Two supercell thunderstorms that occurred in Del Rio and Burkburnett, Texas in the spring of 2020 produced greater than 5 cm diameter hailstones. Fourteen hailstones, eight from Del Rio and six from Burkburnett, were cut in half along the maximum principal axis. Half the hailstone was dissected at 0.5 cm increments, providing 198 subsamples of meltwater (0.25 ml) for stable isotope analysis (delta H-2 and delta O-18) using a Picarro L2130-i water analyzer and half used to create a thin section for identification and analysis of ice crystal morphology. Deuterium serves as a proxy for relative altitude, providing a record of hailstone formation and growth trajectory, while crystallographic analysis differentiates wet and dry growth. Results indicate significant variations in isotopic composition and ice crystal morphology, reflecting distinct growth trajectories in the supercell environment, providing evidence that hailstone formation and growth occurred at varying altitudes and reflect recycling in the updraft. The formation of a hailstone embryo at higher altitudes challenges the paradigm of embryo formation only at the cloud base or lower altitudes compared to the outer growth layers. Dendritic crystal morphologies observed in Del Rio hailstones are consistent with rapid freezing in the presence of elevated supercooled liquid water content. Mass ratio analysis provided that most of the hailstones approached the density of bubble free ice while the sphericity index reflected that most hailstones approached 0.80 sphericity but can more accurately be described as triaxial ellipsoids. The goal of this research was to provide better information on hailstone growth and trajectory through supercell convective thunderstorms.

Landfast ice, sea ice fastened to coastal land or ice shelves, generally undergoes distinctive thermodynamic growth and less active dynamic deformation due to its prolonged attachment to the land, resulting in a thicker and smoother surface compared to drifting pack ice. In 2019, large landfast ice floes were detached from the Ronne Ice Shelf, and the broken pieces started to drift into the Weddell Sea. This study employs a random forest (RF) machine learning model to detect these ex-fast ice floes using six key variables from the ICESat-2 ATL10 sea ice freeboard product: freeboard, Gaussian width of photon height distribution, standard deviation of freeboard, floe length, modal freeboard, and sea ice concentration. The RF model achieves an overall accuracy of 99 % in detecting ex-fast ice, effectively capturing the drift, freeboard distribution, and size distribution of ex-fast ice floes across the western Weddell Sea in 2019. Among six variables, freeboard, standard deviation of freeboard, and Gaussian width of photon height distribution contribute over 94 % to the model performance. Furthermore, the detection of ex-fast ice improves the quantification of sea ice topographical features derived from ICESat-2, including modal freeboard, ridge fraction, and surface roughness. This study highlights the effectiveness of discriminating heterogeneous ex-fast ice from typical pack ice to enhance sea ice measurements using ICESat-2 satellite altimeter data.

The sea ice extent in the Weddell Sea exhibited a positive trend from the start of satellite observations in 1978 until 2016 but has shown a decreasing trend since then. This study analyzes seasonal and interannual variations in sea ice thickness using ICESat-2 laser altimetry data over the Weddell Sea from 2019 to 2022. Sea ice thickness was calculated from ICESat-2’s ATL10 freeboard product using the Improved Buoyancy Equation. Seasonal variability in ice thickness, characterized by an increase from February to September, is more pronounced in the eastern Weddell sector, while interannual variability is more evident in the western Weddell sector. The results were compared with field data obtained between 2019 and 2022, showing a general agreement in ice thickness distributions around predominantly level ice. A decreasing trend in sea ice thickness was observed when compared to measurements from 2003 to 2017. Notably, the spring of 2021 and summer of 2022 saw significant decreases in Sea Ice Extent (SIE). Although the overall mean sea ice thickness remained unchanged, the northwestern Weddell region experienced a noticeable decrease in ice thickness.

