We analyze sea-ice conditions along seven segments of the Northern Sea Route (NSR) over four time periods. We researched sea ice by segment, using data from the satellite microwave sensors SMMR, SSM/I and AMSR-E. The four analysis periods (periods I–IV: 1980–88, 1989–2001, 2002–06 and 2007–11, respectively) were determined based on changes in the extent of minimum sea ice throughout the Arctic Ocean. Sea ice showed a decreasing tendency from period I to period IV. For example, sea-ice area in period IV decreased compared to previous periods in the eastern East Siberian Sea and around Severnaya Zemlya, areas that had very high sea-ice concentrations in period I. Sea-ice area in the eastern East Siberian Sea decreased sharply during period III, whereas the Severnaya Zemlya area maintained a high ice concentration. In period IV, sea-ice coverage around Severnaya Zemlya was low, although it remained at 25% in the area east of Severnaya Zemlya, which is a key area for navigation. The proportion of multi-year (MY) ice drastically decreased after winter 2002, and only a small amount of MY ice existed in the winters of 2003–06. MY ice disappeared from the eastern East Siberian Sea after 2007. On the other hand, around Severnaya Zemlya the proportion of MY ice showed cyclic ups and downs between 1997 and 2008. Thus, the persistence of various types of sea ice varies according to region. The persistence of various types of sea ice around Severnaya Zemlya also varied each year.

This study investigated the stable water isotopic (δ18O and δD) evolution of two snowpacks located at the same elevation on a north-facing (Nf) and a south-facing (Sf) slope within the Tarfala research catchment, northern Sweden, and the potential impact of pre-freshet snowpack melt on streamflow. Our results indicate that over the study period the Sf snowpack showed a more enriched isotopic composition, especially in the top of the profile, and contributed more snowmelt to streamflow than the Nf slope. The Sf snowpack also showed a significantly higher variability in snowpack δ18O levels and snowpack snow water equivalent (SWE) over time. Comparing snowpack and snowmelt isotopic values it was seen that the Sf slope experienced earlier snowmelt from upslope positions due to greater insolation that subsequently enhanced the meltwater flux at the base of downslope snowpacks. In contrast, the Nf slope primarily underwent changes within the snowpack and experienced relatively minimal melt. Detailed field-based isotopic snowmelt studies such as this highlight the potential importance of incorporating spatio-temporal runoff generation concepts into distributed energy-balance models, which could allow for more accurate prediction with regard to the spatio-temporal dynamics associated with the snowmelt ion pulse.

To evaluate water and anion movement in an agricultural field in different frost conditions, a paired-plot field experiment was conducted at an agricultural site in northern Japan where a significant decrease in the frost depth has occurred during the past 20 years. Snow cover was removed to enhance soil freezing in one plot (treated plot), whereas natural conditions were maintained in a control plot. The maximum frost depth was 0.43 m in the treated plot and 0.11 m in the control plot, which induced substantial differences in water movement throughout the winter. A substantial amount of water moved upward before the onset of snowmelt. However, nitrate did not move markedly before the snowmelt period in either plot. The amount of snowmelt infiltration in the control plot was larger than in the treated plot. Correspondingly, the peak of nitrate content in the control plot was deeper than that in the treated plot after the snowmelt period. Soil freezing, snow accumulation and snowmelt processes were simulated reasonably well using a one-dimensional numerical model: Simultaneous Heat and Water (SHAW). Nevertheless, the model performed poorly for simulating soil thawing and soil water movement, suggesting a need for improvement.

Knowledge of snow properties across Antarctica is important in estimating how climate could potentially influence the mass balance of the Antarctic ice sheet. However, measuring these variables has proven to be challenging because appropriate techniques have not yet been developed and extensive datasets of field estimates are lacking. The goal of this study was to estimate the relationship between field-observed snow particle-size parameters from across the East Antarctic ice sheet and a suite of spatial datasets (i.e. topography, remote-sensing data) using a principal component analysis (PCA). Five snow particle-size parameters were correlated to spatial datasets of the following five groups: (1) relief properties such as elevation and slope; (2) remote-sensing data from Moderate Resolution Imaging Spectroradiometer (MODIS) and synthetic aperture radar (SAR) sensors; (3) spatially interpolated data (i.e. 10 m maps of temperature and approximate snow accumulation in kgm−2 ä−1); (4) field-retrieved data on surface roughness; and (5) in situ elevation and distance from the coast. The results show that the relief parameter slope correlated best with the snow particle length and area (r=0.76, r=0.80). Further, the PCA indicated that the different remote-sensing parameters correlated differently with the size parameters and that the most common parameter in visual analysis, particle length (grain diameter), is not always the optimal parameter to characterize the snow particle size as, for example, area correlates better to slope and aspect than length.

