Sea Ice Backscatter Research Articles

Mapping icebergs in sea ice: An analysis of seasonal SAR backscatter at C- and L-band

Icebergs in the Arctic can pose a threat to maritime traffic and offshore installations and influence the properties of the upper ocean layer. While icebergs in open water are regularly monitored using C-band SAR satellites, less attention has been paid to icebergs in regions with a high areal fraction of sea ice, where detection using traditional methods is more difficult. In this study, we compare the capability of C- and L-band SAR to detect icebergs in level and deformed fast sea ice across various seasons. To this end we use a timeseries of SAR images acquired at HH- and HV-polarization in 2019 and 2020, covering respectively 301 and 356 icebergs. As reference data for validation, we used iceberg polygons derived from Sentinel-2 images. Our results reveal that compared to C-band, L-band SAR is significantly better at separating the backscatter of icebergs and sea ice and thus is preferable for detecting icebergs in ocean regions with a high sea ice concentration. It is further shown that L-band SAR is less affected by melting conditions, suggesting that it can be used for iceberg detections in both summer and winter.

Polar Sea Ice Detection Using a Rotating Fan Beam Scatterometer

Scatterometers are dedicated to monitoring sea surface wind vectors, but they also provide valuable data for polar applications. As a new type of scatterometer, the rotating fan beam scatterometer delivers a higher diversity of incidence angles and more azimuth sampling. The paper takes the first rotating fan beam scatterometer, the China France Oceanography Satellite scatterometer (CSCAT), as an example to explore the effectiveness of this new type of scatterometer in polar sea ice detection. In this paper, a Bayesian method with consideration of geometric characteristics of CSCAT is developed for sea ice detection. The implementation of this method includes the definition of CSCAT backscatter space, an estimation of the sea ice Physical Model Function (GMF), a calculation of the sea ice backscatter distance to the sea ice GMF, a probability distribution function (PDF) estimation of the square distance to the GMF (sea ice GMF and wind GMF), and a calculation of the sea ice Bayesian posterior probability. This algorithm was used to generate a daily CSCAT polar sea ice mask during the CSCAT mission period (2019–2022) (by setting a 55% threshold on the Bayesian posterior probability). The sea ice masks were validated against passive microwaves by quantitatively comparing the sea ice edges and extents. The validation suggests that the CSCAT sea ice edge and extent show good agreement with the sea ice concentration distribution (i.e., sea ice concentration ≥ 15%) of the Advanced Microwave Scanning Radiometer 2 (AMSR2). The average Euclidean distance of the sea ice edges was basically less than 12.5 km, and the deviation of the sea ice extents was less than 0.3 × 106 km2.

Intercomparison of Arctic Sea Ice Backscatter and Ice Type Classification Using Ku-Band and C-Band Scatterometers

As a result of global warming, multiyear ice (MYI) is being replaced by first-year ice (FYI) in the Arctic. Microwave scatterometers in the Ku-band and C-band can provide daily observations of sea ice type. However, their comparative capabilities in mapping ice type have not been thoroughly evaluated. We present a systematic intercomparison of the backscatter signature in VV polarization ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\sigma }_{\mathrm {vv}}^{\mathrm {o}}$ </tex-math></inline-formula> ) and the sea ice classification from three scatterometer systems using the same ice classification approach. The systems are the Ku-band quick scatterometer (QSCAT) and the newly launched Chinese rotating fan-beam scatterometer (RFSCAT) and the C-band advanced scatterometer (ASCAT). Three freezing seasons are used, i.e., 2007/08 and 2008/09 for the QSCAT/ASCAT comparison and 2019/20 for the RFSCAT/ASCAT comparison. With reference to ASCAT, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\sigma }_{\mathrm {vv}}^{\mathrm {o}}$ </tex-math></inline-formula> bias between QSCAT and RFSCAT results from their different incidence angles. A continuous declining trend of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\sigma }_{\mathrm {vv}}^{\mathrm {o}}$ </tex-math></inline-formula> from MYI and FYI is observed during winter, with a greater difference between MYI and FYI in the Ku-band. The MYI and FYI extent derived from QSCAT/RFSCAT is highly consistent with that derived from ASCAT, with a difference less than 7% and 3% for MYI and FYI, respectively. The overall accuracy (OA) is around 77% and 80% for the RFSCAT results and ASCAT results, respectively, compared with Sentinel-1 SAR images. The classification results show high consistency (81%–89%) with ice charts from the Canadian Ice Service. The incorporation of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\mathrm {Tb}}_{36\mathrm {h}}$ </tex-math></inline-formula> from AMSR-E/AMSR2 improves the OA of the classification when using ASCAT or RFSCAT by 7%–11%.

