Search

Spaceborne radar sounders are high frequency (HF)/very high frequency (VHF) nadir-looking sensors devoted to subsurface investigations. Their data interpretation can be severely hindered by off-nadir surface clutter. Recent literature showed that the clutter suppression capabilities of this class of systems can be greatly enhanced by deploying an array of orbiting sensors in formation flight synthesizing a narrow radar antenna beam. In this article, we assess the capability of distributed radar sounding to discriminate clutter from subsurface returns by exploiting direction of arrival (DOA) estimation techniques. This is achieved by first outlining an approach for designing and evaluating the distributed radar sounder DOA estimation performance as function of the radar system parameters (e.g., intersensor distance) and external noise factors such as ionospheric scintillations. Then, the theory is complemented by radar simulations of several acquisitions over Greenland assuming a variety of subsurface geometries. The simulations confirm that clutter discrimination through DOA estimation is a viable approach to further improve the array capability in disambiguation of subsurface echoes from surface ones. [ABSTRACT FROM AUTHOR]

During the past 20 years, multi-channel radar emerged as a key tool for deciphering an ice sheet's internal architecture. To assign ages to radar reflections and connect them over large areas in the ice sheet, the layer genesis has to be understood on a microphysical scale. Synthetic radar trace modelling based on the dielectric profile of ice cores allows for the assignation of observed physical properties' variations on the decimetre scale to radar reflectors extending from the coring site to a regional or even whole-ice-sheet scale. In this paper we rely on the available dielectric profiling data of the northern Greenland deep ice cores: NGRIP, NEEM and EGRIP. The three records are well suited for assigning an age model to the stratigraphic radar-mapped layers, and linking up the reflector properties to observations in the cores. Our modelling results show that the internal reflections are mainly due to conductivity changes. Furthermore, we deduce fabric characteristics at the EGRIP drill site from two-way-travel-time differences of along and across-flow polarized radarwave reflections of selected horizons (below 980 m). These indicate in deeper parts of the ice column an across-flow concentrated c-axis fabric. [ABSTRACT FROM AUTHOR]

The earliest airborne geophysical campaigns over Antarctica and Greenland in the 1960s and 1970s collected ice penetrating radar data on 35 mm optical film. Early subglacial topographic and englacial stratigraphic analyses of these data were foundational to the field of radioglaciology. Recent efforts to digitize and release these data have resulted in geometric and ice-thickness analysis that constrain subsurface change over multiple decades but stop short of radiometric interpretation. The primary challenge for radiometric analysis is the poorly-characterized compression applied to Z-scope records and the sparse sampling of A-scope records. Here, we demonstrate the information richness and radiometric interpretability of Z-scope records. Z-scope pixels have uncalibrated fast-time, slow-time, and intensity scales. We develop approaches for mapping each of these scales to physical units (microseconds, seconds, and signal to noise ratio). We then demonstrate the application of this calibration and analysis approach to a flight in the interior of East Antarctica with subglacial lakes and to a reflight of an East Antarctic ice shelf that was observed by both archival and modern radar. These results demonstrate the potential use of Z-scope signals to extend the baseline of radiometric observations of the subsurface by decades. [ABSTRACT FROM AUTHOR]

An intrinsic challenge in the geophysical interpretation of low-frequency synthetic aperture radar (SAR) imagery of semitransparent media, such as ice sheets, is the position ambiguity of the scattering structures within the glacial volume. Commonly tackled by applying interferometric and tomographic techniques, their spaceborne implementation exhibits by orders higher complexity compared to missions relying on single SAR images, making them cost expensive or, in the context of planetary missions, even impossible due to limited navigation capability. Besides, even these sophisticated techniques are commonly biased due to inaccurate permittivity estimates, leading to geometric distortions up to several meters. We present a novel inversion procedure to estimate volume parameters of ice sheets, namely, the depth of the scattering layer within the glacial volume and the dielectric permittivity of the ice, based on single-image single-polarization SAR acquisitions. The information is inherent in the processed SAR data as phase errors on the azimuth signals resulting from uncompensated nonlinear propagation of the radar echoes through ice. We suggest a local map-drift autofocus approach to quantify and spatially resolve the phase errors and an inversion model to relate them to the penetration depth and permittivity. Testing the proposed technique using P-band SAR data acquired using DLR’s airborne sensor F-SAR during the ARCTIC15 campaign in Greenland shows promising results and good agreement with tomographic products of the analyzed test site. [ABSTRACT FROM AUTHOR]

Glacier calving is a key dynamical process of the Greenland Ice Sheet and a major driver of its increasing mass loss. Calving waves, generated by the sudden detachment of ice from the glacier terminus, can reach tens of meters in height and provide very valuable insights into quantifying calving activity. In this study, we present a new method for the detection of source location, timing, and magnitude of calving waves using a terrestrial radar interferometer. This method was applied to 11 500 1 min interval acquisitions from Eqip Sermia, West Greenland, in July 2018. Over 7 d, more than 2000 calving waves were detected, including waves generated by submarine calving, which are difficult to observe with other methods. Quantitative assessment with a wave power index (WPI) yields a higher wave activity (+49 %) and higher temporally cumulated WPI (+34 %) in deep water than under shallow conditions. Subglacial meltwater plumes, occurring 2.3 times more often in the deep sector, increase WPI and the number of waves by a factor of 1.8 and 1.3, respectively, in the deep and shallow sector. We therefore explain the higher calving activity in the deep sector by a combination of more frequent meltwater plumes and more efficient calving enhancement linked with better connections to warm deep ocean water. [ABSTRACT FROM AUTHOR]