The major factor shaping the coast of the micro-tidal Baltic Sea is wave activity, the impact of which is limited by the presence or absence of sea ice. Existing studies have revealed almost no correlation between the annual mean wave height and the duration of the ice season. We attempt to evaluate the correlation between ice season duration and bulk wave energy flux (wave power), mean energy and average wave height over the ice-free season for three segments of the Estonian coast (Vilsandi, Pakri and Narva-Joesuu). Statistically significant correlation at the 95% confidence level exists between the duration of the ice season at Vilsandi (a site fully open to the predominant winds) and bulk wave energy flux derived from both observed and modelled wave properties. Similarly strong correlation exists between the mean wave energy and the duration of the ice season at the largely sheltered Narva-Joesuu site.

Regions of Japan facing the Sea of Japan have some of the world’s deepest seasonal snowpack, although air temperatures averaged over the coldest winter months (January-February) are above 0°C in this mid-altitude zone (35–40˚ N). Therefore, even a slight temperature increase is likely to drastically decrease the snow depth in this area by changing the precipitation form from snow to rain. In this study, to enhance understanding of winter precipitation conditions, we introduce the winter precipitation curve (WPC), which plots the distribution of winter precipitation (December-February) against the corresponding air temperature using hourly data. We then examine the relationship between the WPC and regional climate conditions. The WPC was classified into two types in the study: the single-peak (SP) type, having one sharp peak near 0°C, and the multiple-peaks (MP) type, having several small peaks. Here ‘peak’ indicates that precipitation frequently occurs at a certain temperature. Peaks of the MP type changed position with fluctuations in mean winter temperature, but the peak of the SP type moved only slightly despite annual fluctuations in winter mean temperature. One of the multiple peaks of the MP type appeared at ∼0°C and this peak also remained nearly stationary with fluctuations in winter mean temperature. The peak of the WPC appearing near 0°C in both the SP and MP types likely resulted from the development of the 0°C isothermal layer due to the cooling effect of melting snow particles.

A study on effective thermal conductivity (ETC) of natural ice in a reservoir with <3% gas content and ∼300 mg L−1 dissolved matter content in the parent water was conducted in a laboratory. Ice sections were prepared to obtain the ice crystal structure, gas bubble content and ice density. Profiles of horizontal and vertical thermal conductivity of ice samples were determined with attention to relatively high temperature spanning 0 to –10°C. A detailed comparative analysis was conducted of the effects on ETC of direction, ice crystal structure, gas bubbles and temperature. Gas inclusions had little impact on the thermal conductivity of natural reservoir ice due to their quite low content (<3%). At high ice temperature the ETC decreases remarkably as ice temperature approaches the melting point, as the ETC of sea ice does due to its brine volume at ice temperature approaching its melting point. The measured conductivities for temperatures higher than –4°C were less than expected from previous work on saline ice. Whether this was due to the measurement techniques or actual properties of the reservoir ice is not clear. The present findings could lead to better understanding of the thermal processes of ice in natural freshwater bodies.

Spring snow cover across Arctic lands has, on average, retreated ∼5 days earlier since the late 1980s compared to the previous 20 years. However, it appears that since about the late 1980s the date the snowline first retreats north during the spring has changed only slightly: in the last ∼20 years snow has not been disappearing significantly earlier. Snowmelt changes observed since the late 1980s have been step-like, unlike the more continuous downward trend seen in Arctic sea-ice extent. At 70° N, several longitudinal segments (of 10°) show significant (negative) trends, while only two longitudinal segments at 60° N show significant trends, one positive and one negative. These variations appear to be related to variations in the Arctic Oscillation (AO). When the springtime AO is strongly positive, snow melts earlier. When it is strongly negative, snow disappears later in the spring. The winter AO is less straightforward. At higher latitudes (70° N), a positive AO during the winter months is correlated with later snowmelt, but at lower latitudes (50° N and 60° N) a positive wintertime AO is correlated with earlier snowmelt. If the AO during the winter months is negative, the reverse is true. Similar stepwise changes (since the late 1980s) have been noted in sea surface temperatures and in phytoplankton abundance as well as in snow cover.