Errata for “Bayesian Sea Ice Detection With the ERS Scatterometer and Sea Ice Backscatter Model at C-Band” [Apr 18 2248-2254

In [1] , (16) in Section IV titled “Sea ice backscatter model at C-band” contains a typographical error. It currently reads as

Predicting Melt Pond Fraction on Landfast Snow Covered First Year Sea Ice from Winter C-Band SAR Backscatter Utilizing Linear, Polarimetric and Texture Parameters

Early-summer melt pond fraction is predicted using late-winter C-band backscatter of snow-covered first-year sea ice. Aerial photographs were acquired during an early-summer 2012 field campaign in Resolute Passage, Nunavut, Canada, on smooth first-year sea ice to estimate the melt pond fraction. RADARSAT-2 Synthetic Aperture Radar (SAR) data were acquired over the study area in late winter prior to melt onset. Correlations between the melt pond fractions and late-winter linear and polarimetric SAR parameters and texture measures derived from the SAR parameters are utilized to develop multivariate regression models that predict melt pond fractions. The results demonstrate substantial capability of the regression models to predict melt pond fractions for all SAR incidence angle ranges. The combination of the most significant linear, polarimetric and texture parameters provide the best model at far-range incidence angles, with an R 2 of 0.62 and a pond fraction RMSE of 0.09. Near- and mid- range incidence angle models provide R 2 values of 0.57 and 0.61, respectively, with an RMSE of 0.11. The strength of the regression models improves when SAR parameters are combined with texture parameters. These predictions also serve as a proxy to estimate snow thickness distributions during late winter as higher pond fractions evolve from thinner snow cover.

A scatterometer record of sea ice extents and backscatter: 1992–2016

Abstract. This paper presents the first long-term climate data record of sea ice extents and backscatter derived from intercalibrated satellite scatterometer missions (ERS, QuikSCAT and ASCAT) extending from 1992 to the present date (Verhoef et al., 2018). This record provides a valuable independent account of the evolution of Arctic and Antarctic sea ice extents, one that is in excellent agreement with the passive microwave records during the fall and winter months but shows higher sensitivity to lower concentration and melting sea ice during the spring and summer months. The scatterometer record also provides a depiction of sea ice backscatter at C- and Ku-bands, allowing the separation of seasonal and perennial sea ice in the Arctic and further differentiation between second-year (SY) and older multiyear (MY) ice classes, revealing the emergence of SY ice as the dominant perennial ice type after the historical sea ice loss in 2007 and bearing new evidence on the loss of multiyear ice in the Arctic over the last 25 years. The relative good agreement between the backscatter-based sea ice (FY, SY and older MY) classes and the ice thickness record from Cryosat suggests its applicability as a reliable proxy in the historical reconstruction of sea ice thickness in the Arctic.

Bayesian Sea Ice Detection With the ERS Scatterometer and Sea Ice Backscatter Model at C-Band

This paper describes the adaptation of a Bayesian sea ice detection algorithm for the scatterometer on-board the European Remote Sensing (ERS) satellites (ERS-1 and ERS-2). The algorithm is based on statistics of distances to ocean wind and sea ice geophysical model functions (GMFs) and its performance is validated against coincident active and passive microwave data. We furthermore propose a new model for sea ice backscatter at the C-band in vertical polarization based on the sea ice GMFs derived from ERS and advanced scatterometer data. The model characterizes the dependence of sea ice backscatter on the incidence angle and the sea ice type, allowing a more precise incidence angle correction than afforded by the usual linear transformation. The resulting agreement between the ERS, QuikSCAT, and special sensor microwave imager sea ice extents during the year 2000 is high during the fall and winter seasons, with an estimated ice edge accuracy of about 20 km, but shows persistent biases between scatterometer and radiometer extents during the melting period, with scatterometers being more sensitive to summer (lower concentration and rotten) sea ice types.