We present continuous estimates of snow and firn density, layer depth and accumulation from a multi-channel, multi-offset, ground-penetrating radar traverse. Our method uses the electromagnetic velocity, estimated from waveform travel-times measured at common-midpoints between sources and receivers. Previously, common-midpoint radar experiments on ice sheets have been limited to point observations. We completed radar velocity analysis in the upper ~2 m to estimate the surface and average snow density of the Greenland Ice Sheet. We parameterized the Herron and Langway (1980) firn density and age model using the radar-derived snow density, radar-derived surface mass balance (2015–2017) and reanalysis-derived temperature data. We applied structure-oriented filtering to the radar image along constant age horizons and increased the depth at which horizons could be reliably interpreted. We reconstructed the historical instantaneous surface mass balance, which we averaged into annual and multidecadal products along a 78 km traverse for the period 1984–2017. We found good agreement between our physically constrained parameterization and a firn core collected from the dry snow accumulation zone, and gained insights into the spatial correlation of surface snow density. [ABSTRACT FROM AUTHOR]

Perennial water storage in firn aquifers has been observed within the lower percolation zone of the southeast Greenland ice sheet. Spatially distributed seismic and radar observations, made ~50 km upstream of the Helheim Glacier terminus, reveal spatial variations of seismic velocity within a firn aquifer. From 1.65 to 1.8 km elevation, shear‐wave velocity (Vs) is 1,290 ± 180 m/s in the unsaturated firn, decreasing below the water table (~15 m depth) to 1,130 ± 250 m/s. Below 1.65 km elevation, Vs in the saturated firn is 1,270 ± 220 m/s. The compressional‐to‐shear velocity ratio decreases in the downstream saturated zone, from 2.30 ± 0.54 to 2.01 ± 0.46, closer to its value for pure ice (2.00). Consistent with colocated firn cores, these results imply an increasing concentration of ice in the downstream sites, reducing the porosity and storage potential of the firn likely caused by episodic melt and freeze during the evolution of the aquifer. Plain Language Summary: An integrated geophysical analysis of seismic, radar, and borehole measurements has been completed over a firn aquifer in southeast Greenland. We show the stiffness of the aquifer increases at lower elevations, closer to sea level, which leads to a decrease in pore space for the meltwater to be stored. This corresponds to an increase in ice content within the firn at lower elevations, as observed in borehole measurements, and likely caused by the meltwater refreezing within and below the aquifer. Key Points: Characterizing the storage potential of firn aquifers is important for predicting water budgets in a warming worldMultiple geophysical methods are combined to evaluate the elastic properties and porosity of the firn aquiferSeismic velocities detect an increase in ice content in the aquifer downslope, reducing the porosity and storage potential of the firn [ABSTRACT FROM AUTHOR]

Greenland Ice Sheet surface melting has increased since the 1990s, affecting the rheology and scattering properties of the near‐surface firn. We combine firn cores and modeled firn densities with 7 years of CryoVEx airborne Ku‐band (13.5 GHz) radar profiles to quantify the impact of melting on microwave radar penetration in West Central Greenland. Although annual layers are present in the Ku‐band radar profiles to depths up to 15 m below the ice sheet surface, fluctuations in summer melting strongly affect the degree of radar penetration. The extreme melting in 2012, for example, caused an abrupt 6.2 ± 2.4 m decrease in Ku‐band radar penetration. Nevertheless, retracking the radar echoes mitigates this effect, producing surface heights that agree to within 13.9 cm of coincident airborne laser measurements. We also examine 2 years of Ka‐band (34.5 GHz) airborne radar data and show that the degree of penetration is half that of coincident Ku‐band. Plain Language Summary: Radar waves emitted by satellites can be used to measure changes in surface elevation of the Greenland Ice Sheet. However, they do not reflect off the ice sheet surface itself but penetrate into the snow to a depth of about 15 m for radar wavelengths of 2.3 cm. When the snow melts, meltwater can percolate into the snow or refreeze at the surface. Layers of refrozen ice sharply reduce the degree of radar penetration and may be mistaken for an elevation increase in radar measurements. Here, we combine firn cores and modeled firn densities with 7 years of airborne radar data collected during field campaigns in West Central Greenland to quantify this effect. We identify internal layers corresponding to annual stratigraphy within the snowpack, and we show that more melt means less radar penetration into the firn. The unprecedented surface melting which occurred across Greenland in 2012 caused a sharp reduction in the degree of radar penetration, from 11.5 to 5.3 m. However, if the effects of penetration are corrected for, radar altimeters can accurately measure the surface elevation of the ice sheet. Key Points: We use airborne radar, firn cores, and firn models to confirm that summer melting is the principal source of radar penetration fluctuationsThe largest fluctuations are recorded after the extreme melt event of 2012, which caused a 6.2 ± 2.4 m reduction in radar penetrationEcho retracking compensates for radar penetration fluctuations, leading to surface heights that agree with laser data to within 13.9 cm [ABSTRACT FROM AUTHOR]

