An Xbox Kinect sensor was used as a close-range surface scanner to produce the first accurate 3D surface models of spikes of snow and ice (known as penitentes) that develop in cold, dry, sunny conditions. The data collected show how penitentes develop over time and how they affect the surface roughness of a glacier. These surface models are useful inputs to modelling studies of how penitentes alter energy exchanges between the atmosphere and the surface and how this affects meltwater production.

The way that ice moves over rough ground at the base of an ice sheet is important to understand and predict the behaviour of ice sheets. Here, I argue that if basal ice is at the melting temperature, as is locally the case below the Greenland Ice Sheet, this basal motion is easier and faster than hitherto thought. A thick (tens of metres) layer of ice at the melting temperature may better explain some ice streams and needs to be taken into account when modelling future ice sheet behaviour.

We present recent (1958–2015) mass balance time series for the Greenland ice sheet. We show that recent mass loss is caused by a combination of increased surface meltwater runoff and solid ice discharge. Most meltwater above 2000 m a.s.l. refreezes in the cold firn and does not leave the ice sheet, but this goes at the expense of firn heating and densifying. In spite of a temporary rebound in 2013, it appears that the ice sheet remains in a state of persistent mass loss.

In hydrological models it is important to properly simulate the spatial distribution of snow water equivalent (SWE) for the timing of spring melt floods and the accounting of energy fluxes. This paper describes a method for the spatial distribution of SWE which is parameterised from observed spatial variability of precipitation and has hence no calibration parameters. Results show improved simulation of SWE and the evolution of snow-free areas when compared with the standard method.

We investigate Greenland Ice Sheet mass change from 2003–2012 by comparing observations from GRACE with state-of-the-art atmospheric and ice sheet model simulations. We find that the largest discrepancies (in the northwest and southeast) are likely controlled by errors in modeled surface climate as well as ice–ocean interaction and hydrological processes (not included in the models). Models should consider such processes at monthly to seasonal resolutions in order to improve future projections.

Along a traverse through North Greenland in May 2015 we collected snow cores up to 2 m in depth and analyzed their properties (e.g., density). A new technique for this sampling and an adapted algorithm for comparing data sets from different positions and aligning stratigraphic features are presented. We find good agreement of the density layering in the snowpack over hundreds of kilometers. This allows the construction of a representative density profile that is statistically validated.

We use CryoSat-2 satellite data to provide the first near-real-time (NRT) measurements of absolute sea ice thickness across the entire Northern Hemisphere. We analyse our NRT sea-ice-thickness data for one sea ice growth season, from October 2014 to April 2015. Over that time period a NRT thickness measurement was delivered, on average, within 14, 7 and 6 km of each location in the Arctic every 2, 14 and 28 days respectively.

We investigate capillary barriers and preferential flow in layered snow during nine cold laboratory experiments. The dynamics of each sample were replicated solving Richards equation within the 1-D multi-layer physically based SNOWPACK model. Results show that both processes affect the speed of water infiltration in stratified snow and are marked by a high degree of spatial variability at cm scale and complex 3-D patterns.

Polar bears, found in 19 distinct regions of the Arctic, depend on sea ice as a platform for hunting. Using satellite data of sea ice for the period 1979–2014, we found that the date of sea-ice retreat in spring is arriving weeks earlier, and the date of sea-ice advance in fall is arriving weeks later, in all 19 regions. We calculated several other measures of sea ice, which all show declines. These measures were designed to be useful for management agencies to assess polar bear habitat.

MIT was reported to have calved subsequent to being rammed by a large iceberg. However from remote sensing, the ice fronts being rammed did not move out first which led us to detect the influence of seafloor on instability of MIT. Using Firn Air Content extracted from slightly grounded icebergs, laser altimetry, remote sensing, and seafloor topography data, grounding of the MIT caused by Mertz Bank is extracted. Mertz Bank is confirmed to control calving of the MIT at a cycle of ~70 years.

We study chemical reactions in snow and ice by freezing solutions in the laboratory. Although it is important to know where these chemicals are in the frozen sample (at the surface or buried in the ice), we do not understand this well. In this paper, we used X-rays to look at the chemical location in frozen samples. We found chemical location is sensitive to freezing method, sample container, and chemical characteristics, requiring careful experimental design and interpretation of results.

Ice layers in snowpacks introduce uncertainty in satellite-derived estimates of snow water equivalent, have ecological impacts on plants and animals, and change the thermal and vapour transport properties of the snowpack. Here we present a new field method for measuring the density of ice layers. The method was used in the Arctic and mid-latitudes; the mean measured ice layer density was significantly higher than values typically used in the literature.