Spatial degree-day factors (DDFs) are required for spatial snowmelt modeling over large areas by the degree-day method. We propose a method to obtain DDFs by incorporating snow disappearance dates (SDDs), derived from 10 day composites of Satellite Pour l’Observation de la Terre (SPOT)/VEGETATION data, into the degree-day method. This approach allowed determination of DDFs for each gridpoint so as to better reflect regional characteristics than use of spatially constant DDFs obtained from point measurements. Simulations at six observation sites successfully accounted for variations in snow water equivalent (SWE), even at elevations different from the closest measurement site. These results suggest that incorporating satellite-derived SDDs into the degree-day method decreases spatial uncertainty compared with the use of spatially constant DDFs. Application of our method to Japanese cold regions revealed that gridded DDFs were negatively correlated with accumulated positive degree-days (APDDs) and were high only when APDDs were low. These results imply that high DDFs resulted from the dominant contribution of solar radiation to snowmelt at low temperatures and that low DDFs resulted from a relatively high contribution of sensible heat flux at high temperatures. The proposed method seems to adequately account for the main energetic components of snowmelt during the snow-cover season over large areas.

High-density snow layers deteriorate grazing conditions for reindeer during winter. We compare two different methods for identifying past winters with difficult grazing conditions for reindeer in Kautokeino, northern Norway. A long-term climate analysis based on monthly values of precipitation and temperature (1900–2011) demonstrated that the reported winters were difficult to identify systematically. The processes leading to hard layers or ground-ice layers occur on daily, not monthly, timescales, and whether or not specific conditions are problematic depends on the development throughout the winter, not just on single values. To better analyse the weather conditions and development over time that favour the formation of high-density snow layers, we apply the multilayer model SNOWPACK over the period 1956–2010. We simulate the evolution of the snowpack by forcing the model with 6 hour interval meteorological data. The model output was analysed by summing up the vertical extent of the simulated high-density snow layers (>350 kg m–3) for each winter. These results were compared with historical records of difficult winter grazing conditions reported in the period 1956–2010. In particular, the heavy losses of reindeer during the catastrophic 1967/68 winter were caused by the occurrence of ground ice together with long snow-cover duration. This unfavourable coincidence is well reproduced by our model results, together with eight of the ten reported difficult winters.

Sea-ice thicknesses observed in Canadian coastal waters with helicopter-borne electromagnetic–laser sensors show large interannual variability caused by atmospheric fluctuations in two years for two areas where surveys were repeated, one in the Amundsen Gulf of the Canadian Beaufort Sea and one over the Labrador Shelf. For the Amundsen Gulf, the bimodal ice thickness peaks shifted by 40 cm to thinner thicknesses for the warmer winter of 2008 compared with 2004. The thinner ice in 2008 can be explained partially by reduced thermodynamic ice growth during the warmer winter of 2008. In addition, winds from the east were more persistent throughout the winter of 2008, increasing ice export from the Amundsen Gulf and thereby creating open-water areas where new ice growth in late winter produced the thinner ice classes. For the Labrador Shelf, the mean ice thicknesses of the warmer winter of 2011 (0.71 m) were much less than those of the near-normal winter of 2009 (1.60 m). Again the difference can be explained by the fact that along the entire Labrador Shelf the winter of 2011 was much warmer, reducing ice growth and resulting in thinner ice locally and thinner ice being transported into the survey region from northern latitudes. In addition, northwesterly winds occurred less frequently during the winter of 2011, reducing the transport of relatively thicker ice into the survey area from northern latitudes.

We introduce a method to construct an ice thickness chart using two satellite datasets and a one-dimensional high-resolution thermodynamic snow and sea-ice model (HIGHTSI). Thin-ice thickness up to 40 cm is retrieved using MODIS ice surface temperature data. For thicker ice, thickness is retrieved from a combination of SAR image and background level ice thickness chart Hi provided by HIGHTSI. Because Hi is inherently static, we utilize passive microwave data based sea-ice concentration charts to introduce mesoscale ice dynamics into Hi. In order to create the thickness chart, the value of modified Hi is scaled by a factor which depends on the magnitude of backscattering coefficient σ0. The proposed method is effective in cold winter conditions. We calculated a series of ice thickness charts covering the Kara Sea, Arctic Russia, during winter 2008/09. The ice thickness charts were validated against Russian ice charts. Our method gave realistic results in the thickness range 30–120 cm and was uncertain for the detection of thinner-ice areas. Due to the limitations in our reference dataset, more validation work is needed to establish the accuracy in more detail. Our thickness charts can be used for operational purposes and in climate studies.