Calving is a crucial process for the recently observed dynamic mass loss changes of the Greenland ice sheet. Despite its importance for global sea level change, major limitations in understanding the process of calving remain. This study presents high-resolution calving event data and statistics recorded with a terrestrial radar interferometer at the front of Eqip Sermia, a marine-terminating outlet glacier in Greenland. The derived digital elevation models with a spatial resolution of several metres recorded at 1 min intervals were processed to provide source areas and volumes of 906 individual calving events during a 6 d period. The calving front can be divided into sectors ending in shallow and deep water with different calving statistics and styles. For the shallow sector, characterized by an inclined and very high front, calving events are more frequent and larger than for the vertical ice cliff of the deep sector. We suggest that the calving volume deficiency of 90 % relative to the estimated ice flux in our observations of the deep sector is removed by oceanic melt, subaquatic calving, and small aerial calving events. Assuming a similar ice thickness for both sectors implies that subaqueous mass loss must be substantial for this sector with a contribution of up to 65 % to the frontal mass loss. The size distribution of the shallow sector is represented by a log-normal model, while for the deep sector the log-normal and power-law model fit well, but none of them are significantly better. Variations in calving activity and style between the sectors seem to be controlled by the bed topography and the front geometry. Within the short observation period no simple relationship between environmental forcings and calving frequency or event volume could be detected. [ABSTRACT FROM AUTHOR]

Englacial layers in Antarctica and Greenland are indicators of the dynamic, rheological and subglacial configuration of the ice sheets. Airborne radar sounder data is the primary remote sensing solution for directly observing englacial layers and structures at the glacier-catchment to ice-sheet scale. However, when traditional along-track synthetic aperture radar (SAR) processing is applied, steep layers can disappear, limiting the detectability and interpretability of englacial layer geometry. This study provides a reconstruction algorithm to address the problem of destructive phase interference during the radargram formation. We develop and apply a novel SAR processor optimized for layer detection that enhances the Signal-to-Noise ratio (SNR) of specular reflectors. The algorithm also enables the automatic estimation of layer slope. We demonstrate the algorithm using data acquired at the Institute Ice Stream, West Antarctica. [ABSTRACT FROM AUTHOR]

Ice streams are bounded by abrupt transitions in speed called shear margins. Some shear margins are fixed by subglacial topography, but others are thought to be self‐organizing, evolving by thermal feedback to ice viscosity and basal drag which govern the stress balance of ice sheets. Resistive stresses (and properties governing shear‐margin formation) manifest nonuniquely at the surface, motivating the use of subsurface observations to constrain ice sheet models. In this study, we use ice‐penetrating radar data to evaluate three 3‐D thermomechanical models of the Northeast Greenland Ice Stream, focusing on model reproductions of ice temperature (a primary control on viscosity) and subsurface velocity. Data/model agreement indicates elevated temperatures in the Northeast Greenland Ice Stream margins, with depth‐averaged temperatures between 2 °C and 6 °C warmer in the southeast margin compared to ice in streaming flow, driven by vertical heat transport rather than shear heating. This work highlights complexity in ice divergence across stagnant/streaming transitions. Plain Language Summary: Ice sheet models used to project future sea level rise are calibrated using modern observations of ice flow at the ice sheet surface. However, the subsurface ice and rock properties that ultimately control the patterns of ice flow cannot be uniquely determined using observations of the surface alone. In this study, we use the structural and electromagnetic characteristics of the Greenland Ice Sheet (determined from ice‐penetrating radar data) to evaluate the subsurface performance of three different ice‐flow models of the Northeast Greenland Ice Stream. We show that fast flow in Northeast Greenland is, in part, controlled by softer, warmer ice, and that correctly modeling heat transport at the boundaries of ice streams is critical for realistic projections of their future behavior. Ultimately, we provide insight into a sensitive region of Greenland together with a new approach to geophysical data use in model evaluation, with the goal of reducing the range of plausible models projecting the future of the Greenland and Antarctic Ice Sheets. Key Points: Thermal softening of ice is present in the Northeast Greenland Ice Stream (NEGIS) shear margins, despite low strain ratesVertical advection of heat dominates the shear‐margin temperature structure here, validated by radar reflectivity and isochron geometryRadar data can be used to constrain ice temperature and subsurface velocity to evaluate ice sheet model spin‐up and inversions [ABSTRACT FROM AUTHOR]