Incidence Angle Dependence of First-Year Sea Ice Backscattering Coefficient in Sentinel-1 SAR Imagery Over the Kara Sea

We have studied the incidence angle ( $\theta _{0}$ ) dependence of the sea ice backscattering coefficient ( $\sigma$ °) for Sentinel-1 (S-1) extra wide (EW) mode dual-polarization (HH/HV) synthetic aperture radar (SAR) imagery acquired over the Kara Sea under winter and summer melting conditions. The determination of the $\sigma$ ° versus $\theta _{0}$ dependence was based on SAR image pairs acquired on ascending and descending orbits over the same sea ice area with a short time difference. The SAR noise floor was subtracted from the HV images. From the image pairs 1.1 by 1.1 km windows representing level first-year ice (LFYI) and deformed first-year ice (DFYI) were manually selected, and a linear regression was fit between the resulting $\sigma$ ° and $\theta _{0}$ differences of the windows to estimate the slope ${b}_{1}$ (dB/1°) between $\sigma$ ° and $\theta _{0}$ . For example, under winter condition ${b}_{1}$ for DFYI at HH- and HV-polarizations was found to be −0.24 and −0.16 dB/1°, respectively, and ${b}_{1}$ for LFYI at HH-polarization was −0.25 dB/1°. It was not possible to determine a reliable ${b}_{1}$ for LFYI at HV due to a contamination effect of the S-1 noise floor. The ${b}_{1}$ values at HH compared well with previous studies. They can be used to compensate the $\sigma$ ° incidence angle variation in the S-1 EW SAR images with good accuracy. The HH ${b}_{1}$ values are applicable to other S-1 imaging modes and other C-band SAR sensors like RADARSAT-2. Unfortunately, the HV ${b}_{1}$ values are specific to the S-1 EW mode due to the noise floor problem.

Multiyear Arctic Sea Ice Classification Using QuikSCAT

Long-term trends in Arctic sea ice are of particular interest in studies of global temperature, climate change, and industrial application. This paper analyzes intra-annual and interannual trends in Ku-band backscatter over first-year (FY) and multiyear (MY) sea ice to develop a new sea-ice-type classification method. Histograms of backscatter are derived from high-resolution backscatter images created using the scatterometer image reconstruction (SIR) algorithm applied to measurements obtained by the SeaWinds instrument aboard QuikSCAT. The backscatter of FY and MY sea ice are clearly identifiable and are observed to vary seasonally. Using an average of the annual backscatter trends obtained from QuikSCAT, a classification of MY ice is obtained, which uses a time-dependent threshold value. Validation of the classification method is done using regional ice charts from the Canadian Ice Service. Differences in ice classification are found to be less than 6% for the winters of 2006-2007 and 2007-2008, and the end of 2008. Anomalies in the distribution of sea ice backscatter from year to year suggest a reduction in MY ice cover between 2003 and 2009 and an approximately equivalent increase in FY ice cover.

Arctic sea ice mapping with satellite radars

The drastic reduction of Arctic sea ice in recent years demands ice monitoring over various spatial and temporal scales. Sea ice backscatter signatures from field measurements and from model analyses are obtained at L-band and C-band frequencies. Based on these signatures, capabilities for Arctic sea ice mapping are determined for current and future satellite active microwave sensors including synthetic aperture radars (SAR) and scatterometers. This study includes L-band and C-band radars such as the ERS (European Remote Sensing), Envisat (Environmental Satellite), RADARSAT-1 and 2, ALOS (Advanced Land Observing Satellite), and DESDnyl (Deformation, Ecosystem Structure, and Dynamics of Ice). SARs with resolutions from 10 to 100 m, and the SMAP (Soil Moisture Active-Passive) scatterometer with resolutions from 1 to 10 km.

New data set of onset of annual snowmelt on Antarctic sea ice

The annual onset of snowmelt on sea ice is essential for climate monitoring since it triggers a decrease in surface albedo that feeds back into a stronger absorption of shortwave radiation—a process known as the snowmelt‐albedo feedback—and thus strongly modifies the surface energy balance during summer [Curry et al., 1995]. Algorithms designed for the detection of snowmelt on Arctic sea ice and based on longterm passive‐microwave data [Anderson, 1997; Drobot and Anderson, 2001] revealed the melt season in the Arctic from 1979 to 1998 to be significantly elongated and the onset of melt to be shifted toward earlier dates [Drobot and Anderson, 2001; Belchansky et al., 2004].In the Antarctic, however, little effort has been made so far in detecting the length of the summer melt season on sea ice by means of satellite microwave data. This results from the fact that surface melting in the Antarctic differs significantly from corresponding processes in the Arctic [Nicolaus et al., 2006]. The hemispheric differences are supported by extensive field measurements [Massom et al., 2001; Haas et al., 2001 ] and find expression in a reversal of the general surface radar backscatter and brightness temperature (TB) tendencies during summer [Haas, 2001; Kern and Heygster, 2001] : In the Antarctic, sea ice backscatter increases and TB decreases when summer approaches, contrary to the Arctic. Hence, algorithms developed for Arctic sea ice are not applicable on its southern counterpart. As summer air temperatures in the Antarctic rarely rise above 0°C, classical surface melt ponds have never been observed to the extent they appear in the Arctic and the sea ice surface typically remains snow‐covered year‐round. Drinkwater and Liu [2000] investigate snowmelt on Antarctic sea ice based on a method that identifies a decrease in surface radar backscatter. However, they detect melt to be lasting for only some days and exclusively on first‐year ice. Presumably, the backscatter decrease they observe is due to flooding of the snow before the ice underneath finally deteriorates.

