Search

Accurate simulations of snow emission in surface-sensitive microwave channels are needed to separate snow from atmospheric information essential for numerical weather prediction. Measurements from a field campaign in Trail Valley Creek, Inuvik, Canada, during March 2018 were used to evaluate the Snow Microwave Radiative Transfer (SMRT) model at 89 GHz and, for the first time, frequencies between 118 and 243 GHz. In situ data from 29 snow pits, including snow specific surface area, were used to calculate exponential correlation lengths to represent the snow microstructure and to initialize snowpacks for simulation with SMRT. Measured variability in snowpack properties was used to estimate uncertainty in the simulations. SMRT was coupled with the Atmospheric Radiative Transfer Simulator to account for the directionally dependent emission and attenuation of radiation by the atmosphere. This is a major developmental step needed for top-of-atmosphere simulations of microwave brightness temperature at atmosphere-sensitive frequencies with SMRT. Nadir-simulated brightness temperatures at 89, 118, 157, 183 and 243 GHz were compared with airborne measurements and with ground-based measurements at 89 GHz. Inclusion of anisotropic atmospheric radiance in SMRT had the greatest impact on brightness temperature simulations at 183 GHz and the least impact at 89 GHz. Medians of simulations compared well with medians of observations, with a root mean squared difference of 14 K across five frequencies and two flights (n=10). However, snow pit measurements did not capture the observed variability fully as simulations and airborne observations formed statistically different distributions. Topographical differences in simulated brightness temperature between sloped, valley and plateau areas diminished with increasing frequency as the penetration depth within the snow decreased and less emission from the underlying ground contributed to the airborne observations. Observed brightness temperature differences between flights were attributed to the deposition of a thin layer of very-low-density snow. This illustrates the need to account for both temporal and spatial variabilities in surface snow microstructure at these frequencies. Sensitivity to snow properties and the ability to reflect changes in observed brightness temperature across the frequency range for different landscapes, as demonstrated by SMRT, are necessary conditions for inclusion of atmospheric measurements at surface-sensitive frequencies in numerical weather prediction.

The surface temperature of snow cover is a key variable, as it provides information about the current state of the snowpack, helps predict its future evolution, and enhances estimations of the snow water equivalent. Although satellites are often used to measure the surface temperature despite the difficulty of retrieving accurate surface temperatures from space, calibration–validation datasets over snow-covered areas are scarce. We present a dataset of extensive measurements of the surface radiative temperature of snow acquired with an uncooled thermal-infrared (TIR) camera. The set accuracy goal is 0.7 K, which is the radiometric accuracy of the TIR sensor of the future CNES/ISRO TRISHNA mission. TIR images have been acquired over two winter seasons, November 2021 to May 2022 and February to May 2023, at the Col du Lautaret, 2057 m a.s.l. in the French Alps. During the first season, the camera operated in the off-the-shelf configuration with rough thermal regulation (7–39 °C). An improved setup with a stabilized internal temperature was developed for the second campaign, and comprehensive laboratory experiments were carried out in order to characterize the physical properties of the components of the TIR camera and its calibration. Thorough processing, including radiometric processing, orthorectification, and a filter for poor-visibility conditions due to fog or snowfall, was performed. The result is two winter season time series of 130 019 maps of the surface radiative temperature of snow with meter-scale resolution over an area of 0.5 km2. The validation was performed against precision TIR radiometers. We found an absolute accuracy (mean absolute error, MAE) of 1.28 K during winter 2021–2022 and 0.67 K for spring 2023. The efforts to stabilize the internal temperature of the TIR camera therefore led to a notable improvement of the accuracy. Although some uncertainties persist, particularly the temperature overestimation during melt, this dataset represents a major advance in the capacity to monitor and map surface temperature in mountainous areas and to calibrate–validate satellite measurements over snow-covered areas of complex topography. The complete dataset is provided at https://doi.org/10.57932/8ed8f0b2-e6ae-4d64-97e5-1ae23e8b97b1 (Arioli et al., 2024a) and https://doi.org/10.57932/1e9ff61f-1f06-48ae-92d9-6e1f7df8ad8c (Arioli et al., 2024b).

While the processes controlling pore closure are broadly understood, the physical mechanisms driving the associated elemental fractionation remains ambiguous. Previous studies have shown that the pore closure process leads to a depletion in small-sized molecules (e.g. H2, O2, Ar, Ne, He) in ice core bubbles relative to larger-sized molecules like N2. This size-dependent fractionation, identified using ice core δ(O2/N2) records, exhibits a clear anti-correlation with local summer solstice insolation, making δ(O2/N2) a valuable ice core dating tool. Mechanisms controlling this relationship are attributed to the physical properties of deep firn. In this study, we compile δ(O2/N2) records from 15 polar ice cores and show a new additional link between δ(O2/N2) and local surface temperature and/or accumulation rate. Using the Crocus snowpack model, we perform sensitivity tests to identify the response of near-surface snow properties to changes in insolation intensity, accumulation rate, and air temperature. These tests support a mechanism linked to firn grain size, such that the larger the grain size for a given density, the stronger the pore closure fractionation and, hence, the lower the δ(O2/N2) values archived in the ice. Based on both snowpack model outputs and data compilation, our findings suggest that local accumulation rate and temperature should be considered when interpreting δ(O2/N2) as a local insolation proxy.

In late March 2011, landfast sea ice (hereafter, “fast ice”) formed in the northern Larsen B embayment and persisted continuously as multi-year fast ice until January 2022. In the 11 years of fast-ice presence, the northern Larsen B glaciers slowed significantly, thickened in their lower reaches, and developed extensive mélange areas, leading to the formation of ice tongues that extended up to 16 km from the 2011 ice fronts. In situ measurements of ice speed on adjacent ice shelf areas spanning 2011 to 2017 show that the fast ice provided significant resistive stress to ice flow. Fast-ice breakout began in late January 2022 and was closely followed by retreat and breakup of both the fast-ice mélange and the glacier ice tongues. We investigate the probable triggers for the loss of fast ice and document the initial upstream glacier responses. The fast-ice breakup is linked to the arrival of a strong ocean swell event (>1.5 m amplitude; wave period waves >5 s) originating from the northeast. Wave propagation to the ice front was facilitated by a 12-year low in sea ice concentration in the northwestern Weddell Sea, creating a near-ice-free corridor to the open ocean. Remote sensing data in the months following the fast-ice breakout reveals an initial ice flow speed increase (>2-fold), elevation loss (9 to 11 m), and rapid calving of floating and grounded ice for the three main embayment glaciers Crane (11 km), Hektoria (25 km), and Green (18 km).

Monitoring of lake ice is important to maintain transportation routes, but in recent decades the number of in situ observations have declined. Remote sensing has worked to fill this gap in observations, with active microwave sensors, particularly synthetic-aperture radar (SAR), being a crucial technology. However, the impact of wet conditions on radar and how interactions change under these conditions have been largely ignored. It is important to understand these interactions as warming conditions are likely to lead to an increase in the occurrence of slush layers. This study works to address this gap using the Snow Microwave Radiative Transfer (SMRT) model to conduct forward-modelling experiments of backscatter for Lake Oulujärvi in Finland. Experiments were conducted under dry conditions, under moderate wet conditions, and under saturated conditions. These experiments reflected field observations during the 2020–2021 ice season. Results of the dry-snow experiments support the dominance of surface scattering from the ice–water interface. However, conditions where layers of wet snow are introduced show that the primary scattering interface changes depending on the location of the wet layer. The addition of a saturated layer at the ice surface results in the highest backscatter values due to the larger dielectric contrast created between the overlying dry snow and the slush layer. Improving the representation of these conditions in SMRT can also aid in more accurate retrievals of lake ice properties such as roughness, which is key for inversion modelling of other properties such as ice thickness.

This study explores the seasonal variations in snow grain size on the East Antarctic Plateau, where dry metamorphism occurs, by using microwave radiometer observations from 2000 to 2022. Local meteorological conditions and large-scale atmospheric phenomena have been considered in order to explain some peculiar changes in the snow grains. We find that the highest ice divide is the region with the largest grain size in the summer, mainly because the wind speed is low. Moreover, some extreme grain size values with respect to the average (over +3σ) were identified. In these cases, the ERA5 reanalysis revealed a high-pressure blocking close to the onsets of the summer increase in the grain size. It channels moisture intrusions from the mid-latitudes, through atmospheric rivers that cause major snowfall events over the plateau. If conditions of weak wind and low temperature occur during the following weeks, dry snow metamorphism is facilitated, leading to grain growth. This determines anomalous high maximums of the snow grain size at the end of summer. These phenomena confirm the importance of moisture intrusion events in East Antarctica and their impact on the physical properties of the ice sheet surface, with a co-occurrence of atmospheric rivers and seasonal changes in the grain size with a significance of over 95 %.

Forests strongly modify the accumulation, metamorphism and melting of snow in midlatitude and high-latitude regions. Recently, snow routines in hydrological and land surface models were improved to incorporate more accurate representations of forest snow processes, but model intercomparison projects have identified deficiencies, partly due to incomplete knowledge of the processes controlling snow cover in forests. The Snow Under Forest (SnoUF) project was initiated to enhance knowledge of the complex interactions between snow and vegetation. Two field campaigns, during the winters 2016–2017 and 2017–2018, were conducted in a coniferous forest bordering the snow study at Col de Porte (1325 m a.s.l., French Alps) to document the snow accumulation and ablation processes. This paper presents the field site, the instrumentation and the collection and postprocessing methods. The observations include distributed forest characteristics (tree inventory, lidar measurements of forest structure, subcanopy hemispherical photographs), meteorology (automatic weather station and an array of radiometers), snow cover and depth (snow pole transect and laser scan) and snow interception by the canopy during precipitation events. The weather station installed under dense canopy during the first campaign has been maintained since then and has provided continuous measurements throughout the year since 2018. Data are publicly available from the repository of the Observatoire des Sciences de l'Univers de Grenoble (OSUG) data center at https://doi.org/10.17178/SNOUF.2022 (Sicart et al., 2022).

