The transformation of Arctic sea ice from mainly perennial, multi-year ice to a seasonal, first-year ice is believed to have been accompanied by a reduction of the roughness of the ice cover surface.

This smoothening effect has been shown to (i) modify the momentum and heat transfer between the atmosphere and ocean, (ii) to alter the ice thickness distribution which in turn controls the snow and melt pond repartition over the ice cover, and (iii) to bias airborne and satellite remote sensing measurements that depend on the scattering and reflective characteristics over the sea ice surface topography.

The value of characterising surface roughness of sea ice is therefore clear. Recent research has seen the development of surface roughness proxies that can be used alongside additional datasets to monitor the declining Arctic sea ice trend. Whilst particle-sized ice may be forward scattering (focused away from the sun), at a larger scale rougher surfaces will exhibit stronger backward scattering allowing for sea ice classification based on this. Surface texture type will impact the spectral reflectance and reflected radiance angles.

We will review existing and novel remote sensing methodologies proposed to estimate sea ice roughness, ranging from airborne LIDAR measurement (ie Operation IceBridge), to backscatter coefficients from scatterometers (ASCAT, QUICKSCAT), to multi angle maging spectroradiometer (MISR), and to laser (Icesat) and radar altimeters (Envisat, Cryosat, Altika, Sentinel-3).

We will show that by comparing and cross-calibrating these different products we can offer a consistent multi-mission, multi-decadal view of the declining sea ice roughness. Implications for sea ice physics, climate and remote sensing will also be discussed.

Figure caption: Airborne and satellite remote sensing sea ice roughness proxies used in this study over the 1991 to present. The measurements from the LIDAR instrument onboard Operation IceBridge will act here as ground truth for other satellite products.

Lacking information of snow depth introduces one of the largest uncertainties in the error budget in altimetry derived sea ice thicknesses (Giles et al. 2008). Techniques using data from passive microwave radiometers have shown to perform well in Antarctica (Markus and Cavalieri (2003) and Comiso et al. (2003)), where the sea ice consists primarily of first-year ice. A similar approach is unreliable in the Arctic in areas covered by multi-year ice, and related Arctic snow depth products do not provide snow depths in areas covered by multi-year ice. Today most Arctic space derived sea ice thickness products rely on the Warren snow climatology (Warren et al., 1999) based on snow measurements, which are not living up to the present state of the sea ice cover and do not provide inter-annual changes. Thus, it is important to find a solution to obtain reliable Arctic wide snow depth products in support of space derived sea ice thickness maps from altimetry. Combined measurements of Ku/Ka-band from CryoSat-2 and SARAL/AltiKa altimetry missions has shown promising snow depth results (Guirreira et al., 2017), but lacks information north of 81.5N.

To further exploit the potential for using the dual-frequency concept to estimate snow depth the ESA CryoSat-2 validation experiment (CryoVEx) 2016 airborne fall campaign was the first to fly combined Ka- and Ku-band radars on the same platform. Where the CryoVEx 2016 fall campaign was dedicated to the Greenland Ice Sheet following campaigns in spring 2017 (Arctic) and the southern hemisphere Summer 2017/18 (Antarctic) were covering sea ice in both hemispheres in different regions representing different sea ice types and settings.

The airborne instruments consist of the ESA Ku-band radar altimeter (ASIRAS), MetaSensing Ka-band radar altimeter (KAREN), together near-infrared Airborne Laser Scanner (ALS). Even though some studies has indicated that ASIRAS reflects from the air-snow surface and do not penetrate down to the snow-ice surface, as typically assumed by Ku-band radars in dry cold snow conditions, the waveform turns out to be more complex, and often show dual peaked waveforms, potentially reflecting from the air-snow and snow-ice surface. The Ka-band radar is expected to reflect from the air-snow surface. Both radars are tied together with the ALS observations of the air-snow surface.

Here we present an analysis of waveforms from the Ka/Ku-band airborne radars in different sea ice regions, to investigate the potential for estimating snow depths using airborne data from the CryoVEx 2017 Arctic and 2017/18 Antarctic campaigns. Where available near coincident in situ measurement and snow depths from NASA Operation IceBridge will be used as validation.


Giles, K. A., S.W. Laxon, D.J. Wingham, D.W. Wallis, W.B. Krabill, C.J. Leuschen, D. McAdoo, S.S. Manizade, R.K. Raney (2007), Combined airborne laser and radar altimeter measurements over the Fram Strait in May 2002. Remote Sensing of Environment 111, 182-194.

Guerreiro, K., S. Fleury, E. Zakharova, F. Rémy, and A. Kouraev (2016): Potential for estimation of snow depth on Arctic sea ice from CryoSat-2 and SARAL/AltiKa missions. Remote Sensing of Environment, 186, 339–349, doi:10.1016/j.rse.2016.07.013

Markus, T. and D.J. Cavalieri, Snow depth distribution over sea ice in the Southern Ocean from satellite passive microwave data, in Antarctic Sea Ice Physical Processes, Interactions and Variability, Antarctic Research Series, 74, edited by M.O. Jeffries, pp.19-40, AGU, Washington, D.C., 1998.

Comiso, J.C., D.J. Cavalieri, and T. Markus, Sea ice concentration, ice temperature, and snow depth using AMSR-E data, IEEE Trans. Geoscience and Remote Sensing, 41(2), 243-252, 2003.