For more than two decades, radar altimetry missions have provided continuous elevation estimates of the Greenland ice sheet (GrIS). Here, we propose a method for using such data to estimate ice-sheet-wide surface elevation changes (SECs). The final data set will be based on observations acquired from the European Space Agency’s Environmental Satellite (ENVISAT), European Remote Sensing (ERS)-1 and -2, CryoSat-2, and, in the longer term, Sentinel-3 satellites. In order to find the best-performing method, an intercomparison exercise has been carried out in which the scientific community was asked to provide their best SEC estimates as well as feedback sheets describing the applied method. Due to the hitherto few radar-based SEC analyses as well as the higher accuracy of laser data, the participants were asked to use either ENVISAT radar or ICESat (Ice, Cloud, and land Elevation Satellite) laser altimetry over the Jakobshavn Isbræ drainage basin. The submissions were validated against airborne laser-scanner data, and intercomparisons were carried out to analyse the potential of the applied methods and to find whether the two altimeters were capable of resolving the same signal. The analyses found great potential of the applied repeat-track and cross-over techniques, and, for the first time over Greenland, that repeat-track analyses from radar altimetry agreed well with laser data. Since topography-related errors can be neglected in cross-over analyses, it is expected that the most accurate, ice-sheet-wide SEC estimates are obtained by combining the cross-over and repeat-track techniques. It is thus possible to exploit the high accuracy of the former and the large spatial data coverage of the latter. Based on CryoSat’s different operation modes, and the increased spatial and temporal data coverage, this shows good potential for a future inclusion of CryoSat-2 and Sentinel-3 data to continuously obtain accurate SEC estimates both in the interior and margin ice sheet. [ABSTRACT FROM AUTHOR]

We analyze the internal stratigraphy in radio-echo sounding data of the northeast Greenland ice stream to infer past and present ice dynamics. In the upper reaches of the ice stream, we propose that shear-margin steady-state folds in internal reflecting horizons (IRHs) form due to the influence of ice flow over spatially varying basal lubrication. IRHs are generally lower in the ice stream than outside, likely because of greater basal melting in the ice stream from enhanced geothermal flux and heat of sliding. Strain-rate modeling of IRHs deposited during the Holocene indicates no recent major changes in ice-stream vigor or extent in this region. Downstream of our survey, IRHs are disrupted as the ice flows into a prominent overdeepening. When combined with additional data from other studies, these data suggest that upstream portions of the ice stream are controlled by variations in basal lubrication whereas downstream portions are confined by basal topography. [ABSTRACT FROM AUTHOR]

Glaciers in polar regions are sensitive to climate and ocean changes and can thin rapidly as a consequence of global warming. Digital Elevation Models (DEMs) from remote sensing observations have been widely used to detect changes in polar glaciers. DEMs from Terrestrial Radar Interferometer (TRI) have recently been used for high frequency glacier change and glacier-ocean interaction studies. However, it is unclear whether TRI DEM over a large study area can be combined directly with remote sensing observations to investigate glacier changes as well as the accuracy of TRI DEM at far range. In this study, we deployed a TRI close to Helheim Glacier, East Greenland and generated DEMs using TRI and satellite laser altimetry. We analyzed the accuracy of the TRI DEM using theoretical calculations, comparisons based on repeat observations, and comparisons with a high accurate ArcticDEM. The validation results suggest that for stable ground surfaces, the uncertainty (standard deviation) is <5 m at range < 9.8 km. Averaging across time (e.g. one hour) decreases the uncertainty almost linearly with range, over 0.5 m to 1.2 m when the range increases from 7.0 km to 10.0 km. Increasing the correlation coefficient threshold for phase unwrapping does not significantly reduce uncertainty. TRI DEMs are influenced by systematic error at far range primarily due to coarse azimuth resolution and phase unwrapping difficulties in discontinuous interferograms. As the absolute accuracy of TRI DEMs is not uniformly distributed in the range direction (farther points have worse uncertainty), our findings indicate that TRI DEMs within range of 10 km can reach <5 m uncertainty, which can be compared with DEMs obtained from remote sensing satellites to detect glacier thinning. • A method to generate TRI DEM referring to WGS-84 ellipsoid is presented. • Averaging across time decreases the DEM uncertainty almost linearly with range. • Increasing correlation coefficient threshold does not significantly improve TRI DEM. • Large uncertainty at far range is caused by coarser azimuth resolution and discontinuous interferogram. • TRI DEMs with range < 10 km have <5 m uncertainty and can match satellite observations. [ABSTRACT FROM AUTHOR]

Accurate measurements of fast flowing outlet glaciers in Greenland and Antarctica are of vital importance to improve ice-sheet models that predict the ice-sheets’ contribution to sea-level rise over the next century. Radars with high sensitivity and advanced processing capabilities are required to sound ice in fast-flowing glaciers. We developed the multichannel coherent radar depth sounder/imager (MCoRDS/I) for ice thickness measurements and 3-D imaging of bedrock over important areas of Greenland and Antarctica. Radio frequency interference (RFI) degrades the sensitivity of the MCoRDS/I system, thus affecting the quality of the data collected. Data from the 2010 Greenland field season exhibited a degraded signal-to-noise ratio (SNR), requiring extensive RFI analysis, investigation, and mitigation for the MCoRDS/I system. Measurements were taken in an electromagnetic interference (EMI) chamber on individual sections of the MCoRDS/I system to isolate the sources of RFI. Then, RFI mitigation techniques were implemented for the offending sections and EMI chamber measurements verified the integrity of the solution prior to the 2011 Greenland deployment. Recorded data from the 2011 Greenland field season also verified that the RFI mitigation resulted in more than 20 dB improvement in the SNR compared to the 2010 Greenland data. The reduction of EMI emissions has also been beneficial to data collection in subsequent deployments. [ABSTRACT FROM AUTHOR]