Both regional climate models (RCMs) and remote sensing (RS) data are essential tools in understanding the response of polar regions to climate change. RCMs can simulate how certain climate variables, such as surface melt, runoff and snowfall, are likely to change in response to different climate scenarios but are subject to biases and errors. RS data can assist in reducing and quantifying model uncertainties by providing indirect observations of the modeled variables on the present climate. In this work, we improve on an existing scheme to assimilate RS wet snow occurrence data with the “Modèle Atmosphérique Régional” (MAR) RCM and investigate the sensitivity of the RCM to the parameters of the scheme. The assimilation is performed by nudging the MAR snowpack temperature to match the presence of liquid water observed by satellites. The sensitivity of the assimilation method is tested by modifying parameters such as the depth to which the MAR snowpack is warmed or cooled, the quantity of water required to qualify a MAR pixel as “wet” (0.1 % or 0.2 % of the snowpack mass being water), and assimilating different RS datasets. Data assimilation is carried out on the Antarctic Peninsula for the 2019–2021 period. The results show an increase in meltwater production (+66.7 % on average, or +95 Gt), along with a small decrease in surface mass balance (SMB) (−4.5 % on average, or −20 Gt) for the 2019–2020 melt season after assimilation. The model is sensitive to the tested parameters, albeit with varying orders of magnitude. The prescribed warming depth has a larger impact on the resulting surface melt production than the liquid water content (LWC) threshold due to strong refreezing occurring within the top layers of the snowpack. The values tested for the LWC threshold are lower than the LWC for typical melt days (approximately 1.2 %) and impact results mainly at the beginning and end of the melting period. The assimilation method will allow for the estimation of uncertainty in MAR meltwater production and will enable the identification of potential issues in modeling near-surface snowpack processes, paving the way for more accurate simulations of snow processes in model projections.

Surface melting is one of the primary drivers of ice shelf collapse in Antarctica and is expected to increase in the future as the global climate continues to warm because there is a statistically significant positive relationship between air temperature and melting. Enhanced surface melt will impact the mass balance of the Antarctic Ice Sheet (AIS) and, through dynamic feedbacks, induce changes in global mean sea level (GMSL). However, the current understanding of surface melt in Antarctica remains limited in terms of the uncertainties in quantifying surface melt and understanding the driving processes of surface melt in past, present and future contexts. Here, we construct a novel grid-cell-level spatially distributed positive degree-day (PDD) model, forced with 2 m air temperature reanalysis data and spatially parameterized by minimizing the error with respect to satellite estimates and surface energy balance (SEB) model outputs on each computing cell over the period 1979 to 2022. We evaluate the PDD model by performing a goodness-of-fit test and cross-validation. We assess the accuracy of our parameterization method, based on the performance of the PDD model when considering all computing cells as a whole, independently of the time window chosen for parameterization. We conduct a sensitivity experiment by adding ±10 % to the training data (satellite estimates and SEB model outputs) used for PDD parameterization and a sensitivity experiment by adding constant temperature perturbations (+1, +2, +3, +4 and +5 ∘C) to the 2 m air temperature field to force the PDD model. We find that the PDD melt extent and amounts change analogously to the variations in the training data with steady statistically significant correlations and that the PDD melt amounts increase nonlinearly with the temperature perturbations, demonstrating the consistency of our parameterization and the applicability of the PDD model to warmer climate scenarios. Within the limitations discussed, we suggest that an appropriately parameterized PDD model can be a valuable tool for exploring Antarctic surface melt beyond the satellite era.

The assimilation of data from Earth observation satellites into numerical models is considered to be the path forward to estimate snow cover distribution in mountain catchments, providing accurate information on the mountainous snow water equivalent (SWE). The land surface temperature (LST) can be observed from space, but its potential to improve SWE simulations remains underexplored. This is likely due to the insufficient temporal or spatial resolution offered by the current thermal infrared (TIR) missions. However, three planned missions will provide global-scale TIR data at much higher spatiotemporal resolution in the coming years. To investigate the value of TIR data to improve SWE estimation, we developed a synthetic data assimilation (DA) experiment at five snow-dominated sites covering a latitudinal gradient in the Northern Hemisphere. We generated synthetic true LST and SWE series by forcing an energy balance snowpack model with the ERA5-Land reanalysis. We used this synthetic true LST to recover the synthetic true SWE from a degraded version of ERA5-Land. We defined different observation scenarios to emulate the revisiting times of Landsat 8 (16 d) and the Thermal infraRed Imaging Satellite for High-resolution Natural resource Assessment (TRISHNA) (3 d) while accounting for cloud cover. We replicated the experiments 100 times at each experimental site to assess the robustness of the assimilation process with respect to cloud cover under both revisiting scenarios. We performed the assimilation using two different approaches: a sequential scheme (particle filter) and a smoother (particle batch smoother). The results show that LST DA using the smoother reduced the normalized root mean square error (nRMSE) of the SWE simulations from 61 % (open loop) to 17 % and 13 % for 16 d revisit and 3 d revisit respectively in the absence of clouds. We found similar but higher nRMSE values by removing observations due to cloud cover but with a substantial increase in the standard deviation of the nRMSE of the replicates, highlighting the importance of revisiting times in the stability of the assimilation performance. The smoother largely outperformed the particle filter algorithm, suggesting that the capability of a smoother to propagate the information along the season is key to exploit LST information for snow modelling. Finally, we have compared the benefit of assimilating LST with synthetic observations of fractional snow cover area (FSCA). LST DA performed better than FSCA DA in all the study sites, suggesting that the information provided by LST is not limited to the duration of the snow season. These results suggest that the LST data assimilation has an underappreciated potential to improve snowpack simulations and highlight the value of upcoming TIR missions to advance snow hydrology.

The grain size of the superficial snow layer is a key determinant of the surface albedo in Antarctica. Its evolution is the result of multiple interacting processes, such as dry and wet metamorphism, melt, snow drift, and precipitation. Among them, snow drift has the least known and least predictable impact. The goal of this study is to relate the variations in surface snow grain size to these processes in a windy location of the Antarctic coast. For this, we retrieved the daily grain size from 5-year-long in situ observations of the spectral albedo recorded by a new multi-band albedometer, unique in terms of autonomy and described here for the first time. An uncertainty assessment and a comparison with satellite-retrieved grain size were carried out to verify the reliability of the instrument, and an RMSE up to 0.16 mm in the observed grain size was found. By relating these in situ measurements to time series of snow drift, surface temperature, snow surface height and snowfall, we established that the evolution of the grain size in the presence of snow drift is complex and follows two possible pathways: (1) a decrease in the grain size (about half of our measurements) resulting from the deposition of small grains advected by the wind (surprisingly, this decrease is often – 2/3 of the cases– associated with a decrease in the surface height, i.e., a net erosion over the drift episode), (2) an increase in the grain size (the other half) due to either the removal of the surface layer or metamorphism. However, we note that this increase is often limited with respect to the increase predicted by a theoretical metamorphism model, suggesting that a concomitant deposition of small grains is likely. At last, we found that wind also completely impedes the deposition of snowfall during half of the observed precipitation events. When this happens, the grain size evolves as if precipitation were not occurring. As a result of all these processes, we conclude that the grain size in a windy area remains more stable than it would be in the absence of snow drift, hence limiting the variations in the albedo and in the radiative energy budget.

Surface melting on the Antarctic Ice Sheet has been monitored by satellite microwave radiometry for over 40 years. Despite this long perspective, our understanding of the microwave emission from wet snow is still limited, preventing the full exploitation of these observations to study supraglacial hydrology. Using the Snow Microwave Radiative Transfer (SMRT) model, this study investigates the sensitivity of microwave brightness temperature to snow liquid water content at frequencies from 1.4 to 37 GHz. We first determine the snowpack properties for eight selected coastal sites by retrieving profiles of density, grain size and ice layers from microwave observations when the snowpack is dry during wintertime. Second, a series of brightness temperature simulations is run with added water. The results show that (i) a small quantity of liquid water (≈0.5 kg m−2) can be detected, but the actual quantity cannot be retrieved out of the full range of possible water quantities; (ii) the detection of a buried wet layer is possible up to a maximum depth of 1 to 6 m depending on the frequency (6–37 GHz) and on the snow properties (grain size, density) at each site; (iii) surface ponds and water-saturated areas may prevent melt detection, but the current coverage of these waterbodies in the large satellite field of view is presently too small in Antarctica to have noticeable effects; and (iv) at 1.4 GHz, while the simulations are less reliable, we found a weaker sensitivity to liquid water and the maximal depth of detection is relatively shallow ( m) compared to the typical radiation penetration depth in dry firn (≈1000 m) at this low frequency. These numerical results pave the way for the development of improved multi-frequency algorithms to detect melt intensity and the depth of liquid water below the surface in the Antarctic snowpack.

