Fringe 2017 > Session details
Paper 29 - Session title: Ice and snow
15:20 Meltdown of Ice Bridges and Emergence of New Islands in the Barents-Kara Region Observed by Sentinel-1 INSAR
Sharov, Aleksey (1); Nikolskiy, Dmitry (2) 1: JOANNEUM RESEARCH, Austria; 2: Sovzond, Russian Federation
Natural ice bridges formed by glaciers stretching across water bodies and connecting isolated tracts of land to the mainland belong to relatively uncommon, rapidly vanishing and very attractive objects of present-day glaciation. Under current climate warming and essential ice loss due to calving, surface ablation and basal melting, low-lying ice bridges melt down much faster than inland glacier parts. Rapid disintegration of ice bridges in the Arctic reveals new, as yet uncharted islands, bays and capes formerly covered by glacier ice. Publications periodically announce the appearance of new typically small islands along Arctic ice coasts. Some of these findings turned out to be drying shoals, unlithified morainic remnants or grounded icebergs.
Satellite radar interferometry (INSAR) helps a lot in identifying subglacial straits, studying rheology of ice bridges, mapping and forecasting their changes. The present paper discusses the use of Sentinel-1AB INSAR data series for determining rheological characteristics of the largest ice bridges and their parental glaciation and for mapping new large islands emerged in the Franz Josef Land, Novaya Zemlya and Svalbard archipelagos due to glacial retreat. Wide terrestrial coverage and relatively short repetition interval of Sentinel-1 IW data with accordingly high quality and detail of SAR interferograms obtained in the period of X.2015 – XI.2016 allowed several case studies to be carried out on the use of Sentinel-1 interferometric products for documenting breakups of ice bridges in three different parts of the Barents-Kara glaciation:
1) observational study of glacier changes and mapping of new islands appeared in north Novaya Zemlya in 2014-2016,
2) detailed study of the ice bridge breakup at Hall Island and the emergence of Littrow Island and other islets in the eastern part of Franz Josef Land,
3) in-depth study of the rheology and evolution of Hornbreen-Hambergbreen icy isthmus (35 km²) in Hornsund, south Spitsbergen, the largest ice bridge in the study region.
Our Sentinel-1 IW data set included
Sentinel-1AA INSAR pair of 19.09 - 01.10.2015 and Sentinel-1BB INSAR pair of 25.10 – 06.11.2016 representing the entire Northern Island of Novaya Zemlya;
Sentinel-1AA and -1BB INSAR pairs of 12.10 – 24.10.2015 and 24.10 – 05.11.2016 taken over the Franz Josef Land archipelago;
Sentinel-1AA INSAR pair of 20.01 – 01.02.2015 and Sentinel-1AB INSAR pair of 31.10 - 06.11.2016 obtained over south Spitsbergen.
The interpretation of Sentinel-1 data was supported with previously processed ERS-1/2 INSAR time series of 1993-2004 and geodetic, oceanographic and glaciological data obtained during field surveys in the 2000s. The impact of long-term glacioclimatic and oceanographic trends, seasonal effects, hydrometeorological conditions and inherent limitations of satellite INSAR on the validity and worth of our research was discussed.
The main results and conclusions from the research were summarized as follows:
1. In total, 14 new islands with the overall land area of 73 km² formerly attached to the larger lands by glaciers along the Barents- and Kara coasts of Novaya Zemlya and Franz Josef Land were discovered and mapped; their geodetic coordinates and main topographic characteristics were determined and validated.
2. So, Littrow Island in Franz Josef Land specified as a peninsula in contemporary maps, was – for the second time after the American Wellman polar expedition (1898) – discovered to be separated from Hall Island by the Nordenskjöld Channel. The present width of Nordenskjöld Channel was measured as 1 km and the total area of Littrow Island was given as 60.0 km². This is the largest island emerged in the study region because of the ice bridge collapse.
3. The Hornbreen - Hambergbreen icy isthmus with the maximum height of 130 m asl (2006) is still intact although its width decreased from 8.8 km (2004) to 5.4 km (2016). Currently, the entire area of the ice bridge is set in motion as a result of glacier flow, tidal effects and basal melting. The unilateral gradient of phase fringes on either side of the ice divide observed in winter Sentinel-1 SAR interferograms proves the prevalence of vertical motions in the central part of the ice bridge, which complies with the results of dGPS surveys carried out in situ in 2006. The intensity of motions increased drastically in the course of past 10 years. The overall length of calving fronts along both bridge sides was measured as 8.2 km, which is equal to or even longer than the total width of all outlets feeding the icy isthmus. We conclude that Sörkapp Land in South Spitsbergen becomes a separate island with a total land area of 1200 km² due to the disintegration of the ice bridge.
4. The breakup of ice bridges led to the increase of ice flow velocities on tributary glaciers. Several formerly quiescent ice streams conveying ice into newly opened water channels “woke up” and their flow velocities increased essentially. The regular patterns of transverse crevasses detected with the aid of Sentinel-1 coherence images on the surface of tributary glaciers, which were released from the buttressing resistance of icy lintels, indicate the extended character of glacier motion.