This study presents a multi-temporal dataset of geodetically derived elevation changes on debris-free and debris-covered glaciers in the Langtang valley, Nepalese Himalaya. Overall, we observe accelerated glacier wastage, but highly heterogeneous spatial patterns and temporal trends across glaciers. Accelerations in thinning correlate with the presence of supraglacial cliffs and lakes, whereas thinning rates remained constant or declined on stagnating debris-covered glacier areas.

The upper 50–100 m of the world's ice sheets consists of the firn layer, a porous layer of snow that is slowly compacted by overlying snow. Understanding air movement inside the firn is critical for ice core climate reconstructions. Buizert and Severinghaus identify and describe a new mechanism of firn air movement. High- and low-pressure systems force air movement in the firn that drives strong mixing, called dispersion. Dispersion is the main mechanism for air mixing in the deep firn.

This article presents a set of spectro-goniometric measurements of different water ice samples and their comparison with an approximated radiative transfer model using a Bayesian approach. Two kinds of experiments were conducted: the specular spot was investigated, and then the diffuse radiation. We show that the approximated model is able to reproduce the spectro-radiogoniometric data satisfactorily, and that the inverted parameters are compatible with independent measurements.

Satellite data are often the only source of information on mountain glaciers. We show that data from ICESat laser satellite can accurately reflect glacier volume development in 2003–2008, also for individual years. We detect a spatially varying elevation bias in commonly used data sets, and provide a correction that strongly increases the significance of the glacier change estimates – a crucial driver of climate-induced meltwater changes that directly affect the life of lowland populations.

Our paper presents the Conceptual Cryosphere Hydrology Framework (CCHF), a tool to enable more rapid development and intercomparison of cryosphere process representations. Using the CCHF, we demonstrate that some common existing degree index cryosphere models are not well suited for assessing impacts across climates, even though these models appear to perform well under a common evaluation strategy. We show that more robust models can be formulated without increasing data input requirements.

The fate of ikaite precipitation within sea ice is poorly understood. In this study, we estimated ikaite precipitation of up to 167 µmol kg^-1 within sea ice, while its export and dissolution into the underlying seawater was responsible for a TA increase of 64–66 μmol kg^-1. We estimated that more than half of the total ikaite precipitated was still contained in the ice when sea ice began to melt. The dissolution of the ikaite crystals in the water column kept the seawater pCO₂ undersaturated.

Analysis reveals that “Arctic amplification”, lower tropospheric winter temperature rise in the Arctic, is due to sea ice melting and the resulting increase in the amount of turbulent heat flux from the ocean. As a result of increased turbulent heat flux, lower atmosphere warms up, resulting in increased downward longwave radiation. A detailed physical mechanism is presented together with an explanation why this positive feedback process is currently possible in the Barents–Kara seas.

Here we utilize declassified spy satellite imagery to quantify ice volume loss of glaciers in the eastern Himalayas over approximately the last three decades. Clean-ice and debris-covered glaciers show similar magnitudes of ice loss, while calving glaciers are contributing a disproportionately large amount to total ice loss. Results highlight important physical processes affecting the ice mass budget and associated water resources in the Himalayas.

Sea ice, frozen seawater floating on polar oceans, is covered by meltwater puddles, so-called melt ponds, during summer. Methods used to compute Arctic sea-ice concentration (SIC) from microwave satellite data are influenced by melt ponds. We apply eight such methods to one microwave dataset and compare SIC with visible data. We conclude all methods fail to distinguish melt ponds from leads between ice floes; SIC biases are negative (positive) for ponded (non-ponded) sea ice and can exceed 20 %.

Microtopography of the low-gradient polygonal tundra plays a critical role in these ecosystem; however, patterns and drivers are poorly understood. A modeling-based approach was developed in this study to characterize and represent the permafrost soils in the model and simulate the thermal dynamics using a mechanistic high-resolution model. Results shows the ability of the model to simulate the patterns and variability of thermal regimes and improve our understanding of polygonal tundra.

The EUMETSAT sea ice climate record (ESICR) is based on the Nimbus 7 SMMR (1978–1987), the SSM/I (1987–2009), and the SSMIS (2003–today) microwave radiometer data. It uses a combination of two sea ice concentration algorithms with dynamical tie points, explicit atmospheric correction using numerical weather prediction data for error reduction and it comes with spatially and temporally varying uncertainty estimates describing the residual uncertainties.

Bryophyte and lichen cover on the forest floor at high latitudes insulates the ground and thus decreases soil temperature. This can protect permafrost soil, stabilising it against global warming. To quantify the insulating effect, we integrate a novel, process-based model of bryophyte and lichen growth into the global land surface model JSBACH. We find an average cooling effect of the bryophyte and lichen cover of 2.7 K, which implies a significant impact on soil temperature at high latitudes.