Melt on the surface of Antarctic ice shelves can potentially lead to their disintegration, accelerating the flow of grounded ice to the ocean and raising global sea levels. However, the current understanding of the processes driving surface melt is incomplete, increasing uncertainty in predictions of ice shelf stability and thus of Antarctica's contribution to sea-level rise. Previous studies of surface melt in Antarctica have usually focused on either a process-level understanding of melt through energy-balance investigations or used metrics such as the annual number of melt days to quantify spatiotemporal variability in satellite observations of surface melt. Here, we help bridge the gap between work at these two scales. Using daily passive microwave observations from the AMSR-E and AMSR-2 sensors and the machine learning approach of a self-organising map, we identify nine representative spatial distributions (“patterns”) of surface melt on the Shackleton Ice Shelf in East Antarctica from 2002/03–2020/21. Combined with output from the RACMO2.3p3 regional climate model and surface topography from the REMA digital elevation model, our results point to a significant role for surface air temperatures in controlling the interannual variability in summer melt and also reveal the influence of localised controls on melt. In particular, prolonged melt along the grounding line shows the importance of katabatic winds and surface albedo. Our approach highlights the necessity of understanding both local and large-scale controls on surface melt and demonstrates that self-organising maps can be used to investigate the variability in surface melt on Antarctic ice shelves.

Microwave remote sensing of the cryosphere demands a formulation of the scattering coefficient which can be applied over the entire range of relevant densities, from fresh snow to bubbly ice, at all frequencies and for any grain size and snow type. Most challenging are intermediate densities (450–550 kg m−3) and high frequencies (or coarse-grained snow) where current scattering formulations break down. In this brief communication we demonstrate that the strong contrast expansion method recently developed for heterogeneous, dielectric media can be applied to microwave scattering in snow, firn and ice to solve these problems.

The energy budget of Arctic sea ice is strongly affected by the snow cover. Intensive sampling of snow properties was conducted near Qikiqtarjuaq in Baffin Bay on typical landfast sea ice during two melt seasons in 2015 and 2016. The sampling included stratigraphy, vertical profiles of snow specific surface area (SSA), density and irradiance, and spectral albedo (300–1100 nm). Both years featured four main phases: (I) dry snow cover, (II) surface melting, (III) ripe snowpack, and (IV) melt pond formation. Each phase was characterized by distinctive physical and optical properties. A high SSA value of 49.3 m2 kg−1 was measured during phase I on surface wind slabs together with a corresponding broadband albedo (300–3000 nm) of 0.87. Phase II was marked by alternating episodes of surface melting, which dramatically decreased the SSA below 3 m2 kg−1, and episodes of snowfall re-establishing pre-melt conditions. Albedo was highly time-variable, with minimum broadband values around 0.70. In phase III, continued melting led to a fully ripe snowpack composed of clustered rounded grains. Albedo began to decrease in the visible as snow thickness decreased but remained steady at longer wavelengths. Moreover, significant spatial variability appeared for the first time following snow depth heterogeneity. Spectral albedo was simulated by radiative transfer using measured SSA and density vertical profiles and estimated impurity contents based on limited measurements. Simulations were most of the time within 1 % of measurements in the visible and within 2 % in the infrared. Simulations allowed the calculations of albedo and of the spectral flux at the snow–ice interface. These showed that photosynthetically active radiation fluxes at the bottom of the snowpack durably exceeded 5 W m−2 (∼9.2 µmol m−2 s−1) only when the snowpack thickness started to decrease at the end of phase II.

The surface temperature controls the temporal evolution of the snowpack, playing a key role in metamorphism and snowmelt. It shows large spatial variations in mountainous areas because the surface energy budget is affected by the topography, for instance because of the modulation of the short-wave irradiance by the local slope and the shadows and the short-wave and long-wave re-illumination of the surface from surrounding slopes. These topographic effects are often neglected in large-scale models considering the surface to be flat and smooth. Here we aim at estimating the surface temperature of snow-covered mountainous terrain in clear-sky conditions in order to evaluate the relative importance of the different processes that control the spatial variations. For this, a modelling chain is implemented to compute the surface temperature in a kilometre-wide area from local radiometric and meteorological measurements at a single station. The first component of this chain is the Rough Surface Ray-Tracing (RSRT) model. Based on a photon transport Monte Carlo algorithm, this model quantifies the incident and reflected short-wave radiation on every facet of the mesh describing the snow-covered terrain. The second component is a surface scheme that estimates the terms of the surface energy budget from which the surface temperature is eventually estimated. To assess the modelling chain performance, we use in situ measurements of surface temperature and satellite thermal observations (Landsat 8) in the Col du Lautaret area, in the French Alps. The results of the simulations show (i) an agreement between the simulated and measured surface temperature at the station for a diurnal cycle in winter within 0.2 ∘C; (ii) that the spatial variations in surface temperature are on the order of 5 to 10 ∘C in the domain and are well represented by the model; and (iii) that the topographic effects ranked by importance are the modulation of solar irradiance by the local slope, followed by the altitudinal variations in air temperature (lapse rate), the re-illumination by long-wave thermal emission from surrounding terrain, and the spectral dependence of snow albedo. The changes in the downward long-wave flux because of variations in altitude and the absorption enhancement due to multiple bounces of photons in steep terrain play a less significant role. These results show the necessity of considering the topography to correctly assess the energy budget and the surface temperature of snow-covered complex terrain.

The warming-induced expansion of shrubs in the Arctic is transforming snowpacks into a mixture of snow, impurities and buried branches. Because snow is a translucent medium into which light penetrates up to tens of centimetres, buried branches may alter the snowpack radiation budget with important consequences for the snow thermal regime and microstructure. To characterize the influence of buried branches on radiative transfer in snow, irradiance profiles were measured in snowpacks with and without shrubs near Umiujaq in the Canadian Low Arctic (56.5∘ N, 76.5∘ W) in November and December 2015. Using the irradiance profiles measured in shrub-free snowpacks in combination with a Monte Carlo radiative transfer model revealed that the dominant impurity type was black carbon (BC) in variable concentrations up to 185 ng g−1. This allowed the separation of the radiative effects of impurities and buried branches. Irradiance profiles measured in snowpacks with shrubs showed that the impact of buried branches was local (i.e. a few centimetres around branches) and only observable in layers where branches were also visible in snowpit photographs. The local-effect hypothesis was further supported by observations of localized melting and depth hoar pockets that formed in the vicinity of branches. Buried branches therefore affect snowpack properties, with possible impacts on Arctic flora and fauna and on the thermal regime of permafrost. Lastly, the unexpectedly high BC concentrations in snow are likely caused by nearby open-air waste burning, suggesting that cleaner waste management plans are required for northern community and ecosystem protection.

In the 2019/2020 austral summer, the surface melt duration and extent on the northern George VI Ice Shelf (GVIIS) was exceptional compared to the 31 previous summers of distinctly lower melt. This finding is based on analysis of near-continuous 41-year satellite microwave radiometer and scatterometer data, which are sensitive to meltwater on the ice shelf surface and in the near-surface snow. Using optical satellite imagery from Landsat 8 (2013 to 2020) and Sentinel-2 (2017 to 2020), record volumes of surface meltwater ponding were also observed on the northern GVIIS in 2019/2020, with 23 % of the surface area covered by 0.62 km3 of ponded meltwater on 19 January. These exceptional melt and surface ponding conditions in 2019/2020 were driven by sustained air temperatures ≥0 ∘C for anomalously long periods (55 to 90 h) from late November onwards, which limited meltwater refreezing. The sustained warm periods were likely driven by warm, low-speed (≤7.5 m s−1) northwesterly and northeasterly winds and not by foehn wind conditions, which were only present for 9 h total in the 2019/2020 melt season. Increased surface ponding on ice shelves may threaten their stability through increased potential for hydrofracture initiation; a risk that may increase due to firn air content depletion in response to near-surface melting.

The presence of light-absorbing particles (LAPs) in snow leads to a decrease in short-wave albedo affecting the surface energy budget. However, the understanding of the impacts of LAPs is hampered by the lack of dedicated datasets, as well as the scarcity of models able to represent the interactions between LAPs and snow metamorphism. The present study aims to address both these limitations by introducing a survey of LAP concentrations over two snow seasons in the French Alps and an estimation of their impacts based on the Crocus snowpack model that represents the complex interplays between LAP dynamics and snow metamorphism. First, a unique dataset collected at Col du Lautaret (2058 m a.s.l., above sea level, French Alps) for the two snow seasons 2016–2017 and 2017–2018 is presented. This dataset consists of spectral albedo measurements, vertical profiles of snow specific surface area (SSA), density and concentrations of different LAP species. Spectral albedos are processed to estimate SSA and LAP absorption-equivalent concentrations near the surface of the snowpack. These estimates are then compared to chemical measurements of LAP concentrations and SSA measurements. Our dataset highlights, among others, large discrepancies between two measurement techniques of black carbon (BC) concentrations in snow (namely thermal-optical and laser-induced incandescence). Second, we present ensemble snowpack simulations of the multi-physics version of the detailed snowpack model Crocus, forced with in situ meteorological data, as well as dust and BC deposition fluxes from an atmospheric model. The temporal variations of near-surface LAP concentrations and SSA are most of the time correctly simulated. The simulated seasonal radiative forcing of LAPs is 33 % higher for the 2017–2018 snow season than for the 2016–2017 one, highlighting a strong variability between these two seasons. However, the shortening of the snow season caused by LAPs is similar with 10 ± 5 and 11 ± 1 d for the first and the second snow seasons, respectively. This counter-intuitive result is attributed to two small snowfalls at the end of the first season and highlights the importance in accounting for meteorological conditions to correctly predict the impact of LAPs. The strong variability of season shortening caused by LAPs in the multi-physics ensemble for the first season (10 ± 5 d) also points out the sensitivity of model-based estimations of LAP impact on modelling uncertainties of other processes. Finally, the indirect impact of LAPs (i.e. the enhancement of energy absorption due to the acceleration of the metamorphism by LAPs) is negligible for the 2 years considered here, which is contrary to what was found in previous studies for other sites.