During the summer months in the Antarctic, perennial and seasonal sea ice floes flood. Flooding is caused by snow at the surface weighing down the ice, causing a negative freeboard and flooding the basal snow layer with seawater. This creates a brine-slush layer. Alternatively, or simultaneously, meltwater can percolate through the snow and flood the surface of the ice floe. The appearance of these flooded ice floes changes dramatically in synthetic aperture radar (SAR) scenes with season and as the dielectric constant changes with brine content. In addition to this, the incident look angle of the radar imager affects the returned backscatter intensity across the scene. The Sentinel-1 instrument began collecting data with its S1A instrument in 2014 and later S1B in 2016 and continues to acquire SAR data across the globe. Sentinel-1 supplies an unprecedented, dual-band look at sea ice in the North and South poles to understand the dynamics of sea ice processes during polar nighttime. The satellite instrument provides a unique opportunity to study the signal attenuations and the subsequent backscatter intensities in the SAR scene that change with seasonal ice flooding. This paper uses the Sentinel-1 radar data to understand the changes in backscatter intensity in flooded floes in the Amundsen, whose changes in floe flooding show spatial and spectral changes throughout the seasons.

Atmospheric circulation has significant impacts on sea ice drifting patterns and mass balance, as wind drag induces pressure ridges and leads on the sea ice surface. In this study, the spatiotemporal distributions of these dynamic sea ice deformation features in the Ross Sea are examined using ICESat-2 (IS2) ATL10 freeboard data (2019-2022). The temporal variation of the modal sea ice thickness (SIT), caused by thermodynamic ice growth and sea ice advection, varies from 0.7-1.0 m in April to 1.0-1.6 m in July-September and decreases thereafter in the northwest (NW) and northeast (NE) sectors. This temporal variation of modal SIT agrees with the air temperature (correlation coefficients >0.5). The southwest (SW) sector shows a consistently low modal SIT (<1.0 m) because of the production of new ice in polynyas and continuous northward sea ice drift. Meanwhile, the southeast (SE) sector shows the thickest ice in Octobers 2019 and 2020 because of the advection of thick ice from the Amundsen Sea, which was reduced in 2021 and 2022. In terms of dynamic sea ice deformation, the SE sector shows the largest deformation because of the wind-driven convergence of sea ice movement. However, such intense deformation in the SE sector diminished in 2021 and 2022 due to the dominance of strong southerly wind associated with the Amundsen Sea Low (ASL). This study emphasizes the potential of IS2 sea ice products to assess the role of atmospheric driving forces on thermodynamic and dynamic sea ice changes.

An airborne microwave wide-band radiometer (500–2000 MHz) was operated for the first time in Antarctica to better understand the emission properties of sea ice, outlet glaciers and the interior ice sheet from Terra Nova Bay to Dome C. The different glaciological regimes were revealed to exhibit unique spectral signatures in this portion of the microwave spectrum. Generally, the brightness temperatures over a vertically homogeneous ice sheet are warmest at the lowest frequencies, consistent with models that predict that those channels sensed the deeper, warmer parts of the ice sheet. Vertical heterogeneities in the ice property profiles can alter this basic interpretation of the signal. Spectra along the lengths of outlet glaciers were modulated by the deposition and erosion of snow, driven by strong katabatic winds. Similar to previous experiments in Greenland, the brightness temperatures across the frequency band were low in crevasse areas. Variations in brightness temperature were consistent with spatial changes in sea ice type identified in satellite imagery and in situ ground-penetrating radar data. The results contribute to a better understanding of the utility of microwave wide-band radiometry for cryospheric studies and also advance knowledge of the important physics underlying existing L-band radiometers operating in space.

