Improving InSAR elevation models in Antarctica using laser altimetry, accounting for ice motion, orbital errors and atmospheric delays

• Published in: Remote Sensing of Environment
• Main Authors: Zhou, Y, Zhou, C, Deng, F, E, D, Liu, H, Wen, Y
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: Elsevier 2016
• Subjects: Antarctic; Grove Mountains; Antarc*; Antarctica
• Online Access: https://doi.org/10.1016/j.rse.2015.01.017; https://ora.ox.ac.uk/objects/uuid:2bf48a4c-6526-4150-aeb3-46a9f8bc1dfc

Ice dynamics is closely related to climate change. However, a lack of digital elevation models (DEMs) of sufficient horizontal resolution and vertical precision has been a major issue in dynamic studies for years. Existing Antarctic DEMs are derived from satellite radar and laser altimetry as well as limited terrestrial data; these sparse datasets result in poor spatial resolution (hundreds of metres to 1 km). In this paper, we propose a method for generating high-accuracy and high-resolution DEMs using interferometric synthetic aperture radar (InSAR) and Ice, Cloud and land Elevation Satellite (ICESat) laser altimetry by reducing the influence of ice motion, satellite orbital errors and topographically correlated atmospheric delays. A case study in the Grove Mountains area shows that the InSAR DEM has a root-mean-square (RMS) error of 5.9 m, which is better than the Radarsat Antarctic Mapping Project (RAMP) (21.1 m) and Bamber (8.5 m) DEMs. The new DEM is used to calculate ice velocity over the Grove Mountains area where GPS measurements have been collected for comparison. Compared to the NASA Making Earth System Data Records for Use in Research Environments (MEaSUREs) InSAR-Based Antarctica Ice Velocity Map, ice velocity estimates from long-baseline interferogram and the new DEM have a smaller RMS error; this suggests that with the new DEM, baseline length is no longer a limiting factor for the accuracy of InSAR-based ice velocity mapping.