The monitoring of snow-covered surfaces on Earth is largely facilitated by the wealth of satellite data available, with increasing spatial resolution and temporal coverage over the last few years. Yet to date, retrievals of snow physical properties still remain complicated in mountainous areas, owing to the complex interactions of solar radiation with terrain features such as multiple scattering between slopes, exacerbated over bright surfaces. Existing physically based models of solar radiation across rough scenes are either too complex and resource-demanding for the implementation of systematic satellite image processing, not designed for highly reflective surfaces such as snow, or tied to a specific satellite sensor. This study proposes a new formulation, combining a forward model of solar radiation over rugged terrain with dedicated snow optics into a flexible multi-sensor tool that bridges a gap in the optical remote sensing of snow-covered surfaces in mountainous regions. The model presented here allows one to perform rapid calculations over large snow-covered areas. Good results are obtained even for extreme cases, such as steep shadowed slopes or, on the contrary, strongly illuminated sun-facing slopes. Simulations of Sentinel-3 OLCI (Ocean and Land Colour Instrument) scenes performed over a mountainous region in the French Alps allow us to reduce the bias by up to a factor of 6 in the visible wavelengths compared to methods that account for slope inclination only. Furthermore, the study underlines the contribution of the individual fluxes to the total top-of-atmosphere radiance, highlighting the importance of reflected radiation from surrounding slopes which, in midwinter after a recent snowfall (13 February 2018), accounts on average for 7 % of the signal at 400 nm and 16 % at 1020 nm (on 13 February 2018), as well as of coupled diffuse radiation scattered by the neighbourhood, which contributes to 18 % at 400 nm and 4 % at 1020 nm. Given the importance of these contributions, accounting for slopes and reflected radiation between terrain features is a requirement for improving the accuracy of satellite retrievals of snow properties over snow-covered rugged terrain. The forward formulation presented here is the first step towards this goal, paving the way for future retrievals.

Surface albedo is an essential variable to determine the Earth's surface energy budget, in particular for snow-covered areas where it is involved in one of the most powerful positive feedback loops of the climate system. In situ measurements of broadband and spectral albedo are therefore common. However they are subject to several artefacts. Here we investigate the sensitivity of spectral albedo measurements to surface slope, and we propose simple correction algorithms to retrieve the intrinsic albedo of a slope from measurements, as if it were flat. For this, we first derive the analytical equations relating albedo measured on a slope to intrinsic direct and diffuse albedo, the apportionment between diffuse and direct incoming radiation, and slope inclination and aspect. The theory accounts for two main slope effects. First, the slope affects the proportion of solar radiation intercepted by the surface relative to that intercepted by the upward-looking, horizontal, sensor. Second, the upward- and downward-looking sensors receive reduced radiation from the sky and the surface respectively and increased radiation from neighbouring terrain. Using this theory, we show that (i) slope has a significant effect on albedo (over 0.01) from as little as a ≈1∘ inclination, causing distortions of the albedo spectral shape; (ii) the first-order slope effect is sufficient to fully explain measured albedo up to ≈15∘, which we designate “small-slope approximation”; and (iii) for larger slopes, the theory depends on the neighbouring slope geometry and land cover, leading to much more complex equations. Next, we derive four correction methods from the small-slope approximation, to be used depending on whether (1) the slope inclination and orientation are known or not, (2) the snow surface is free of impurities or dirty, and (3) a single or a time series of albedo measurements is available. The methods applied to observations taken in the Alps on terrain with up to nearly 20∘ slopes prove the ability to recover intrinsic albedo with a typical accuracy of 0.03 or better. From this study, we derive two main recommendations for future field campaigns: first, sloping terrain requires more attention because it reduces the measurement accuracy of albedo even for almost invisible slopes (1–2∘). Second, while the correction of the slope effect is possible, it requires additional information such as the spectral diffuse and direction partitioning and if possible the actual slope inclination and aspect, especially when the absence of impurities can not be assumed.

Most models simulating snow albedo assume a flat and smooth surface, neglecting surface roughness. However, the presence of macroscopic roughness leads to a systematic decrease in albedo due to two effects: (1) photons are trapped in concavities (multiple reflection effect) and (2) when the sun is low, the roughness sides facing the sun experience an overall decrease in the local incidence angle relative to a smooth surface, promoting higher absorption, whilst the other sides have weak contributions because of the increased incidence angle or because they are shadowed (called the effective-angle effect here). This paper aims to quantify the impact of surface roughness on albedo and to assess the respective role of these two effects, with (1) observations over varying amounts of surface roughness and (2) simulations using the new rough surface ray-tracing (RSRT) model, based on a Monte Carlo method for photon transport calculation. The observations include spectral albedo (400–1050 nm) over manually created roughness surfaces with multiple geometrical characteristics. Measurements highlight that even a low fraction of surface roughness features (7 % of the surface) causes an albedo decrease of 0.02 at 1000 nm when the solar zenith angle (θs) is larger than 50∘. For higher fractions (13 %, 27 % and 63 %), and when the roughness orientation is perpendicular to the sun, the decrease is of 0.03–0.04 at 700 nm and of 0.06–0.10 at 1000 nm. The impact is 20 % lower when roughness orientation is parallel to the sun. The observations are subsequently compared to RSRT simulations. Accounting for surface roughness improves the model observation agreement by a factor of 2 at 700 and 1000 nm (errors of 0.03 and 0.04, respectively) compared to simulations considering a flat smooth surface. The model is used to explore the albedo sensitivity to surface roughness with varying snow properties and illumination conditions. Both multiple reflections and the effective-angle effect have a greater impact with low specific surface area (SSA;  m2 kg−1). The effective-angle effect also increases rapidly with θs at large θs. This latter effect is larger when the overall slope of the surface is facing away from the sun and has a roughness orientation perpendicular to the sun. For a snowpack where artificial surface roughness features were created, we showed that a broadband albedo decrease of 0.05 may cause an increase in the net shortwave radiation of 80 % (from 15 to 27 W m−2). This paper highlights the necessity of considering surface roughness in the estimation of the surface energy budget and opens the way for considering natural rough surfaces in snow modelling.

Melt occurrence in Antarctica is derived from L-band observations from the Soil Moisture and Ocean Salinity (SMOS) satellite between the austral summer 2010–2011 and 2017–2018. The detection algorithm is adapted from a threshold method previously developed for 19 GHz passive microwave measurements from the special sensor microwave imager (SSM/I) and special sensor microwave imager sounder (SSMIS). The comparison of daily melt occurrence retrieved from 1.4 and 19 GHz observations shows an overall close agreement, but a lag of few days is usually observed by SMOS at the beginning of the melt season. To understand the difference, a theoretical analysis is performed using a microwave emission radiative transfer model. It shows that the sensitivity of 1.4 GHz signal to liquid water is significantly weaker than at 19 GHz if the water is only present in the uppermost tens of centimetres of the snowpack. Conversely, 1.4 GHz measurements are sensitive to water when spread over at least 1 m and when present in depths up to hundreds of metres. This is explained by the large penetration depth in dry snow and by the long wavelength (21 cm). We conclude that SMOS and higher-frequency radiometers provide interesting complementary information on melt occurrence and on the location of the water in the snowpack.

The Green Edge initiative was developed to investigate the processes controlling the primary productivity and fate of organic matter produced during the Arctic phytoplankton spring bloom (PSB) and to determine its role in the ecosystem. Two field campaigns were conducted in 2015 and 2016 at an ice camp located on landfast sea ice southeast of Qikiqtarjuaq Island in Baffin Bay (67.4797∘ N, 63.7895∘ W). During both expeditions, a large suite of physical, chemical and biological variables was measured beneath a consolidated sea-ice cover from the surface to the bottom (at 360 m depth) to better understand the factors driving the PSB. Key variables, such as conservative temperature, absolute salinity, radiance, irradiance, nutrient concentrations, chlorophyll a concentration, bacteria, phytoplankton and zooplankton abundance and taxonomy, and carbon stocks and fluxes were routinely measured at the ice camp. Meteorological and snow-relevant variables were also monitored. Here, we present the results of a joint effort to tidy and standardize the collected datasets, which will facilitate their reuse in other Arctic studies. The dataset is available at https://doi.org/10.17882/59892 (Massicotte et al., 2019a).

Snow albedo schemes in regional climate models often lack a sophisticated radiation penetration scheme and generally compute only a broadband albedo. Here, we present the Spectral-to-NarrOWBand ALbedo module (SNOWBAL, version 1.2) to couple effectively a spectral albedo model with a narrowband radiation scheme. Specifically, the Two-streAm Radiative TransfEr in Snow model (TARTES) is coupled with the European Centre for Medium-Range Weather Forecasts (ECMWF) Integrated Forecast System (IFS), cycle 33R1, atmospheric radiation scheme based on the Rapid Radiation Transfer Model, which is embedded in the Regional Atmospheric Climate Model version 2.3p2 (RACMO2). This coupling allows to explicitly account for the effect of clouds, water vapor, snow impurities and snow metamorphism on albedo. Firstly, we present a narrowband albedo method to project the spectral albedos of TARTES onto the 14 spectral bands of the IFS shortwave radiation scheme using a representative wavelength (RW) for each band. Using TARTES and spectral downwelling surface irradiance derived with the DIScrete Ordinate Radiative Transfer atmospheric model, we show that RWs primarily depend on the solar zenith angle (SZA), cloud content and water vapor. Secondly, we compare the TARTES narrowband albedo, using offline RACMO2 results for south Greenland, with the broadband albedo parameterizations of Gardner and Sharp (2010), currently implemented in RACMO2, and the multi-layered parameterization of Kuipers Munneke et al. (2011, PKM). The actual absence of radiation penetration in RACMO2 leads on average to a higher albedo compared with TARTES narrowband albedo. Furthermore, large differences between the TARTES narrowband albedo and PKM and RACMO2 are observed for high SZA and clear-sky conditions, and after melt events when the snowpack is very inhomogeneous. This highlights the importance of accounting for spectral albedo and radiation penetration to simulate the energy budget of the Greenland ice sheet.