Paper 42 - Session title: Ice and snow
14:00 Measuring Strain and Rotation using InSAR. Example of a Glacier Flow
Parizzi, Alessandro; Abdel Jaber, Wael Remote Sensing Technology Institute DLR, Germany
Interferometric phase derived from the SAR coherent signal can provide displacement measurements much more accurate than correlation techniques. Nevertheless, due to the ambiguity of the interferometric phase, such measurements are not absolute, but they have to be referenced to a point in the scene assumed to be zero. This fits perfectly with many geophysical applications having as aim the estimation of the deformation on different scales, from tens of meters like in glaciology up to hundreds of kilometers like in tectonic. However the sensitivity of InSAR phase to the motion is limited to the radar range direction (LoS) therefore only having various available geometries it will be possible to resolve the different directions of the relative displacements. In and several others papers the reconstruction of the 3D motion from different LoS has been presented and analyzed.
Aim of this work is to consider the problem in terms of displacements gradients and discuss reconstruction of the 3D gradient tensor having different InSAR geometries. It will be shown how the gradient tensor permits to distinguish between deformation and rotation phenomena. In order to provide an example of the framework the case of a glacier flow has been studied. This case is particularly interesting because as far as the movement is quite big w.r.t the scale of interest it is possible to derive deformation measurements avoiding to resolve the phase ambiguities (phase unwrapping) .
The full abstract submission with discussion and Figures is available in the attached pdf.
Paper 60 - Session title: Ice and snow
14:40 Snow Water Equivalent (SWE) Retrieval By Sentinel-1 SAR Data
Conde, Vasco (1); Nico, Giovanni (2); Mateus, Pedro (1); Catalao, Joao (1); Kontu, Anna (3); Gritsevich, Maria (4,5) 1: Universidade de Lisboa, Instituto Dom Luiz, Lisbon, Portugal; 2: Consiglio Nazionale delle Ricerche (CNR), Istituto per le Applicazioni del Calcolo (IAC), Bari, Italy; 3: Finnish Meteorological Institute (FMI), Sodankylä, Finland; 4: University of Helsinki, Department of Physics, Helsinki, Finland; 5: Finnish Geospatial Research Institute (FGI), Masala, Finland
In this work we investigate the use of SAR interferometry (InSAR) to derive SWE maps using spaceborne C-band SAR data. The main objective is to reduce the impact of model inversion on the SWE estimation  and, at the same time, provide an alternative procedure to derive the SWE using SAR data.
The physical principle used in SAR interferometry relies on phase delay occurring due to the propagation in a non-dispersive medium. This implies that the snow is assumed to be dry in order to allow the propagation of the SAR signal. Furthermore, the fact that the SWE estimation is based on the measurement of a phase delay implies that phase contributions due to topography and propagation in the atmosphere should be properly identified and corrected. A precise Digital Elevation Model (DEM) of the area is used to model and remove the phase delay due to the topography modulation of the interferometric signal. The mitigation of atmospheric phase delay can be done by using external data such as Numerical Weather Models (NWMs)  or delay measurements provided by the Global Navigation Satellite System (GNSS) receivers or passive satellite sensors . Concerning the occurrence of terrain displacements, due to different geological phenomena, it is assumed that they are negligible within the short temporal baseline between the two SAR images used to generate the interferograms. The proposed methodology provides a direct estimate of the snow depth which is then used to derive the SWE. The knowledge of the SWE at several reference locations is needed, provided by either in-situ measurements or other remote sensing techniques. These in-situ data are used to calibrate InSAR estimates of SWE since the phase measurements are needed to be unwrapped both in time and space. In this work we present the first results obtained using C-band Sentinel-1 SAR images acquired over Finland between November 2015 and May 2016. Results obtained by SAR interferometry are compared to both in-situ measurements and estimates obtained by amplitude images in the VV and VH polarimetric channels.
 H. Rott, C. Duguay, P. Etchevers, R. Essery, I. Hajnsek, G.Macelloni, E. Malnes, J. Pulliainen, “Report for Mission Selection CoReH20”, European Space Agency, Nordwijk, The Netherlands, 2012.
 G. Nico, R. Tomé, J. Catalão, and P. Miranda, “On the use of the WRF model to mitigate tropospheric phase delay effects in SAR interferograms,” IEEE Transactions on Geoscience and Remote Sensing, 49(12), 4970–4976, 2011.
 P. Mateus, G. Nico, R. Tomé, J. Catalão, and P. Miranda, “Experimental study on the atmospheric delay based on GPS, SAR interferometry and numerical weather model data,” IEEE Transactions on Geoscience and Remote Sensing, 51(1), 6–11, 2013.
