present examples of mass losses due to widespread, intensifying glacier dynamic thinning in northwest Greenland, but local ablation rates in the northeast that are unchanged for decades. Advances in GRACE processing reveal Greenland net ice-sheet mass loss continuing into 2010, at 195 ± 30 Gta-1. A similarly negative trend in the Gulf of Alaska has significant spatial and temporal variation, that highlights the importance of intense summer melting here. Strong summer melt on the Antarctic Peninsula also precipitated recent ice-shelf collapse and prompted rapid dynamic thinning of tributary glaciers at up to 70 m a-1. Thinning continued for years to decades after collapse and propagated far inland. While understanding of the physical mechanisms of change continues to improve, estimates of future behaviour, and in particular the near-future glacial sea-level contribution, still rely on projections from such observations. We introduce the suite of new sensors that will monitor the ice sheets into the future.