Light-absorbing particles (LAPs) such as black carbon or mineral dust are some of the main drivers of snow radiative transfer. Small amounts of LAPs significantly increase snowpack absorption in the visible wavelengths where ice absorption is particularly weak, impacting the surface energy budget of snow-covered areas. However, linking measurements of LAP concentration in snow to their actual radiative impact is a challenging issue which is not fully resolved. In the present paper, we point out a new method based on spectral irradiance profile (SIP) measurements which makes it possible to identify the radiative impact of LAPs on visible light extinction in homogeneous layers of the snowpack. From this impact on light extinction it is possible to infer LAP concentrations present in each layer using radiative transfer theory. This study relies on a unique dataset composed of 26 spectral irradiance profile measurements in the wavelength range 350–950 nm with concomitant profile measurements of snow physical properties and LAP concentrations, collected in the Alps over two snow seasons in winter and spring conditions. For 55 homogeneous snow layers identified in our dataset, the concentrations retrieved from SIP measurements are compared to chemical measurements of LAP concentrations. A good correlation is observed for measured concentrations higher than 5 ng g−1 (r2=0.81) despite a clear positive bias. The potential causes of this bias are discussed, underlining a strong sensitivity of our method to LAP optical properties and to the relationship between snow microstructure and snow optical properties used in the theory. Additional uncertainties such as artefacts in the measurement technique for SIP and chemical contents along with LAP absorption efficiency may explain part of this bias. In addition, spectral information on LAP absorption can be retrieved from SIP measurements. We show that for layers containing a unique absorber, this absorber can be identified in some cases (e.g. mineral dust vs. black carbon). We also observe an enhancement of light absorption between 350 and 650 nm in the presence of liquid water in the snowpack, which is discussed but not fully elucidated. A single SIP acquisition lasts approximately 1 min and is hence much faster than collecting a profile of chemical measurements. With the recent advances in modelling LAP–snow interactions, our method could become an attractive alternative to estimate vertical profiles of LAP concentrations in snow.

Snow accumulation is the main positive component of the mass balance in Antarctica. In contrast to the major efforts deployed to estimate its overall value on a continental scale – to assess the contribution of the ice sheet to sea level rise – knowledge about the accumulation process itself is relatively poor, although many complex phenomena occur between snowfall and the definitive settling of the snow particles on the snowpack. Here we exploit a dataset of near-daily surface elevation maps recorded over 3 years at Dome C using an automatic laser scanner sampling 40–100 m2 in area. We find that the averaged accumulation is relatively regular over the 3 years at a rate of +8.7 cm yr−1. Despite this overall regularity, the surface changes very frequently (every 3 d on average) due to snow erosion and heterogeneous snow deposition that we call accumulation by “patches”. Most of these patches (60 %–85 %) are ephemeral but can survive a few weeks before being eroded. As a result, the surface is continuously rough (6–8 cm root-mean-square height) featuring meter-scale dunes aligned along the wind and larger, decameter-scale undulations. Additionally, we deduce the age of the snow present at a given time on the surface from elevation time series and find that snow age spans over more than a year. Some of the patches ultimately settle, leading to a heterogeneous internal structure which reflects the surface heterogeneity, with many snowfall events missing at a given point, whilst many others are overrepresented. These findings have important consequences for several research topics including surface mass balance, surface energy budget, photochemistry, snowpack evolution, and the interpretation of the signals archived in ice cores.

Surface snow density is an important variable for the surface mass balance and energy budget. It evolves according to meteorological conditions, in particular, snowfall, wind, and temperature, but the physical processes governing atmospheric influence on snow are not fully understood. A reason is that no systematic observation is available on a continental scale. Here, we use the passive microwave observations from AMSR-E satellite to retrieve the surface snow density at Dome C on the East Antarctic Plateau. The retrieval method is based on the difference of surface reflections between horizontally and vertically polarized brightness temperatures at 37 GHz, highlighted by the computation of the polarization ratio, which is related to surface snow density. The relationship has been obtained with a microwave emission radiative transfer model (DMRT-ML). The retrieved density, approximately representative of the topmost 3 cm of the snowpack, compares well with in situ measurements. The difference between mean in situ measurements and mean retrieved density is 26.2 kg m−3, which is within typical in situ measurement uncertainties. We apply the retrieval method to derive the time series over the period 2002–2011. The results show a marked and persistent pluri-annual decrease of about 10 kg m−3 yr−1, in addition to atmosphere-related seasonal, weekly, and daily density variations. This trend is confirmed by independent active microwave observations from the ENVISAT and QuikSCAT satellites, though the link to the density is more difficult to establish. However, no related pluri-annual change in meteorological conditions has been found to explain such a trend in snow density. Further work will concern the extension of the method to the continental scale.

We propose a system of analytical equations to retrieve snow grain size and absorption coefficient of pollutants from snow reflectance or snow albedo measurements in the visible and near-infrared regions of the electromagnetic spectrum, where snow single-scattering albedo is close to 1.0. It is assumed that ice grains and impurities (e.g., dust, black and brown carbon) are externally mixed, and that the snow layer is semi-infinite and vertically and horizontally homogeneous. The influence of close-packing effects on reflected light intensity are assumed to be small and ignored. The system of nonlinear equations is solved analytically under the assumption that impurities have the spectral absorption coefficient, which obey the Ångström power law, and the impurities influence the registered spectra only in the visible and not in the near infrared (and vice versa for ice grains). The theory is validated using spectral reflectance measurements and albedo of clean and polluted snow at various locations (Antarctica Dome C, European Alps). A technique to derive the snow albedo (plane and spherical) from reflectance measurements at a fixed observation geometry is proposed. The technique also enables the simulation of hyperspectral snow reflectance measurements in the broad spectral range from ultraviolet to the near infrared for a given snow surface if the actual measurements are performed at a restricted number of wavelengths (two to four, depending on the type of snow and the measurement system).

The Snow Microwave Radiative Transfer (SMRT) thermal emission and backscatter model was developed to determine uncertainties in forward modeling through intercomparison of different model ingredients. The model differs from established models by the high degree of flexibility in switching between different electromagnetic theories, representations of snow microstructure, and other modules involved in various calculation steps. SMRT v1.0 includes the dense media radiative transfer theory (DMRT), the improved Born approximation (IBA), and independent Rayleigh scatterers to compute the intrinsic electromagnetic properties of a snow layer. In the case of IBA, five different formulations of the autocorrelation function to describe the snow microstructure characteristics are available, including the sticky hard sphere model, for which close equivalence between the IBA and DMRT theories has been shown here. Validation is demonstrated against established theories and models. SMRT was used to identify that several former studies conducting simulations with in situ measured snow properties are now comparable and moreover appear to be quantitatively nearly equivalent. This study also proves that a third parameter is needed in addition to density and specific surface area to characterize the microstructure. The paper provides a comprehensive description of the mathematical basis of SMRT and its numerical implementation in Python. Modularity supports model extensions foreseen in future versions comprising other media (e.g., sea ice, frozen lakes), different scattering theories, rough surface models, or new microstructure models.

To evaluate the impact of vapor diffusion on isotopic composition variations in snow pits and then in ice cores, we introduced water isotopes in the detailed snowpack model Crocus. At each step and for each snow layer, (1) the initial isotopic composition of vapor is taken at equilibrium with the solid phase, (2) a kinetic fractionation is applied during transport, and (3) vapor is condensed or snow is sublimated to compensate for deviation to vapor pressure at saturation.We study the different effects of temperature gradient, compaction, wind compaction, and precipitation on the final vertical isotopic profiles. We also run complete simulations of vapor diffusion along isotopic gradients and of vapor diffusion driven by temperature gradients at GRIP, Greenland and at Dome C, Antarctica over periods of 1 or 10 years. The vapor diffusion tends to smooth the original seasonal signal, with an attenuation of 7 to 12 % of the original signal over 10 years at GRIP. This is smaller than the observed attenuation in ice cores, indicating that the model attenuation due to diffusion is underestimated or that other processes, such as ventilation, influence attenuation. At Dome C, the attenuation is stronger (18 %), probably because of the lower accumulation and stronger δ18O gradients.

Spaceborne radar altimeters are a valuable tool for observing the Antarctic Ice Sheet. The radar wave interaction with the snow provides information on both the surface and the subsurface of the snowpack due to its dependence on the snow properties. However, the penetration of the radar wave within the snowpack also induces a negative bias on the estimated surface elevation. Empirical corrections of this space- and time-varying bias are usually based on the backscattering coefficient variability. We investigate the spatial and seasonal variations of the backscattering coefficient at the S (3.2 GHz ∼ 9.4 cm), Ku (13.6 GHz ∼ 2.3 cm) and Ka (37 GHz ∼ 0.8 cm) bands. We identified that the backscattering coefficient at Ku band reaches a maximum in winter in part of the continent (Region 1) and in the summer in the remaining (Region 2), while the evolution at other frequencies is relatively uniform over the whole continent. To explain this contrasting behavior between frequencies and between regions, we studied the sensitivity of the backscattering coefficient at three frequencies to several parameters (surface snow density, snow temperature and snow grain size) using an electromagnetic model. The results show that the seasonal cycle of the backscattering coefficient at Ka frequency is dominated by the volume echo and is mainly driven by snow temperature evolution everywhere. In contrast, at S band, the cycle is dominated by the surface echo. At Ku band, the seasonal cycle is dominated by the volume echo in Region 1 and by the surface echo in Region 2. This investigation provides new information on the seasonal dynamics of the Antarctic Ice Sheet surface and provides new clues to build more accurate corrections of the radar altimeter surface elevation signal in the future.