The Moderate Resolution Imaging Spectroradiometer (MODIS), Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP), and CloudSat Cloud Profiling Radar (CPR) set of sensors, all in the Afternoon Constellation (A-Train), has been regarded as among the most powerful tools for characterizing the cloud cover. While providing good complementary information, the authors also observed that, at least for the Arctic region, the different sensors provide significantly different statistics about cloud cover characteristics. Data in 2007 and 2010 were analyzed, and the annual averages of cloud cover in the Arctic region were found to be 66.8%, 78.4%, and 63.3% as derived from MODIS, CALIOP, and CPR, respectively. A large disagreement between MODIS and CALIOP over sea ice and Greenland is observed, with a cloud percentage difference of 30.9% and 31.5%, respectively. In the entire Arctic, the average disagreement between MODIS and CALIOP increased from 13.1% during daytime to 26.7% during nighttime. Furthermore, the MODIS cloud mask accuracy has a high seasonal dependence, in that MODIS-CALIOP disagreement is the lowest during summertime at 10.7% and worst during winter at 28.0%. During nighttime the magnitude of the bias is higher because cloud detection is limited to the use of infrared bands. The clouds not detected by MODIS are typically low-level (top height <2 km) and high-level clouds (top height >6 km) and, especially, those that are geometrically thin (<2 km). Geometrically thin clouds (<2 km) accounted for about 95.5% of all clouds that CPR misses. As reported in a similar study, very low and thin clouds (<0.3 km) over sea ice that are detected by MODIS are sometimes not observed by CPR and misclassified by CALIOP. [ABSTRACT FROM AUTHOR]

Despite containing only 14% of the Greenland ice sheet by area, the southeastern sector has the highest accumulation rates, and hence receives ~30% of the total snow accumulation. We present accumulation rates obtained during our 2010 Arctic Circle Traverse derived from three 50m firn cores dated using geochemical analysis. We tracked continuous internal reflection horizons between the firn cores using a 400MHz ground-penetrating radar (GPR). GPR data combined with depth-age scales from the firn cores provide accumulation rates along a 70 km transect. We followed an elevation gradient from ~2350 to ~1830m to understand how progressive surface melt may affect the ability to chemically date the firn cores and trace the internal layers with GPR. From the firn cores, we find a 52% (~0.43m w.e. a-1) increase in average snow accumulation and greater interannual variability at the lower site than the upper site. The GPR profiling reveals that accumulation rates are influenced by topographic undulations on the surface, with up to 23% variability over 7 km. These measurements confirm the presence of high accumulation rates in the southeast as predicted by the calibrated regional climate model Polar MM5. [ABSTRACT FROM AUTHOR]

Abstract: Seasonal glacier velocity changes across the High Arctic, including the Greenland Ice Sheet, have been observed and have attracted significant attention over the past decade. However, it remains uncertain how much short-term variability exists in other polythermal glaciers, particularly those in High Asia. Here we report satellite radar image analyses that reveal diverse glacier surface velocities and their evolution in West Kunlun Shan (WKS), NW Tibet, where little is known about glacier dynamics. On the basis of radar images obtained from 2003 to 2011, we examined 36 glaciers, and classified them into two classes according to their multi-temporal velocity profiles: 25 as normal-flow type (surface velocity reaches maxima around the middle part, and gradually approaches zero toward downstream and upstream), and four as surging type (surface velocities are greater than 150m/yr, and/or the terminus advance is recognized from the radar images). Seven other glaciers do not fit the former two classes, and reveal stagnant velocity profiles that are nearly zero in the lower part but are similar to those of the normal type in the upper part. Although these glaciers could be just stagnant tongues indicative of receding normal type glaciers, given the temporal evolution at the Zhongfeng Glacier, the stagnant type possibly represents a quiescent phase of the surging type glaciers. While glacier surfaces are mostly clean with limited debris-cover, except near the termini, surge-type glaciers might be common in WKS. The observed short-term velocity changes provide us with evidence for efficient basal slip even at the high-elevation polythermal glaciers. This study demonstrates that frequent radar image acquisitions are helpful to understand short-term velocity changes at remote glaciers in detail. [Copyright &y& Elsevier]

Abstract: Ground and airborne radar depth-sounding of the Greenland and Antarctic ice sheets have been used for many years to remotely determine characteristics such as ice thickness, subglacial topography, and mass balance of large bodies of ice. Ice coring efforts have supported these radar data to provide ground truth for validation of the state (wet or frozen) of the interface between the bottom of the ice sheet and the underlying bedrock. Subglacial state governs the friction, flow speed, transport of material, and overall change of the ice sheet. In this paper, we utilize machine learning and classifier combination to model water presence from airborne polar radar data acquired on Greenland in 1999 and 2007. The underlying method results in radar independence, allowing model transfer from 1999 to 2007 radar data to produce water presence maps of the Greenland ice sheet with differing radars. We focus on how to construct a successful set of classifiers capable of high classification accuracy. Utilizing multiple machine learning algorithms is shown to be successful for this classification problem, achieving 86% classification accuracy in the best case. Several heuristics are presented for constructing teams of multiple classifiers for predicting subglacial water presence. The presented methodology could also be applied to radar data acquired over the Antarctic ice sheet. [Copyright &y& Elsevier]

