Icebergs around Antarctica: size distributions, mass fluxes, and local calving rates by using remote sensing and drift modeling
Final Report Abstract
Antarctica’s coastline is surrounded by ice shelves and glacier tongues, which were the birthplaces of icebergs. During the Antarctic Mapping Mission-1 (AMM-1) of the Radarsat-1 Antarctic Mapping Project (RAMP), a mosaic compiled of high resolution SAR images was generated, covering the entire Antarctic continent and adjacent ocean regions. Within the framework of this project, this mosaic was used to investigate the ice shelves/ ice tongues and the icebergs around Antarctica. First, the mosaic was used to map the surface features on the ice shelves around Antarctica. For this purpose edge detection was applied on all 183 ice shelves larger than 20 km2. Studies on the nature of icebergs calved from Svalbard’s ice cliffs showed a dependency of the iceberg shape on the surface feature distribution close to the ice front. Therefore, the surface feature mapping was utilized to classify the Antarctic coastline in terms of the orientation of surface features close to calving front relative to the main course of the calving front. There were three main classes of surface features: (C1) parallel features, (C2) orthogonal features, and (C3) a combination of (C1) and (C2) called ice stream (IS) pattern. A visual inspection of the calving front showed, that the three classes lead to icebergs with a certain shape: for C1 rectangular shaped icebergs, for C2 trapezoidal or triangular shaped icebergs, and for C3 small quadratic shaped icebergs in case for small ice shelves carrying the IS pattern or, in case of glacier tongues, giant icebergs that still show the IS pattern at the surface. Series of satellite images allow calving front change detection. An example of the calving front evolution of the Ekström Ice Shelf over 4.5 years was shown in this report. Two larger calving events were detected, as well as an area with nearly no or only small changes. Erosion or break offs are not the only processes that can be noticed by satellite imagery. As ice shelves expand at the seaward edge, accretion of ice shelves is also visible and could be found at the Ekström Ice Shelf. Besides the investigation of ice shelves of Antarctica, the RAMP AMM-1 mosaic offered the opportunity to investigate the iceberg distribution in the near-coastal ocean regions. For this purpose, we applied the automated iceberg detection method presented in Wesche & Dierking (2012) on the mosaic. A following visual inspection and correction of the result lead to 6912 detected icebergs with sizes between 0.3 to 4717.6 km2. Due to the visual inspection, the iceberg distribution is highly reliable, because 77 % of the ice mass (all icebergs > 5km2) was corrected if necessary and obvious false detections were removed. The iceberg drift modeling was another task during this project. Two different approaches were used. The first based on the complex Finite Elemente Sea ice Ocean Model (FESOM) for iceberg drift studies in the entire Southern Ocean over several years and the second a more simple using a wind-driven analytical approach (S-IB) for regional short-term iceberg drift modeling to close potential temporal gaps between satellite images that cover the icebergs that should be tracked. The FESOM approach included the thermodynamics of icebergs, which enables the calculation of the freshwater input due to iceberg melting. A first study had been done using an artificial iceberg data set to test the general model setup. In a next step, the iceberg distribution mentioned earlier will be used to model the drift paths and the freshwater input over a period of at least five years. The wind-driven model (S-IB) is based on the equations published by Crépon et al. (1988). It is planned to utilize S-IB to predict iceberg positions within few days, to narrow the search radius for new satellite images. To test S-IB the drift results were compared to iceberg buoy data in three small regions of the Weddell Sea. The results are promising, but some more investigations are needed. This is work in progress.
Publications
- (2013), Calving Fronts of Antarctica: Mapping and Classification, Remote Sensing, 5(12), pp. 6305-6322
Wesche, C., Jansen, D. and Dierking, W.
(See online at https://doi.org/10.3390/rs5126305) - (2013), Iceberg Drift in the Eastern Weddell Sea: Observed and Modeled, ESA Living Planet Symposium, Edinburgh, Scotland, 9 September - 13 September 2013
Wesche, C., Rackow, T. and Dierking, W.
- (2013), Modelling Southern Ocean iceberg drift and decay with FESOM-IB, European Geophysical Union, Vienna, Austria, 7 April 2013 - 12 April 2013
Rackow, T., Wesche, C. , Timmermann, R. and Juricke, S.
- (2014), C-Band Radar Polarimetry - Useful for detection of icebergs in sea ice?, IEEE Transactions on Geoscience and Remote Sensing, 52(1), pp. 25-37
Dierking, W. and Wesche, C.
(See online at https://doi.org/10.1109/TGRS.2012.2234756) - (2014), From ice shelves to icebergs: Classification of calving fronts, iceberg monitoring and drift simulation, Proceedings of the IEEE International Geoscience and Remote Sensing Symposium (IGARSS) in Québec City, Canada
Wesche, C. and Dierking, W.
(See online at https://doi.org/10.1109/IGARSS.2014.6946410) - (2015), Near-coastal circum-Antarctic iceberg size distributions determined from Synthetic Aperture Radar images, Remote Sensing of Environment, 156(1), pp. 561-569
Wesche, C. and Dierking, W.
(See online at https://doi.org/10.1016/j.rse.2014.10.025)