The oldest ice core records are obtained from the East Antarctic Plateau. Water isotopes are key proxies to reconstructing past climatic conditions over the ice sheet and at the evaporation source. The accuracy of climate reconstructions depends on knowledge of all processes affecting water vapour, precipitation and snow isotopic compositions. Fractionation processes are well understood and can be integrated in trajectory-based Rayleigh distillation and isotope-enabled climate models. However, a quantitative understanding of processes potentially altering snow isotopic composition after deposition is still missing. In low-accumulation sites, such as those found in East Antarctica, these poorly constrained processes are likely to play a significant role and limit the interpretability of an ice core's isotopic composition.By combining observations of isotopic composition in vapour, precipitation, surface snow and buried snow from Dome C, a deep ice core site on the East Antarctic Plateau, we found indications of a seasonal impact of metamorphism on the surface snow isotopic signal when compared to the initial precipitation. Particularly in summer, exchanges of water molecules between vapour and snow are driven by the diurnal sublimation–condensation cycles. Overall, we observe in between precipitation events modification of the surface snow isotopic composition. Using high-resolution water isotopic composition profiles from snow pits at five Antarctic sites with different accumulation rates, we identified common patterns which cannot be attributed to the seasonal variability of precipitation. These differences in the precipitation, surface snow and buried snow isotopic composition provide evidence of post-deposition processes affecting ice core records in low-accumulation areas.

Light-absorbing impurities (LAIs) decrease snow albedo, increasing the amount of solar energy absorbed by the snowpack. Its most intuitive and direct impact is to accelerate snowmelt. Enhanced energy absorption in snow also modifies snow metamorphism, which can indirectly drive further variations of snow albedo in the near-infrared part of the solar spectrum because of the evolution of the near-surface snow microstructure. New capabilities have been implemented in the detailed snowpack model SURFEX/ISBA-Crocus (referred to as Crocus) to account for impurities' deposition and evolution within the snowpack and their direct and indirect impacts. Once deposited, the model computes impurities' mass evolution until snow melts out, accounting for scavenging by meltwater. Taking advantage of the recent inclusion of the spectral radiative transfer model TARTES (Two-stream Analytical Radiative TransfEr in Snow model) in Crocus, the model explicitly represents the radiative impacts of light-absorbing impurities in snow. The model was evaluated at the Col de Porte experimental site (French Alps) during the 2013–2014 snow season against in situ standard snow measurements and spectral albedo measurements. In situ meteorological measurements were used to drive the snowpack model, except for aerosol deposition fluxes. Black carbon (BC) and dust deposition fluxes used to drive the model were extracted from simulations of the atmospheric model ALADIN-Climate. The model simulates snowpack evolution reasonably, providing similar performances to our reference Crocus version in terms of snow depth, snow water equivalent (SWE), near-surface specific surface area (SSA) and shortwave albedo. Since the reference empirical albedo scheme was calibrated at the Col de Porte, improvements were not expected to be significant in this study. We show that the deposition fluxes from the ALADIN-Climate model provide a reasonable estimate of the amount of light-absorbing impurities deposited on the snowpack except for extreme deposition events which are greatly underestimated. For this particular season, the simulated melt-out date advances by 6 to 9 days due to the presence of light-absorbing impurities. The model makes it possible to apportion the relative importance of direct and indirect impacts of light-absorbing impurities on energy absorption in snow. For the snow season considered, the direct impact in the visible part of the solar spectrum accounts for 85 % of the total impact, while the indirect impact related to accelerated snow metamorphism decreasing near-surface specific surface area and thus decreasing near-infrared albedo accounts for 15 % of the total impact. Our model results demonstrate that these relative proportions vary with time during the season, with potentially significant impacts for snowmelt and avalanche prediction.

Ice is a highly transparent material in the visible. According to the most widely used database (IA2008; Warren and Brandt, 2008), the ice absorption coefficient reaches values lower than 10−3 m−1 around 400 nm. These values were obtained from a vertical profile of spectral radiance measured in a single snow layer at Dome C in Antarctica. We reproduced this experiment using an optical fiber inserted in the snow to record 56 profiles from which 70 homogeneous layers were identified. Applying the same estimation method on every layer yields 70 ice absorption spectra. They present a significant variability but absorption coefficients are overall larger than IA2008 by 1 order of magnitude at 400–450 nm. We devised another estimation method based on Bayesian inference that treats all the profiles simultaneously. It reduces the statistical variability and confirms the higher absorption, around 2 × 10−2 m−1 near the minimum at 440 nm. We explore potential instrumental artifacts by developing a 3-D radiative transfer model able to explicitly account for the presence of the fiber in the snow. The simulation shows that the radiance profile is indeed perturbed by the fiber intrusion, but the error on the ice absorption estimate is not larger than a factor of 2. This is insufficient to explain the difference between our new estimate and IA2008. The same conclusion applies regarding the plausible contamination by black carbon or dust, concentrations reported in the literature are insufficient. Considering the large number of profiles acquired for this study and other estimates from the Antarctic Muon and Neutrino Detector Array (AMANDA), we nevertheless estimate that ice absorption values around 10−2 m−1 at the minimum are more likely than under 10−3 m−1. A new estimate in the range 400–600 nm is provided for future modeling of snow, cloud, and sea-ice optical properties. Most importantly, we recommend that modeling studies take into account the large uncertainty of the ice absorption coefficient in the visible and that future estimations of the ice absorption coefficient should also thoroughly account for the impact of the measurement method.

Snowpack is a multiphase (photo)chemical reactor that strongly influences the air composition in polar and snow-covered regions. Snowpack plays a special role in the nitrogen cycle, as it has been shown that nitrate undergoes numerous recycling stages (including photolysis) in the snow before being permanently buried in the ice. However, the current understanding of these physicochemical processes remains very poor. Several modelling studies have attempted to reproduce (photo)chemical reactions inside snow grains, but these have relied on strong assumptions to characterise snow reactive properties, which are not well defined. Air–snow exchange processes such as adsorption, solid-state diffusion, or co-condensation also affect snow chemical composition. Here, we present a physically based model of these processes for nitrate. Using as input a 1-year-long time series of atmospheric nitrate concentration measured at Dome C, Antarctica, our model reproduces with good agreement the nitrate measurements in the surface snow. By investigating the relative importance of the main exchange processes, this study shows that, on the one hand, the combination of bulk diffusion and co-condensation allows a good reproduction of the measurements (correlation coefficient r = 0.95), with a correct amplitude and timing of summer peak concentration of nitrate in snow. During winter, nitrate concentration in surface snow is mainly driven by thermodynamic equilibrium, whilst the peak observed in summer is explained by the kinetic process of co-condensation. On the other hand, the adsorption of nitric acid on the surface of the snow grains, constrained by an already existing parameterisation for the isotherm, fails to fit the observed variations. During winter and spring, the modelled concentration of adsorbed nitrate is respectively 2.5 and 8.3-fold higher than the measured one. A strong diurnal variation driven by the temperature cycle and a peak occurring in early spring are two other major features that do not match the measurements. This study clearly demonstrates that co-condensation is the most important process to explain nitrate incorporation in snow undergoing temperature gradient metamorphism. The parameterisation developed for this process can now be used as a foundation piece in snowpack models to predict the inter-relationship between snow physical evolution and snow nitrate chemistry.

Although both the temporal and spatial variations of the snow depth are usually of interest for numerous applications, available measurement techniques are either space-oriented (e.g. terrestrial laser scans) or time-oriented (e.g. ultrasonic ranging probe). Because of snow heterogeneity, measuring depth in a single point is insufficient to provide accurate and representative estimates. We present a cost-effective automatic instrument to acquire spatio-temporal variations of snow depth. The device comprises a laser meter mounted on a 2-axis stage and can scan ≈ 200 000 points over an area of 100–200 m2 in 4 h. Two instruments, installed in Antarctica (Dome C) and the French Alps (Col de Porte), have been operating continuously and unattended over 2015 with a success rate of 65 and 90 % respectively. The precision of single point measurements and long-term stability were evaluated to be about 1 cm and the accuracy to be 5 cm or better. The spatial variability in the scanned area reached 7–10 cm (root mean square) at both sites, which means that the number of measurements is sufficient to average out the spatial variability and yield precise mean snow depth. With such high precision, it was possible for the first time at Dome C to (1) observe a 3-month period of regular and slow increase of snow depth without apparent link to snowfalls and (2) highlight that most of the annual accumulation stems from a single event although several snowfall and strong wind events were predicted by the ERA-Interim reanalysis. Finally the paper discusses the benefit of laser scanning compared to multiplying single-point sensors in the context of monitoring snow depth.