This paper focuses on automating the task of estimating Polar ice thickness from airborne radar data acquired over Greenland and Antarctica. This process involves the identification and accurate selection of the ice sheet's surface location and interface between the ice sheet and the underlying bedrock for each measurement. Identifying the surface and bedrock locations in the radar imagery enables the computation of ice sheet thickness, which is important for the study of ice sheets, their volume, and how they may contribute to global climate change. The time-consuming manual approach requires sparse hand-selection of surface and bedrock interfaces by several human experts, and interpolating between the selections to save time. Two primary methods have been studied in this paper, namely, edge-based and active contour. Results are evaluated and presented using the metrics of time requirements and accuracy. Automated ice thickness estimation results from 2006 and 2007 Greenland field campaigns illustrate that the edge-based approach offers faster processing (seconds compared to minutes), but suffers from a lack of continuity and smoothness aspects that active contours provide. The active contour approach is more accurate when compared to ground truth selections provided by human experts, and has proven to be more robust to image artifacts. It is shown that both techniques offer advantages which could be integrated to yield a more effective system. [ABSTRACT FROM AUTHOR]

Climate change is extensively affecting ice sheets resulting in accelerating mass loss in recent decades. Assessment of this reduction and its causes is required to project future ice mass loss. Annual snow accumulation is an important component of the surface mass balance of ice sheets. While in situ snow accumulation measurements are temporally and spatially limited due to their high cost, airborne radar sounders can achieve ice sheet wide coverage by capturing and tracking annual snow layers in the radar images or echograms. In this paper, we use deep learning to uniquely identify the position of each annual snow layer in the Snow Radar echograms taken across different regions over the Greenland ice sheet. We train with more than 15,000 images generated from radar echograms and estimate the thickness of each snow layer within a mean absolute error of 0.54 to 7.28 pixels, depending on dataset. A highly precise snow layer thickness can help improve weather models and, thus, support glaciological studies. Such a well-trained deep learning model can be used with ever-growing datasets to aid in the accurate assessment of snow accumulation on the dynamically changing ice sheets. [ABSTRACT FROM AUTHOR]

An extended Kalman filter (EKF) to estimate the ionospheric total electron content (TEC) along the radar line-of-sight has been developed. By varying the carrier frequency of consecutive radar pulses, it is shown that the TEC along the line-of-sight becomes observable and can be estimated in real-time to a high degree of accuracy. It is shown that the filtering formulation presented here optimally reduces the randomly varying range bias caused by the irregular electron density structure in the upper atmosphere. The performance of the tracking filter is examined with a Monte Carlo simulation analysis. It is shown that for a radar located in Thule, Greenland and operating at a nominal frequency of 435 MHz (UHF), an improvement of approximately 50% in the range estimation accuracy is achieved with a frequency separation of 18 MHz between two consecutive pulses. if the frequency separation is increased to 60 MHz, the increase in accuracy is 75%. It is also shown that in the absence of frequency separation, the performance of the tracking filter presented here reduces to that of a conventional six-state EKF that does not account for the ionospheric bias. [ABSTRACT FROM AUTHOR]

We provide an overview of a project to develop and demonstrate radars for imaging the ice–bed interface, mapping shallow and deep internal layers, and measuring ice thickness. We developed an ultra-wideband very high-frequency (VHF) radar (120 to 300 MHz) for simultaneously measuring ice thickness, mapping internal layers at depth, and imaging the ice–bed interface, and an ultra high-frequency (UHF) ultra-wideband radar (500 to 2000 MHz) for fine-resolution mapping of near-surface internal layers. We used these radars to collect data at Summit Camp, Greenland, during July 2005, and the West Antarctic Ice Sheet (WAIS) deep drill core site during January 2006. These results provide us with an unprecedented view of basal characteristics and internal layers. We also developed an outreach program, a communications system, and rovers to facilitate geophysical measurements in the future. [Copyright &y& Elsevier]

Snow accumulation in remote regions, such as Green- land and Antarctica, is a key factor for estimating the Earth's ice mass balance. In situ data are sparse; hence, they are useful to derive snow accumulation from remote sensing observations, such as microwave thermal emission and radar brightness. These data are usually interpreted using electromagnetic models in which volume scattering is the dominant mechanism. The main limitation of this approach is that microwave brightness is not well related to backscatter if the ice sheet is layered. Because larger grain size and thicker annual layers both increase radar image brightness, with the first corresponding to lower accumulation rate and the second to higher accumulation rate, models of radar brightness alone cannot accurately reflect accumulation. Consideration of correlation measurements can also resolve this ambiguity. We introduce an interferometric ice scattering model that relates the interferometric synthetic aperture radar correlation and radar brightness to both ice grain size and hoar layer spacing in the dry-snow zone of Greenland. We use this model and the European Remote Sensing satellite radar observations to derive several parameters related to snow accumulation rates in a small area in the dry-snow zone. These parameters show agreement with four in situ core accumulation rate measurements in this area, whereas models using only radar brightness data do not match the observed variation in accumulation rates. [ABSTRACT FROM AUTHOR]

