Problems with the retrieval of glacier net surface balance from SAR imagery

Abstract

ABSTRACT. There are relatively few comparisons between synthetic aperture radar (SAR) observations and glacier mass-balance measurements. More typically, SAR has been deployed to identify changes in the end-of-summer snowline and other facies boundaries. In this paper, we analyze the geophysical processes affecting SAR amplitude data over two Arctic glacier systems in northern Scandinavia to assess the potential of SAR observations for the retrieval of surface balance parameters. Using a backscatter model and in situ data, we identify the controls on SAR imagery in terms of mass-balance measurement and discuss the glaciological parameters which can reasonably be derived from multi-temporal SAR data. Our results show that amplitude SAR imagery, in the absence of in situ measurements, is not capable of providing meaningful mass-balance data. We show that backscatter from temperate glaciers is affected primarily by snow grain-size and density, and therefore processes such as firnification or depth-hoar formation can complicate the analysis of imagery. We conclude that SAR imagery over temperate glaciers can provide valuable proxy information but not direct mass-balance terms.

Using SAR satellite data time series for regional glacier mapping

“…Backscatter signals show not only surface reflectivity but also signals from below the surface (volume scattering), allowing for differentiation between SAR glacier zones (e.g., Fahnestock et al, 1993;Rau et al, 2000). Studies since the 1980s have been used to investigate snow and ice, particularly using summer SAR images for exploring and tracking transient snow lines (TSLs) (Rott, 1984;Hall et al, 2000;Rees et al, 1995;Casey and Kelly, 2010;Callegari et al, 2016). Synergistic use of SAR and optical satellite imagery for mapping snow and ice has been evolving since the 1990s (Rott and Strobl, 1992;Rott, 1994;Shi et al, 1994;Sephton et al, 1995).…”

Investigation of the spatial variations in synthetic aperture radar backscatter in western Dronning Maud Land, Antarctica

“…This may suggest that coastal climate processes, for example, wind roughening of the ice shelf surface, affected scattering between the two acquisitions. This is certainly within the range of possible variations in surface scattering intensity according to surface scattering model sensitivity tests [19,21]. A further expression of the strong similarity between the images is the high interferometric coherence found between the scenes.…”

“…The Rayleigh backscatter model implies that grain size can strongly influence scattering [19,21]. Noted objections to the applicability model are specific to the dry snow zone [9,10].…”