Water stable isotopes in central Antarctic ice cores are critical to quantify past temperature changes. Accurate temperature reconstructions require one to understand the processes controlling surface snow isotopic composition. Isotopic fractionation processes occurring in the atmosphere and controlling snowfall isotopic composition are well understood theoretically and implemented in atmospheric models. However, post-deposition processes are poorly documented and understood. To quantitatively interpret the isotopic composition of water archived in ice cores, it is thus essential to study the continuum between surface water vapour, precipitation, surface snow and buried snow. Here, we target the isotopic composition of water vapour at Concordia Station, where the oldest EPICA Dome C ice cores have been retrieved. While snowfall and surface snow sampling is routinely performed, accurate measurements of surface water vapour are challenging in such cold and dry conditions. New developments in infrared spectroscopy enable now the measurement of isotopic composition in water vapour traces. Two infrared spectrometers have been deployed at Concordia, allowing continuous, in situ measurements for 1 month in December 2014–January 2015. Comparison of the results from infrared spectroscopy with laboratory measurements of discrete samples trapped using cryogenic sampling validates the relevance of the method to measure isotopic composition in dry conditions. We observe very large diurnal cycles in isotopic composition well correlated with temperature diurnal cycles. Identification of different behaviours of isotopic composition in the water vapour associated with turbulent or stratified regime indicates a strong impact of meteorological processes in local vapour/snow interaction. Even if the vapour isotopic composition seems to be, at least part of the time, at equilibrium with the local snow, the slope of δD against δ18O prevents us from identifying a unique origin leading to this isotopic composition.

Spectral albedo of the snow surface in the visible/near-infrared range has been measured for 3 years by an automatic spectral radiometer installed at Dome C (75° S, 123° E) in Antarctica in order to retrieve the specific surface area (SSA) of superficial snow. This study focuses on the uncertainties of the SSA retrieval due to instrumental and data processing limitations. We find that when the solar zenith angle is high, the main source of uncertainties is the imperfect angular response of the light collectors. This imperfection introduces a small spurious wavelength-dependent trend in the albedo spectra which greatly affects the SSA retrieval. By modeling this effect, we show that for typical snow and illumination conditions encountered at Dome C, retrieving SSA with an accuracy better than 15 % (our target) requires the difference of response between 400 and 1100 nm to not exceed 2 %. Such a small difference can be achieved only by (i) a careful design of the collectors, (ii) an ad hoc correction of the spectra using the actual measured angular response of the collectors, and (iii) for solar zenith angles less than 75°. The 3-year time series of retrieved SSA features a 3-fold decrease every summer which is significantly larger than the estimated uncertainties. This highlights the high dynamics of near-surface SSA at Dome C.

This paper examines the ability of optical reflectance data assimilation to improve snow depth and snow water equivalent simulations from a chain of models with the SAFRAN meteorological model driving the detailed multilayer snowpack model Crocus now including a two-stream radiative transfer model for snow, TARTES. The direct use of reflectance data, allowed by TARTES, instead of higher level snow products, mitigates uncertainties due to commonly used retrieval algorithms.Data assimilation is performed with an ensemble-based method, the Sequential Importance Resampling Particle filter, to represent simulation uncertainties. In snowpack modeling, uncertainties of simulations are primarily assigned to meteorological forcings. Here, a method of stochastic perturbation based on an autoregressive model is implemented to explicitly simulate the consequences of these uncertainties on the snowpack estimates.Through twin experiments, the assimilation of synthetic spectral reflectances matching the MODerate resolution Imaging Spectroradiometer (MODIS) spectral bands is examined over five seasons at the Col du Lautaret, located in the French Alps. Overall, the assimilation of MODIS-like data reduces by 45 % the root mean square errors (RMSE) on snow depth and snow water equivalent. At this study site, the lack of MODIS data on cloudy days does not affect the assimilation performance significantly. The combined assimilation of MODIS-like reflectances and a few snow depth measurements throughout the 2010/2011 season further reduces RMSEs by roughly 70 %. This work suggests that the assimilation of optical reflectances has the potential to become an essential component of spatialized snowpack simulation and forecast systems. The assimilation of real MODIS data will be investigated in future works.

Orbital tuning is central for ice core chronologies beyond annual layer counting, available back to 60 ka (i.e. thousands of years before 1950) for Greenland ice cores. While several complementary orbital tuning tools have recently been developed using δ18Oatm, δO2⁄N2 and air content with different orbital targets, quantifying their uncertainties remains a challenge. Indeed, the exact processes linking variations of these parameters, measured in the air trapped in ice, to their orbital targets are not yet fully understood. Here, we provide new series of δO2∕N2 and δ18Oatm data encompassing Marine Isotopic Stage (MIS) 5 (between 100 and 160 ka) and the oldest part (340–800 ka) of the East Antarctic EPICA Dome C (EDC) ice core. For the first time, the measurements over MIS 5 allow an inter-comparison of δO2∕N2 and δ18Oatm records from three East Antarctic ice core sites (EDC, Vostok and Dome F). This comparison highlights some site-specific δO2∕N2 variations. Such an observation, the evidence of a 100 ka periodicity in the δO2∕N2 signal and the difficulty to identify extrema and mid-slopes in δO2∕N2 increase the uncertainty associated with the use of δO2∕N2 as an orbital tuning tool, now calculated to be 3–4 ka. When combining records of δ18Oatm and δO2∕N2 from Vostok and EDC, we find a loss of orbital signature for these two parameters during periods of minimum eccentricity (∼ 400 ka, ∼ 720–800 ka). Our data set reveals a time-varying offset between δO2∕N2 and δ18Oatm records over the last 800 ka that we interpret as variations in the lagged response of δ18Oatm to precession. The largest offsets are identified during Terminations II, MIS 8 and MIS 16, corresponding to periods of destabilization of the Northern polar ice sheets. We therefore suggest that the occurrence of Heinrich–like events influences the response of δ18Oatm to precession.

This study aims to better understand and quantify the uncertainties in microwave snow emission models using the Dense Media Radiative Theory Multi-Layer model (DMRT-ML) with in situ measurements of snow properties. We use surface-based radiometric measurements at 10.67, 19 and 37 GHz in boreal forest and subarctic environments and a new in situ data set of measurements of snow properties (profiles of density, snow grain size and temperature, soil characterization and ice lens detection) acquired in the James Bay and Umiujaq regions of Northern Québec, Canada. A snow excavation experiment – where snow was removed from the ground to measure the microwave emission of bare frozen ground – shows that small-scale spatial variability (less than 1 km) in the emission of frozen soil is small. Hence, in our case of boreal organic soil, variability in the emission of frozen soil has a small effect on snow-covered brightness temperature (TB). Grain size and density measurement errors can explain the errors at 37 GHz, while the sensitivity of TB at 19 GHz to snow increases during the winter because of the snow grain growth that leads to scattering. Furthermore, the inclusion of observed ice lenses in DMRT-ML leads to significant improvements in the simulations at horizontal polarization (H-pol) for the three frequencies (up to 20 K of root mean square error). However, representation of the spatial variability of TB remains poor at 10.67 and 19 GHz at H-pol given the spatial variability of ice lens characteristics and the difficulty in simulating snowpack stratigraphy related to the snow crust. The results also show that, in our study with the given forest characteristics, forest emission reflected by the snow-covered surface can increase the TB up to 40 K. The forest contribution varies with vegetation characteristics and a relationship between the downwelling contribution of vegetation and the proportion of pixels occupied by vegetation (trees) in fisheye pictures was found. We perform a comprehensive analysis of the components that contribute to the snow-covered microwave signal, which will help to develop DMRT-ML and to improve the required field measurements. The analysis shows that a better consideration of ice lenses and snow crusts is essential to improve TB simulations in boreal forest and subarctic environments.

On the Antarctic Plateau, snow specific surface area (SSA) close to the surface shows complex variations at daily to seasonal scales which affect the surface albedo and in turn the surface energy budget of the ice sheet. While snow metamorphism, precipitation and strong wind events are known to drive SSA variations, usually in opposite ways, their relative contributions remain unclear. Here, a comprehensive set of SSA observations at Dome C is analysed with respect to meteorological conditions to assess the respective roles of these factors. The results show an average 2-to-3-fold SSA decrease from October to February in the topmost 10 cm in response to the increase of air temperature and absorption of solar radiation in the snowpack during spring and summer. Surface SSA is also characterized by significant daily to weekly variations due to the deposition of small crystals with SSA up to 100 m2 kg−1 onto the surface during snowfall and blowing snow events. To complement these field observations, the detailed snowpack model Crocus is used to simulate SSA, with the intent to further investigate the previously found correlation between interannual variability of summer SSA decrease and summer precipitation amount. To this end, some Crocus parameterizations have been adapted to Dome C conditions, and the model was forced by ERA-Interim reanalysis. It successfully matches the observations at daily to seasonal timescales, except for the few cases when snowfalls are not captured by the reanalysis. On the contrary, the interannual variability of summer SSA decrease is poorly simulated when compared to 14 years of microwave satellite data sensitive to the near-surface SSA. A simulation with disabled summer precipitation confirms the weak influence in the model of the precipitation on metamorphism, with only 6 % enhancement. However, we found that disabling strong wind events in the model is sufficient to reconciliate the simulations with the observations. This suggests that Crocus reproduces well the contributions of metamorphism and precipitation on surface SSA, but snow compaction by the wind might be overestimated in the model.

The description of snow microstructure in microwave models is often simplified to facilitate electromagnetic calculations. Within dense media radiative transfer (DMRT), the microstructure is commonly described by sticky hard spheres (SHS). An objective mapping of real snow onto SHS is however missing which prevents measured input parameters from being used for DMRT. In contrast, the microwave emission model of layered snowpacks (MEMLS) employs a conceptually different approach, based on the two-point correlation function which is accessible by tomography. Here we show the equivalence of both electromagnetic approaches by reformulating their microstructural models in a common framework. Using analytical results for the two-point correlation function of hard spheres, we show that the scattering coefficient in both models only differs by a factor which is close to unity, weakly dependent on ice volume fraction and independent of other microstructural details. Additionally, our analysis provides an objective retrieval method for the SHS parameters (diameter and stickiness) from tomography images. For a comprehensive data set we demonstrate the variability of stickiness and compare the SHS diameter to the optical equivalent diameter. Our results confirm the necessity of a large grain-size scaling when relating both diameters in the non-sticky case, as previously suggested by several authors.