Dependence between standard deviation and measurement length for C-band backscattering signatures of the Baltic Sea ice

This paper studies whether the standard deviation (std) of the Baltic Sea ice backscattering coefficient (sigmadeg) depends on the length of measurement (l). For many kinds of surfaces, especially for a fractal one, this is the case. The study was conducted using one-dimensional C-band helicopter-borne scatterometer data and ENVISAT synthetic aperture radar (SAR) images. The results with both data sets indicate mostly a strong linear dependence between ln(l) and ln(std(sigmadeg)) up to a distance of at least a few kilometers. Based on the analysis of empirical and simulated data (fractal and nonfractal profiles), it seems that sea ice sigmadeg as a function of l is not completely described either by fractional Brownian motion or by a process with a single-scale autocorrelation function. Neither can the values of sigmadeg be regarded as samples from only one probability distribution. The regression coefficients describing the dependency of ln(l) versus ln(std(sigmadeg)) do not discriminate various ice types better than just mean and std of sigmadeg. However, the use of regression coefficients instead of mean and std is preferred due to their scale-invariant comparability with the results of other studies. The dependence of std(sigmadeg) on l should also be taken generally into account in the data analysis, e.g., when constructing classifiers for sea ice SAR data

Classification of fully polarimetric single- and dual-frequency SAR data of sea ice using the Wishart statistics

Information on the extent and composition of sea ice is important for shipping and offshore operations. Single-polarization spaceborne synthetic aperture radar (SAR) data are an important information source for ice centres around the world. Next-generation SAR satellites will have the capability to collect fully polarimetric SAR data. In this paper we analyze the differences between polarimetric signatures of sea ice and those of land, test several applications of an unsupervised segmentation based on the Wishart distribution of polarimetric data, and give recommendations for modifications to the segmentation method. Airborne C- and L-band sea ice data are investigated. Surface scattering largely dominates sea ice backscatter. A volume component is present; it only dominates for some regions of selected ice types depending on the radar frequency. Dihedral scattering rarely dominates. The application of speckle filters further reduces the number of pixels with predominantly double bounce. The separation of scattering characteristics using the Freeman–Durden model before image classification provides little additional information. The initialization of the segmentation method with a scattering strength based classifier is very efficient and leads to good results.

C‐band SAR backscatter characteristics of Arctic sea and land ice during winter

Synthetic Aperture Radar (SAR) data has become an important tool for studies of polar regions, due to high spatial resolution even during the polar night and under cloudy skies. We have studied the temporal variation of sea and land ice backscatter of twenty‐four SAR images from the European Remote Sensing satellite (ERS‐1) covering an area in Lady Ann Strait and Jones Sound, Nunavut, from January to March 1992. The presence of fast ice in Jones Sound and glaciers and ice caps on the surrounding islands provides an ideal setting for temporal backscatter studies of ice surfaces. Sample regions for eight different ice types were selected and the temporal backscatter variation was studied. The observed backscatter values for each ice type characterize the radar signatures of the ice surfaces. This time series of twenty‐four SAR images over a 3‐month period provides new insights into the degree of temporal variability of each surface. Ice caps exhibit the highest backscatter value of ‐3.9 dB with high temporal variability. Valley glacier ice backscatter values decrease with decreasing altitude, and are temporally the most stable, with standard deviations of 0.08–0.10 dB over the 90‐day period. First‐year ice and lead ice show a negative trend in backscatter values in time and a positive correlation of up to 0.59 with air temperature over the 90‐day period. For first‐year ice and lead ice, episodes of large temperature fluctuations (±12°C) are associated with rapid changes in backscatter values (±2 dB). We attribute the backscatter increase to a temperature‐induced increase in brine volume at the base of the snow pack. Multi‐year ice, conglomerate ice and shore ice are relatively stable over the 3‐month period, with a backscatter variation of only a few dBs. An observed lag time of up to three days between backscatter increase/decrease and air temperature can be attributed to the insulation effect of the snow cover over sea ice. The net range of the backscatter values observed on the most temporally stable surface, valley glacier ice, of about 0.30 dB indicates that the ERS‐1 SAR instrument exceeds the 1 dB calibration accuracy specified for the Alaska SAR Facility processor for the three winter months.