The ICESat-2 (IS2) ATL07 sea ice height product provides precise and accurate measurements of sea ice height for the polar regions. Although the original ATL07 product classifies the sea ice surface into snow-covered ice and open water (dark leads and specular leads), it has two critical issues: (1) it does not distinguish thin ice (gray ice) from thick/snow-covered ice or open water and (2) it involves uncertainties in the dark lead determination. To address these issues and obtain more accurate lead fraction and freeboard estimation, in this study, we assess a data-driven machine learning approach for sea ice surface type classification from the ATL07 product. A total of 17 IS2 tracks covering the Ross Sea in February, March, September, October, and November 2019 are used in this study. For training and testing the neural network (NN) models, we label the ATL07 data into three surface types: thick/snow-covered sea ice, thin/gray sea ice, and open water based on the coincident Sentinel-2 optical images. We use six variables from the ATL07 data: background rate, photon rate, relative surface height, width of height distribution, number of laser pulses, and difference in mean and median height, with the first three variables found to be the most significant in determining surface types. Our spatiotemporal test shows that the NN models can be applied to the ATL07 dataset with similar to 99% accuracy in the Ross Sea, and the beam test shows that the difference in the accuracy for three beams is negligible. While the current ATL07 product captures only similar to 5% of thin ice and similar to 81% of open water correctly, our NN models capture similar to 90% of both open water and thin ice leads successfully. When we qualitatively assess the surface classification of the NN models by using the coincident optical and radar images, our NN models do not show significant misclassification issues both in the daytime and nighttime and both in the winter and summer seasons. Compared to the IS2 ATL10 sea ice freeboard product, the ability of our NN model to separate thin ice and open water from snow-covered/thick ice is significant in freeboard estimation, especially for winter and highly packed sea ice areas. Open water and thin ice classes also improve mapping lead fraction and floe size distribution.

Seasonal formation of Dense Shelf Water (DSW) in the Ross Sea is a direct precursor to Antarctic Bottom Water, which fills the deep ocean with atmospheric gases in what composes the southern limb of the solubility pump. Measurements of seawater noble gas concentrations during katabatic wind events in two Ross Sea polynyas reveal the physical processes that determine the boundary value properties for DSW. This decomposition reveals 5-6 g kg(-1) of glacial meltwater in DSW and sea-ice production rates of up to 14 m yr(-1) within the Terra Nova Bay polynya. Despite winds upwards of 35 m s(-1 )during the observations, air bubble injection had a minimal contribution to gas exchange, accounting for less than 0.01 mu mols kg(-1 )of argon in seawater. This suggests the slurry of frazil ice and seawater at the polynya surface inhibits air-sea exchange. Most noteworthy is the revelation that sea-ice formation and glacial melt contribute significantly to the ventilation of DSW, restoring 10% of the gas deficit for krypton, 24% for argon, and 131% for neon, while diffusive gas exchange contributes the remainder. These measurements reveal a cryogenic component to the solubility pump and demonstrate that while sea ice blocks air-sea exchange, sea ice formation and glacial melt partially offset this effect via addition of gases. While polynyas are a small surface area, they represent an important ventilation site within the southern-overturning cell, suggesting that ice processes both enhance and hinder the solubility pump.

Limitations of access have long restricted exploration and investigation of the cavities beneath ice shelves to a small number of drillholes. Studies of sea-ice underwater morphology are limited largely to scientific utilization of submarines. Remotely operated vehicles, tethered to a mother ship by umbilical cable, have been deployed to investigate tidewater-glacier and ice-shelf margins, but their range is often restricted. The development of free-flying autonomous underwater vehicles (AUVs) with ranges of tens to hundreds of kilometres enables extensive missions to take place beneath sea ice and floating ice shelves. Autosub2 is a 3600 kg, 6.7m long AUV, with a 1600m operating depth and range of 400 km, based on the earlier Autosub1 which had a 500m depth limit. A single direct-drive d.c. motor and five-bladed propeller produce speeds of 1–2m s. Rear-mounted rudder and stern-plane control yaw, pitch and depth. The vehicle has three sections. The front and rear sections are freeflooding, built around aluminium extrusion space-frames covered with glass-fibre reinforced plastic panels. The central section has a set of carbon-fibre reinforced plastic pressure vessels. Four tubes contain batteries powering the vehicle. The other three house vehicle-control systems and sensors. The rear section houses subsystems for navigation, control actuation and propulsion and scientific sensors (e.g. digital camera, upward-looking 300 kHz acoustic Doppler current profiler, 200 kHz multibeam receiver). The front section contains forward-looking collision sensor, emergency abort, the homing systems, Argos satellite data and location transmitters and flashing lights for relocation as well as science sensors (e.g. twin conductivity–temperature–depth instruments, multibeam transmitter, subbottom profiler, AquaLab water sampler). Payload restrictions mean that a subset of scientific instruments is actually in place on any given dive. The scientific instruments carried on Autosub are described and examples of observational data collected from each sensor in Arctic or Antarctic waters are given (e.g. of roughness at the underside of floating ice shelves and sea ice).