Rijpfjorden (808 N, 22° E) is a high-Arctic fjord on Nordaustlandet in the Svalbard archipelago. To monitor the thermodynamic change of sea ice in spring, an ice mass-balance buoy (IMB) was deployed for 2.5 months (10 April–26 June 2011), with accompanying in situ measurements, sea-ice sampling on three occasions and ice-core analysis. Uncertainties and sources of error in in situ measurements and IMB data are discussed. The in situ measurements, ice-core analysis and IMB data together depict the development of snow and ice in spring. Snow and ice thickness exhibited large spatial and temporal variability. After relatively stable conditions with only little change in ice thickness and accumulation of snow, a layer of superimposed ice ∼0.06 m thick formed at the snow-ice interface due to refreezing of snow meltwater in late spring. Ice thickness (except for growth of superimposed ice) did not change significantly based on in situ observations. In contrast, the under-ice sonar data from the IMB show reflections from a layer deeper than the underside of the ice during the melting phase. This can be explained as a reflection of the sonar pulses from an interface between a freshwater layer under the ice and more saline water below, or as a false-bottom formation.

Variability of sea-ice and snow conditions on the scale of a few hundred meters is examined using in situ measurements collected in first-year pack ice in the European Arctic north of Svalbard. Snow thickness and surface elevation measurements were performed in the standard manner using a snow stick and a rotating laser. Altogether, 4109 m of measurement lines were surveyed. The snow loading was large, and in many locations the ice freeboard was negative (38.8% of snowline measurements), although the modal ice and snow thickness was 1.8 m. The mean of all the snow thickness measurements was 36 cm, with a standard deviation of 26 cm. The mean freeboard was only 3 cm, with a standard deviation of 23 cm. There were noticeable differences in snow thickness among the measurement sites. Over the undeformed ice areas, the mean snow thickness and freeboard were 23 and 2.4 cm, respectively. Over the ridged ice areas, the mean freeboard was only –0.3 cm due to snow accumulation on the sails of ridges (average thickness 54 cm). These findings imply that retrieval algorithms for converting freeboard to ice thickness should take account of spatial variability of snow cover.

Based on recent studies on sea ice and newly developed technologies, this paper examines the requirements and environmental factors for exploitation of sea ice as a freshwater resource. Considering the ‘mining’ of sea ice, the suitability of locations in the Bohai Sea, China, for exploitation was analysed and evaluated using GIS technology. The ice must be >10 cm thick, the intertidal zone must be narrow and the ‘mining’ area must have a water depth of at least 2 m. The sea areas with feasible resources are located mainly in the nearshore area of northern Liaodong Bay and in a narrow and long area along the coast of Bohai Bay. Most coastal areas of northern Liaodong Bay have favourable conditions, while Bohai Bay is widely constrained by unsuitable conditions. According to the status of the resources and the coastal environment conditions, the eastern and western coasts of northern Liaodong Bay are the most suitable areas for exploitation at the current technological level.

This study presents results from an Airborne Laser Scanning (ALS) mapping survey of snow depth on the mountain plateau Hardangervidda, Norway, in 2008 and 2009 at the approximate time of maximum snow accumulation during the winter. The spatial extent of the survey area is >240 km2. Large variability is found for snow depth at a local scale (2 m2), and similar spatial patterns in accumulation are found between 2008 and 2009. The local snow-depth measurements were aggregated by averaging to produce new datasets at 10, 50, 100, 250 and 500 m2 and 1 km2 resolution. The measured values at 1 km2 were compared with simulated snow depth from the seNorge snow model (www.senorge.no), which is run on a 1 km2 grid resolution. Results show that the spatial variability decreases as the scale increases. At a scale of about 500 m2 to 1 km2 the variability of snow depth is somewhat larger than that modeled by seNorge. This analysis shows that (1) the regional-scale spatial pattern of snow distribution is well captured by the seNorge model and (2) relatively large differences in snow depth between the measured and modeled values are present.