Paper 63 - Session title: Ice and snow
15:00 Measuring Sea Ice Surface Topography With Single-Pass InSAR
Dierking, Wolfgang (1); Lang, Oliver (2); Busche, Thomas (3) 1: Alfred Wegener Institute & University in Tromsø, Germany; 2: Airbus Defence and Space; 3: German Aerospace Center (DLR)
The topography of the sea ice surface is a consequence of ice drift variations on scales of tens of meters to hundreds of kilometers. Parameters describing the topography are needed in different fields such as boundary layer meteorology and geophysics, e. g. in studies of the coupling between the atmosphere and the ice, or of sea ice mechanics and ice mass balance. Furthermore, ship routing profits from detection of potential hazards like ridges. With the advance of single-pass interferometric missions such as TanDEM-X, the retrieval of ice surface topography from space-borne platforms becomes possible. However, the demands on phase and height accuracy are high, since the mean maximum heights of first-year ice ridges reported for different areas of the Arctic range from 1.1 m to 3.3 m, and the individual widths of ridge sails are between 1.8 m and 73.2 m with mean values between 9.6 m and 17.5 m.
We will present the results of a study in which the potential of InSAR measurements is examined for different satellite configurations, considering statistics of heights and widths of sea ice ridges. We define the requirements for retrieving sea ice surface height variations with errors ≤ 0.5 m. It is, e. g., necessary to use very large interferometric baselines that lead to a significant decrease of the interferometric coherence. Another challenge is that the sea ice drift may contribute significantly to the measured interferometric phase. It depends on the instantaneous line-of-sight velocity of the ice, which has to be known for reliable height retrievals. The third problem is the dependence of the achievable interferometric baseline on the satellite orbit configuration. In our presentation we will address these issues and also report on expected signal-to-noise ratios for different sea ice types and their effect on height retrievals.
During the TanDEM-X Science Phase between March and June 2015, 46 bistatic TanDEM-X data takes in different polarizations were acquired over the area of interest at Beaufort Sea (off the coast of Pt. Barrow, Alaska). The normal baselines were typically a few hundred meters to more than one kilometer, and the incidence angle varied within the performance range of the sensor. Hence, this data set provided an optimal opportunity to study the potential and limitations of single-pass INSAR. We will show results of surface height retrievals. In the paper we will discuss the conditions for reliable sea ice topography derivation and provide an outlook on the operational potential of the application.
Paper 142 - Session title: Ice and snow
14:20 InSAR Scattering Phase Centre of Antarctic Snow – An Experimental Study
Rott, Helmut (1); Wuite, Jan (1); Nagler, Thomas (1); Floricioiu, Dana (2); Rizzoli, Paola (3); Helm, Veit (4) 1: ENVEO IT, Innsbruck, Austria; 2: Remote Sensing Technology Institute, DLR, Germany; 3: Microwaves and Radar Institute, DLR, Germany; 4: Alfred-Wegener-Institut, Bremerhaven, Germany
Digital elevation models, derived from single-pass interferometric SAR (SP-InSAR) data, are a key source for monitoring surface elevation and its temporal changes over ice sheets and glaciers. SP-InSAR delivers highly accurate cross-track interferograms that are not affected by temporal decorrelation and variations in atmospheric delay. Over snow and ice effects of signal penetration have to be taken into account. The apparent surface for uncorrected InSAR elevation data refers to the position of the scattering phase centre in the snow volume. Losses due to absorption and scattering in the volume and the position and strength of the scattering sources are the main factors determining the depth of the scattering phase centre at a given radar wavelength. The absorption losses can be deduced from temperature and mass of the medium along the radar propagation path. The scattering sources and losses can be highly variable dependent on snow structure and stratification. We report on studies of factors that are governing signal penetration and the related bias in surface elevation for snow and ice in Antarctica. To this end field measurements on snow structure and stratigraphy will be performed in December 2016 in a test area in the Ellesworth mountains, West Antarctica (80 deg. S). The test area includes bare ice surfaces (blue ice fields) and sites with different accumulation rates, as deduced from the analysis of radar backscatter and coherence images. The main satellite data base for the study are X-band SAR data of the TanDEM-X mission, including amplitude images, SP-InSAR coherence images, across-track interferograms, and DEMs. Several repeat-pass lidar tracks of the ICESat mission between 2002 and 2008 and CryoSat data are used for comparisons. The ICESat time series shows very little temporal change of surface elevation over the study area, thus providing a good basis for inferring the depth of the scattering phase centre. The sites for the field measurements have been selected in order to cover different levels of mismatch between ICESat and TanDEM-X surface elevation, ranging from zero over blue ice fields to 5 m at sites with high accumulation in wind-sheltered locations. The analysis of the satellite data indicates distinct relations between ICESat - TanDEM-X elevation difference and coherence/radar backscatter properties which may be exploited for correcting X-band SAR signal penetration. Based on the field measurements, procedures for penetration correction will be further consolidated and relations to structure and morphology of dry Antarctic snow and firn will be elaborated.