Polynyas are key sites of ice production during the winter and are important sites of biological activity and carbon sequestration during the summer. The Amundsen Sea Polynya (ASP) is the fourth largest Antarctic polynya, has recorded the highest primary productivity, and lies in an embayment of key oceanographic significance. However, knowledge of its dynamics, and of sub-annual variations in its area and ice production, is limited. In this study we primarily utilize Sentinel-1 synthetic aperture radar (SAR) imagery, sea ice concentration products, and climate reanalysis data, along with bathymetric data, to analyze the ASP over the period November 2016–March 2021. Specifically, we analyze (i) qualitative changes in the ASP's characteristics and dynamics, as well as quantitative changes in (ii) summer polynya area, and (iii) winter polynya area and ice production. From our analysis of SAR imagery we find that ice produced by the ASP becomes stuck in the vicinity of the polynya and sometimes flows back into the polynya, contributing to its closure and limiting further ice production. The polynya forms westward off a persistent chain of grounded icebergs that are located at the site of a bathymetric high. Grounded icebergs also influence the outflow of ice and facilitate the formation of a “secondary polynya” at times. Additionally, unlike some polynyas, ice produced by the polynya flows westward after formation, along the coast and into the neighboring sea sector. During the summer and early winter, broader regional sea ice conditions can play an important role in the polynya. The polynya opens in all summers, but record-low sea ice conditions in 2016/17 cause it to become part of the open ocean. During the winter, an average of 78 % of ice production occurs in April–May and September–October, but large polynya events often associated with high, southeasterly or easterly winds can cause ice production throughout the winter. While passive microwave data or daily sea ice concentration products remain key for analyzing variations in polynya area and ice production, we find that the ability to directly observe and qualitatively analyze the polynya at a high temporal and spatial resolution with Sentinel-1 imagery provides important insights about the behavior of the polynya that are not possible with those datasets.

Monitoring Antarctic icebergs helps us understand the interaction between ocean, atmosphere, and sea ice in the Southern Ocean. Although gigantic icebergs have been the objects of many remote sensing studies, medium icebergs in the Southern Ocean have been rarely monitored or traced. In this study, we develop an iceberg detection and tracking tool particularly for medium and large icebergs (0.4-10 km2), by using Python programming language and Sentinel-1 (S1) imagery, based on Google Earth Engine (GEE). To detect icebergs, we employ the simple non-iterative clustering (SNIC) and region adjacency graph (RAG) merging for object-based image segmentation and train/test the support vector machine (SVM) model with 6432 labeled segments of iceberg or non-icebergs from 40 images of S1 (2019-2021). Radar backscatter features and morphological features of those segments are used as the inputs of the SVM model. After icebergs are detected in two image scenes of different dates, we track the displacements of detected icebergs by comparing their 1-D shape signals. Our model shows -99% of accuracy in detecting icebergs and - 93-98% of accuracy in tracking icebergs depending on the day difference between image scenes. When using our tool for the monitoring of icebergs in the Amundsen Sea, we find that the iceberg fraction varies from 2% to 8% in 2021 and most of icebergs move westward with a speed of <0.2 km/day.

Snow cover is a primary control on Antarctic sea ice mass balance as it controls basal ice growth and snow ice formation. It is also a primary control on the surface energy budget, partitioning of solar radiation, and sea ice biological communities. Finally, knowledge of its distribution is critical for accurate estimation of sea ice thickness from satellite altimeters. The floe-scale distribution of snow is highly variable, driven by wind redistribution over complex sea ice surface topography. Yet, our understanding of the seasonal evolution of snow depth distribution is poor and its representation in models is simple or non-existent. We present observations of the three-dimensional distribution of snow depth, ice thickness, and surface topography from a suite of cruises in the Weddell, Bellingshausen, Ross, and East Antarctic Seas that span the full growth season – from autumn, through winter, to late spring. The distribution of snow depth changes from a right-skewed distribution in autumn as snow initially accumulates around ridges to a gaussian by spring as snow deepens and ice surface topography roughens. While the distribution is spatially complex, the spectral distribution of snow features is similar across seasons. Using these data we construct a simple statistical model for the seasonal evolution of floe-scale snow depth distribution. We also compare our results to prior observations from drilling transects and larger-scale airborne observations from NASA’s Operation IceBridge. For the latter we use a convolutional neural network to demonstrate that the surface topography can be used as a reliable predictor of the snow depth distribution at regional scales.