Intercomparison of backscatter maps over Arctic sea ice from NSCAT and the ERS scatterometer

C band and Ku band Arctic Ocean sea ice backscatter maps at 40° incidence angle, produced from the ERS 2 active microwave instrument operating in scatterometer mode and Advanced Earth Observing System NASA scatterometers, respectively, are presented and compared. The noise level on these maps is estimated from the comparison of successive maps and is found comparable, although larger at C band than at Ku band. In both cases the relative noise level increases as backscatter level decreases. Backscatter shows a greater dynamic range at Ku band than at C band, going from first‐year to multiyear ice; moreover, the Ku band to C band backscatter ratio varies considerably as a function of sea ice type. Surface thawing, in June, leads to a brutal decrease in backscatter in both frequency bands, as well as in the Ku band to C band ratio. The study of the time evolution of backscatter over five different regions of the Arctic Ocean demonstrates the influence of both advection and local evolution on this parameter.

Diurnal thermal cycling effects on microwave signatures of thin sea ice

To investigate effects of diurnal thermal cycles on C-band polarimetric backscatter and millimeter-wave emission from sea ice, the authors carried out a winter experiment at the outdoor geophysical research facility (GRF) in the cold regions research and engineering laboratory (CRREL), the ice sheet grew from open sea water to a thickness of 10 cm in 2.5 days, during which they took polarimetric backscatter data with a C-band scatterometer, interlaced with brightness temperature measurements at 90 GHz in conjunction with meteorological and sea ice characterizations. The initial ice growth in the late morning was slow due to high insolation. As the air temperature dropped during the night, the growth rate increased significantly. Air temperature changed drastically from about -12 to -36/spl deg/C between day and night, the diurnal thermal cycle repeated itself the next day and the growth rate varied in the same manner. Ice temperature profiles clearly show the diurnal response in the ice sheet with a lag of 2.5 h behind the time of the maximum short-wave incident solar radiation. The diurnal cycles are also evident in the millimeter-wave brightness temperature data, measured sea ice backscatter revealed substantial diurnal variations up to 6 dB with repeatable cycles in synchronization with the temperature cycles and the brightness temperature modulations, the diurnal cycles in backscatter indicate that the dominant scattering mechanism related to thermodynamic processes in sea ice is reversible, a diurnal backscatter model based on sea ice electrodynamics and thermodynamics explains the observed diurnal signature. This work shows that diurnal effects are important for inversion algorithms to retrieve sea ice geophysical parameters from remote sensing data acquired with a satellite synthetic aperture radar (SAR) or scatterometer on Sun-synchronous orbits.

Arctic summer sea-ice SAR signatures, melt-season characteristics, and melt-pond fractions

AbstractVariations in multiyear sea-ice backscatter from the synthetic aperture radar (SAR) aboard the ERS-1 satellite are interpreted in terms of melt-season characteristics (onset of melt in spring and of freeze-up in autumn, and the duration of the snow-decay period, the melt season, and the melt-pond season) from late winter to early autumn 1992 in two regions of the Arctic Ocean: the northeastern Beaufort Sea adjacent to the Queen Elizabeth Islands in the Canadian high Arctic and the western Beaufort Sea north of Alaska. In the northeastern Beaufort Sea, the onset of melt occurs later, and the periods of snow-cover decay and the occurrence of melt ponds are shorter than in the western Beaufort Sea. These melt-season characteristics of each area are consistent with previous observations that the northeastern Beaufort Sea has one of the most severe summer climates in the Arctic Ocean. A model, which assumes that the backscatter from multiyear floes is the sum of backscatter from bare ice and melt ponds, is used to derive the melt-pond fraction during the summer. The results show that melt-pond fractions decrease from an early-summer maximum of about 60% to a late-summer minimum around 10%. The magnitude of the melt-pond fractions and their decline during the summer is consistent with previous, more qualitative data. The SAR model, which gives melt-pond fractions with lower variability and less uncertainty than previous data, offers an improved approach to the reliable estimation of the areal extent of water on ice floes. Suggestions for further improvement of the model include accounting for the consequences of wind-speed variations, summer snowfall, and freeze/thaw cycles and their effects on melt-pond and ice-surface roughness.