Using satellite radar interferometry observations of Greenland, we detected widespread glacier acceleration below 66° north between 1996 and 2000, which rapidly expanded to 70° north in 2005. Accelerated ice discharge in the west and particularly in the east doubled the ice sheet mass deficit in the last decade from 90 to 220 cubic kilometers per year. As more glaciers accelerate farther north, the contribution of Greenland to sea-level rise will continue to increase. [ABSTRACT FROM AUTHOR]

Characterization of the microwave signature of the Greenland snow surface enables delineation of the different snow facies and is a tool for tracking the effects of climate change. A new empirical observation model is introduced that uses a limited number of parameters to characterize the snow surface based on the dependence of radar backscatter on incidence angle, azimuth angle, spatial gradient, and temporal rate of change. The individual model parameters are discussed in depth with examples using data from the NASA Scatterometer (NSCEF) and from the C-band European Remote Sensing satellite (ERS) Advanced Microwave Instrument in scatterometer mode. The contribution of each model term to the overall accuracy of the model is evaluated. The relative contributions of the modeled dependencies vary by region. Two studies illustrating applications of the model are included. First, interannual changes over the Greenland ice sheet are investigated using nine years of ERS data. Surface changes are observed as anomalies in the σ° model parameters. Second, intra- annual variations of the surface are investigated. These changes are observed in the average backscatter normalized to a given observation geometry. The results indièate that the model can be used to obtain a more complete understanding of multiyear change and to obtain low-variance high temporal resolution Observations of intraannual changes. The model may be applied for increased accuracy in scatterometer, synthetic aperture radar (SAR), and wide- angle SAR studies. [ABSTRACT FROM AUTHOR]

Sea-ice edge detection is an essential task at the different national ice services to secure navigation in ice-covered seas. Comparison between the Remund and Long ice mask image from enhanced-resolution QuikScat/Sea Winds (QS) products and the analyzed ice edge from high-resolution RADARSAT synthetic aperture radar has shown that the automatically determined QS ice mask underestimates the Arctic ice extent. QS data was statistically analyzed by calculating the data with ice charts around Greenland and with the National Aeronautics and Space Administration Team's Special Sensor Microwave/Imager (SSMII) ice concentration algorithm over the whole Arctic region. All variables, i.e., the backscatter in vertical and horizontal polarization, the active polarization ratio (APR) and the daily standard deviation, are sensitive to ice types and are strongly correlated with ice concentration when the relationship is expressed in exponential form. Our study showed that the APR is especially suitable for ice-ocean separation, and a threshold of -0.02 was determined. An ice edge algorithm based on this APR threshold was developed using the other variables with conservative season-dependent thresholds to eliminate additional ocean noise. Also, the history of the ice cover is considered in order to detect single ice fields that are separated from the main Arctic pack ice. Validation with RADARSAT 1 and with the Advanced Very High Resolution Radiometer showed that the new algorithm successfully detects very low ice concentrations of about 10% during the entire year. The validity of the detected ice edge for near-real-time issues is also discussed in relation to the ice motion in the Marginal Ice Zone and the integration time necessary to produce the enhanced-resolution images. The new algorithm improves the automatic global ice edge resolution by a factor of two when compared to SSM/I products and could be used in both model initialization and data assimilation. [ABSTRACT FROM AUTHOR]

Snow accumulation rate is an important parameter in determining the mass balance of polar ice sheets. Accumulation rate is currently determined by analyzing ice cores and snow pits. Inadequate sampling of the spatial variations in the ice sheet accumulation has resulted in accumulation rate uncertainties as large as 24%. We designed and developed a 600-900-MHz airborne radar system for high-resolution mapping of the near-surface internal layers for estimating the accumulation rate of polar ice sheets. Our radar system can provide improved spatial and temporal coverage by mapping a continuous profile of the isochronous layers in the ice sheet. During the 2002 field season in Greenland, we successfully mapped the near-surface layers to a depth of 200 m in the dry-snow zone, 120 m in the percolation zone, and 20 m in the melt zone. We determined the water equivalent accumulation rate at the NASA-U_1 site to be 34.9 ± 5.1 cm/year from 1964 to 1992. This is in close agreement with the ice-core derived accumulation rate of 34.6 cm/year for the same period. [ABSTRACT FROM AUTHOR]