Moderate-Resolution Imaging spectroradiometer (MODIS) land surface temperatures in Antarctica were processed in order to produce a gridded data set at 25 km resolution, spanning the period 2000–2011 at an hourly time step. The Aqua and Terra orbits and MODIS swath width, combined with frequent clear-sky conditions, lead to very high availability of quality-controlled observations: on average, hourly data are available 14 h per day at the grid points around the South Pole and more than 9 h over a large area of the Antarctic Plateau. Processed MODIS land surface temperatures, referred to hereinafter as MODIS Ts values, were compared with in situ hourly measurements of surface temperature collected over the entirety of the year 2009 by seven stations from the Baseline Surface Radiation Network (BSRN) and automatic weather stations (AWSs). In spite of an occasional failure in the detection of clouds, MODIS Ts values exhibit a good performance, with a bias ranging from −1.8 to 0.1 °C and errors ranging from 2.2 to 4.8 °C root mean square at the five stations located on the plateau. These results show that MODIS Ts values can be used as a precise and accurate reference to test other surface temperature data sets. Here, we evaluate the performance of surface temperature in the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis known as ERA-Interim reanalysis. During conditions detected as cloud free by MODIS, ERA-Interim shows a widespread warm bias in Antarctica in every season, ranging from +3 to +6 °C on the plateau. This confirms a recent study which showed that the largest discrepancies in 2 m air temperature between ERA-Interim and the global temperature data set HadCRUT4 compiled by the Met Office Hadley Centre and the University of East Anglia's Climatic Research Unit occur in Antarctica. A comparison with in situ surface temperature shows that this bias is not strictly limited to clear-sky conditions. A detailed comparison with stand-alone simulations by the Crocus snowpack model, forced by ERA-Interim, and with the ERA-Interim/land simulations, shows that the warm bias may be due primarily to an overestimation of the surface turbulent fluxes in very stable conditions. Numerical experiments with Crocus show that a small change in the parameterization of the effects of stability on the surface exchange coefficients can significantly impact the snow surface temperature. The ERA-Interim warm bias appears to be likely due to an overestimation of the surface exchange coefficients under very stable conditions.

Space-borne passive microwave radiometers are widely used to retrieve information in snowy regions by exploiting the high sensitivity of microwave emission to snow properties. For the Antarctic Plateau, many studies presenting retrieval algorithms or numerical simulations have assumed, explicitly or not, that the subpixel-scale heterogeneity is negligible and that the retrieved properties were representative of whole pixels. In this paper, we investigate the spatial variations of brightness temperature over a range of a few kilometers in the Dome C area. Using ground-based radiometers towed by a vehicle, we collected brightness temperature at 11, 19 and 37 GHz at horizontal and vertical polarizations along transects with meter resolution. The most remarkable observation was a series of regular undulations of the signal with a significant amplitude reaching 10 K at 37 GHz and a quasi-period of 30–50 m. In contrast, the variability at longer length scales seemed to be weak in the investigated area, and the mean brightness temperature was close to SSM/I and WindSat satellite observations for all the frequencies and polarizations. To establish a link between the snow characteristics and the microwave emission undulations, we collected detailed snow grain size and density profiles at two points where opposite extrema of brightness temperature were observed. Numerical simulations with the DMRT-ML microwave emission model revealed that the difference in density in the upper first meter explained most of the brightness temperature variations. In addition, we found that these variations of density near the surface were linked to snow hardness. Patches of hard snow – probably formed by wind compaction – were clearly visible and covered as much as 39% of the investigated area. Their brightness temperature was higher than in normal areas. This result implies that the microwave emission measured by satellites over Dome C is more complex than expected and very likely depends on the year-to-year areal proportion of the two different types of snow.

In the SURFEX/ISBA-Crocus multi-layer snowpack model, the snow microstructure has up to now been characterised by the grain size and by semi-empirical shape variables which cannot be measured easily in the field or linked to other relevant snow properties. In this work we introduce a new formulation of snow metamorphism directly based on equations describing the rate of change of the optical diameter (dopt). This variable is considered here to be equal to the equivalent sphere optical diameter, which is inversely proportional to the specific surface area (SSA). dopt thus represents quantitatively some of the geometric characteristics of a porous medium. Different prognostic rate equations of dopt, including a re-formulation of the original Crocus scheme and the parameterisations from Taillandier et al. (2007) and Flanner and Zender (2006), were evaluated by comparing their predictions to field measurements carried out at Summit Camp (Greenland) in May and June 2011 and at Col de Porte (French Alps) during the 2009/10 and 2011/12 winter seasons. We focused especially on results in terms of SSA. In addition, we tested the impact of the different formulations on the simulated density profile, the total snow height, the snow water equivalent (SWE) and the surface albedo. Results indicate that all formulations perform well, with median values of the RMSD between measured and simulated SSA lower than 10 m2 kg−1. Incorporating the optical diameter as a fully fledged prognostic variable is an important step forward in the quantitative description of the snow microstructure within snowpack models, because it opens the way to data assimilation of various electromagnetic observations.

The energy budget and the photochemistry of a snowpack depend greatly on the penetration of solar radiation in snow. Below the snow surface, spectral irradiance decreases exponentially with depth with a decay constant called the asymptotic flux extinction coefficient. As with the albedo of the snowpack, the asymptotic flux extinction coefficient depends on snow grain shape. While representing snow by a collection of spherical particles has been successful in the numerical computation of albedo, such a description poorly explains the decrease of irradiance in snow with depth. Here we explore the limits of the spherical representation. Under the assumption of geometric optics and weak absorption by snow, the grain shape can be simply described by two parameters: the absorption enhancement parameter B and the geometric asymmetry factor gG. Theoretical calculations show that the albedo depends on the ratio B/(1-gG) and the asymptotic flux extinction coefficient depends on the product B(1-gG). To understand the influence of grain shape, the values of B and gG are calculated for a variety of simple geometric shapes using ray tracing simulations. The results show that B and (1-gG) generally covary so that the asymptotic flux extinction coefficient exhibits larger sensitivity to the grain shape than albedo. In particular it is found that spherical grains propagate light deeper than any other investigated shape. In a second step, we developed a method to estimate B from optical measurements in snow. A multi-layer, two-stream, radiative transfer model, with explicit grain shape dependence, is used to retrieve values of the B parameter of snow by comparing the model to joint measurements of reflectance and irradiance profiles. Such measurements were performed in Antarctica and in the Alps yielding estimates of B between 0.8 and 2.0. In addition, values of B were estimated from various measurements found in the literature, leading to a wider range of values (1.0–9.9) which may be partially explained by the limited accuracy of the data. This work highlights the large variety of snow microstructure and experimentally demonstrates that spherical grains, with B = 1.25, are inappropriate to model irradiance profiles in snow, an important result that should be considered in further studies dedicated to subsurface absorption of short-wave radiation and snow photochemistry.

Hoar crystals episodically cover the snow surface in Antarctica and affect the roughness and reflective properties of the air–snow interface. However, little is known about their evolution and the processes responsible for their development and disappearance despite a probable influence on the surface mass balance and energy budget. To investigate hoar evolution, we use continuous observations of the surface by in situ near-infrared photography and by passive microwave remote sensing at Dome C in Antarctica. From the photography data, we retrieved a daily indicator of the presence/absence of hoar crystals using a texture analysis algorithm. The analysis of this 2 yr long time series shows that Dome C surface is covered almost half of the time by hoar. The development of hoar crystals takes a few days and seems to occur whatever the meteorological conditions. In contrast, the disappearance of hoar is rapid (a few hours) and coincident with either strong winds or with moderate winds associated with a change in wind direction from southwest (the prevailing direction) to southeast. From the microwave satellite data, we computed the polarisation ratio (i.e. horizontal over vertical polarised brightness temperatures), an indicator known to be sensitive to hoar in Greenland. Photography data and microwave polarisation ratio are correlated, i.e. high values of polarisation ratio which theoretically correspond to low snow density values near the surface are associated with the presence of hoar crystals in the photography data. Satellite data over nearly ten years (2002–2011) confirm that a strong decrease of the polarisation ratio (i.e. signature of hoar disappearance) is associated with an increase of wind speed or a change in wind direction from the prevailing direction. The photography data provides, in addition, evidence of interactions between hoar and snowfall. Further adding the combined influence of wind speed and wind direction results in a complex picture of the snow–atmosphere interactions in Antarctica which deserves further quantification and modelling.

DMRT-ML is a physically based numerical model designed to compute the thermal microwave emission of a given snowpack. Its main application is the simulation of brightness temperatures at frequencies in the range 1–200 GHz similar to those acquired routinely by space-based microwave radiometers. The model is based on the Dense Media Radiative Transfer (DMRT) theory for the computation of the snow scattering and extinction coefficients and on the Discrete Ordinate Method (DISORT) to numerically solve the radiative transfer equation. The snowpack is modeled as a stack of multiple horizontal snow layers and an optional underlying interface representing the soil or the bottom ice. The model handles both dry and wet snow conditions. Such a general design allows the model to account for a wide range of snow conditions. Hitherto, the model has been used to simulate the thermal emission of the deep firn on ice sheets, shallow snowpacks overlying soil in Arctic and Alpine regions, and overlying ice on the large ice-sheet margins and glaciers. DMRT-ML has thus been validated in three very different conditions: Antarctica, Barnes Ice Cap (Canada) and Canadian tundra. It has been recently used in conjunction with inverse methods to retrieve snow grain size from remote sensing data. The model is written in Fortran90 and available to the snow remote sensing community as an open-source software. A convenient user interface is provided in Python.