Surface and radiative characteristics of the summer Arctic sea ice cover from multisensor satellite observations

Accurate quantification and characterization of the Arctic summer ice cover are needed for mass balance, heat flux, and modeling studies in the region. A general assessment of the state and basic characteristics of the ice cover can best be done in summer because it is when the perennial component is fully revealed. The main source of summer ice information has been passive microwave and to a lesser degree active microwave data. However, the emissivity and backscatter of sea ice are abnormal and difficult to resolve during this time period, causing large uncertainties in the interpretation of satellite data. In this study we examined the state of the sea ice cover by using special scanning microwave imager (SSM/I), synthetic aperture radar (SAR), and advanced very high resolution radiometer (AVHRR) satellite data synergistically. The surface and radiative characteristics of the summer ice cover were evaluated in the context of three special events: onset of melt, melt ponding, and freeze‐up. These events affect the emissivity and backscatter and may alter the albedo and ice structure. Onset of melt is readily detectable and is shown to migrate rapidly to the north in June. Melt ponding is not directly observable but is postulated to be the main cause of the decreases in brightness temperatures and large discrepancies between the SSM/I and SAR ice concentration results in many areas. In these areas, SAR and AVHRR results show concentrations near 100%, while the SSM/I data were as low as 70%. During freeze‐up the ice signatures are still quite different from those of midwinter ice, but the ice concentrations from SSM/I generally agree well with those from SAR data. Our results show that, generally, the average ice concentration within the pack is usually greater than 90% during the summer, which is substantially larger than that inferred previously from passive microwave data. The use of combined SAR and SSM/I data may also provide melt‐ponding fraction and first‐order estimate of albedo in the Arctic region.

Observation of autumn freeze‐up in the Beaufort and Chukchi Seas using the ERS 1 synthetic aperture radar

We present observations of the transition in sea ice backscattering signatures from late summer into early winter in the Beaufort and Chukchi Seas during 1991 and 1992, using data from the ERS 1 synthetic aperture radar (SAR). We employ both analyzed surface temperature fields and direct observations of near‐surface temperature from drifting buoys. Consistent with previous surface‐based observations, backscattering from sea ice surviving at the end of summer increases strongly and rapidly when temperatures fall below freezing for the final time in the autumn, apparently because of increased volume scattering from bubbles in the upper part of the ice. Areally averaged backscattering sometimes overshoots typical winter multiyear ice values by 1–2 dB, before settling back over a period of approximately a week. The backscattering‐temperature link forms a sufficient basis for an algorithm to estimate retrospectively, from time series of SAR images, the date of freeze‐up (defined here to be the date on which all liquid water in the bubbly upper layer of surviving sea ice freezes and remains frozen for the duration of the ensuing autumn and winter). We present a prototype algorithm and use it to estimate freeze‐up dates in nine Lagrangian cells in the Beaufort Sea during the autumn of 1992. We observe a 12‐day spread in freeze‐up dates between latitudes of approximately 73°N and 82°N, with dates in the northernmost cells of August 29–30 and those in the southernmost cells of September 7. Two cells at latitudes of 75°N–77°N appear to freeze‐up earlier than cells further north, but present uncertainties limit the significance of this observation.

Backscatter characteristics of the winter ice cover in the Beaufort Sea

The microwave backscatter statistics of the sea ice cover in the winter Beaufort Sea are characterized using C band synthetic aperture radar (SAR) data collected by the European Earth Resources Satellite 1. Sea ice backscatter signatures were sampled from SAR image data collected during the winter of 1991–1992 and winter of 1992–1993. The spatial and temporal variabilities of the backscatter signatures of different ice types are discussed. The results show considerable seasonal stability of the backscatter signature of multiyear ice as well as of first‐year ice. Small amplitude regional variations of the multiyear ice backscatter can be observed. Consistent contrast between multiyear ice and first‐year ice is maintained throughout the season. The highest radiometric variability is observed in sea ice in leads. On the basis of these observations the backscatter from the principal ice types (multiyear and first‐year) are consistent with scatterometer observations and can be easily identified under winter conditions. Correlations of the regional variations of multiyear ice signatures to physical processes are suggested.