AbstractOver the Ross Sea shelf, annual primary production is limited by dissolved iron (DFe) supply. Here, a major source of DFe to surface waters is thought to be vertical resupply from the benthos, which is assumed most prevalent during winter months when katabatic winds drive sea ice formation and convective overturn in coastal polynyas, although the impact of these processes on water‐column DFe distributions has not been previously documented. We collected hydrographic data and water‐column samples for trace metals analysis in the Terra Nova Bay and Ross Ice Shelf polynyas during April–May 2017 (late austral fall). In the Terra Nova Bay polynya, we observed intense katabatic wind events, and surface mixed layer depths varied from ∼250 to ∼600 m over lateral distances <10 km; there vertical mixing was just starting to excavate the dense, iron‐rich Shelf Waters, and there was also evidence of DFe inputs at shallower depths in the water column. In the Ross Ice Shelf polynya, wind speeds were lower, mixed layers were <300 m deep, and DFe distributions were similar to previous, late‐summer observations, with concentrations elevated near the seafloor. Corresponding measurements of dissolved manganese and zinc, and particulate iron, manganese, and aluminum, suggest that deep DFe maxima and some mid‐depth DFe maxima primarily reflect sedimentary inputs, rather than remineralization. Our data and model simulations imply that vertical resupply of DFe in the Ross Sea occurs mainly during mid‐late winter, and may be particularly sensitive to changes in the timing and extent of sea ice production.

The increasing concern regarding climate change continues to motivate research in the Arctic. Within the delicate cycle of the cryosphere, leads are an important kinematic feature that regulates heat balances and gaseous exchanges in the Arctic. Therefore, it is necessary to quantify the genesis of leads over time to identify when and where changes are occurring. The use of learning techniques is one such tool that is used to identify the characteristics of seasonal ice variability. This paper utilizes the Sentinel-1 RADAR imagery with a support vector machine learning scheme through Google Earth Engine to classify lead types in the Beaufort Sea.

The fine spatial resolution of the ICESat-2 (IS2) satellite altimeter allows monitoring the evolution of sea ice thickness with detailed dynamic information (e.g. ridges and leads). In this study, we first assess the ability of IS2 to estimate thermodynamic ice growth and dynamic thickening during the ice-growing season in the central Arctic Ocean. As an indicator of the thermodynamic ice growth, we use 10 thermistor string-based sea ice mass balance array (SIMBA) buoys deployed at a scale of -50 km from the Icebreaker Polarstern during the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition. We collect IS2 data within 20 km buffer distance from the individual buoys, and calculate the mode, median, and mean of the IS2-derived ice thickness. The IS2 modal thickness shows the least bias (-0.169 m) with the buoy ice thickness, representing level ice thickness. In addition, the increasing rate of the IS2 modal thickness is close to the thermodynamic ice growth with a small bias of -0.054 cm/day. However, the increasing rates of the IS2 median and mean thickness are greater than the thermodynamic ice growth by about 0.114 cm/day and 0.198 cm/day, respectively, because they also include ice growth caused by thickness redistribution during dynamic deformation. The dynamic contributions may account for 26.1 +/- 10.3% and 34.4 +/- 10.1% of the total increase of the IS2 median and mean thickness, respectively. Within a - 50 km radius area from the MOSAiC Central Observatory, IS2 measurements exhibit that the ridge fraction increased from <2% in November to -4% in March (-0.029%/day of average increasing rate) and ridge height increased about 0.047 cm/day during the same period. However, lead formation does not show significant contributions to the dynamic ice thickening because leads are temporary features lasting only 2-3 days. Although there are considerable uncertainties in IS2 ice thickness estimation and IS2-buoy thickness comparison, this study emphasizes the importance of combining measurements by IS2 and SIMBA buoys to explain the regional sea ice mass balance with separating the thermodynamic and dynamic contributions.