Monitoring ice shelf velocities from repeat MODIS and Landsat data – a method study on the Larsen C ice shelf, Antarctic Peninsula, and 10 other ice shelves around Antarctica

Abstract. We investigate the velocity field of the Larsen C ice shelf, Antarctic Peninsula, over the periods 2002–2006 and 2006–2009 based on repeat optical satellite data. The velocity field of the entire ice shelf is measured using repeat low resolution MODIS data (250 m spatial resolution). The measurements are validated for two ice shelf sections against repeat medium resolution Landsat 7 ETM + pan data (15 m spatial resolution). Horizontal surface velocities are obtained through image matching using both orientation correlation operated in the frequency domain and normalized crosscorrelation operated in the spatial domain, and the two methods compared. The uncertainty in the displacement measurements turns out to be about one fourth of the pixel size for the MODIS derived data, and about one pixel for the Landsat derived data. The difference between MODIS and Landsat based speeds is −15.4 m a⁻¹ and 13.0 m a⁻¹, respectively, for the first period for the two different validation sections on the ice shelf, and −26.7 m a⁻¹ and 27.9 m a⁻¹ for the second period for the same sections. This leads us to conclude that repeat MODIS images are well suited to measure ice shelf velocity fields and monitor their changes over time. Orientation correlation seems better suited for this purpose because it produces fewer mismatches, is able to match images with regular noise and data voids, and is faster. Since it can match images with regular data voids it is possible to match Landsat 7 ETM+ images even after the 2003 failure of the Scan Line Corrector (SLC off) that leaves significant image stripes with no data. Image matching based on the original 12-bit radiometric resolution MODIS data produced slightly better results than using the 8-bit version of the same images. Streamline interpolation from the obtained surface velocity field on Larsen C indicates ice travel times of up to 450 to 550 years between the inland boundary and the ice shelf edge. In a second step of the study we test our method successfully on 10 other ice shelves around Antarctica demonstrating that the approach presented could in fact be used for large scale monitoring of ice shelf dynamics.