To identify the regional characteristics of mountain snowpacks related to avalanches in the central mountains of Japan, snow-pit observations were carried out over a period of between two and five winters at six study sites in three major mountain ranges: the north, central and south alps. The relationship between snowpack characteristics and meteorological properties was then examined to suggest indicators that could determine the snowpack characteristics in the region using meteorological data. On the basis of the dominant type of persistent weakness found within snowpacks, the study sites were divided into three regions: (1) the mountains on the Sea of Japan side - the persistent weakness was caused mainly by the presence of melt-freeze crust (MFcr) layers; (2) the inland mountains - the persistent weakness was caused mainly by the presence of weak layers consisting of near-surface faceted crystals (FCsf); and (3) the mountains on the Pacific Ocean side - the persistent weakness was caused mainly by the formation of thick weak layers consisting of depth hoar (DH) near the ground. The results also suggest that certain meteorological elements are useful indicators of the tendency of a persistent weakness to form within the snowpack: (1) DH layers: TG (°C m−1) = |average air temperature |/average snow depth (notable weakness in DH layers is found in areas where the TG exceeds 10°C m−1, but in such areas rainfall prevents the development of DH layers); (2) FCsf layers: FCsf index (CC m) = average diurnal range of air temperature × snow depth (notable weakness in FCsf layers is found in areas where the FCsf index is ∼10°C m); (3) MFcr layers: MFcr index (day m) = number of days that exceed 0°C × snow depth (notable weakness in MFcr layers is found in areas where the MFcr index exceeds 40 day m).

Dynamic ice models use stress tensor to describe the forces arising from internal ice friction. The model stress values are typically one to two magnitudes smaller than values measured by stressmeters deployed on ice floes. The synthesis of the pack-ice stress state from the measurements has been complicated by the peaky character of stress records, and the means to connect them with spatial stress distribution of the floe system have been lacking. Here a reanalysis of Arctic Sea Ice Mechanics Initiative (SIMI) data is made in terms of extreme value statistics. The basic quantity is the maximum stress observed during a time period. The records exhibit self-affine scaling. The statistics are then determined by two parameters, the Hurst exponent H and a reference stress level. Similar analysis is possible for the kinematic data. This establishes the comparability of stress records with each other and with kinematic records. The results suggest that the exponent is related to the stress state of the regional floe system, while the stress level is determined by local floe characteristics. Based on this a characterization of spatial distribution of pack-ice stresses is given.

The major river systems of India, i.e. the Indus, Ganga and Brahmaputra river systems originating in the Himalayan region, are considered the lifeline of the Indian subcontinent. The main sources maintaining the flow of the Himalayan rivers are snow/glacial melt runoff, rainfall runoff and base flow. The Beas River originates from Beas Kund Glacier in the Himalayan region and flows down to join the Sutlej River, which is a tributary of the Indus River system. In the present study two approaches, namely hydrologic modelling and isotope analysis, have been applied to estimate the contribution of snow and glacier melt. Samples of streamflow, rainfall and snow for isotopic analysis were collected daily from April to September and weekly from October to March during 2010 and 2011. The isotope analysis of samples reveals that the snow/glacier melt contribution to the Beas River at Manali is 50% of the total flow during these 2 years. Snowmelt runoff modelling has been carried out using the SNOWMOD model, and the snow/glacier melt runoff contribution is calculated to be 52% of the total flow during the same period. These findings indicate that the results obtained from the two approaches are similar.

A good estimate of the spatial probability density function (PDF) of snow water equivalent (SWE) provides the mean of the snow reservoir, but also enables modelling of the changes in snow-covered area (SCA), which is crucial for the runoff dynamics in spring. The spatial PDF of accumulated SWE is here modelled as a sum of correlated gamma-distributed variables, called units. The spatial variance of accumulated SWE is evaluated by the covariance matrix of the units. For accumulation events, there are only positive elements in the covariance matrix, whereas for melting events there are both positive and negative elements. The negative elements dictate that the correlation between melt and SWE is negative. After accumulation and melting events, the changes in the spatial moments are weighted by changes in SCA. Results from the model are in good agreement with observed spatial moments of SWE and SCA and found to provide better estimates of the spatial variability than the current model for snow distribution used in the Norwegian version of the Swedish rainfall–runoff model HBV. The parameters in the distribution model are estimated from observed historical precipitation, so no calibration parameters are introduced.