We explore the effects of melting of the East Antarctic Ice Sheet on the Southern Ocean. Using a climate model, we find that melting changes the ocean circulation and causes warming of more than 1 °C at depth. We also discover the potential existence of a "domino effect", whereby the initial warming spreads westwards around the Antarctic continent. Melting of just one sector could therefore destabilise the wider Antarctic Ice Sheet, leading to substantial increases in global sea level.

We analyse two methods of freeboard retrieval from ICESat satellite data that were used to derive the two widely used Arctic sea ice thickness products. We show that although different factors result in significant local differences between freeboards, they roughly compensate each other with respect to overall freeboard estimation. Thus the difference found between the sea ice thickness datasets should be attributed to different parameters used in the freeboard-to-thickness conversion.

This study demonstrates a robust procedure for accumulating precipitation gauge measurements and provides an analysis of bias corrections of precipitation measurements across experimental sites in different ecoclimatic regions of western Canada. It highlights the need for and importance of precipitation bias corrections at both research sites and operational networks for water balance assessment and the validation of global/regional climate–hydrology models.

We present a 1 km resolution data set (1958–2015) of daily Greenland ice sheet surface mass balance (SMB), statistically downscaled from the data of RACMO2.3 at 11 km using elevation dependence, precipitation and bare ice albedo corrections. The data set resolves Greenland narrow ablation zones and local outlet glaciers, and shows more realistic SMB patterns, owing to enhanced runoff at the ice sheet margins. An evaluation of the product against SMB measurements shows improved agreement.

Despite the crucial role of polar regions in the global climate system, the limited availability of observations on the ground hampers a detailed understanding of their energy budget. Here we develop a method to use satellites to fill these observational gaps. We show that by sampling satellite observations in a smart way, coverage is greatly enhanced. We conclude that this method might help improve our understanding of the polar energy budget, and ultimately its effects on the global climate.

This is the first study of recent glacier change for the whole of the Bolivian Cordillera Oriental from 1986 to 2014. These glaciers have shrunk by ~ 43 %, which is a concern for regional water resources. We provide the first quantification of meltwater lake development across the Bolivian Andes as glaciers have receded. These lakes have increased markedly in number and area. We identify 25 lakes as potential outburst flood risks to downstream communities. These lakes require further monitoring.

Here, we focus on the Concordia Station, central East Antarctic plateau, providing a multi-year record (2008–2010) of daily precipitation types identified from crystal morphologies, precipitation amounts and isotopic composition. Relationships between local meteorological data and precipitation oxygen isotope composition are investigated. Our dataset is available for in-depth model evaluation at the synoptic scale.

The anomaly correlation analysis between the decadal hindcast and observed sea ice suggests that in the Arctic, the areas showing significant predictive skill become broader associated with increasing lead times. This area expansion is largely because nearly all the models are capable of predicting the observed decreasing Arctic sea ice cover. Antarctic sea ice decadal hindcasts do not show broad predictive skill at any timescales.

Integrated snow products were used to analyze the temporal and spatial variation of the snow cover in China during 2000–2014. The results indicated that the overall annual number of snow-covered days and average snow depth in China have increased in the past 14 years, but differences in the snow cover variation in China were significant. Overall, the snow cover increased significantly in south and northeast China, but decreased significantly in Xinjiang and the Tibetan Plateau.

Fourteen years of satellite observations are used to monitor the albedo of Brewster Glacier, New Zealand and estimate annual and seasonal balances. This confirms the governing role of the summer balance in the annual balance and allows the reconstruction of the annual balance to 1977 using a photographic record of the snowline. The longest mass balance record for a New Zealand glacier shows negative balances after 2008, yielding a loss of 35 % of the gain accumulated over the previous 30 years.

The permafrost in eastern Siberia contains large amounts of carbon frozen in soils and sediments. Continuing global warming is thawing the permafrost and releasing carbon to the Arctic Ocean. We used pyrolysis-GCMS, a chemical fingerprinting technique, to study the types of carbon being deposited on the continental shelf. We found large amounts of permafrost-sourced carbon being deposited up to 200 km offshore.

The reconstruction of past snow accumulation rates is crucial in the context of recent climate change and sea level rise. We measured ~ 250 years of snow accumulation using a 120 m ice core drilled in coastal East Antarctica, where such long records are very scarce. This study is the first to show an increase in snow accumulation, beginning in the 20th and particularly marked in the last 50 years, thereby confirming model predictions of increased snowfall associated with climate change.

Permafrost temperatures in Alaska are increasing, yet in many areas we lack data needed to assess future changes and potential risks. In this paper we show that classifying the landscape into landcover types is an effective way to scale up permafrost temperature data collected from field monitoring sites. Based on these results, a map of mean annual ground temperature ranges at 1 m depth was produced. The map should be useful for land use decision making and identifying potential risk areas.