Warren, S.G., I.G. Rigor, N. Untersteiner, V.F. Radionov, N.N. Bryazgin, Y.I. Aleksandrov, and R. Colony, R., (1999). Snow Depth on Arctic Sea Ice. J. Climate, 12, 1814–1829.

Snow on sea ice plays fundamental role in climate processes. Further, incomplete knowledge of snow depth and density on sea ice remains one of the largest sources of uncertainty in ice thickness retrievals from laser or radar altimetry. In this study we produce the first daily, pan-Arctic snow depth and density estimates using a sophisticated new snow model (SnowModel) developed specifically for sea ice applications and forced by atmospheric reanalysis. SnowModel is implemented in a Lagrangian framework, where snow accumulates on each ice parcel as it moves around the Arctic. Validation of our derived snow depth estimates is performed through comparisons of snow depths from Operation IceBridge (OIB), ice mass balance buoys (IMBs), snow buoys, magnaprobes and rulers. We further compare our modeled snow depths with those from two climatologies and snow depth estimates from passive microwave over first-year ice. The model framework is tested with and without data assimilation.

In the last decade advances in both satellite and airborne remote sensing have revolutionized our ability to observe and understand the changing Arctic sea ice cover. Satellite altimeter, scatterometer, and SAR observations, from CryoSat-2, ICESat-2, ASCAT, Sentinel-1 and Sentinel-3 provide routine monitoring, and the opportunity to track both seasonal and interannual change. Extensive lidar and radar observations and visible imagery from airborne platforms provide a novel, high-resolution perspective. Advancing our understanding of how the sea ice cover varies in space and time is required to characterize the geophysical processes governing change and to improve sea ice forecasts.

Here we explore novel sea ice data products developed at the NOAA Laboratory for Satellite Altimetry that are derived from laser and radar instruments deployed on both satellite and airborne platforms. We exploit scatterometer and SAR observations to monitor the location and extent of Arctic multi-year sea ice and track sea ice dynamics, while inter-annual variability in sea ice growth rates and thickness are derived from satellite altimeter data. High-resolution observations from multi-instrumented IceBridge airborne surveys provide a decade-long time series of dynamical and morphological sea ice features, including sea ice pressure ridges, melt ponds, and snow on sea ice. ICESat-2, launched in September 2018, carries a photon-counting laser altimeter that provides very high-resolution measurements of sea ice elevation and surface morphology, across basin scales. ICESat-2 thus bridges the gap in resolution between traditional satellite radar altimeters and the more detailed airborne measurements. We present the very first results emerging from the ICESat-2 mission, describing sea ice freeboard, surface roughness, and pressure ridge sail height.

Together these remote sensing observations are used to gain further insights into the deformation characteristics of the Arctic ice cover, and to develop multi-sensor data products that advance capabilities to model and predict future change, addressing stakeholder needs.

In February and March 2018 a unique polynya of open water had opened in the Wandel Sea north of Greenland due to unusually strong southerly winds. The polynya subsequently refroze and the area of the growing young ice strongly decreased by ice convergence due to northerly winds, constituting a natural, well-constrained, full-scale ice deformation experiment. We have carried out an airborne electromagnetic ice thickness survey of the young ice one month after the polynya began to close. It showed a thickness distribution with a modal ice thickness of 1 m and mean thickness of 2 m, representative of the contributions of thermodynamic and dynamic growth in the one month since the ice had begun to grow. We used time series of Sentinel 1 SAR images to back-track the surveyed young ice to its location of initial formation. Results showed that the area of young ice approximately halved between the time of its formation and the time of the ice thickness survey. This is in good qualitative agreement with the result of the thickness survey, showing that its mean thickness was twice its modal thickness, i.e. mean ice thickness doubled while the region's area halved. These results and the SAR observed deformation processes provide valuable information for improving representation of ice rheology and thickness redistribution in sea ice models.

Theoretical and field studies demonstrate that low frequency microwave radiometry (i.e. lower than 1 GHz) can surpass current limitations in observing sea-ice thickness. In particular, the expected SMOS (1.4 GHz) derived maximum thickness of 0.5 m can be increased to 1 m. In order to test this concept an airborne, ultra-wideband radiometer was used to measure microwave brightness temperature spectra over Arctic Sea Ice in the Lincoln Sea near the coast of Greenland. The experiment occurred on September 13, 2017 near the seasonal point of minimum sea ice extent. Spectra over the 0.5 to 2 GHz band were compared to thermal infra-red images collected during the airborne campaign and also compared to nearly concurrent Sentinel C-band SAR data. Results prove that locations with low radar backscatter spectral signatures (less than -15 dB for HH polarization) and high IR temperatures (261 K relative to the uncalibrated mean value of about 257 K) were associated with thin ice. Locations with low relative IR temperatures (between 190 and 210 K) and high backscatter (-10 dB) were associated with multiyear ice. Brightness temperature differences between the lowest and highest frequency channels were about 15 K for multiyear ice and about 25 K for first year or possibly thinner ice. A radiative transfer model was then used to invert for sea ice thickness and salinity. Inferred thicknesses ranged between about 0.2 to 3 m (depending somewhat on the radiative transfer model) and are consistent with expectations for ice floes in the Lincoln Sea. Salinities are higher than expected (4- 20 ppt) and may be a consequence of ignoring surface and volume scattering in the models and uncertainties in the sea ice permittivity. While the proposed method is not yet fully validated the obtained results are very promising and preliminary studies to move the concept to space are now in progress. Moreover, a new campaign with in situ data will carried out in Antarctica in December 2018.