In this paper, we build on past efforts with regard to the implementation of an efficient feature tracking algorithm for the mass processing of satellite images. This generic open-source feature tracking routine can be applied to any type of imagery to measure sub-pixel displacements between images. The routine consists of a feature tracking module (autoRIFT) that enhances computational efficiency and a geocoding module (Geogrid) that mitigates problems found in existing geocoding algorithms. When applied to satellite imagery, autoRIFT can run on a grid in the native image coordinates (such as radar or map) and, when used in conjunction with the Geogrid module, on a user-defined grid in geographic Cartesian coordinates such as Universal Transverse Mercator or Polar Stereographic. To validate the efficiency and accuracy of this approach, we demonstrate its use for tracking ice motion by using ESA's Sentinel-1A/B radar data (seven pairs) and NASA's Landsat-8 optical data (seven pairs) collected over Greenland's Jakobshavn Isbræ glacier in 2017. Feature-tracked velocity errors are characterized over stable surfaces, where the best Sentinel-1A/B pair with a 6 day separation has errors in X/Y of 12 m/year or 39 m/year, compared to 22 m/year or 31 m/year for Landsat-8 with a 16-day separation. Different error sources for radar and optical image pairs are investigated, where the seasonal variation and the error dependence on the temporal baseline are analyzed. Estimated velocities were compared with reference velocities derived from DLR's TanDEM-X SAR/InSAR data over the fast-moving glacier outlet, where Sentinel-1 results agree within 4% compared to 3–7% for Landsat-8. A comprehensive apples-to-apples comparison is made with regard to runtime and accuracy between multiple implementations of the proposed routine and the widely-used "dense ampcor" program from NASA/JPL's ISCE software. autoRIFT is shown to provide two orders of magnitude of runtime improvement with a 20% improvement in accuracy. [ABSTRACT FROM AUTHOR]

C-band radar backscatter generally seems to be inversely related to accumulation rate across Greenland's dry snow zone. A coupled snow metamorphosis-backscatter model was constructed in order to investigate the observation. Accumulation rate and mean annual temperature were the coupled-model inputs. In a local area where accumulation rate is known very accurately the model predicts the backscatter level along several transects within the calibration uncertainties of the ERS-1 SAR data. The analysis was applied to the remainder of the dry snow zone. The model showed that both accumulation rate and temperature can modulate surface backscatter across Greenland's dry snow zone with accumulation rate being the primary influence. [ABSTRACT FROM AUTHOR]

Using time-variable gravity measurements from the Gravity Recovery and Climate Experiment (GRACE) satellite mission, we estimate ice mass changes over Greenland during the period April 2002 to November 2005. After correcting for the effects of spatial filtering and limited resolution of GRACE data, the estimated total ice melting rate over Greenland is -239 ± 23 cubic kilometers per year, mostly from East Greenland. This estimate agrees remarkably well with a recent assessment of -224 ± 41 cubic kilometers per year, based on satellite radar interferometry data. GRACE estimates in southeast Greenland suggest accelerated melting since the summer of 2004, consistent with the latest remote sensing measurements. [ABSTRACT FROM AUTHOR]

The warming climate is changing the surface dynamics of the Greenland Ice Sheet, including the balance between snowfall and melt. Increasing surface melt impacts the structure of the relatively porous near-surface layer known as firn. Camp Century, a base abandoned in 1967, now comprises a subsurface debris field within the firn in Northwest Greenland. We collected 80 km of 100 or 250 MHz radar data in nested grids over this subsurface debris field. Here, we present a comprehensive analysis and interpretation of this ice-penetrating radar survey. The vast majority (95%) of subsurface reflectors are located at depths of greater than 32 m. The tunnel network, as well as an overlying layer associated with historical surface activities, is readily visible in the radar data. This subsurface debris field is approximately circular with a radius of less than 1 km. Local downwarping of clear internal layers – likely annual accumulation layers - identifies now-collapsed liquid sumps. Analysis of radar signal polarity suggests that liquid hydrocarbons are likely present in one of these sumps. The radar data and a geo-referenced site map of Camp Century are freely accessible at www.campcenturyclimate.dk. • A dense ground-penetrating radar survey has been conducted over the former Camp Century station area • Analysis of the radar data reveals the exact location of camp debris and successfully identifies the infrastructure • Areas of layer down-welling indicate sumps of liquid material which is interpreted as either water or glycol [ABSTRACT FROM AUTHOR]

Crystal Orientation Fabric (COF) of c-axes in ice cores reveals information about deformation within ice sheets. While this is a well established analysis technique for deep ice cores from ice divides, information about COF in ice streams is just now becoming available: the EastGRIP ice core is situated inside the largest ice stream in Greenland, the North East Greenland Ice Stream (NEGIS). With the ongoing analysis of samples from the EastGRIP ice core, COF is now available down to 1714 m, revealing an extremely more rapid evolution of COF anisotropy with depth compared to all other ice cores. This enables us to study the ability of polarimetric radar measurements to infer an overall pattern of COF from measurements conducted at the surface. Depending on whether the COF is isotropic or anisotropic, a radar signal is reflected differently in terms of angle dependence and polarization. We conducted these polarimetric measurements around the EastGRIP drill site and we compare them to COF data obtained from 778 thin sections, prepared and measured at EastGRIP drill site. We investigate the hypothesis that the same pattern of COF can be retrieved from the polarimetric measurements as is available from the ice core. If confirmed, this would provide an addition constraint on the (an)isotropy at locations where no ice core is available. This would potentially provide quasi spatial coverage and greatly improve our understanding of the evolution of anisotropy over from ice divides to outlet glaciers. [ABSTRACT FROM AUTHOR]